diff --git "a/sft_full/smoe_perturbed/trainer_state.json" "b/sft_full/smoe_perturbed/trainer_state.json" new file mode 100644--- /dev/null +++ "b/sft_full/smoe_perturbed/trainer_state.json" @@ -0,0 +1,93184 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.0, + "eval_steps": 500, + "global_step": 13306, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 7.515406583496167e-05, + "grad_norm": 20.468459265587903, + "learning_rate": 0.0, + "loss": 1.5843, + "step": 1 + }, + { + "epoch": 0.00015030813166992335, + "grad_norm": 13.441950571113182, + "learning_rate": 4.6275642631951835e-07, + "loss": 1.692, + "step": 2 + }, + { + "epoch": 0.000225462197504885, + "grad_norm": 12.547891654016523, + "learning_rate": 7.334515826841693e-07, + "loss": 1.6105, + "step": 3 + }, + { + "epoch": 0.0003006162633398467, + "grad_norm": 14.08007257096055, + "learning_rate": 9.255128526390367e-07, + "loss": 1.5526, + "step": 4 + }, + { + "epoch": 0.00037577032917480833, + "grad_norm": 12.388105754495733, + "learning_rate": 1.0744871473609632e-06, + "loss": 1.5925, + "step": 5 + }, + { + "epoch": 0.00045092439500977, + "grad_norm": 13.68767725163484, + "learning_rate": 1.1962080090036876e-06, + "loss": 1.5714, + "step": 6 + }, + { + "epoch": 0.0005260784608447317, + "grad_norm": 13.22384941994748, + "learning_rate": 1.2991215311418868e-06, + "loss": 1.4503, + "step": 7 + }, + { + "epoch": 0.0006012325266796934, + "grad_norm": 9.580226848042845, + "learning_rate": 1.3882692789585548e-06, + "loss": 1.37, + "step": 8 + }, + { + "epoch": 0.000676386592514655, + "grad_norm": 8.412817902814608, + "learning_rate": 1.4669031653683387e-06, + "loss": 1.5099, + "step": 9 + }, + { + "epoch": 0.0007515406583496167, + "grad_norm": 12.99951170957971, + "learning_rate": 1.5372435736804818e-06, + "loss": 1.3597, + "step": 10 + }, + { + "epoch": 0.0008266947241845784, + "grad_norm": 2.609535075320156, + "learning_rate": 1.6008742129373428e-06, + "loss": 0.9571, + "step": 11 + }, + { + "epoch": 0.00090184879001954, + "grad_norm": 6.224454937094173, + "learning_rate": 1.658964435323206e-06, + "loss": 1.4301, + "step": 12 + }, + { + "epoch": 0.0009770028558545017, + "grad_norm": 5.127906161421229, + "learning_rate": 1.7124022597777776e-06, + "loss": 1.4212, + "step": 13 + }, + { + "epoch": 0.0010521569216894633, + "grad_norm": 3.508685506751856, + "learning_rate": 1.761877957461405e-06, + "loss": 1.2688, + "step": 14 + }, + { + "epoch": 0.0011273109875244252, + "grad_norm": 5.135429934264629, + "learning_rate": 1.8079387300451324e-06, + "loss": 1.3217, + "step": 15 + }, + { + "epoch": 0.0012024650533593868, + "grad_norm": 3.239898388293572, + "learning_rate": 1.8510257052780734e-06, + "loss": 1.2745, + "step": 16 + }, + { + "epoch": 0.0012776191191943484, + "grad_norm": 3.0077286349408037, + "learning_rate": 1.891499697130832e-06, + "loss": 1.3359, + "step": 17 + }, + { + "epoch": 0.00135277318502931, + "grad_norm": 3.4676082897260128, + "learning_rate": 1.929659591687857e-06, + "loss": 1.3346, + "step": 18 + }, + { + "epoch": 0.0014279272508642717, + "grad_norm": 2.5391230649327126, + "learning_rate": 1.9657557553855114e-06, + "loss": 1.2042, + "step": 19 + }, + { + "epoch": 0.0015030813166992333, + "grad_norm": 2.590340268969583, + "learning_rate": 2e-06, + "loss": 1.1651, + "step": 20 + }, + { + "epoch": 0.0015782353825341952, + "grad_norm": 3.3271219237368137, + "learning_rate": 2.032573113826056e-06, + "loss": 1.2231, + "step": 21 + }, + { + "epoch": 0.0016533894483691568, + "grad_norm": 3.259270577298688, + "learning_rate": 2.063630639256861e-06, + "loss": 1.1083, + "step": 22 + }, + { + "epoch": 0.0017285435142041184, + "grad_norm": 2.005480766780503, + "learning_rate": 2.093307365019873e-06, + "loss": 1.1884, + "step": 23 + }, + { + "epoch": 0.00180369758003908, + "grad_norm": 2.282999178063758, + "learning_rate": 2.1217208616427245e-06, + "loss": 1.2749, + "step": 24 + }, + { + "epoch": 0.0018788516458740417, + "grad_norm": 2.2316498042925605, + "learning_rate": 2.1489742947219264e-06, + "loss": 1.2457, + "step": 25 + }, + { + "epoch": 0.0019540057117090033, + "grad_norm": 2.125927666525449, + "learning_rate": 2.175158686097296e-06, + "loss": 1.0765, + "step": 26 + }, + { + "epoch": 0.002029159777543965, + "grad_norm": 1.8548124096144114, + "learning_rate": 2.200354748052508e-06, + "loss": 1.1195, + "step": 27 + }, + { + "epoch": 0.0021043138433789266, + "grad_norm": 2.5055325325441222, + "learning_rate": 2.2246343837809235e-06, + "loss": 1.1665, + "step": 28 + }, + { + "epoch": 0.0021794679092138887, + "grad_norm": 2.4832967661612466, + "learning_rate": 2.2480619244333726e-06, + "loss": 1.1458, + "step": 29 + }, + { + "epoch": 0.0022546219750488503, + "grad_norm": 1.7619944016349265, + "learning_rate": 2.270695156364651e-06, + "loss": 1.1224, + "step": 30 + }, + { + "epoch": 0.002329776040883812, + "grad_norm": 2.1492364680933718, + "learning_rate": 2.2925861798799734e-06, + "loss": 1.2799, + "step": 31 + }, + { + "epoch": 0.0024049301067187736, + "grad_norm": 2.178665194849819, + "learning_rate": 2.3137821315975918e-06, + "loss": 1.1449, + "step": 32 + }, + { + "epoch": 0.002480084172553735, + "grad_norm": 1.890218580715651, + "learning_rate": 2.334325795621512e-06, + "loss": 1.1097, + "step": 33 + }, + { + "epoch": 0.002555238238388697, + "grad_norm": 2.3427088141423185, + "learning_rate": 2.3542561234503503e-06, + "loss": 1.1166, + "step": 34 + }, + { + "epoch": 0.0026303923042236585, + "grad_norm": 1.9846185850090332, + "learning_rate": 2.3736086785028504e-06, + "loss": 1.249, + "step": 35 + }, + { + "epoch": 0.00270554637005862, + "grad_norm": 1.94615265501745, + "learning_rate": 2.392416018007375e-06, + "loss": 1.1841, + "step": 36 + }, + { + "epoch": 0.0027807004358935817, + "grad_norm": 1.592436080560082, + "learning_rate": 2.41070802255664e-06, + "loss": 0.9196, + "step": 37 + }, + { + "epoch": 0.0028558545017285434, + "grad_norm": 2.257039079556827, + "learning_rate": 2.4285121817050297e-06, + "loss": 1.2075, + "step": 38 + }, + { + "epoch": 0.002931008567563505, + "grad_norm": 3.2092478220849707, + "learning_rate": 2.445853842461947e-06, + "loss": 1.1348, + "step": 39 + }, + { + "epoch": 0.0030061626333984666, + "grad_norm": 2.618876717717686, + "learning_rate": 2.4627564263195183e-06, + "loss": 1.1768, + "step": 40 + }, + { + "epoch": 0.0030813166992334287, + "grad_norm": 2.440488401685906, + "learning_rate": 2.4792416194780364e-06, + "loss": 1.0871, + "step": 41 + }, + { + "epoch": 0.0031564707650683903, + "grad_norm": 1.8359232151206237, + "learning_rate": 2.4953295401455745e-06, + "loss": 1.1649, + "step": 42 + }, + { + "epoch": 0.003231624830903352, + "grad_norm": 1.744591324724666, + "learning_rate": 2.511038886149501e-06, + "loss": 1.1869, + "step": 43 + }, + { + "epoch": 0.0033067788967383136, + "grad_norm": 2.756026859722828, + "learning_rate": 2.526387065576379e-06, + "loss": 1.1328, + "step": 44 + }, + { + "epoch": 0.0033819329625732752, + "grad_norm": 1.8548527709032043, + "learning_rate": 2.5413903127293017e-06, + "loss": 1.041, + "step": 45 + }, + { + "epoch": 0.003457087028408237, + "grad_norm": 1.7665248367070878, + "learning_rate": 2.5560637913393917e-06, + "loss": 1.2099, + "step": 46 + }, + { + "epoch": 0.0035322410942431985, + "grad_norm": 1.7849356684058084, + "learning_rate": 2.5704216866765804e-06, + "loss": 1.1076, + "step": 47 + }, + { + "epoch": 0.00360739516007816, + "grad_norm": 1.9044692680381092, + "learning_rate": 2.584477287962243e-06, + "loss": 1.2239, + "step": 48 + }, + { + "epoch": 0.0036825492259131218, + "grad_norm": 1.9276399444362924, + "learning_rate": 2.5982430622837735e-06, + "loss": 1.1172, + "step": 49 + }, + { + "epoch": 0.0037577032917480834, + "grad_norm": 1.6286821462765082, + "learning_rate": 2.6117307210414448e-06, + "loss": 1.0843, + "step": 50 + }, + { + "epoch": 0.003832857357583045, + "grad_norm": 4.589773104509514, + "learning_rate": 2.624951279815001e-06, + "loss": 1.0887, + "step": 51 + }, + { + "epoch": 0.003908011423418007, + "grad_norm": 1.8095845419030396, + "learning_rate": 2.6379151124168143e-06, + "loss": 1.135, + "step": 52 + }, + { + "epoch": 0.003983165489252968, + "grad_norm": 2.0249456139313673, + "learning_rate": 2.650631999796137e-06, + "loss": 1.1699, + "step": 53 + }, + { + "epoch": 0.00405831955508793, + "grad_norm": 3.3810679670541015, + "learning_rate": 2.6631111743720262e-06, + "loss": 0.9961, + "step": 54 + }, + { + "epoch": 0.004133473620922892, + "grad_norm": 2.883414022239092, + "learning_rate": 2.675361360298306e-06, + "loss": 1.1241, + "step": 55 + }, + { + "epoch": 0.004208627686757853, + "grad_norm": 1.9871589775935983, + "learning_rate": 2.6873908101004422e-06, + "loss": 1.1867, + "step": 56 + }, + { + "epoch": 0.004283781752592815, + "grad_norm": 1.7754708619236523, + "learning_rate": 2.6992073380696804e-06, + "loss": 1.1486, + "step": 57 + }, + { + "epoch": 0.004358935818427777, + "grad_norm": 1.8453019932408203, + "learning_rate": 2.710818350752891e-06, + "loss": 1.0997, + "step": 58 + }, + { + "epoch": 0.004434089884262739, + "grad_norm": 1.6491509651771799, + "learning_rate": 2.7222308748360397e-06, + "loss": 1.0583, + "step": 59 + }, + { + "epoch": 0.004509243950097701, + "grad_norm": 1.8118517143803337, + "learning_rate": 2.733451582684169e-06, + "loss": 1.1229, + "step": 60 + }, + { + "epoch": 0.004584398015932662, + "grad_norm": 1.8392388064028415, + "learning_rate": 2.744486815770336e-06, + "loss": 1.1301, + "step": 61 + }, + { + "epoch": 0.004659552081767624, + "grad_norm": 1.8364817360526235, + "learning_rate": 2.755342606199492e-06, + "loss": 1.135, + "step": 62 + }, + { + "epoch": 0.0047347061476025855, + "grad_norm": 2.551877310604906, + "learning_rate": 2.766024696510225e-06, + "loss": 1.0844, + "step": 63 + }, + { + "epoch": 0.004809860213437547, + "grad_norm": 3.4208916074100384, + "learning_rate": 2.7765385579171097e-06, + "loss": 1.1576, + "step": 64 + }, + { + "epoch": 0.004885014279272509, + "grad_norm": 2.858376917678217, + "learning_rate": 2.7868894071387408e-06, + "loss": 1.1018, + "step": 65 + }, + { + "epoch": 0.00496016834510747, + "grad_norm": 1.638967857864937, + "learning_rate": 2.79708222194103e-06, + "loss": 1.1346, + "step": 66 + }, + { + "epoch": 0.005035322410942432, + "grad_norm": 4.873225940785661, + "learning_rate": 2.807121755511699e-06, + "loss": 1.1601, + "step": 67 + }, + { + "epoch": 0.005110476476777394, + "grad_norm": 1.790649110322178, + "learning_rate": 2.8170125497698686e-06, + "loss": 1.1186, + "step": 68 + }, + { + "epoch": 0.005185630542612355, + "grad_norm": 1.6694744380697435, + "learning_rate": 2.826758947704043e-06, + "loss": 1.105, + "step": 69 + }, + { + "epoch": 0.005260784608447317, + "grad_norm": 1.9640036392799736, + "learning_rate": 2.8363651048223687e-06, + "loss": 1.1251, + "step": 70 + }, + { + "epoch": 0.005335938674282279, + "grad_norm": 2.9608295994727993, + "learning_rate": 2.8458349997907386e-06, + "loss": 1.1333, + "step": 71 + }, + { + "epoch": 0.00541109274011724, + "grad_norm": 1.4894132539268274, + "learning_rate": 2.8551724443268935e-06, + "loss": 0.9023, + "step": 72 + }, + { + "epoch": 0.005486246805952202, + "grad_norm": 1.7825984201104776, + "learning_rate": 2.8643810924121057e-06, + "loss": 1.1308, + "step": 73 + }, + { + "epoch": 0.0055614008717871635, + "grad_norm": 1.6616757608366988, + "learning_rate": 2.8734644488761585e-06, + "loss": 1.2039, + "step": 74 + }, + { + "epoch": 0.005636554937622125, + "grad_norm": 3.9615134999782917, + "learning_rate": 2.882425877406096e-06, + "loss": 1.1325, + "step": 75 + }, + { + "epoch": 0.005711709003457087, + "grad_norm": 2.0729672091918196, + "learning_rate": 2.891268608024548e-06, + "loss": 1.2151, + "step": 76 + }, + { + "epoch": 0.005786863069292048, + "grad_norm": 1.349684319851223, + "learning_rate": 2.8999957440792298e-06, + "loss": 0.8944, + "step": 77 + }, + { + "epoch": 0.00586201713512701, + "grad_norm": 1.7327859145550963, + "learning_rate": 2.9086102687814654e-06, + "loss": 1.0001, + "step": 78 + }, + { + "epoch": 0.005937171200961972, + "grad_norm": 1.585240478684354, + "learning_rate": 2.9171150513282156e-06, + "loss": 1.116, + "step": 79 + }, + { + "epoch": 0.006012325266796933, + "grad_norm": 1.4778775610893107, + "learning_rate": 2.9255128526390366e-06, + "loss": 1.1466, + "step": 80 + }, + { + "epoch": 0.006087479332631896, + "grad_norm": 1.7168056934811555, + "learning_rate": 2.9338063307366773e-06, + "loss": 1.1567, + "step": 81 + }, + { + "epoch": 0.006162633398466857, + "grad_norm": 2.002222495216034, + "learning_rate": 2.9419980457975543e-06, + "loss": 1.0501, + "step": 82 + }, + { + "epoch": 0.006237787464301819, + "grad_norm": 2.1227045564065343, + "learning_rate": 2.9500904648961173e-06, + "loss": 1.0303, + "step": 83 + }, + { + "epoch": 0.006312941530136781, + "grad_norm": 2.069755508063496, + "learning_rate": 2.958085966465093e-06, + "loss": 1.1318, + "step": 84 + }, + { + "epoch": 0.006388095595971742, + "grad_norm": 2.602645419525882, + "learning_rate": 2.965986844491795e-06, + "loss": 1.1557, + "step": 85 + }, + { + "epoch": 0.006463249661806704, + "grad_norm": 2.144488113608742, + "learning_rate": 2.973795312469019e-06, + "loss": 1.1513, + "step": 86 + }, + { + "epoch": 0.006538403727641666, + "grad_norm": 2.0798511649948113, + "learning_rate": 2.981513507117542e-06, + "loss": 1.1615, + "step": 87 + }, + { + "epoch": 0.006613557793476627, + "grad_norm": 1.7025601911617716, + "learning_rate": 2.989143491895898e-06, + "loss": 1.1137, + "step": 88 + }, + { + "epoch": 0.006688711859311589, + "grad_norm": 2.023122402091681, + "learning_rate": 2.9966872603118436e-06, + "loss": 1.0988, + "step": 89 + }, + { + "epoch": 0.0067638659251465505, + "grad_norm": 1.7300883751275085, + "learning_rate": 3.00414673904882e-06, + "loss": 1.0485, + "step": 90 + }, + { + "epoch": 0.006839019990981512, + "grad_norm": 2.1158601886516846, + "learning_rate": 3.0115237909196643e-06, + "loss": 1.1396, + "step": 91 + }, + { + "epoch": 0.006914174056816474, + "grad_norm": 1.9068741323843563, + "learning_rate": 3.01882021765891e-06, + "loss": 1.1073, + "step": 92 + }, + { + "epoch": 0.006989328122651435, + "grad_norm": 1.8025197142053335, + "learning_rate": 3.0260377625641433e-06, + "loss": 1.1589, + "step": 93 + }, + { + "epoch": 0.007064482188486397, + "grad_norm": 3.9397388935224247, + "learning_rate": 3.033178112996099e-06, + "loss": 1.069, + "step": 94 + }, + { + "epoch": 0.007139636254321359, + "grad_norm": 2.0568157820637305, + "learning_rate": 3.0402429027464746e-06, + "loss": 1.1373, + "step": 95 + }, + { + "epoch": 0.00721479032015632, + "grad_norm": 1.696315205163402, + "learning_rate": 3.047233714281761e-06, + "loss": 1.1262, + "step": 96 + }, + { + "epoch": 0.007289944385991282, + "grad_norm": 2.0463563684257635, + "learning_rate": 3.0541520808708106e-06, + "loss": 1.122, + "step": 97 + }, + { + "epoch": 0.0073650984518262436, + "grad_norm": 1.982433901595293, + "learning_rate": 3.0609994886032923e-06, + "loss": 1.0103, + "step": 98 + }, + { + "epoch": 0.007440252517661205, + "grad_norm": 1.4020132881492582, + "learning_rate": 3.067777378305681e-06, + "loss": 1.0623, + "step": 99 + }, + { + "epoch": 0.007515406583496167, + "grad_norm": 1.6424301389239444, + "learning_rate": 3.0744871473609635e-06, + "loss": 1.0666, + "step": 100 + }, + { + "epoch": 0.0075905606493311285, + "grad_norm": 0.8801821454608585, + "learning_rate": 3.081130151437802e-06, + "loss": 0.8185, + "step": 101 + }, + { + "epoch": 0.00766571471516609, + "grad_norm": 2.073701898294034, + "learning_rate": 3.0877077061345193e-06, + "loss": 1.009, + "step": 102 + }, + { + "epoch": 0.007740868781001052, + "grad_norm": 0.8145082304171625, + "learning_rate": 3.0942210885428814e-06, + "loss": 0.8506, + "step": 103 + }, + { + "epoch": 0.007816022846836013, + "grad_norm": 0.9120029820343254, + "learning_rate": 3.1006715387363326e-06, + "loss": 0.8558, + "step": 104 + }, + { + "epoch": 0.007891176912670975, + "grad_norm": 1.9612439284707808, + "learning_rate": 3.107060261187019e-06, + "loss": 1.0502, + "step": 105 + }, + { + "epoch": 0.007966330978505937, + "grad_norm": 2.225289070982099, + "learning_rate": 3.113388426115655e-06, + "loss": 1.0749, + "step": 106 + }, + { + "epoch": 0.008041485044340898, + "grad_norm": 2.421513684300461, + "learning_rate": 3.119657170778007e-06, + "loss": 1.0466, + "step": 107 + }, + { + "epoch": 0.00811663911017586, + "grad_norm": 1.5443069256515152, + "learning_rate": 3.1258676006915446e-06, + "loss": 1.105, + "step": 108 + }, + { + "epoch": 0.008191793176010822, + "grad_norm": 7.529310807785423, + "learning_rate": 3.1320207908055525e-06, + "loss": 1.0512, + "step": 109 + }, + { + "epoch": 0.008266947241845783, + "grad_norm": 2.6601886047098566, + "learning_rate": 3.138117786617824e-06, + "loss": 1.1679, + "step": 110 + }, + { + "epoch": 0.008342101307680745, + "grad_norm": 1.512736369807768, + "learning_rate": 3.144159605240809e-06, + "loss": 1.0661, + "step": 111 + }, + { + "epoch": 0.008417255373515706, + "grad_norm": 0.7882095107675242, + "learning_rate": 3.1501472364199597e-06, + "loss": 0.8391, + "step": 112 + }, + { + "epoch": 0.008492409439350668, + "grad_norm": 2.3470170710771345, + "learning_rate": 3.156081643506813e-06, + "loss": 1.1089, + "step": 113 + }, + { + "epoch": 0.00856756350518563, + "grad_norm": 2.459852191237184, + "learning_rate": 3.161963764389199e-06, + "loss": 1.1838, + "step": 114 + }, + { + "epoch": 0.008642717571020593, + "grad_norm": 1.5488105763310698, + "learning_rate": 3.167794512380837e-06, + "loss": 1.0867, + "step": 115 + }, + { + "epoch": 0.008717871636855555, + "grad_norm": 0.8319760872907396, + "learning_rate": 3.1735747770724093e-06, + "loss": 0.8672, + "step": 116 + }, + { + "epoch": 0.008793025702690516, + "grad_norm": 1.626620804825464, + "learning_rate": 3.179305425146116e-06, + "loss": 1.0725, + "step": 117 + }, + { + "epoch": 0.008868179768525478, + "grad_norm": 1.6596168988308746, + "learning_rate": 3.184987301155558e-06, + "loss": 1.0662, + "step": 118 + }, + { + "epoch": 0.00894333383436044, + "grad_norm": 1.7667404972099925, + "learning_rate": 3.190621228272719e-06, + "loss": 0.9914, + "step": 119 + }, + { + "epoch": 0.009018487900195401, + "grad_norm": 1.755701518362099, + "learning_rate": 3.1962080090036873e-06, + "loss": 1.2045, + "step": 120 + }, + { + "epoch": 0.009093641966030363, + "grad_norm": 1.9425464252437559, + "learning_rate": 3.2017484258746856e-06, + "loss": 1.1037, + "step": 121 + }, + { + "epoch": 0.009168796031865325, + "grad_norm": 1.9174702107131072, + "learning_rate": 3.207243242089855e-06, + "loss": 1.1712, + "step": 122 + }, + { + "epoch": 0.009243950097700286, + "grad_norm": 5.11744672977692, + "learning_rate": 3.212693202162205e-06, + "loss": 1.1213, + "step": 123 + }, + { + "epoch": 0.009319104163535248, + "grad_norm": 1.5719950757131544, + "learning_rate": 3.2180990325190106e-06, + "loss": 1.06, + "step": 124 + }, + { + "epoch": 0.00939425822937021, + "grad_norm": 11.597724007254074, + "learning_rate": 3.22346144208289e-06, + "loss": 1.0653, + "step": 125 + }, + { + "epoch": 0.009469412295205171, + "grad_norm": 1.8650912085108473, + "learning_rate": 3.2287811228297436e-06, + "loss": 1.0932, + "step": 126 + }, + { + "epoch": 0.009544566361040133, + "grad_norm": 2.931751226261455, + "learning_rate": 3.2340587503246298e-06, + "loss": 1.1588, + "step": 127 + }, + { + "epoch": 0.009619720426875094, + "grad_norm": 1.812469482164954, + "learning_rate": 3.239294984236628e-06, + "loss": 1.1456, + "step": 128 + }, + { + "epoch": 0.009694874492710056, + "grad_norm": 2.098074134389625, + "learning_rate": 3.24449046883367e-06, + "loss": 1.0444, + "step": 129 + }, + { + "epoch": 0.009770028558545018, + "grad_norm": 2.2246254509268795, + "learning_rate": 3.249645833458259e-06, + "loss": 0.9896, + "step": 130 + }, + { + "epoch": 0.00984518262437998, + "grad_norm": 3.401477515123196, + "learning_rate": 3.2547616929849703e-06, + "loss": 1.1776, + "step": 131 + }, + { + "epoch": 0.00992033669021494, + "grad_norm": 1.5961480144998754, + "learning_rate": 3.2598386482605483e-06, + "loss": 1.1009, + "step": 132 + }, + { + "epoch": 0.009995490756049902, + "grad_norm": 2.3196854177489015, + "learning_rate": 3.2648772865273986e-06, + "loss": 0.9942, + "step": 133 + }, + { + "epoch": 0.010070644821884864, + "grad_norm": 2.1204445198355084, + "learning_rate": 3.269878181831217e-06, + "loss": 1.0595, + "step": 134 + }, + { + "epoch": 0.010145798887719826, + "grad_norm": 2.3289766738005198, + "learning_rate": 3.274841895413471e-06, + "loss": 1.1791, + "step": 135 + }, + { + "epoch": 0.010220952953554787, + "grad_norm": 1.9112773951680895, + "learning_rate": 3.279768976089387e-06, + "loss": 1.1193, + "step": 136 + }, + { + "epoch": 0.010296107019389749, + "grad_norm": 1.7642749587481872, + "learning_rate": 3.2846599606121004e-06, + "loss": 1.0484, + "step": 137 + }, + { + "epoch": 0.01037126108522471, + "grad_norm": 1.0019792400920182, + "learning_rate": 3.289515374023561e-06, + "loss": 0.8875, + "step": 138 + }, + { + "epoch": 0.010446415151059672, + "grad_norm": 1.690720134885921, + "learning_rate": 3.2943357299927686e-06, + "loss": 1.0966, + "step": 139 + }, + { + "epoch": 0.010521569216894634, + "grad_norm": 4.139089853981904, + "learning_rate": 3.2991215311418867e-06, + "loss": 1.0456, + "step": 140 + }, + { + "epoch": 0.010596723282729596, + "grad_norm": 3.1268753528156346, + "learning_rate": 3.30387326936075e-06, + "loss": 1.0806, + "step": 141 + }, + { + "epoch": 0.010671877348564557, + "grad_norm": 1.4805493496416264, + "learning_rate": 3.308591426110257e-06, + "loss": 1.0486, + "step": 142 + }, + { + "epoch": 0.010747031414399519, + "grad_norm": 1.8277566000623977, + "learning_rate": 3.3132764727151197e-06, + "loss": 1.0203, + "step": 143 + }, + { + "epoch": 0.01082218548023448, + "grad_norm": 2.878248121380431, + "learning_rate": 3.317928870646412e-06, + "loss": 1.1111, + "step": 144 + }, + { + "epoch": 0.010897339546069442, + "grad_norm": 1.9476065839363546, + "learning_rate": 3.3225490717943362e-06, + "loss": 1.1308, + "step": 145 + }, + { + "epoch": 0.010972493611904404, + "grad_norm": 1.625847228808962, + "learning_rate": 3.327137518731624e-06, + "loss": 1.1357, + "step": 146 + }, + { + "epoch": 0.011047647677739365, + "grad_norm": 2.4147742743665668, + "learning_rate": 3.3316946449679425e-06, + "loss": 1.1238, + "step": 147 + }, + { + "epoch": 0.011122801743574327, + "grad_norm": 1.9731345626024168, + "learning_rate": 3.336220875195677e-06, + "loss": 1.1835, + "step": 148 + }, + { + "epoch": 0.011197955809409289, + "grad_norm": 1.8663156940116952, + "learning_rate": 3.3407166255274344e-06, + "loss": 1.1057, + "step": 149 + }, + { + "epoch": 0.01127310987524425, + "grad_norm": 1.6230514287597821, + "learning_rate": 3.345182303725614e-06, + "loss": 1.0232, + "step": 150 + }, + { + "epoch": 0.011348263941079212, + "grad_norm": 1.4955580427052444, + "learning_rate": 3.3496183094243384e-06, + "loss": 1.1147, + "step": 151 + }, + { + "epoch": 0.011423418006914174, + "grad_norm": 2.199853293989325, + "learning_rate": 3.3540250343440664e-06, + "loss": 1.1623, + "step": 152 + }, + { + "epoch": 0.011498572072749135, + "grad_norm": 2.597445325964439, + "learning_rate": 3.35840286249917e-06, + "loss": 1.19, + "step": 153 + }, + { + "epoch": 0.011573726138584097, + "grad_norm": 1.5146992796008358, + "learning_rate": 3.3627521703987477e-06, + "loss": 1.138, + "step": 154 + }, + { + "epoch": 0.011648880204419058, + "grad_norm": 1.9387680334239983, + "learning_rate": 3.367073327240937e-06, + "loss": 1.0596, + "step": 155 + }, + { + "epoch": 0.01172403427025402, + "grad_norm": 2.0242404596481522, + "learning_rate": 3.3713666951009833e-06, + "loss": 1.1196, + "step": 156 + }, + { + "epoch": 0.011799188336088982, + "grad_norm": 1.8826434427554382, + "learning_rate": 3.375632629113298e-06, + "loss": 1.0953, + "step": 157 + }, + { + "epoch": 0.011874342401923943, + "grad_norm": 1.8235559995403137, + "learning_rate": 3.3798714776477344e-06, + "loss": 1.1057, + "step": 158 + }, + { + "epoch": 0.011949496467758905, + "grad_norm": 1.8712484339603657, + "learning_rate": 3.3840835824803065e-06, + "loss": 1.1486, + "step": 159 + }, + { + "epoch": 0.012024650533593867, + "grad_norm": 1.922684909772807, + "learning_rate": 3.388269278958555e-06, + "loss": 0.9706, + "step": 160 + }, + { + "epoch": 0.01209980459942883, + "grad_norm": 0.9190038486043147, + "learning_rate": 3.3924288961617605e-06, + "loss": 0.8694, + "step": 161 + }, + { + "epoch": 0.012174958665263792, + "grad_norm": 2.6246098728011975, + "learning_rate": 3.3965627570561953e-06, + "loss": 1.0006, + "step": 162 + }, + { + "epoch": 0.012250112731098753, + "grad_norm": 2.480187203355865, + "learning_rate": 3.4006711786456036e-06, + "loss": 1.054, + "step": 163 + }, + { + "epoch": 0.012325266796933715, + "grad_norm": 1.9069619647125187, + "learning_rate": 3.404754472117073e-06, + "loss": 1.0605, + "step": 164 + }, + { + "epoch": 0.012400420862768676, + "grad_norm": 2.4368619817409027, + "learning_rate": 3.408812942982475e-06, + "loss": 1.1122, + "step": 165 + }, + { + "epoch": 0.012475574928603638, + "grad_norm": 5.812584332910557, + "learning_rate": 3.4128468912156357e-06, + "loss": 1.1156, + "step": 166 + }, + { + "epoch": 0.0125507289944386, + "grad_norm": 2.1306374886265083, + "learning_rate": 3.4168566113853806e-06, + "loss": 1.019, + "step": 167 + }, + { + "epoch": 0.012625883060273561, + "grad_norm": 2.4426221581544953, + "learning_rate": 3.420842392784611e-06, + "loss": 1.0799, + "step": 168 + }, + { + "epoch": 0.012701037126108523, + "grad_norm": 1.96109257423038, + "learning_rate": 3.424804519555555e-06, + "loss": 1.0574, + "step": 169 + }, + { + "epoch": 0.012776191191943485, + "grad_norm": 1.7035925687064901, + "learning_rate": 3.4287432708113135e-06, + "loss": 1.1223, + "step": 170 + }, + { + "epoch": 0.012851345257778446, + "grad_norm": 1.5397348242505946, + "learning_rate": 3.4326589207538503e-06, + "loss": 1.0507, + "step": 171 + }, + { + "epoch": 0.012926499323613408, + "grad_norm": 2.0619491072626315, + "learning_rate": 3.436551738788537e-06, + "loss": 0.9284, + "step": 172 + }, + { + "epoch": 0.01300165338944837, + "grad_norm": 2.8835367259647993, + "learning_rate": 3.440421989635386e-06, + "loss": 1.1311, + "step": 173 + }, + { + "epoch": 0.013076807455283331, + "grad_norm": 1.5360621140805468, + "learning_rate": 3.44426993343706e-06, + "loss": 1.1177, + "step": 174 + }, + { + "epoch": 0.013151961521118293, + "grad_norm": 2.7945557253829287, + "learning_rate": 3.4480958258638136e-06, + "loss": 1.1016, + "step": 175 + }, + { + "epoch": 0.013227115586953254, + "grad_norm": 3.970424509101099, + "learning_rate": 3.4518999182154156e-06, + "loss": 1.0868, + "step": 176 + }, + { + "epoch": 0.013302269652788216, + "grad_norm": 1.4478811172768453, + "learning_rate": 3.4556824575202087e-06, + "loss": 1.0982, + "step": 177 + }, + { + "epoch": 0.013377423718623178, + "grad_norm": 2.116032922542716, + "learning_rate": 3.4594436866313616e-06, + "loss": 1.0881, + "step": 178 + }, + { + "epoch": 0.01345257778445814, + "grad_norm": 2.6088159384845673, + "learning_rate": 3.463183844320436e-06, + "loss": 1.1314, + "step": 179 + }, + { + "epoch": 0.013527731850293101, + "grad_norm": 2.665228803735887, + "learning_rate": 3.4669031653683388e-06, + "loss": 0.9712, + "step": 180 + }, + { + "epoch": 0.013602885916128063, + "grad_norm": 1.542136902447741, + "learning_rate": 3.4706018806537624e-06, + "loss": 1.1424, + "step": 181 + }, + { + "epoch": 0.013678039981963024, + "grad_norm": 1.7602656899744276, + "learning_rate": 3.4742802172391827e-06, + "loss": 1.0283, + "step": 182 + }, + { + "epoch": 0.013753194047797986, + "grad_norm": 1.7757619672971097, + "learning_rate": 3.4779383984545055e-06, + "loss": 1.1934, + "step": 183 + }, + { + "epoch": 0.013828348113632948, + "grad_norm": 2.127716459924063, + "learning_rate": 3.481576643978429e-06, + "loss": 1.0334, + "step": 184 + }, + { + "epoch": 0.01390350217946791, + "grad_norm": 1.6072015993796747, + "learning_rate": 3.485195169917603e-06, + "loss": 1.0246, + "step": 185 + }, + { + "epoch": 0.01397865624530287, + "grad_norm": 1.9194440897186633, + "learning_rate": 3.4887941888836612e-06, + "loss": 1.0706, + "step": 186 + }, + { + "epoch": 0.014053810311137832, + "grad_norm": 1.845285196772569, + "learning_rate": 3.4923739100681745e-06, + "loss": 1.0426, + "step": 187 + }, + { + "epoch": 0.014128964376972794, + "grad_norm": 1.7967772084038145, + "learning_rate": 3.4959345393156175e-06, + "loss": 1.1329, + "step": 188 + }, + { + "epoch": 0.014204118442807756, + "grad_norm": 4.783087283905041, + "learning_rate": 3.4994762791943946e-06, + "loss": 1.0861, + "step": 189 + }, + { + "epoch": 0.014279272508642717, + "grad_norm": 2.318848118929597, + "learning_rate": 3.502999329065993e-06, + "loss": 1.0442, + "step": 190 + }, + { + "epoch": 0.014354426574477679, + "grad_norm": 2.3145783853234785, + "learning_rate": 3.506503885152319e-06, + "loss": 0.9882, + "step": 191 + }, + { + "epoch": 0.01442958064031264, + "grad_norm": 2.1885187229247185, + "learning_rate": 3.5099901406012796e-06, + "loss": 1.15, + "step": 192 + }, + { + "epoch": 0.014504734706147602, + "grad_norm": 2.1770312897760147, + "learning_rate": 3.513458285550655e-06, + "loss": 1.1009, + "step": 193 + }, + { + "epoch": 0.014579888771982564, + "grad_norm": 1.6757743497408564, + "learning_rate": 3.516908507190329e-06, + "loss": 1.023, + "step": 194 + }, + { + "epoch": 0.014655042837817525, + "grad_norm": 2.518687782749711, + "learning_rate": 3.5203409898229102e-06, + "loss": 1.0276, + "step": 195 + }, + { + "epoch": 0.014730196903652487, + "grad_norm": 2.30422958738795, + "learning_rate": 3.52375591492281e-06, + "loss": 1.043, + "step": 196 + }, + { + "epoch": 0.014805350969487449, + "grad_norm": 0.8203522808783303, + "learning_rate": 3.527153461193815e-06, + "loss": 0.8988, + "step": 197 + }, + { + "epoch": 0.01488050503532241, + "grad_norm": 1.8823869786270535, + "learning_rate": 3.5305338046251994e-06, + "loss": 1.0904, + "step": 198 + }, + { + "epoch": 0.014955659101157372, + "grad_norm": 1.7506284947549406, + "learning_rate": 3.533897118546427e-06, + "loss": 1.1824, + "step": 199 + }, + { + "epoch": 0.015030813166992334, + "grad_norm": 1.898903557982585, + "learning_rate": 3.5372435736804815e-06, + "loss": 1.1426, + "step": 200 + }, + { + "epoch": 0.015105967232827295, + "grad_norm": 1.8938003231198575, + "learning_rate": 3.5405733381958684e-06, + "loss": 1.0372, + "step": 201 + }, + { + "epoch": 0.015181121298662257, + "grad_norm": 1.6407011468563977, + "learning_rate": 3.5438865777573207e-06, + "loss": 1.1145, + "step": 202 + }, + { + "epoch": 0.015256275364497219, + "grad_norm": 1.820066443582367, + "learning_rate": 3.5471834555752594e-06, + "loss": 1.1626, + "step": 203 + }, + { + "epoch": 0.01533142943033218, + "grad_norm": 2.332461012405062, + "learning_rate": 3.5504641324540377e-06, + "loss": 1.1489, + "step": 204 + }, + { + "epoch": 0.015406583496167142, + "grad_norm": 1.5759012398830679, + "learning_rate": 3.5537287668389996e-06, + "loss": 1.0808, + "step": 205 + }, + { + "epoch": 0.015481737562002103, + "grad_norm": 1.664092945662101, + "learning_rate": 3.5569775148623998e-06, + "loss": 1.0845, + "step": 206 + }, + { + "epoch": 0.015556891627837067, + "grad_norm": 3.246811696703037, + "learning_rate": 3.5602105303882114e-06, + "loss": 1.0256, + "step": 207 + }, + { + "epoch": 0.015632045693672027, + "grad_norm": 2.4430756987415463, + "learning_rate": 3.563427965055851e-06, + "loss": 1.0377, + "step": 208 + }, + { + "epoch": 0.01570719975950699, + "grad_norm": 1.539659800542594, + "learning_rate": 3.566629968322854e-06, + "loss": 1.1123, + "step": 209 + }, + { + "epoch": 0.01578235382534195, + "grad_norm": 1.469864361661973, + "learning_rate": 3.5698166875065377e-06, + "loss": 1.0127, + "step": 210 + }, + { + "epoch": 0.015857507891176913, + "grad_norm": 1.6128327344287736, + "learning_rate": 3.5729882678246694e-06, + "loss": 1.0982, + "step": 211 + }, + { + "epoch": 0.015932661957011873, + "grad_norm": 1.884057610702055, + "learning_rate": 3.5761448524351738e-06, + "loss": 0.914, + "step": 212 + }, + { + "epoch": 0.016007816022846837, + "grad_norm": 1.5767800671535424, + "learning_rate": 3.579286582474908e-06, + "loss": 1.1374, + "step": 213 + }, + { + "epoch": 0.016082970088681797, + "grad_norm": 2.0846416434378297, + "learning_rate": 3.582413597097526e-06, + "loss": 1.0611, + "step": 214 + }, + { + "epoch": 0.01615812415451676, + "grad_norm": 2.462951126989078, + "learning_rate": 3.5855260335104637e-06, + "loss": 1.1144, + "step": 215 + }, + { + "epoch": 0.01623327822035172, + "grad_norm": 1.938130429975917, + "learning_rate": 3.588624027011063e-06, + "loss": 1.0811, + "step": 216 + }, + { + "epoch": 0.016308432286186683, + "grad_norm": 1.1173601884023459, + "learning_rate": 3.5917077110218606e-06, + "loss": 0.9977, + "step": 217 + }, + { + "epoch": 0.016383586352021643, + "grad_norm": 2.1389928010852404, + "learning_rate": 3.5947772171250713e-06, + "loss": 1.0804, + "step": 218 + }, + { + "epoch": 0.016458740417856606, + "grad_norm": 2.0652826207245334, + "learning_rate": 3.597832675096275e-06, + "loss": 1.0882, + "step": 219 + }, + { + "epoch": 0.016533894483691566, + "grad_norm": 1.673353308118977, + "learning_rate": 3.600874212937343e-06, + "loss": 1.0719, + "step": 220 + }, + { + "epoch": 0.01660904854952653, + "grad_norm": 1.5832968088588528, + "learning_rate": 3.603901956908609e-06, + "loss": 1.0512, + "step": 221 + }, + { + "epoch": 0.01668420261536149, + "grad_norm": 1.7140381469665698, + "learning_rate": 3.6069160315603275e-06, + "loss": 1.1442, + "step": 222 + }, + { + "epoch": 0.016759356681196453, + "grad_norm": 1.5425412901757194, + "learning_rate": 3.6099165597634083e-06, + "loss": 1.1317, + "step": 223 + }, + { + "epoch": 0.016834510747031413, + "grad_norm": 1.8009195336725363, + "learning_rate": 3.6129036627394785e-06, + "loss": 1.1171, + "step": 224 + }, + { + "epoch": 0.016909664812866376, + "grad_norm": 2.1454093538941157, + "learning_rate": 3.615877460090265e-06, + "loss": 1.0289, + "step": 225 + }, + { + "epoch": 0.016984818878701336, + "grad_norm": 2.1745164574872864, + "learning_rate": 3.6188380698263314e-06, + "loss": 1.1263, + "step": 226 + }, + { + "epoch": 0.0170599729445363, + "grad_norm": 1.5778331675604778, + "learning_rate": 3.6217856083951765e-06, + "loss": 1.1357, + "step": 227 + }, + { + "epoch": 0.01713512701037126, + "grad_norm": 1.7975520972924453, + "learning_rate": 3.6247201907087175e-06, + "loss": 1.0921, + "step": 228 + }, + { + "epoch": 0.017210281076206223, + "grad_norm": 1.6556261435159527, + "learning_rate": 3.627641930170173e-06, + "loss": 1.1342, + "step": 229 + }, + { + "epoch": 0.017285435142041186, + "grad_norm": 1.5340833336228876, + "learning_rate": 3.630550938700355e-06, + "loss": 1.0676, + "step": 230 + }, + { + "epoch": 0.017360589207876146, + "grad_norm": 1.631924474427164, + "learning_rate": 3.6334473267633984e-06, + "loss": 1.0003, + "step": 231 + }, + { + "epoch": 0.01743574327371111, + "grad_norm": 1.9177551150551497, + "learning_rate": 3.6363312033919277e-06, + "loss": 1.0893, + "step": 232 + }, + { + "epoch": 0.01751089733954607, + "grad_norm": 1.9107235805147886, + "learning_rate": 3.639202676211685e-06, + "loss": 1.2072, + "step": 233 + }, + { + "epoch": 0.017586051405381033, + "grad_norm": 1.7493272578194028, + "learning_rate": 3.642061851465635e-06, + "loss": 1.1226, + "step": 234 + }, + { + "epoch": 0.017661205471215993, + "grad_norm": 1.503098783839403, + "learning_rate": 3.644908834037544e-06, + "loss": 1.0557, + "step": 235 + }, + { + "epoch": 0.017736359537050956, + "grad_norm": 1.7921441203473851, + "learning_rate": 3.647743727475077e-06, + "loss": 1.1443, + "step": 236 + }, + { + "epoch": 0.017811513602885916, + "grad_norm": 1.7693531207770308, + "learning_rate": 3.650566634012385e-06, + "loss": 1.127, + "step": 237 + }, + { + "epoch": 0.01788666766872088, + "grad_norm": 1.618221965667775, + "learning_rate": 3.653377654592237e-06, + "loss": 1.0812, + "step": 238 + }, + { + "epoch": 0.01796182173455584, + "grad_norm": 5.33917504499832, + "learning_rate": 3.6561768888876717e-06, + "loss": 1.0664, + "step": 239 + }, + { + "epoch": 0.018036975800390802, + "grad_norm": 1.6831991702458473, + "learning_rate": 3.658964435323206e-06, + "loss": 1.0212, + "step": 240 + }, + { + "epoch": 0.018112129866225762, + "grad_norm": 1.7707726038074854, + "learning_rate": 3.6617403910956026e-06, + "loss": 1.0889, + "step": 241 + }, + { + "epoch": 0.018187283932060726, + "grad_norm": 1.4644424092126371, + "learning_rate": 3.6645048521942035e-06, + "loss": 1.1205, + "step": 242 + }, + { + "epoch": 0.018262437997895686, + "grad_norm": 1.4313908836623437, + "learning_rate": 3.667257913420846e-06, + "loss": 1.1053, + "step": 243 + }, + { + "epoch": 0.01833759206373065, + "grad_norm": 1.69512411028893, + "learning_rate": 3.6699996684093732e-06, + "loss": 1.0573, + "step": 244 + }, + { + "epoch": 0.01841274612956561, + "grad_norm": 2.512253106735515, + "learning_rate": 3.6727302096447376e-06, + "loss": 1.1433, + "step": 245 + }, + { + "epoch": 0.018487900195400572, + "grad_norm": 1.7529892595204744, + "learning_rate": 3.6754496284817233e-06, + "loss": 1.1632, + "step": 246 + }, + { + "epoch": 0.018563054261235532, + "grad_norm": 1.7925694505120457, + "learning_rate": 3.678158015163289e-06, + "loss": 1.0594, + "step": 247 + }, + { + "epoch": 0.018638208327070496, + "grad_norm": 2.06973281518147, + "learning_rate": 3.680855458838529e-06, + "loss": 1.0693, + "step": 248 + }, + { + "epoch": 0.018713362392905455, + "grad_norm": 2.0418626989596014, + "learning_rate": 3.6835420475802863e-06, + "loss": 1.0838, + "step": 249 + }, + { + "epoch": 0.01878851645874042, + "grad_norm": 1.8332928378516722, + "learning_rate": 3.686217868402408e-06, + "loss": 1.0107, + "step": 250 + }, + { + "epoch": 0.01886367052457538, + "grad_norm": 2.2569529867418137, + "learning_rate": 3.688883007276652e-06, + "loss": 1.0969, + "step": 251 + }, + { + "epoch": 0.018938824590410342, + "grad_norm": 1.6379579209545865, + "learning_rate": 3.6915375491492623e-06, + "loss": 1.0278, + "step": 252 + }, + { + "epoch": 0.019013978656245302, + "grad_norm": 1.4033598977215214, + "learning_rate": 3.694181577957216e-06, + "loss": 1.1124, + "step": 253 + }, + { + "epoch": 0.019089132722080265, + "grad_norm": 1.8221425038373842, + "learning_rate": 3.6968151766441486e-06, + "loss": 1.0547, + "step": 254 + }, + { + "epoch": 0.019164286787915225, + "grad_norm": 1.6932879898177928, + "learning_rate": 3.699438427175964e-06, + "loss": 1.1252, + "step": 255 + }, + { + "epoch": 0.01923944085375019, + "grad_norm": 1.7764710683788547, + "learning_rate": 3.702051410556147e-06, + "loss": 1.094, + "step": 256 + }, + { + "epoch": 0.01931459491958515, + "grad_norm": 1.8403693418295997, + "learning_rate": 3.7046542068407645e-06, + "loss": 0.9172, + "step": 257 + }, + { + "epoch": 0.019389748985420112, + "grad_norm": 1.808784820811994, + "learning_rate": 3.7072468951531883e-06, + "loss": 1.1524, + "step": 258 + }, + { + "epoch": 0.019464903051255072, + "grad_norm": 2.2239950803866964, + "learning_rate": 3.7098295536985265e-06, + "loss": 1.0448, + "step": 259 + }, + { + "epoch": 0.019540057117090035, + "grad_norm": 2.098464554351672, + "learning_rate": 3.7124022597777775e-06, + "loss": 1.0072, + "step": 260 + }, + { + "epoch": 0.019615211182924995, + "grad_norm": 2.119153113262454, + "learning_rate": 3.7149650898017115e-06, + "loss": 1.0923, + "step": 261 + }, + { + "epoch": 0.01969036524875996, + "grad_norm": 2.32352664163386, + "learning_rate": 3.717518119304489e-06, + "loss": 1.157, + "step": 262 + }, + { + "epoch": 0.01976551931459492, + "grad_norm": 1.888046587187407, + "learning_rate": 3.7200614229570204e-06, + "loss": 1.0065, + "step": 263 + }, + { + "epoch": 0.01984067338042988, + "grad_norm": 1.8736492377755878, + "learning_rate": 3.7225950745800667e-06, + "loss": 1.0922, + "step": 264 + }, + { + "epoch": 0.01991582744626484, + "grad_norm": 2.1684983165154796, + "learning_rate": 3.7251191471571003e-06, + "loss": 1.0736, + "step": 265 + }, + { + "epoch": 0.019990981512099805, + "grad_norm": 1.5576550709293973, + "learning_rate": 3.7276337128469165e-06, + "loss": 1.0505, + "step": 266 + }, + { + "epoch": 0.020066135577934765, + "grad_norm": 3.045228170731449, + "learning_rate": 3.730138842996013e-06, + "loss": 1.0912, + "step": 267 + }, + { + "epoch": 0.020141289643769728, + "grad_norm": 1.845708227402835, + "learning_rate": 3.7326346081507353e-06, + "loss": 1.086, + "step": 268 + }, + { + "epoch": 0.020216443709604688, + "grad_norm": 1.4238962751072861, + "learning_rate": 3.7351210780691997e-06, + "loss": 1.0822, + "step": 269 + }, + { + "epoch": 0.02029159777543965, + "grad_norm": 2.5242676683253182, + "learning_rate": 3.7375983217329894e-06, + "loss": 1.0147, + "step": 270 + }, + { + "epoch": 0.02036675184127461, + "grad_norm": 2.850309917263504, + "learning_rate": 3.7400664073586386e-06, + "loss": 1.0003, + "step": 271 + }, + { + "epoch": 0.020441905907109575, + "grad_norm": 2.087332269931272, + "learning_rate": 3.7425254024089058e-06, + "loss": 1.059, + "step": 272 + }, + { + "epoch": 0.020517059972944535, + "grad_norm": 1.9919275450800495, + "learning_rate": 3.7449753736038338e-06, + "loss": 1.1257, + "step": 273 + }, + { + "epoch": 0.020592214038779498, + "grad_norm": 1.7502994281190092, + "learning_rate": 3.7474163869316188e-06, + "loss": 1.179, + "step": 274 + }, + { + "epoch": 0.020667368104614458, + "grad_norm": 1.865551283512572, + "learning_rate": 3.749848507659269e-06, + "loss": 1.161, + "step": 275 + }, + { + "epoch": 0.02074252217044942, + "grad_norm": 2.3705834934775742, + "learning_rate": 3.752271800343079e-06, + "loss": 1.0816, + "step": 276 + }, + { + "epoch": 0.020817676236284385, + "grad_norm": 2.1703154089450223, + "learning_rate": 3.7546863288389105e-06, + "loss": 1.0451, + "step": 277 + }, + { + "epoch": 0.020892830302119345, + "grad_norm": 2.511810633282434, + "learning_rate": 3.757092156312287e-06, + "loss": 0.9654, + "step": 278 + }, + { + "epoch": 0.020967984367954308, + "grad_norm": 1.5777371373710807, + "learning_rate": 3.759489345248312e-06, + "loss": 1.0618, + "step": 279 + }, + { + "epoch": 0.021043138433789268, + "grad_norm": 1.9113249840506905, + "learning_rate": 3.7618779574614046e-06, + "loss": 1.035, + "step": 280 + }, + { + "epoch": 0.02111829249962423, + "grad_norm": 1.9540106024439838, + "learning_rate": 3.7642580541048654e-06, + "loss": 1.0515, + "step": 281 + }, + { + "epoch": 0.02119344656545919, + "grad_norm": 2.132877670242596, + "learning_rate": 3.766629695680268e-06, + "loss": 1.143, + "step": 282 + }, + { + "epoch": 0.021268600631294154, + "grad_norm": 0.8878687656596707, + "learning_rate": 3.7689929420466896e-06, + "loss": 0.8472, + "step": 283 + }, + { + "epoch": 0.021343754697129114, + "grad_norm": 0.7525011319883836, + "learning_rate": 3.7713478524297754e-06, + "loss": 0.832, + "step": 284 + }, + { + "epoch": 0.021418908762964078, + "grad_norm": 2.813795511902081, + "learning_rate": 3.7736944854306444e-06, + "loss": 1.165, + "step": 285 + }, + { + "epoch": 0.021494062828799038, + "grad_norm": 2.1353027561478486, + "learning_rate": 3.7760328990346385e-06, + "loss": 1.0944, + "step": 286 + }, + { + "epoch": 0.021569216894634, + "grad_norm": 1.888228292357857, + "learning_rate": 3.778363150619923e-06, + "loss": 1.0381, + "step": 287 + }, + { + "epoch": 0.02164437096046896, + "grad_norm": 2.0204735924029436, + "learning_rate": 3.78068529696593e-06, + "loss": 0.9284, + "step": 288 + }, + { + "epoch": 0.021719525026303924, + "grad_norm": 2.0831512244422545, + "learning_rate": 3.782999394261664e-06, + "loss": 1.1171, + "step": 289 + }, + { + "epoch": 0.021794679092138884, + "grad_norm": 2.3713965937416646, + "learning_rate": 3.785305498113854e-06, + "loss": 1.0919, + "step": 290 + }, + { + "epoch": 0.021869833157973848, + "grad_norm": 1.6095954966186334, + "learning_rate": 3.78760366355498e-06, + "loss": 1.1292, + "step": 291 + }, + { + "epoch": 0.021944987223808807, + "grad_norm": 6.667481595140193, + "learning_rate": 3.789893945051143e-06, + "loss": 1.0208, + "step": 292 + }, + { + "epoch": 0.02202014128964377, + "grad_norm": 1.9728911911455251, + "learning_rate": 3.792176396509817e-06, + "loss": 1.0671, + "step": 293 + }, + { + "epoch": 0.02209529535547873, + "grad_norm": 1.7076033761719145, + "learning_rate": 3.7944510712874613e-06, + "loss": 1.0913, + "step": 294 + }, + { + "epoch": 0.022170449421313694, + "grad_norm": 1.8692119907638634, + "learning_rate": 3.796718022197003e-06, + "loss": 1.0335, + "step": 295 + }, + { + "epoch": 0.022245603487148654, + "grad_norm": 2.144765038592867, + "learning_rate": 3.7989773015151948e-06, + "loss": 1.1111, + "step": 296 + }, + { + "epoch": 0.022320757552983617, + "grad_norm": 1.9290921918474002, + "learning_rate": 3.80122896098985e-06, + "loss": 1.0608, + "step": 297 + }, + { + "epoch": 0.022395911618818577, + "grad_norm": 2.8573512852802465, + "learning_rate": 3.803473051846953e-06, + "loss": 1.0055, + "step": 298 + }, + { + "epoch": 0.02247106568465354, + "grad_norm": 2.0862809373539917, + "learning_rate": 3.805709624797651e-06, + "loss": 1.1869, + "step": 299 + }, + { + "epoch": 0.0225462197504885, + "grad_norm": 2.258026218529017, + "learning_rate": 3.8079387300451326e-06, + "loss": 1.0919, + "step": 300 + }, + { + "epoch": 0.022621373816323464, + "grad_norm": 0.8182743000975866, + "learning_rate": 3.8101604172913872e-06, + "loss": 0.8455, + "step": 301 + }, + { + "epoch": 0.022696527882158424, + "grad_norm": 3.179683750909568, + "learning_rate": 3.8123747357438563e-06, + "loss": 1.0116, + "step": 302 + }, + { + "epoch": 0.022771681947993387, + "grad_norm": 2.2006594883199706, + "learning_rate": 3.814581734121971e-06, + "loss": 1.0272, + "step": 303 + }, + { + "epoch": 0.022846836013828347, + "grad_norm": 1.4374686527980705, + "learning_rate": 3.816781460663585e-06, + "loss": 1.1034, + "step": 304 + }, + { + "epoch": 0.02292199007966331, + "grad_norm": 2.0245484354900634, + "learning_rate": 3.818973963131299e-06, + "loss": 1.1101, + "step": 305 + }, + { + "epoch": 0.02299714414549827, + "grad_norm": 2.824360402038491, + "learning_rate": 3.821159288818688e-06, + "loss": 1.1283, + "step": 306 + }, + { + "epoch": 0.023072298211333234, + "grad_norm": 3.789859648549097, + "learning_rate": 3.823337484556417e-06, + "loss": 1.0466, + "step": 307 + }, + { + "epoch": 0.023147452277168194, + "grad_norm": 2.1733573410642824, + "learning_rate": 3.825508596718266e-06, + "loss": 1.1509, + "step": 308 + }, + { + "epoch": 0.023222606343003157, + "grad_norm": 4.6616094071044065, + "learning_rate": 3.827672671227052e-06, + "loss": 1.0466, + "step": 309 + }, + { + "epoch": 0.023297760408838117, + "grad_norm": 1.979391915068699, + "learning_rate": 3.8298297535604554e-06, + "loss": 0.9601, + "step": 310 + }, + { + "epoch": 0.02337291447467308, + "grad_norm": 2.819531361618737, + "learning_rate": 3.831979888756763e-06, + "loss": 1.0705, + "step": 311 + }, + { + "epoch": 0.02344806854050804, + "grad_norm": 2.3447624562863316, + "learning_rate": 3.834123121420502e-06, + "loss": 1.0899, + "step": 312 + }, + { + "epoch": 0.023523222606343003, + "grad_norm": 35.08170395161041, + "learning_rate": 3.836259495727992e-06, + "loss": 1.0394, + "step": 313 + }, + { + "epoch": 0.023598376672177963, + "grad_norm": 1.9333709896739344, + "learning_rate": 3.838389055432816e-06, + "loss": 1.0813, + "step": 314 + }, + { + "epoch": 0.023673530738012927, + "grad_norm": 2.0426104433789813, + "learning_rate": 3.840511843871188e-06, + "loss": 1.1522, + "step": 315 + }, + { + "epoch": 0.023748684803847887, + "grad_norm": 2.1848029812954843, + "learning_rate": 3.842627903967253e-06, + "loss": 1.0053, + "step": 316 + }, + { + "epoch": 0.02382383886968285, + "grad_norm": 1.709146895174153, + "learning_rate": 3.844737278238285e-06, + "loss": 0.9798, + "step": 317 + }, + { + "epoch": 0.02389899293551781, + "grad_norm": 2.0117531190176994, + "learning_rate": 3.846840008799825e-06, + "loss": 1.1551, + "step": 318 + }, + { + "epoch": 0.023974147001352773, + "grad_norm": 2.414743963060061, + "learning_rate": 3.848936137370715e-06, + "loss": 0.99, + "step": 319 + }, + { + "epoch": 0.024049301067187733, + "grad_norm": 0.7161579246656221, + "learning_rate": 3.851025705278073e-06, + "loss": 0.8301, + "step": 320 + }, + { + "epoch": 0.024124455133022697, + "grad_norm": 2.1717757607160975, + "learning_rate": 3.853108753462177e-06, + "loss": 1.0185, + "step": 321 + }, + { + "epoch": 0.02419960919885766, + "grad_norm": 0.755491379060669, + "learning_rate": 3.855185322481279e-06, + "loss": 0.8408, + "step": 322 + }, + { + "epoch": 0.02427476326469262, + "grad_norm": 2.0153778870974826, + "learning_rate": 3.857255452516343e-06, + "loss": 1.0784, + "step": 323 + }, + { + "epoch": 0.024349917330527583, + "grad_norm": 3.3009603779057857, + "learning_rate": 3.859319183375714e-06, + "loss": 1.0637, + "step": 324 + }, + { + "epoch": 0.024425071396362543, + "grad_norm": 1.9804681480752906, + "learning_rate": 3.861376554499704e-06, + "loss": 1.1505, + "step": 325 + }, + { + "epoch": 0.024500225462197506, + "grad_norm": 2.0540262180724005, + "learning_rate": 3.863427604965122e-06, + "loss": 1.0752, + "step": 326 + }, + { + "epoch": 0.024575379528032466, + "grad_norm": 2.0410000887192776, + "learning_rate": 3.865472373489722e-06, + "loss": 1.0325, + "step": 327 + }, + { + "epoch": 0.02465053359386743, + "grad_norm": 2.306684036556987, + "learning_rate": 3.8675108984365914e-06, + "loss": 1.0982, + "step": 328 + }, + { + "epoch": 0.02472568765970239, + "grad_norm": 1.7637471027742486, + "learning_rate": 3.869543217818467e-06, + "loss": 1.0414, + "step": 329 + }, + { + "epoch": 0.024800841725537353, + "grad_norm": 1.809552723222158, + "learning_rate": 3.871569369301993e-06, + "loss": 1.1076, + "step": 330 + }, + { + "epoch": 0.024875995791372313, + "grad_norm": 2.0602942723203177, + "learning_rate": 3.873589390211904e-06, + "loss": 1.0828, + "step": 331 + }, + { + "epoch": 0.024951149857207276, + "grad_norm": 2.133504343132567, + "learning_rate": 3.875603317535154e-06, + "loss": 1.0379, + "step": 332 + }, + { + "epoch": 0.025026303923042236, + "grad_norm": 1.832040880255206, + "learning_rate": 3.877611187924979e-06, + "loss": 1.1357, + "step": 333 + }, + { + "epoch": 0.0251014579888772, + "grad_norm": 1.8635860809554736, + "learning_rate": 3.879613037704899e-06, + "loss": 0.9661, + "step": 334 + }, + { + "epoch": 0.02517661205471216, + "grad_norm": 1.8999328375540507, + "learning_rate": 3.881608902872662e-06, + "loss": 0.9705, + "step": 335 + }, + { + "epoch": 0.025251766120547123, + "grad_norm": 12.524496870157536, + "learning_rate": 3.88359881910413e-06, + "loss": 1.0964, + "step": 336 + }, + { + "epoch": 0.025326920186382083, + "grad_norm": 3.575935593470937, + "learning_rate": 3.885582821757098e-06, + "loss": 0.9215, + "step": 337 + }, + { + "epoch": 0.025402074252217046, + "grad_norm": 2.356236068724876, + "learning_rate": 3.887560945875073e-06, + "loss": 1.0772, + "step": 338 + }, + { + "epoch": 0.025477228318052006, + "grad_norm": 0.7343214705417113, + "learning_rate": 3.889533226190982e-06, + "loss": 0.8098, + "step": 339 + }, + { + "epoch": 0.02555238238388697, + "grad_norm": 1.8927919564243096, + "learning_rate": 3.891499697130832e-06, + "loss": 1.0101, + "step": 340 + }, + { + "epoch": 0.02562753644972193, + "grad_norm": 2.4518995200474043, + "learning_rate": 3.893460392817316e-06, + "loss": 1.1322, + "step": 341 + }, + { + "epoch": 0.025702690515556893, + "grad_norm": 1.7664362301279497, + "learning_rate": 3.895415347073368e-06, + "loss": 1.0687, + "step": 342 + }, + { + "epoch": 0.025777844581391852, + "grad_norm": 3.8737877210791245, + "learning_rate": 3.89736459342566e-06, + "loss": 1.0427, + "step": 343 + }, + { + "epoch": 0.025852998647226816, + "grad_norm": 2.0484452376230435, + "learning_rate": 3.899308165108055e-06, + "loss": 1.1685, + "step": 344 + }, + { + "epoch": 0.025928152713061776, + "grad_norm": 1.9582557121747755, + "learning_rate": 3.9012460950650064e-06, + "loss": 1.0596, + "step": 345 + }, + { + "epoch": 0.02600330677889674, + "grad_norm": 1.7852003229823452, + "learning_rate": 3.903178415954904e-06, + "loss": 1.0826, + "step": 346 + }, + { + "epoch": 0.0260784608447317, + "grad_norm": 1.6947914412704845, + "learning_rate": 3.905105160153384e-06, + "loss": 1.143, + "step": 347 + }, + { + "epoch": 0.026153614910566662, + "grad_norm": 0.9170345016320448, + "learning_rate": 3.907026359756579e-06, + "loss": 0.8668, + "step": 348 + }, + { + "epoch": 0.026228768976401622, + "grad_norm": 1.8150133924804586, + "learning_rate": 3.908942046584326e-06, + "loss": 1.0783, + "step": 349 + }, + { + "epoch": 0.026303923042236586, + "grad_norm": 2.6798034287002075, + "learning_rate": 3.910852252183332e-06, + "loss": 1.1384, + "step": 350 + }, + { + "epoch": 0.026379077108071546, + "grad_norm": 0.7855185816149596, + "learning_rate": 3.912757007830285e-06, + "loss": 0.9036, + "step": 351 + }, + { + "epoch": 0.02645423117390651, + "grad_norm": 3.1390069997465773, + "learning_rate": 3.914656344534934e-06, + "loss": 1.0858, + "step": 352 + }, + { + "epoch": 0.02652938523974147, + "grad_norm": 2.3117165140371285, + "learning_rate": 3.916550293043113e-06, + "loss": 1.0427, + "step": 353 + }, + { + "epoch": 0.026604539305576432, + "grad_norm": 1.7621221313150135, + "learning_rate": 3.9184388838397275e-06, + "loss": 1.0097, + "step": 354 + }, + { + "epoch": 0.026679693371411392, + "grad_norm": 1.4241156058907753, + "learning_rate": 3.9203221471517014e-06, + "loss": 1.0091, + "step": 355 + }, + { + "epoch": 0.026754847437246355, + "grad_norm": 1.3537462266497802, + "learning_rate": 3.92220011295088e-06, + "loss": 1.1168, + "step": 356 + }, + { + "epoch": 0.026830001503081315, + "grad_norm": 2.076568938762709, + "learning_rate": 3.924072810956888e-06, + "loss": 1.0436, + "step": 357 + }, + { + "epoch": 0.02690515556891628, + "grad_norm": 1.8065933798721154, + "learning_rate": 3.925940270639954e-06, + "loss": 1.0311, + "step": 358 + }, + { + "epoch": 0.02698030963475124, + "grad_norm": 1.6859860163116676, + "learning_rate": 3.927802521223695e-06, + "loss": 1.0363, + "step": 359 + }, + { + "epoch": 0.027055463700586202, + "grad_norm": 2.1707636782355713, + "learning_rate": 3.929659591687857e-06, + "loss": 1.0611, + "step": 360 + }, + { + "epoch": 0.027130617766421162, + "grad_norm": 2.1721628121883056, + "learning_rate": 3.931511510771023e-06, + "loss": 1.0498, + "step": 361 + }, + { + "epoch": 0.027205771832256125, + "grad_norm": 1.565956641000098, + "learning_rate": 3.93335830697328e-06, + "loss": 1.1629, + "step": 362 + }, + { + "epoch": 0.027280925898091085, + "grad_norm": 1.6272657444389014, + "learning_rate": 3.935200008558854e-06, + "loss": 1.0477, + "step": 363 + }, + { + "epoch": 0.02735607996392605, + "grad_norm": 1.720579595385877, + "learning_rate": 3.937036643558701e-06, + "loss": 0.9706, + "step": 364 + }, + { + "epoch": 0.02743123402976101, + "grad_norm": 1.7476367661162109, + "learning_rate": 3.938868239773069e-06, + "loss": 1.0311, + "step": 365 + }, + { + "epoch": 0.027506388095595972, + "grad_norm": 1.75596154990277, + "learning_rate": 3.940694824774024e-06, + "loss": 1.109, + "step": 366 + }, + { + "epoch": 0.02758154216143093, + "grad_norm": 1.7640446029309351, + "learning_rate": 3.942516425907938e-06, + "loss": 1.0313, + "step": 367 + }, + { + "epoch": 0.027656696227265895, + "grad_norm": 4.853317361078774, + "learning_rate": 3.944333070297947e-06, + "loss": 1.0717, + "step": 368 + }, + { + "epoch": 0.02773185029310086, + "grad_norm": 2.1499718198481945, + "learning_rate": 3.946144784846375e-06, + "loss": 1.021, + "step": 369 + }, + { + "epoch": 0.02780700435893582, + "grad_norm": 3.3720527778830234, + "learning_rate": 3.947951596237121e-06, + "loss": 0.9826, + "step": 370 + }, + { + "epoch": 0.02788215842477078, + "grad_norm": 2.7471517204321354, + "learning_rate": 3.949753530938024e-06, + "loss": 1.0053, + "step": 371 + }, + { + "epoch": 0.02795731249060574, + "grad_norm": 1.8008542656318172, + "learning_rate": 3.951550615203179e-06, + "loss": 1.0599, + "step": 372 + }, + { + "epoch": 0.028032466556440705, + "grad_norm": 1.8766451929375945, + "learning_rate": 3.953342875075244e-06, + "loss": 1.023, + "step": 373 + }, + { + "epoch": 0.028107620622275665, + "grad_norm": 2.593703487852447, + "learning_rate": 3.955130336387693e-06, + "loss": 0.9555, + "step": 374 + }, + { + "epoch": 0.028182774688110628, + "grad_norm": 1.9521234614733871, + "learning_rate": 3.956913024767059e-06, + "loss": 1.0562, + "step": 375 + }, + { + "epoch": 0.028257928753945588, + "grad_norm": 1.6488319395121047, + "learning_rate": 3.958690965635136e-06, + "loss": 1.0764, + "step": 376 + }, + { + "epoch": 0.02833308281978055, + "grad_norm": 2.5344388678578236, + "learning_rate": 3.96046418421115e-06, + "loss": 0.9704, + "step": 377 + }, + { + "epoch": 0.02840823688561551, + "grad_norm": 21.114226172772707, + "learning_rate": 3.962232705513913e-06, + "loss": 1.0296, + "step": 378 + }, + { + "epoch": 0.028483390951450475, + "grad_norm": 2.8280159945055905, + "learning_rate": 3.963996554363933e-06, + "loss": 1.1149, + "step": 379 + }, + { + "epoch": 0.028558545017285435, + "grad_norm": 1.759513269550556, + "learning_rate": 3.965755755385512e-06, + "loss": 1.0724, + "step": 380 + }, + { + "epoch": 0.028633699083120398, + "grad_norm": 4.408798454760352, + "learning_rate": 3.9675103330088e-06, + "loss": 0.9764, + "step": 381 + }, + { + "epoch": 0.028708853148955358, + "grad_norm": 1.6818235078582036, + "learning_rate": 3.969260311471838e-06, + "loss": 1.0771, + "step": 382 + }, + { + "epoch": 0.02878400721479032, + "grad_norm": 2.252948841927948, + "learning_rate": 3.971005714822564e-06, + "loss": 1.0507, + "step": 383 + }, + { + "epoch": 0.02885916128062528, + "grad_norm": 1.7037481165516575, + "learning_rate": 3.9727465669207975e-06, + "loss": 1.0732, + "step": 384 + }, + { + "epoch": 0.028934315346460245, + "grad_norm": 2.632977941731432, + "learning_rate": 3.974482891440193e-06, + "loss": 0.992, + "step": 385 + }, + { + "epoch": 0.029009469412295204, + "grad_norm": 1.7888378018863307, + "learning_rate": 3.976214711870174e-06, + "loss": 1.0114, + "step": 386 + }, + { + "epoch": 0.029084623478130168, + "grad_norm": 1.5815751536793439, + "learning_rate": 3.977942051517839e-06, + "loss": 1.0528, + "step": 387 + }, + { + "epoch": 0.029159777543965128, + "grad_norm": 1.5997865832569649, + "learning_rate": 3.979664933509847e-06, + "loss": 1.0546, + "step": 388 + }, + { + "epoch": 0.02923493160980009, + "grad_norm": 2.6177353015203204, + "learning_rate": 3.9813833807942695e-06, + "loss": 0.9888, + "step": 389 + }, + { + "epoch": 0.02931008567563505, + "grad_norm": 1.6384122913435988, + "learning_rate": 3.9830974161424286e-06, + "loss": 1.0634, + "step": 390 + }, + { + "epoch": 0.029385239741470014, + "grad_norm": 1.9989502058978807, + "learning_rate": 3.984807062150705e-06, + "loss": 1.0959, + "step": 391 + }, + { + "epoch": 0.029460393807304974, + "grad_norm": 2.243372906207832, + "learning_rate": 3.986512341242329e-06, + "loss": 1.0899, + "step": 392 + }, + { + "epoch": 0.029535547873139938, + "grad_norm": 8.312349475845789, + "learning_rate": 3.98821327566914e-06, + "loss": 1.0251, + "step": 393 + }, + { + "epoch": 0.029610701938974898, + "grad_norm": 2.3297226307687184, + "learning_rate": 3.989909887513334e-06, + "loss": 1.131, + "step": 394 + }, + { + "epoch": 0.02968585600480986, + "grad_norm": 3.911243756229224, + "learning_rate": 3.991602198689179e-06, + "loss": 1.0661, + "step": 395 + }, + { + "epoch": 0.02976101007064482, + "grad_norm": 1.8706030470977537, + "learning_rate": 3.993290230944718e-06, + "loss": 1.1196, + "step": 396 + }, + { + "epoch": 0.029836164136479784, + "grad_norm": 2.5960731673135036, + "learning_rate": 3.994974005863441e-06, + "loss": 1.0438, + "step": 397 + }, + { + "epoch": 0.029911318202314744, + "grad_norm": 2.671596936485749, + "learning_rate": 3.996653544865945e-06, + "loss": 0.9666, + "step": 398 + }, + { + "epoch": 0.029986472268149707, + "grad_norm": 1.7985768096022652, + "learning_rate": 3.9983288692115676e-06, + "loss": 1.0664, + "step": 399 + }, + { + "epoch": 0.030061626333984667, + "grad_norm": 1.9783487202475274, + "learning_rate": 4e-06, + "loss": 1.0866, + "step": 400 + }, + { + "epoch": 0.03013678039981963, + "grad_norm": 2.9276504962655108, + "learning_rate": 3.999999940746171e-06, + "loss": 1.1527, + "step": 401 + }, + { + "epoch": 0.03021193446565459, + "grad_norm": 1.983214752541869, + "learning_rate": 3.999999762984686e-06, + "loss": 1.1123, + "step": 402 + }, + { + "epoch": 0.030287088531489554, + "grad_norm": 2.2298795920739094, + "learning_rate": 3.9999994667155576e-06, + "loss": 1.0241, + "step": 403 + }, + { + "epoch": 0.030362242597324514, + "grad_norm": 1.6962963009659533, + "learning_rate": 3.999999051938802e-06, + "loss": 0.9726, + "step": 404 + }, + { + "epoch": 0.030437396663159477, + "grad_norm": 2.139587819343279, + "learning_rate": 3.9999985186544445e-06, + "loss": 0.9459, + "step": 405 + }, + { + "epoch": 0.030512550728994437, + "grad_norm": 2.581077411044977, + "learning_rate": 3.999997866862515e-06, + "loss": 1.067, + "step": 406 + }, + { + "epoch": 0.0305877047948294, + "grad_norm": 2.0182505222954483, + "learning_rate": 3.9999970965630544e-06, + "loss": 1.1405, + "step": 407 + }, + { + "epoch": 0.03066285886066436, + "grad_norm": 1.606056124493915, + "learning_rate": 3.9999962077561075e-06, + "loss": 1.0719, + "step": 408 + }, + { + "epoch": 0.030738012926499324, + "grad_norm": 1.8819279789253196, + "learning_rate": 3.999995200441726e-06, + "loss": 1.0107, + "step": 409 + }, + { + "epoch": 0.030813166992334284, + "grad_norm": 1.7008472575256772, + "learning_rate": 3.999994074619971e-06, + "loss": 1.1123, + "step": 410 + }, + { + "epoch": 0.030888321058169247, + "grad_norm": 1.8070608938962924, + "learning_rate": 3.999992830290909e-06, + "loss": 1.0413, + "step": 411 + }, + { + "epoch": 0.030963475124004207, + "grad_norm": 1.7970040979864046, + "learning_rate": 3.999991467454612e-06, + "loss": 1.0841, + "step": 412 + }, + { + "epoch": 0.03103862918983917, + "grad_norm": 2.128835022460981, + "learning_rate": 3.999989986111163e-06, + "loss": 1.0114, + "step": 413 + }, + { + "epoch": 0.031113783255674134, + "grad_norm": 2.255129477258013, + "learning_rate": 3.999988386260648e-06, + "loss": 1.0688, + "step": 414 + }, + { + "epoch": 0.031188937321509094, + "grad_norm": 1.8651701431798773, + "learning_rate": 3.999986667903163e-06, + "loss": 1.0058, + "step": 415 + }, + { + "epoch": 0.03126409138734405, + "grad_norm": 1.6638788829383022, + "learning_rate": 3.999984831038811e-06, + "loss": 1.0779, + "step": 416 + }, + { + "epoch": 0.03133924545317902, + "grad_norm": 2.4076994456609375, + "learning_rate": 3.999982875667697e-06, + "loss": 1.0092, + "step": 417 + }, + { + "epoch": 0.03141439951901398, + "grad_norm": 2.3431826095438444, + "learning_rate": 3.999980801789941e-06, + "loss": 1.0237, + "step": 418 + }, + { + "epoch": 0.031489553584848944, + "grad_norm": 1.6269072563968567, + "learning_rate": 3.999978609405662e-06, + "loss": 1.0302, + "step": 419 + }, + { + "epoch": 0.0315647076506839, + "grad_norm": 1.6653555560867737, + "learning_rate": 3.999976298514994e-06, + "loss": 1.0911, + "step": 420 + }, + { + "epoch": 0.03163986171651886, + "grad_norm": 1.7436369609768458, + "learning_rate": 3.999973869118071e-06, + "loss": 1.2069, + "step": 421 + }, + { + "epoch": 0.03171501578235383, + "grad_norm": 1.803909398237664, + "learning_rate": 3.999971321215038e-06, + "loss": 1.1312, + "step": 422 + }, + { + "epoch": 0.03179016984818879, + "grad_norm": 1.804501478507984, + "learning_rate": 3.999968654806046e-06, + "loss": 1.0534, + "step": 423 + }, + { + "epoch": 0.03186532391402375, + "grad_norm": 1.598017793892918, + "learning_rate": 3.999965869891253e-06, + "loss": 0.8723, + "step": 424 + }, + { + "epoch": 0.03194047797985871, + "grad_norm": 0.860175055108587, + "learning_rate": 3.999962966470823e-06, + "loss": 0.9177, + "step": 425 + }, + { + "epoch": 0.03201563204569367, + "grad_norm": 1.731327300291208, + "learning_rate": 3.9999599445449295e-06, + "loss": 1.0804, + "step": 426 + }, + { + "epoch": 0.03209078611152864, + "grad_norm": 1.482079990452187, + "learning_rate": 3.999956804113751e-06, + "loss": 1.0199, + "step": 427 + }, + { + "epoch": 0.03216594017736359, + "grad_norm": 1.6416009815584933, + "learning_rate": 3.999953545177472e-06, + "loss": 1.1257, + "step": 428 + }, + { + "epoch": 0.032241094243198556, + "grad_norm": 1.494598278690196, + "learning_rate": 3.9999501677362885e-06, + "loss": 1.0186, + "step": 429 + }, + { + "epoch": 0.03231624830903352, + "grad_norm": 1.586637821509653, + "learning_rate": 3.9999466717903995e-06, + "loss": 1.0143, + "step": 430 + }, + { + "epoch": 0.03239140237486848, + "grad_norm": 1.6769098069669413, + "learning_rate": 3.999943057340012e-06, + "loss": 1.0786, + "step": 431 + }, + { + "epoch": 0.03246655644070344, + "grad_norm": 1.7311685150470575, + "learning_rate": 3.999939324385339e-06, + "loss": 1.0612, + "step": 432 + }, + { + "epoch": 0.0325417105065384, + "grad_norm": 2.084785171711637, + "learning_rate": 3.999935472926604e-06, + "loss": 1.0353, + "step": 433 + }, + { + "epoch": 0.032616864572373366, + "grad_norm": 1.4666759375076903, + "learning_rate": 3.9999315029640325e-06, + "loss": 1.1322, + "step": 434 + }, + { + "epoch": 0.03269201863820833, + "grad_norm": 2.0517250432443848, + "learning_rate": 3.999927414497862e-06, + "loss": 1.0213, + "step": 435 + }, + { + "epoch": 0.032767172704043286, + "grad_norm": 2.1715591530328733, + "learning_rate": 3.999923207528334e-06, + "loss": 0.9936, + "step": 436 + }, + { + "epoch": 0.03284232676987825, + "grad_norm": 2.315161127534702, + "learning_rate": 3.999918882055698e-06, + "loss": 1.0506, + "step": 437 + }, + { + "epoch": 0.03291748083571321, + "grad_norm": 1.615365370174669, + "learning_rate": 3.9999144380802095e-06, + "loss": 1.1167, + "step": 438 + }, + { + "epoch": 0.032992634901548176, + "grad_norm": 1.9608743969008169, + "learning_rate": 3.999909875602132e-06, + "loss": 1.0349, + "step": 439 + }, + { + "epoch": 0.03306778896738313, + "grad_norm": 0.7695580492339399, + "learning_rate": 3.999905194621737e-06, + "loss": 0.8507, + "step": 440 + }, + { + "epoch": 0.033142943033218096, + "grad_norm": 1.9250318450275354, + "learning_rate": 3.999900395139301e-06, + "loss": 1.0649, + "step": 441 + }, + { + "epoch": 0.03321809709905306, + "grad_norm": 2.427960533355512, + "learning_rate": 3.999895477155108e-06, + "loss": 1.0183, + "step": 442 + }, + { + "epoch": 0.03329325116488802, + "grad_norm": 1.76382313836905, + "learning_rate": 3.9998904406694504e-06, + "loss": 1.1105, + "step": 443 + }, + { + "epoch": 0.03336840523072298, + "grad_norm": 1.8817090485121666, + "learning_rate": 3.999885285682626e-06, + "loss": 1.1427, + "step": 444 + }, + { + "epoch": 0.03344355929655794, + "grad_norm": 1.8454614874687434, + "learning_rate": 3.99988001219494e-06, + "loss": 1.1199, + "step": 445 + }, + { + "epoch": 0.033518713362392906, + "grad_norm": 2.06994187978338, + "learning_rate": 3.999874620206705e-06, + "loss": 1.0664, + "step": 446 + }, + { + "epoch": 0.03359386742822787, + "grad_norm": 1.6453467587402943, + "learning_rate": 3.999869109718242e-06, + "loss": 1.0442, + "step": 447 + }, + { + "epoch": 0.033669021494062826, + "grad_norm": 3.065970342474006, + "learning_rate": 3.999863480729875e-06, + "loss": 1.0487, + "step": 448 + }, + { + "epoch": 0.03374417555989779, + "grad_norm": 1.9100978975022727, + "learning_rate": 3.999857733241938e-06, + "loss": 1.0149, + "step": 449 + }, + { + "epoch": 0.03381932962573275, + "grad_norm": 1.517685737869301, + "learning_rate": 3.999851867254774e-06, + "loss": 1.0403, + "step": 450 + }, + { + "epoch": 0.033894483691567716, + "grad_norm": 1.829284756555358, + "learning_rate": 3.9998458827687286e-06, + "loss": 1.0053, + "step": 451 + }, + { + "epoch": 0.03396963775740267, + "grad_norm": 0.6734855776306719, + "learning_rate": 3.999839779784157e-06, + "loss": 0.8368, + "step": 452 + }, + { + "epoch": 0.034044791823237636, + "grad_norm": 1.7819642759696135, + "learning_rate": 3.999833558301419e-06, + "loss": 0.9961, + "step": 453 + }, + { + "epoch": 0.0341199458890726, + "grad_norm": 1.7984726810411367, + "learning_rate": 3.999827218320886e-06, + "loss": 1.0314, + "step": 454 + }, + { + "epoch": 0.03419509995490756, + "grad_norm": 0.8285597613645415, + "learning_rate": 3.999820759842933e-06, + "loss": 0.8669, + "step": 455 + }, + { + "epoch": 0.03427025402074252, + "grad_norm": 3.023148081724685, + "learning_rate": 3.999814182867941e-06, + "loss": 1.1258, + "step": 456 + }, + { + "epoch": 0.03434540808657748, + "grad_norm": 1.8617805499514082, + "learning_rate": 3.999807487396301e-06, + "loss": 1.1455, + "step": 457 + }, + { + "epoch": 0.034420562152412446, + "grad_norm": 2.5439799886672176, + "learning_rate": 3.999800673428411e-06, + "loss": 1.0029, + "step": 458 + }, + { + "epoch": 0.03449571621824741, + "grad_norm": 2.64001069456882, + "learning_rate": 3.999793740964672e-06, + "loss": 1.0619, + "step": 459 + }, + { + "epoch": 0.03457087028408237, + "grad_norm": 1.7850662308915695, + "learning_rate": 3.999786690005496e-06, + "loss": 1.0764, + "step": 460 + }, + { + "epoch": 0.03464602434991733, + "grad_norm": 2.13161840303179, + "learning_rate": 3.999779520551302e-06, + "loss": 0.9212, + "step": 461 + }, + { + "epoch": 0.03472117841575229, + "grad_norm": 0.8598070619087697, + "learning_rate": 3.9997722326025135e-06, + "loss": 0.8467, + "step": 462 + }, + { + "epoch": 0.034796332481587255, + "grad_norm": 2.76280844852188, + "learning_rate": 3.999764826159562e-06, + "loss": 1.002, + "step": 463 + }, + { + "epoch": 0.03487148654742222, + "grad_norm": 1.720997745089076, + "learning_rate": 3.999757301222887e-06, + "loss": 1.0444, + "step": 464 + }, + { + "epoch": 0.034946640613257175, + "grad_norm": 1.2797056054777927, + "learning_rate": 3.999749657792934e-06, + "loss": 1.0195, + "step": 465 + }, + { + "epoch": 0.03502179467909214, + "grad_norm": 1.6087834280197177, + "learning_rate": 3.999741895870157e-06, + "loss": 1.0178, + "step": 466 + }, + { + "epoch": 0.0350969487449271, + "grad_norm": 1.7020763003306334, + "learning_rate": 3.9997340154550145e-06, + "loss": 1.0402, + "step": 467 + }, + { + "epoch": 0.035172102810762065, + "grad_norm": 1.4332551894214582, + "learning_rate": 3.999726016547974e-06, + "loss": 1.0842, + "step": 468 + }, + { + "epoch": 0.03524725687659702, + "grad_norm": 3.2655839689314514, + "learning_rate": 3.9997178991495105e-06, + "loss": 1.0554, + "step": 469 + }, + { + "epoch": 0.035322410942431985, + "grad_norm": 1.570502426276751, + "learning_rate": 3.9997096632601035e-06, + "loss": 1.0694, + "step": 470 + }, + { + "epoch": 0.03539756500826695, + "grad_norm": 1.3980337657670914, + "learning_rate": 3.999701308880242e-06, + "loss": 1.0105, + "step": 471 + }, + { + "epoch": 0.03547271907410191, + "grad_norm": 1.8206469838379709, + "learning_rate": 3.999692836010419e-06, + "loss": 1.0825, + "step": 472 + }, + { + "epoch": 0.03554787313993687, + "grad_norm": 1.6228087396282467, + "learning_rate": 3.99968424465114e-06, + "loss": 1.1883, + "step": 473 + }, + { + "epoch": 0.03562302720577183, + "grad_norm": 1.916826704503317, + "learning_rate": 3.999675534802911e-06, + "loss": 1.0205, + "step": 474 + }, + { + "epoch": 0.035698181271606795, + "grad_norm": 1.5288337795603884, + "learning_rate": 3.99966670646625e-06, + "loss": 0.9448, + "step": 475 + }, + { + "epoch": 0.03577333533744176, + "grad_norm": 2.116129931284438, + "learning_rate": 3.999657759641679e-06, + "loss": 1.0704, + "step": 476 + }, + { + "epoch": 0.035848489403276715, + "grad_norm": 2.5877860569970483, + "learning_rate": 3.999648694329729e-06, + "loss": 1.1318, + "step": 477 + }, + { + "epoch": 0.03592364346911168, + "grad_norm": 1.2778995249746676, + "learning_rate": 3.9996395105309365e-06, + "loss": 1.0261, + "step": 478 + }, + { + "epoch": 0.03599879753494664, + "grad_norm": 1.8749457669362137, + "learning_rate": 3.999630208245846e-06, + "loss": 0.9784, + "step": 479 + }, + { + "epoch": 0.036073951600781605, + "grad_norm": 1.5970124904022298, + "learning_rate": 3.9996207874750075e-06, + "loss": 1.0804, + "step": 480 + }, + { + "epoch": 0.03614910566661656, + "grad_norm": 1.5168440330811765, + "learning_rate": 3.999611248218982e-06, + "loss": 1.0996, + "step": 481 + }, + { + "epoch": 0.036224259732451525, + "grad_norm": 2.335761124194183, + "learning_rate": 3.999601590478332e-06, + "loss": 0.9153, + "step": 482 + }, + { + "epoch": 0.03629941379828649, + "grad_norm": 1.650007178551057, + "learning_rate": 3.99959181425363e-06, + "loss": 1.1364, + "step": 483 + }, + { + "epoch": 0.03637456786412145, + "grad_norm": 2.1178864736416765, + "learning_rate": 3.999581919545458e-06, + "loss": 1.0375, + "step": 484 + }, + { + "epoch": 0.03644972192995641, + "grad_norm": 1.9517021795028555, + "learning_rate": 3.999571906354399e-06, + "loss": 1.1544, + "step": 485 + }, + { + "epoch": 0.03652487599579137, + "grad_norm": 1.6843260126297668, + "learning_rate": 3.999561774681048e-06, + "loss": 1.1042, + "step": 486 + }, + { + "epoch": 0.036600030061626335, + "grad_norm": 0.8830852256073788, + "learning_rate": 3.999551524526005e-06, + "loss": 0.948, + "step": 487 + }, + { + "epoch": 0.0366751841274613, + "grad_norm": 5.710612526556744, + "learning_rate": 3.9995411558898775e-06, + "loss": 1.015, + "step": 488 + }, + { + "epoch": 0.036750338193296254, + "grad_norm": 4.986764365036457, + "learning_rate": 3.9995306687732795e-06, + "loss": 1.032, + "step": 489 + }, + { + "epoch": 0.03682549225913122, + "grad_norm": 0.9138801588461509, + "learning_rate": 3.9995200631768326e-06, + "loss": 0.9222, + "step": 490 + }, + { + "epoch": 0.03690064632496618, + "grad_norm": 2.85562538312282, + "learning_rate": 3.999509339101166e-06, + "loss": 1.0558, + "step": 491 + }, + { + "epoch": 0.036975800390801145, + "grad_norm": 1.640532790726269, + "learning_rate": 3.999498496546914e-06, + "loss": 1.0358, + "step": 492 + }, + { + "epoch": 0.0370509544566361, + "grad_norm": 1.6885238649660284, + "learning_rate": 3.99948753551472e-06, + "loss": 1.0174, + "step": 493 + }, + { + "epoch": 0.037126108522471064, + "grad_norm": 1.6195910739476533, + "learning_rate": 3.999476456005232e-06, + "loss": 1.0921, + "step": 494 + }, + { + "epoch": 0.03720126258830603, + "grad_norm": 1.6800113099716592, + "learning_rate": 3.999465258019108e-06, + "loss": 0.9464, + "step": 495 + }, + { + "epoch": 0.03727641665414099, + "grad_norm": 1.6862768049209274, + "learning_rate": 3.999453941557011e-06, + "loss": 0.9696, + "step": 496 + }, + { + "epoch": 0.03735157071997595, + "grad_norm": 1.9239055792228714, + "learning_rate": 3.9994425066196105e-06, + "loss": 1.0623, + "step": 497 + }, + { + "epoch": 0.03742672478581091, + "grad_norm": 1.7057553898394784, + "learning_rate": 3.999430953207586e-06, + "loss": 1.0849, + "step": 498 + }, + { + "epoch": 0.037501878851645874, + "grad_norm": 1.6993926562184372, + "learning_rate": 3.999419281321621e-06, + "loss": 1.0632, + "step": 499 + }, + { + "epoch": 0.03757703291748084, + "grad_norm": 8.760651127553873, + "learning_rate": 3.999407490962408e-06, + "loss": 0.9825, + "step": 500 + }, + { + "epoch": 0.037652186983315794, + "grad_norm": 1.582825487239838, + "learning_rate": 3.999395582130644e-06, + "loss": 0.9624, + "step": 501 + }, + { + "epoch": 0.03772734104915076, + "grad_norm": 2.1971916220176944, + "learning_rate": 3.999383554827037e-06, + "loss": 1.0307, + "step": 502 + }, + { + "epoch": 0.03780249511498572, + "grad_norm": 1.8370276278607467, + "learning_rate": 3.999371409052297e-06, + "loss": 1.118, + "step": 503 + }, + { + "epoch": 0.037877649180820684, + "grad_norm": 1.9895336351295723, + "learning_rate": 3.999359144807145e-06, + "loss": 1.0908, + "step": 504 + }, + { + "epoch": 0.03795280324665565, + "grad_norm": 1.666017700050913, + "learning_rate": 3.999346762092307e-06, + "loss": 1.1459, + "step": 505 + }, + { + "epoch": 0.038027957312490604, + "grad_norm": 2.655124560521552, + "learning_rate": 3.999334260908518e-06, + "loss": 1.0297, + "step": 506 + }, + { + "epoch": 0.03810311137832557, + "grad_norm": 1.7541108893484292, + "learning_rate": 3.999321641256519e-06, + "loss": 1.0456, + "step": 507 + }, + { + "epoch": 0.03817826544416053, + "grad_norm": 1.7024516603214321, + "learning_rate": 3.999308903137056e-06, + "loss": 1.0174, + "step": 508 + }, + { + "epoch": 0.038253419509995494, + "grad_norm": 3.7204320613613637, + "learning_rate": 3.999296046550884e-06, + "loss": 1.0, + "step": 509 + }, + { + "epoch": 0.03832857357583045, + "grad_norm": 2.268500877133007, + "learning_rate": 3.999283071498766e-06, + "loss": 1.0646, + "step": 510 + }, + { + "epoch": 0.038403727641665414, + "grad_norm": 1.7283560296643454, + "learning_rate": 3.9992699779814704e-06, + "loss": 1.0535, + "step": 511 + }, + { + "epoch": 0.03847888170750038, + "grad_norm": 1.897468413742187, + "learning_rate": 3.999256765999773e-06, + "loss": 1.0229, + "step": 512 + }, + { + "epoch": 0.03855403577333534, + "grad_norm": 1.5883491107574692, + "learning_rate": 3.999243435554456e-06, + "loss": 1.0383, + "step": 513 + }, + { + "epoch": 0.0386291898391703, + "grad_norm": 2.3967677961544087, + "learning_rate": 3.999229986646311e-06, + "loss": 1.0559, + "step": 514 + }, + { + "epoch": 0.03870434390500526, + "grad_norm": 1.6534664587420382, + "learning_rate": 3.999216419276132e-06, + "loss": 1.1257, + "step": 515 + }, + { + "epoch": 0.038779497970840224, + "grad_norm": 2.2526525121068968, + "learning_rate": 3.999202733444726e-06, + "loss": 1.009, + "step": 516 + }, + { + "epoch": 0.03885465203667519, + "grad_norm": 3.1518484716832673, + "learning_rate": 3.999188929152902e-06, + "loss": 1.0832, + "step": 517 + }, + { + "epoch": 0.038929806102510144, + "grad_norm": 2.5391872903528605, + "learning_rate": 3.999175006401478e-06, + "loss": 1.0092, + "step": 518 + }, + { + "epoch": 0.03900496016834511, + "grad_norm": 1.910634673259133, + "learning_rate": 3.999160965191281e-06, + "loss": 1.039, + "step": 519 + }, + { + "epoch": 0.03908011423418007, + "grad_norm": 1.915145318615641, + "learning_rate": 3.99914680552314e-06, + "loss": 0.9903, + "step": 520 + }, + { + "epoch": 0.039155268300015034, + "grad_norm": 2.3332923022704364, + "learning_rate": 3.999132527397897e-06, + "loss": 1.1208, + "step": 521 + }, + { + "epoch": 0.03923042236584999, + "grad_norm": 1.4620393076359999, + "learning_rate": 3.999118130816395e-06, + "loss": 0.9926, + "step": 522 + }, + { + "epoch": 0.03930557643168495, + "grad_norm": 1.723938549039631, + "learning_rate": 3.999103615779489e-06, + "loss": 1.0306, + "step": 523 + }, + { + "epoch": 0.03938073049751992, + "grad_norm": 0.7049838819419838, + "learning_rate": 3.99908898228804e-06, + "loss": 0.8644, + "step": 524 + }, + { + "epoch": 0.03945588456335488, + "grad_norm": 1.714226057513222, + "learning_rate": 3.999074230342913e-06, + "loss": 1.0493, + "step": 525 + }, + { + "epoch": 0.03953103862918984, + "grad_norm": 1.67055965982253, + "learning_rate": 3.999059359944982e-06, + "loss": 0.998, + "step": 526 + }, + { + "epoch": 0.0396061926950248, + "grad_norm": 2.865002873071425, + "learning_rate": 3.99904437109513e-06, + "loss": 1.0787, + "step": 527 + }, + { + "epoch": 0.03968134676085976, + "grad_norm": 1.5565075979025902, + "learning_rate": 3.999029263794244e-06, + "loss": 1.0664, + "step": 528 + }, + { + "epoch": 0.03975650082669473, + "grad_norm": 1.8301124528822545, + "learning_rate": 3.999014038043219e-06, + "loss": 1.0517, + "step": 529 + }, + { + "epoch": 0.03983165489252968, + "grad_norm": 1.658261283510611, + "learning_rate": 3.9989986938429574e-06, + "loss": 1.1147, + "step": 530 + }, + { + "epoch": 0.03990680895836465, + "grad_norm": 2.2045221971037607, + "learning_rate": 3.9989832311943695e-06, + "loss": 1.0495, + "step": 531 + }, + { + "epoch": 0.03998196302419961, + "grad_norm": 1.5664158390350977, + "learning_rate": 3.99896765009837e-06, + "loss": 1.0066, + "step": 532 + }, + { + "epoch": 0.04005711709003457, + "grad_norm": 1.8319320163814574, + "learning_rate": 3.998951950555883e-06, + "loss": 0.9982, + "step": 533 + }, + { + "epoch": 0.04013227115586953, + "grad_norm": 1.013046983448363, + "learning_rate": 3.998936132567837e-06, + "loss": 0.8546, + "step": 534 + }, + { + "epoch": 0.04020742522170449, + "grad_norm": 2.453464848977306, + "learning_rate": 3.998920196135172e-06, + "loss": 1.0341, + "step": 535 + }, + { + "epoch": 0.040282579287539456, + "grad_norm": 1.4644402113431865, + "learning_rate": 3.998904141258831e-06, + "loss": 1.0673, + "step": 536 + }, + { + "epoch": 0.04035773335337442, + "grad_norm": 2.931532459436528, + "learning_rate": 3.9988879679397644e-06, + "loss": 1.0962, + "step": 537 + }, + { + "epoch": 0.040432887419209376, + "grad_norm": 2.7606381213528604, + "learning_rate": 3.9988716761789324e-06, + "loss": 1.0959, + "step": 538 + }, + { + "epoch": 0.04050804148504434, + "grad_norm": 2.0845261274508866, + "learning_rate": 3.998855265977299e-06, + "loss": 1.063, + "step": 539 + }, + { + "epoch": 0.0405831955508793, + "grad_norm": 1.7601529964140965, + "learning_rate": 3.998838737335837e-06, + "loss": 1.0373, + "step": 540 + }, + { + "epoch": 0.040658349616714266, + "grad_norm": 1.9902911566513652, + "learning_rate": 3.998822090255526e-06, + "loss": 1.0718, + "step": 541 + }, + { + "epoch": 0.04073350368254922, + "grad_norm": 1.9025331751453605, + "learning_rate": 3.9988053247373515e-06, + "loss": 1.0265, + "step": 542 + }, + { + "epoch": 0.040808657748384186, + "grad_norm": 1.9857765114493806, + "learning_rate": 3.998788440782309e-06, + "loss": 1.0205, + "step": 543 + }, + { + "epoch": 0.04088381181421915, + "grad_norm": 1.5973273127532075, + "learning_rate": 3.998771438391396e-06, + "loss": 1.0198, + "step": 544 + }, + { + "epoch": 0.04095896588005411, + "grad_norm": 1.9854255358513562, + "learning_rate": 3.9987543175656214e-06, + "loss": 1.012, + "step": 545 + }, + { + "epoch": 0.04103411994588907, + "grad_norm": 1.6846714332195116, + "learning_rate": 3.998737078306001e-06, + "loss": 1.1143, + "step": 546 + }, + { + "epoch": 0.04110927401172403, + "grad_norm": 1.8874572824861067, + "learning_rate": 3.998719720613554e-06, + "loss": 1.0785, + "step": 547 + }, + { + "epoch": 0.041184428077558996, + "grad_norm": 0.8897120163075101, + "learning_rate": 3.99870224448931e-06, + "loss": 0.9363, + "step": 548 + }, + { + "epoch": 0.04125958214339396, + "grad_norm": 1.995964270048332, + "learning_rate": 3.998684649934305e-06, + "loss": 1.0614, + "step": 549 + }, + { + "epoch": 0.041334736209228916, + "grad_norm": 1.5383137852069755, + "learning_rate": 3.9986669369495805e-06, + "loss": 1.036, + "step": 550 + }, + { + "epoch": 0.04140989027506388, + "grad_norm": 1.8250202648135117, + "learning_rate": 3.998649105536187e-06, + "loss": 1.065, + "step": 551 + }, + { + "epoch": 0.04148504434089884, + "grad_norm": 2.9517357197342844, + "learning_rate": 3.998631155695181e-06, + "loss": 1.0759, + "step": 552 + }, + { + "epoch": 0.041560198406733806, + "grad_norm": 2.34626675583024, + "learning_rate": 3.9986130874276244e-06, + "loss": 1.0545, + "step": 553 + }, + { + "epoch": 0.04163535247256877, + "grad_norm": 2.2645625286907176, + "learning_rate": 3.998594900734591e-06, + "loss": 1.0776, + "step": 554 + }, + { + "epoch": 0.041710506538403726, + "grad_norm": 2.7056940472367152, + "learning_rate": 3.998576595617155e-06, + "loss": 1.0034, + "step": 555 + }, + { + "epoch": 0.04178566060423869, + "grad_norm": 1.7415053738585893, + "learning_rate": 3.998558172076404e-06, + "loss": 1.1527, + "step": 556 + }, + { + "epoch": 0.04186081467007365, + "grad_norm": 2.4055252965541767, + "learning_rate": 3.998539630113427e-06, + "loss": 1.0337, + "step": 557 + }, + { + "epoch": 0.041935968735908616, + "grad_norm": 2.003328414800935, + "learning_rate": 3.998520969729325e-06, + "loss": 1.0919, + "step": 558 + }, + { + "epoch": 0.04201112280174357, + "grad_norm": 1.8251631781380684, + "learning_rate": 3.998502190925202e-06, + "loss": 1.09, + "step": 559 + }, + { + "epoch": 0.042086276867578536, + "grad_norm": 1.4761626494402773, + "learning_rate": 3.998483293702172e-06, + "loss": 1.0175, + "step": 560 + }, + { + "epoch": 0.0421614309334135, + "grad_norm": 0.7164283241523649, + "learning_rate": 3.998464278061353e-06, + "loss": 0.8515, + "step": 561 + }, + { + "epoch": 0.04223658499924846, + "grad_norm": 2.3598036218825706, + "learning_rate": 3.998445144003874e-06, + "loss": 1.0324, + "step": 562 + }, + { + "epoch": 0.04231173906508342, + "grad_norm": 1.534398481573919, + "learning_rate": 3.9984258915308674e-06, + "loss": 1.1015, + "step": 563 + }, + { + "epoch": 0.04238689313091838, + "grad_norm": 0.7129678344720685, + "learning_rate": 3.998406520643475e-06, + "loss": 0.798, + "step": 564 + }, + { + "epoch": 0.042462047196753346, + "grad_norm": 2.3103593764728054, + "learning_rate": 3.998387031342843e-06, + "loss": 1.0358, + "step": 565 + }, + { + "epoch": 0.04253720126258831, + "grad_norm": 1.707489999543558, + "learning_rate": 3.998367423630127e-06, + "loss": 1.049, + "step": 566 + }, + { + "epoch": 0.042612355328423265, + "grad_norm": 1.8548356753052961, + "learning_rate": 3.9983476975064885e-06, + "loss": 1.0801, + "step": 567 + }, + { + "epoch": 0.04268750939425823, + "grad_norm": 1.8031864281410168, + "learning_rate": 3.998327852973098e-06, + "loss": 1.0362, + "step": 568 + }, + { + "epoch": 0.04276266346009319, + "grad_norm": 1.6041268265305872, + "learning_rate": 3.998307890031129e-06, + "loss": 1.0706, + "step": 569 + }, + { + "epoch": 0.042837817525928155, + "grad_norm": 1.774745039023067, + "learning_rate": 3.998287808681766e-06, + "loss": 1.0784, + "step": 570 + }, + { + "epoch": 0.04291297159176311, + "grad_norm": 0.7832670875363225, + "learning_rate": 3.998267608926198e-06, + "loss": 0.9036, + "step": 571 + }, + { + "epoch": 0.042988125657598075, + "grad_norm": 1.8441855999020236, + "learning_rate": 3.998247290765623e-06, + "loss": 1.1272, + "step": 572 + }, + { + "epoch": 0.04306327972343304, + "grad_norm": 2.034602875259904, + "learning_rate": 3.9982268542012435e-06, + "loss": 1.0015, + "step": 573 + }, + { + "epoch": 0.043138433789268, + "grad_norm": 2.0601635038431145, + "learning_rate": 3.998206299234272e-06, + "loss": 0.9341, + "step": 574 + }, + { + "epoch": 0.04321358785510296, + "grad_norm": 1.8530707674630438, + "learning_rate": 3.998185625865924e-06, + "loss": 1.0967, + "step": 575 + }, + { + "epoch": 0.04328874192093792, + "grad_norm": 2.2200950521448086, + "learning_rate": 3.998164834097428e-06, + "loss": 1.0235, + "step": 576 + }, + { + "epoch": 0.043363895986772885, + "grad_norm": 0.794170915332756, + "learning_rate": 3.998143923930013e-06, + "loss": 0.8758, + "step": 577 + }, + { + "epoch": 0.04343905005260785, + "grad_norm": 1.574736025048003, + "learning_rate": 3.998122895364919e-06, + "loss": 1.082, + "step": 578 + }, + { + "epoch": 0.043514204118442805, + "grad_norm": 2.791090369641866, + "learning_rate": 3.998101748403393e-06, + "loss": 1.037, + "step": 579 + }, + { + "epoch": 0.04358935818427777, + "grad_norm": 1.6992718567715828, + "learning_rate": 3.998080483046687e-06, + "loss": 0.9981, + "step": 580 + }, + { + "epoch": 0.04366451225011273, + "grad_norm": 2.4536920365936545, + "learning_rate": 3.998059099296061e-06, + "loss": 1.0596, + "step": 581 + }, + { + "epoch": 0.043739666315947695, + "grad_norm": 1.8401832348352039, + "learning_rate": 3.9980375971527814e-06, + "loss": 1.1084, + "step": 582 + }, + { + "epoch": 0.04381482038178265, + "grad_norm": 1.7735744877141357, + "learning_rate": 3.998015976618124e-06, + "loss": 1.0867, + "step": 583 + }, + { + "epoch": 0.043889974447617615, + "grad_norm": 1.5481999435426765, + "learning_rate": 3.997994237693369e-06, + "loss": 1.1147, + "step": 584 + }, + { + "epoch": 0.04396512851345258, + "grad_norm": 0.6692551834908902, + "learning_rate": 3.997972380379804e-06, + "loss": 0.8446, + "step": 585 + }, + { + "epoch": 0.04404028257928754, + "grad_norm": 1.646750793985002, + "learning_rate": 3.997950404678726e-06, + "loss": 1.1159, + "step": 586 + }, + { + "epoch": 0.0441154366451225, + "grad_norm": 1.1492779828644473, + "learning_rate": 3.997928310591435e-06, + "loss": 0.9804, + "step": 587 + }, + { + "epoch": 0.04419059071095746, + "grad_norm": 3.1480220397507304, + "learning_rate": 3.997906098119241e-06, + "loss": 1.0552, + "step": 588 + }, + { + "epoch": 0.044265744776792425, + "grad_norm": 1.5844843930131562, + "learning_rate": 3.997883767263461e-06, + "loss": 1.0435, + "step": 589 + }, + { + "epoch": 0.04434089884262739, + "grad_norm": 2.4823645368267466, + "learning_rate": 3.997861318025417e-06, + "loss": 0.9818, + "step": 590 + }, + { + "epoch": 0.044416052908462345, + "grad_norm": 2.055819819477736, + "learning_rate": 3.997838750406439e-06, + "loss": 0.9434, + "step": 591 + }, + { + "epoch": 0.04449120697429731, + "grad_norm": 1.8719195811164309, + "learning_rate": 3.997816064407865e-06, + "loss": 1.05, + "step": 592 + }, + { + "epoch": 0.04456636104013227, + "grad_norm": 1.865204847811118, + "learning_rate": 3.997793260031039e-06, + "loss": 1.0406, + "step": 593 + }, + { + "epoch": 0.044641515105967235, + "grad_norm": 1.8101135854751704, + "learning_rate": 3.997770337277313e-06, + "loss": 1.0658, + "step": 594 + }, + { + "epoch": 0.04471666917180219, + "grad_norm": 0.7876794988935172, + "learning_rate": 3.997747296148044e-06, + "loss": 0.8982, + "step": 595 + }, + { + "epoch": 0.044791823237637154, + "grad_norm": 1.8183911706949236, + "learning_rate": 3.997724136644597e-06, + "loss": 1.0622, + "step": 596 + }, + { + "epoch": 0.04486697730347212, + "grad_norm": 2.160920544252328, + "learning_rate": 3.997700858768346e-06, + "loss": 1.0226, + "step": 597 + }, + { + "epoch": 0.04494213136930708, + "grad_norm": 2.8498898133510684, + "learning_rate": 3.99767746252067e-06, + "loss": 1.0359, + "step": 598 + }, + { + "epoch": 0.045017285435142045, + "grad_norm": 1.7835835346692996, + "learning_rate": 3.997653947902954e-06, + "loss": 1.0765, + "step": 599 + }, + { + "epoch": 0.045092439500977, + "grad_norm": 1.5134502420907083, + "learning_rate": 3.997630314916592e-06, + "loss": 1.1247, + "step": 600 + }, + { + "epoch": 0.045167593566811964, + "grad_norm": 1.6461950299029124, + "learning_rate": 3.9976065635629845e-06, + "loss": 1.0185, + "step": 601 + }, + { + "epoch": 0.04524274763264693, + "grad_norm": 0.739691865148263, + "learning_rate": 3.997582693843539e-06, + "loss": 0.8357, + "step": 602 + }, + { + "epoch": 0.04531790169848189, + "grad_norm": 1.4374198061923347, + "learning_rate": 3.997558705759669e-06, + "loss": 1.0665, + "step": 603 + }, + { + "epoch": 0.04539305576431685, + "grad_norm": 1.4920355434837331, + "learning_rate": 3.9975345993127975e-06, + "loss": 1.0632, + "step": 604 + }, + { + "epoch": 0.04546820983015181, + "grad_norm": 2.1928302673602933, + "learning_rate": 3.997510374504351e-06, + "loss": 1.0824, + "step": 605 + }, + { + "epoch": 0.045543363895986774, + "grad_norm": 2.476035558435986, + "learning_rate": 3.9974860313357665e-06, + "loss": 1.0874, + "step": 606 + }, + { + "epoch": 0.04561851796182174, + "grad_norm": 1.673809394152964, + "learning_rate": 3.997461569808485e-06, + "loss": 0.9814, + "step": 607 + }, + { + "epoch": 0.045693672027656694, + "grad_norm": 0.7403896843225087, + "learning_rate": 3.997436989923957e-06, + "loss": 0.842, + "step": 608 + }, + { + "epoch": 0.04576882609349166, + "grad_norm": 1.657022260167593, + "learning_rate": 3.997412291683639e-06, + "loss": 1.0363, + "step": 609 + }, + { + "epoch": 0.04584398015932662, + "grad_norm": 1.5309912015237275, + "learning_rate": 3.997387475088994e-06, + "loss": 1.0507, + "step": 610 + }, + { + "epoch": 0.045919134225161584, + "grad_norm": 3.256051811293418, + "learning_rate": 3.997362540141493e-06, + "loss": 1.0375, + "step": 611 + }, + { + "epoch": 0.04599428829099654, + "grad_norm": 1.549138734321839, + "learning_rate": 3.997337486842612e-06, + "loss": 0.998, + "step": 612 + }, + { + "epoch": 0.046069442356831504, + "grad_norm": 1.8238392992753771, + "learning_rate": 3.997312315193837e-06, + "loss": 0.9444, + "step": 613 + }, + { + "epoch": 0.04614459642266647, + "grad_norm": 1.9604765817673215, + "learning_rate": 3.9972870251966595e-06, + "loss": 1.0052, + "step": 614 + }, + { + "epoch": 0.04621975048850143, + "grad_norm": 1.6678609701528715, + "learning_rate": 3.997261616852578e-06, + "loss": 0.9724, + "step": 615 + }, + { + "epoch": 0.04629490455433639, + "grad_norm": 2.6225176381698954, + "learning_rate": 3.997236090163097e-06, + "loss": 1.0277, + "step": 616 + }, + { + "epoch": 0.04637005862017135, + "grad_norm": 0.6890371002221573, + "learning_rate": 3.997210445129729e-06, + "loss": 0.8828, + "step": 617 + }, + { + "epoch": 0.046445212686006314, + "grad_norm": 1.303216527564973, + "learning_rate": 3.997184681753996e-06, + "loss": 1.0243, + "step": 618 + }, + { + "epoch": 0.04652036675184128, + "grad_norm": 5.270394448967276, + "learning_rate": 3.997158800037422e-06, + "loss": 1.0271, + "step": 619 + }, + { + "epoch": 0.046595520817676234, + "grad_norm": 1.9768891196601452, + "learning_rate": 3.997132799981541e-06, + "loss": 0.9923, + "step": 620 + }, + { + "epoch": 0.0466706748835112, + "grad_norm": 1.4124315667243215, + "learning_rate": 3.997106681587895e-06, + "loss": 0.8973, + "step": 621 + }, + { + "epoch": 0.04674582894934616, + "grad_norm": 1.618528758951053, + "learning_rate": 3.99708044485803e-06, + "loss": 1.0444, + "step": 622 + }, + { + "epoch": 0.046820983015181124, + "grad_norm": 3.34338548045948, + "learning_rate": 3.997054089793501e-06, + "loss": 0.9621, + "step": 623 + }, + { + "epoch": 0.04689613708101608, + "grad_norm": 2.2942661338755137, + "learning_rate": 3.997027616395871e-06, + "loss": 1.0779, + "step": 624 + }, + { + "epoch": 0.046971291146851044, + "grad_norm": 1.7594407064252746, + "learning_rate": 3.997001024666707e-06, + "loss": 0.9767, + "step": 625 + }, + { + "epoch": 0.04704644521268601, + "grad_norm": 33.42609691660282, + "learning_rate": 3.996974314607585e-06, + "loss": 1.0687, + "step": 626 + }, + { + "epoch": 0.04712159927852097, + "grad_norm": 1.649799814507202, + "learning_rate": 3.996947486220088e-06, + "loss": 1.0364, + "step": 627 + }, + { + "epoch": 0.04719675334435593, + "grad_norm": 0.9220418617401981, + "learning_rate": 3.9969205395058064e-06, + "loss": 0.905, + "step": 628 + }, + { + "epoch": 0.04727190741019089, + "grad_norm": 1.9630251406606083, + "learning_rate": 3.996893474466336e-06, + "loss": 1.027, + "step": 629 + }, + { + "epoch": 0.04734706147602585, + "grad_norm": 2.172890953821311, + "learning_rate": 3.99686629110328e-06, + "loss": 1.06, + "step": 630 + }, + { + "epoch": 0.04742221554186082, + "grad_norm": 1.7501412743356053, + "learning_rate": 3.99683898941825e-06, + "loss": 1.0733, + "step": 631 + }, + { + "epoch": 0.04749736960769577, + "grad_norm": 2.4793096762904114, + "learning_rate": 3.996811569412864e-06, + "loss": 1.0377, + "step": 632 + }, + { + "epoch": 0.04757252367353074, + "grad_norm": 1.697356508058339, + "learning_rate": 3.996784031088745e-06, + "loss": 1.0798, + "step": 633 + }, + { + "epoch": 0.0476476777393657, + "grad_norm": 1.5822904398019575, + "learning_rate": 3.996756374447526e-06, + "loss": 1.0607, + "step": 634 + }, + { + "epoch": 0.04772283180520066, + "grad_norm": 1.5703160679714985, + "learning_rate": 3.996728599490847e-06, + "loss": 1.0714, + "step": 635 + }, + { + "epoch": 0.04779798587103562, + "grad_norm": 2.1737817312055725, + "learning_rate": 3.996700706220352e-06, + "loss": 1.0478, + "step": 636 + }, + { + "epoch": 0.04787313993687058, + "grad_norm": 3.604534916816972, + "learning_rate": 3.996672694637694e-06, + "loss": 1.0225, + "step": 637 + }, + { + "epoch": 0.04794829400270555, + "grad_norm": 1.858179878376537, + "learning_rate": 3.996644564744534e-06, + "loss": 1.0485, + "step": 638 + }, + { + "epoch": 0.04802344806854051, + "grad_norm": 1.7761497427385933, + "learning_rate": 3.996616316542537e-06, + "loss": 1.0137, + "step": 639 + }, + { + "epoch": 0.048098602134375466, + "grad_norm": 2.571235004208852, + "learning_rate": 3.996587950033377e-06, + "loss": 1.045, + "step": 640 + }, + { + "epoch": 0.04817375620021043, + "grad_norm": 1.8864342137323178, + "learning_rate": 3.996559465218736e-06, + "loss": 0.9832, + "step": 641 + }, + { + "epoch": 0.04824891026604539, + "grad_norm": 1.976369976573657, + "learning_rate": 3.996530862100302e-06, + "loss": 1.1212, + "step": 642 + }, + { + "epoch": 0.048324064331880356, + "grad_norm": 1.7728174949130482, + "learning_rate": 3.996502140679769e-06, + "loss": 0.9638, + "step": 643 + }, + { + "epoch": 0.04839921839771532, + "grad_norm": 1.6121964398095747, + "learning_rate": 3.996473300958839e-06, + "loss": 1.1478, + "step": 644 + }, + { + "epoch": 0.048474372463550276, + "grad_norm": 1.79291609144486, + "learning_rate": 3.99644434293922e-06, + "loss": 1.0735, + "step": 645 + }, + { + "epoch": 0.04854952652938524, + "grad_norm": 1.6238278783339661, + "learning_rate": 3.99641526662263e-06, + "loss": 1.1133, + "step": 646 + }, + { + "epoch": 0.0486246805952202, + "grad_norm": 1.5815399537347767, + "learning_rate": 3.99638607201079e-06, + "loss": 1.0232, + "step": 647 + }, + { + "epoch": 0.048699834661055166, + "grad_norm": 1.6543672044524966, + "learning_rate": 3.996356759105431e-06, + "loss": 1.0023, + "step": 648 + }, + { + "epoch": 0.04877498872689012, + "grad_norm": 1.8340777128638062, + "learning_rate": 3.996327327908289e-06, + "loss": 0.9552, + "step": 649 + }, + { + "epoch": 0.048850142792725086, + "grad_norm": 1.9565723360816973, + "learning_rate": 3.996297778421109e-06, + "loss": 0.9786, + "step": 650 + }, + { + "epoch": 0.04892529685856005, + "grad_norm": 1.790772066850834, + "learning_rate": 3.996268110645641e-06, + "loss": 0.9707, + "step": 651 + }, + { + "epoch": 0.04900045092439501, + "grad_norm": 1.6496662784981144, + "learning_rate": 3.996238324583643e-06, + "loss": 0.9852, + "step": 652 + }, + { + "epoch": 0.04907560499022997, + "grad_norm": 1.9798844959785866, + "learning_rate": 3.99620842023688e-06, + "loss": 1.0748, + "step": 653 + }, + { + "epoch": 0.04915075905606493, + "grad_norm": 0.7292966772159907, + "learning_rate": 3.996178397607125e-06, + "loss": 0.8444, + "step": 654 + }, + { + "epoch": 0.049225913121899896, + "grad_norm": 1.4860454596321278, + "learning_rate": 3.996148256696155e-06, + "loss": 1.1137, + "step": 655 + }, + { + "epoch": 0.04930106718773486, + "grad_norm": 1.8193977906487837, + "learning_rate": 3.996117997505758e-06, + "loss": 1.1378, + "step": 656 + }, + { + "epoch": 0.049376221253569816, + "grad_norm": 1.8032931270923838, + "learning_rate": 3.996087620037725e-06, + "loss": 1.0236, + "step": 657 + }, + { + "epoch": 0.04945137531940478, + "grad_norm": 0.7804615619927928, + "learning_rate": 3.996057124293857e-06, + "loss": 0.8072, + "step": 658 + }, + { + "epoch": 0.04952652938523974, + "grad_norm": 1.621391250704193, + "learning_rate": 3.996026510275962e-06, + "loss": 1.0427, + "step": 659 + }, + { + "epoch": 0.049601683451074706, + "grad_norm": 1.807486868951779, + "learning_rate": 3.995995777985852e-06, + "loss": 0.9802, + "step": 660 + }, + { + "epoch": 0.04967683751690966, + "grad_norm": 1.5352283878218407, + "learning_rate": 3.995964927425349e-06, + "loss": 1.0216, + "step": 661 + }, + { + "epoch": 0.049751991582744626, + "grad_norm": 2.6561816038413957, + "learning_rate": 3.995933958596282e-06, + "loss": 0.9578, + "step": 662 + }, + { + "epoch": 0.04982714564857959, + "grad_norm": 3.134876665186785, + "learning_rate": 3.995902871500485e-06, + "loss": 1.1779, + "step": 663 + }, + { + "epoch": 0.04990229971441455, + "grad_norm": 1.8852736346821284, + "learning_rate": 3.995871666139799e-06, + "loss": 1.0751, + "step": 664 + }, + { + "epoch": 0.04997745378024951, + "grad_norm": 1.57776418659165, + "learning_rate": 3.995840342516074e-06, + "loss": 1.0926, + "step": 665 + }, + { + "epoch": 0.05005260784608447, + "grad_norm": 1.607742666909636, + "learning_rate": 3.995808900631167e-06, + "loss": 1.0659, + "step": 666 + }, + { + "epoch": 0.050127761911919436, + "grad_norm": 2.2431961608636195, + "learning_rate": 3.99577734048694e-06, + "loss": 1.05, + "step": 667 + }, + { + "epoch": 0.0502029159777544, + "grad_norm": 2.1815566932153954, + "learning_rate": 3.9957456620852636e-06, + "loss": 0.9536, + "step": 668 + }, + { + "epoch": 0.050278070043589355, + "grad_norm": 1.7251363534525437, + "learning_rate": 3.995713865428014e-06, + "loss": 1.0122, + "step": 669 + }, + { + "epoch": 0.05035322410942432, + "grad_norm": 1.5565768316160944, + "learning_rate": 3.995681950517075e-06, + "loss": 1.0397, + "step": 670 + }, + { + "epoch": 0.05042837817525928, + "grad_norm": 1.7518231568029867, + "learning_rate": 3.995649917354339e-06, + "loss": 1.0442, + "step": 671 + }, + { + "epoch": 0.050503532241094246, + "grad_norm": 1.7284335531804542, + "learning_rate": 3.9956177659417036e-06, + "loss": 1.0133, + "step": 672 + }, + { + "epoch": 0.0505786863069292, + "grad_norm": 1.7141275843849304, + "learning_rate": 3.995585496281074e-06, + "loss": 1.0915, + "step": 673 + }, + { + "epoch": 0.050653840372764165, + "grad_norm": 2.4306878977274, + "learning_rate": 3.995553108374362e-06, + "loss": 1.0845, + "step": 674 + }, + { + "epoch": 0.05072899443859913, + "grad_norm": 1.7100273742492436, + "learning_rate": 3.995520602223487e-06, + "loss": 0.9152, + "step": 675 + }, + { + "epoch": 0.05080414850443409, + "grad_norm": 1.5990269384256135, + "learning_rate": 3.995487977830375e-06, + "loss": 1.1148, + "step": 676 + }, + { + "epoch": 0.05087930257026905, + "grad_norm": 1.6110744207231014, + "learning_rate": 3.995455235196959e-06, + "loss": 1.0063, + "step": 677 + }, + { + "epoch": 0.05095445663610401, + "grad_norm": 1.6153590236110804, + "learning_rate": 3.995422374325179e-06, + "loss": 1.0368, + "step": 678 + }, + { + "epoch": 0.051029610701938975, + "grad_norm": 1.825497948634161, + "learning_rate": 3.995389395216983e-06, + "loss": 1.0304, + "step": 679 + }, + { + "epoch": 0.05110476476777394, + "grad_norm": 1.5566926227150948, + "learning_rate": 3.9953562978743244e-06, + "loss": 0.9698, + "step": 680 + }, + { + "epoch": 0.051179918833608895, + "grad_norm": 1.6706197429307397, + "learning_rate": 3.995323082299164e-06, + "loss": 1.0481, + "step": 681 + }, + { + "epoch": 0.05125507289944386, + "grad_norm": 1.528449163398238, + "learning_rate": 3.9952897484934706e-06, + "loss": 0.9705, + "step": 682 + }, + { + "epoch": 0.05133022696527882, + "grad_norm": 1.8357152926117286, + "learning_rate": 3.9952562964592184e-06, + "loss": 1.0556, + "step": 683 + }, + { + "epoch": 0.051405381031113785, + "grad_norm": 1.6830296821674058, + "learning_rate": 3.995222726198391e-06, + "loss": 1.0953, + "step": 684 + }, + { + "epoch": 0.05148053509694874, + "grad_norm": 1.6460911834508498, + "learning_rate": 3.995189037712977e-06, + "loss": 1.0148, + "step": 685 + }, + { + "epoch": 0.051555689162783705, + "grad_norm": 1.685351947285273, + "learning_rate": 3.9951552310049715e-06, + "loss": 1.0712, + "step": 686 + }, + { + "epoch": 0.05163084322861867, + "grad_norm": 1.6139112078505315, + "learning_rate": 3.99512130607638e-06, + "loss": 1.0176, + "step": 687 + }, + { + "epoch": 0.05170599729445363, + "grad_norm": 2.0763608859019227, + "learning_rate": 3.995087262929209e-06, + "loss": 0.9557, + "step": 688 + }, + { + "epoch": 0.051781151360288595, + "grad_norm": 13.831072580446751, + "learning_rate": 3.99505310156548e-06, + "loss": 1.0436, + "step": 689 + }, + { + "epoch": 0.05185630542612355, + "grad_norm": 1.7348555165575255, + "learning_rate": 3.995018821987215e-06, + "loss": 1.0012, + "step": 690 + }, + { + "epoch": 0.051931459491958515, + "grad_norm": 1.968160147330382, + "learning_rate": 3.994984424196445e-06, + "loss": 1.0937, + "step": 691 + }, + { + "epoch": 0.05200661355779348, + "grad_norm": 1.617407668601836, + "learning_rate": 3.994949908195208e-06, + "loss": 1.0259, + "step": 692 + }, + { + "epoch": 0.05208176762362844, + "grad_norm": 0.7159151896510094, + "learning_rate": 3.994915273985551e-06, + "loss": 0.8302, + "step": 693 + }, + { + "epoch": 0.0521569216894634, + "grad_norm": 1.5606564899853126, + "learning_rate": 3.994880521569524e-06, + "loss": 0.9716, + "step": 694 + }, + { + "epoch": 0.05223207575529836, + "grad_norm": 1.6101022947286165, + "learning_rate": 3.994845650949187e-06, + "loss": 1.0116, + "step": 695 + }, + { + "epoch": 0.052307229821133325, + "grad_norm": 0.884282062081904, + "learning_rate": 3.994810662126607e-06, + "loss": 0.9341, + "step": 696 + }, + { + "epoch": 0.05238238388696829, + "grad_norm": 1.7162650437886935, + "learning_rate": 3.994775555103857e-06, + "loss": 1.0294, + "step": 697 + }, + { + "epoch": 0.052457537952803245, + "grad_norm": 1.6065069772311777, + "learning_rate": 3.994740329883016e-06, + "loss": 1.014, + "step": 698 + }, + { + "epoch": 0.05253269201863821, + "grad_norm": 1.771569768325826, + "learning_rate": 3.994704986466172e-06, + "loss": 0.9792, + "step": 699 + }, + { + "epoch": 0.05260784608447317, + "grad_norm": 4.028603564852126, + "learning_rate": 3.99466952485542e-06, + "loss": 1.0327, + "step": 700 + }, + { + "epoch": 0.052683000150308135, + "grad_norm": 1.9427277578519693, + "learning_rate": 3.994633945052861e-06, + "loss": 0.9432, + "step": 701 + }, + { + "epoch": 0.05275815421614309, + "grad_norm": 4.010136159359376, + "learning_rate": 3.994598247060602e-06, + "loss": 1.0356, + "step": 702 + }, + { + "epoch": 0.052833308281978054, + "grad_norm": 2.2158451675762776, + "learning_rate": 3.9945624308807585e-06, + "loss": 1.0839, + "step": 703 + }, + { + "epoch": 0.05290846234781302, + "grad_norm": 3.609515298866182, + "learning_rate": 3.994526496515454e-06, + "loss": 1.0751, + "step": 704 + }, + { + "epoch": 0.05298361641364798, + "grad_norm": 2.631265247871636, + "learning_rate": 3.994490443966818e-06, + "loss": 1.051, + "step": 705 + }, + { + "epoch": 0.05305877047948294, + "grad_norm": 0.7902456798489685, + "learning_rate": 3.994454273236984e-06, + "loss": 0.8146, + "step": 706 + }, + { + "epoch": 0.0531339245453179, + "grad_norm": 2.334990344466641, + "learning_rate": 3.994417984328098e-06, + "loss": 1.0737, + "step": 707 + }, + { + "epoch": 0.053209078611152864, + "grad_norm": 4.814861164250311, + "learning_rate": 3.994381577242309e-06, + "loss": 1.1017, + "step": 708 + }, + { + "epoch": 0.05328423267698783, + "grad_norm": 1.5787479191009024, + "learning_rate": 3.994345051981774e-06, + "loss": 1.0991, + "step": 709 + }, + { + "epoch": 0.053359386742822784, + "grad_norm": 2.5293944351056648, + "learning_rate": 3.994308408548659e-06, + "loss": 1.1393, + "step": 710 + }, + { + "epoch": 0.05343454080865775, + "grad_norm": 1.7826121292050852, + "learning_rate": 3.994271646945133e-06, + "loss": 1.1634, + "step": 711 + }, + { + "epoch": 0.05350969487449271, + "grad_norm": 2.827071441497942, + "learning_rate": 3.994234767173376e-06, + "loss": 0.9906, + "step": 712 + }, + { + "epoch": 0.053584848940327674, + "grad_norm": 1.8144822954968138, + "learning_rate": 3.994197769235572e-06, + "loss": 1.0715, + "step": 713 + }, + { + "epoch": 0.05366000300616263, + "grad_norm": 1.4838177889962678, + "learning_rate": 3.994160653133915e-06, + "loss": 1.0274, + "step": 714 + }, + { + "epoch": 0.053735157071997594, + "grad_norm": 1.316394432309135, + "learning_rate": 3.994123418870603e-06, + "loss": 1.0914, + "step": 715 + }, + { + "epoch": 0.05381031113783256, + "grad_norm": 1.5992689131345468, + "learning_rate": 3.994086066447841e-06, + "loss": 1.0123, + "step": 716 + }, + { + "epoch": 0.05388546520366752, + "grad_norm": 1.0269345240832788, + "learning_rate": 3.994048595867845e-06, + "loss": 0.8607, + "step": 717 + }, + { + "epoch": 0.05396061926950248, + "grad_norm": 1.793446861464666, + "learning_rate": 3.994011007132833e-06, + "loss": 1.0527, + "step": 718 + }, + { + "epoch": 0.05403577333533744, + "grad_norm": 0.697112550147894, + "learning_rate": 3.993973300245034e-06, + "loss": 0.8237, + "step": 719 + }, + { + "epoch": 0.054110927401172404, + "grad_norm": 2.2172724142340607, + "learning_rate": 3.993935475206682e-06, + "loss": 1.0212, + "step": 720 + }, + { + "epoch": 0.05418608146700737, + "grad_norm": 1.8149967754745213, + "learning_rate": 3.993897532020017e-06, + "loss": 0.9841, + "step": 721 + }, + { + "epoch": 0.054261235532842324, + "grad_norm": 2.4226967678722895, + "learning_rate": 3.993859470687288e-06, + "loss": 1.0427, + "step": 722 + }, + { + "epoch": 0.05433638959867729, + "grad_norm": 1.787943594249794, + "learning_rate": 3.993821291210751e-06, + "loss": 1.0483, + "step": 723 + }, + { + "epoch": 0.05441154366451225, + "grad_norm": 1.4393285648278147, + "learning_rate": 3.993782993592667e-06, + "loss": 0.9811, + "step": 724 + }, + { + "epoch": 0.054486697730347214, + "grad_norm": 3.8818551001885506, + "learning_rate": 3.993744577835306e-06, + "loss": 0.9897, + "step": 725 + }, + { + "epoch": 0.05456185179618217, + "grad_norm": 1.584598520978348, + "learning_rate": 3.993706043940945e-06, + "loss": 1.1294, + "step": 726 + }, + { + "epoch": 0.054637005862017134, + "grad_norm": 2.440420601817571, + "learning_rate": 3.993667391911866e-06, + "loss": 1.0389, + "step": 727 + }, + { + "epoch": 0.0547121599278521, + "grad_norm": 2.031227933323648, + "learning_rate": 3.993628621750359e-06, + "loss": 1.034, + "step": 728 + }, + { + "epoch": 0.05478731399368706, + "grad_norm": 1.8615033446514238, + "learning_rate": 3.993589733458723e-06, + "loss": 0.9922, + "step": 729 + }, + { + "epoch": 0.05486246805952202, + "grad_norm": 3.2909995421064737, + "learning_rate": 3.993550727039261e-06, + "loss": 0.9552, + "step": 730 + }, + { + "epoch": 0.05493762212535698, + "grad_norm": 1.8157510211431718, + "learning_rate": 3.993511602494285e-06, + "loss": 0.958, + "step": 731 + }, + { + "epoch": 0.055012776191191944, + "grad_norm": 1.8118921115536992, + "learning_rate": 3.993472359826112e-06, + "loss": 1.0096, + "step": 732 + }, + { + "epoch": 0.05508793025702691, + "grad_norm": 2.2857791410339043, + "learning_rate": 3.993432999037068e-06, + "loss": 1.0021, + "step": 733 + }, + { + "epoch": 0.05516308432286186, + "grad_norm": 1.66981443539154, + "learning_rate": 3.993393520129487e-06, + "loss": 1.0623, + "step": 734 + }, + { + "epoch": 0.05523823838869683, + "grad_norm": 2.2658612358317898, + "learning_rate": 3.993353923105705e-06, + "loss": 1.0556, + "step": 735 + }, + { + "epoch": 0.05531339245453179, + "grad_norm": 0.8172037345484406, + "learning_rate": 3.993314207968071e-06, + "loss": 0.9285, + "step": 736 + }, + { + "epoch": 0.05538854652036675, + "grad_norm": 0.7047317584025806, + "learning_rate": 3.993274374718938e-06, + "loss": 0.801, + "step": 737 + }, + { + "epoch": 0.05546370058620172, + "grad_norm": 1.7940093700287567, + "learning_rate": 3.9932344233606634e-06, + "loss": 1.0538, + "step": 738 + }, + { + "epoch": 0.05553885465203667, + "grad_norm": 1.907236625617493, + "learning_rate": 3.993194353895618e-06, + "loss": 1.0374, + "step": 739 + }, + { + "epoch": 0.05561400871787164, + "grad_norm": 2.280779228233073, + "learning_rate": 3.9931541663261756e-06, + "loss": 1.0456, + "step": 740 + }, + { + "epoch": 0.0556891627837066, + "grad_norm": 1.7798181099058283, + "learning_rate": 3.993113860654715e-06, + "loss": 1.0255, + "step": 741 + }, + { + "epoch": 0.05576431684954156, + "grad_norm": 1.937521657515044, + "learning_rate": 3.993073436883627e-06, + "loss": 1.0125, + "step": 742 + }, + { + "epoch": 0.05583947091537652, + "grad_norm": 1.6273453733057943, + "learning_rate": 3.993032895015304e-06, + "loss": 1.0181, + "step": 743 + }, + { + "epoch": 0.05591462498121148, + "grad_norm": 1.9309718996568552, + "learning_rate": 3.992992235052152e-06, + "loss": 1.0849, + "step": 744 + }, + { + "epoch": 0.05598977904704645, + "grad_norm": 1.5087030972678377, + "learning_rate": 3.992951456996578e-06, + "loss": 1.1061, + "step": 745 + }, + { + "epoch": 0.05606493311288141, + "grad_norm": 1.417425382557547, + "learning_rate": 3.9929105608509984e-06, + "loss": 1.0949, + "step": 746 + }, + { + "epoch": 0.056140087178716366, + "grad_norm": 1.5947366540125818, + "learning_rate": 3.9928695466178375e-06, + "loss": 0.9512, + "step": 747 + }, + { + "epoch": 0.05621524124455133, + "grad_norm": 1.5960296334166288, + "learning_rate": 3.992828414299524e-06, + "loss": 0.9732, + "step": 748 + }, + { + "epoch": 0.05629039531038629, + "grad_norm": 1.73882609197444, + "learning_rate": 3.9927871638984955e-06, + "loss": 1.0708, + "step": 749 + }, + { + "epoch": 0.056365549376221256, + "grad_norm": 2.2377176956744718, + "learning_rate": 3.992745795417198e-06, + "loss": 1.0495, + "step": 750 + }, + { + "epoch": 0.05644070344205621, + "grad_norm": 3.2532052593609415, + "learning_rate": 3.99270430885808e-06, + "loss": 1.0651, + "step": 751 + }, + { + "epoch": 0.056515857507891176, + "grad_norm": 1.6454927438108264, + "learning_rate": 3.992662704223602e-06, + "loss": 1.051, + "step": 752 + }, + { + "epoch": 0.05659101157372614, + "grad_norm": 1.769959118042283, + "learning_rate": 3.992620981516228e-06, + "loss": 1.0471, + "step": 753 + }, + { + "epoch": 0.0566661656395611, + "grad_norm": 1.4238562561521417, + "learning_rate": 3.9925791407384304e-06, + "loss": 1.0921, + "step": 754 + }, + { + "epoch": 0.05674131970539606, + "grad_norm": 1.4821500466151032, + "learning_rate": 3.9925371818926884e-06, + "loss": 1.0799, + "step": 755 + }, + { + "epoch": 0.05681647377123102, + "grad_norm": 0.7255871286292587, + "learning_rate": 3.992495104981489e-06, + "loss": 0.8795, + "step": 756 + }, + { + "epoch": 0.056891627837065986, + "grad_norm": 2.010857555160787, + "learning_rate": 3.992452910007325e-06, + "loss": 0.9975, + "step": 757 + }, + { + "epoch": 0.05696678190290095, + "grad_norm": 2.3282968342495476, + "learning_rate": 3.992410596972696e-06, + "loss": 1.1599, + "step": 758 + }, + { + "epoch": 0.057041935968735906, + "grad_norm": 1.4826923671411454, + "learning_rate": 3.99236816588011e-06, + "loss": 1.0747, + "step": 759 + }, + { + "epoch": 0.05711709003457087, + "grad_norm": 1.4320325952368544, + "learning_rate": 3.992325616732081e-06, + "loss": 0.9893, + "step": 760 + }, + { + "epoch": 0.05719224410040583, + "grad_norm": 1.56122834910643, + "learning_rate": 3.992282949531129e-06, + "loss": 1.0189, + "step": 761 + }, + { + "epoch": 0.057267398166240796, + "grad_norm": 1.778778168059171, + "learning_rate": 3.992240164279785e-06, + "loss": 1.1249, + "step": 762 + }, + { + "epoch": 0.05734255223207575, + "grad_norm": 2.0290900851086744, + "learning_rate": 3.9921972609805815e-06, + "loss": 1.0836, + "step": 763 + }, + { + "epoch": 0.057417706297910716, + "grad_norm": 1.6040333366306896, + "learning_rate": 3.992154239636062e-06, + "loss": 1.0904, + "step": 764 + }, + { + "epoch": 0.05749286036374568, + "grad_norm": 0.7691642544489183, + "learning_rate": 3.992111100248775e-06, + "loss": 0.8586, + "step": 765 + }, + { + "epoch": 0.05756801442958064, + "grad_norm": 1.8080974252116524, + "learning_rate": 3.992067842821277e-06, + "loss": 1.0507, + "step": 766 + }, + { + "epoch": 0.0576431684954156, + "grad_norm": 1.6544697388607716, + "learning_rate": 3.992024467356132e-06, + "loss": 0.9736, + "step": 767 + }, + { + "epoch": 0.05771832256125056, + "grad_norm": 1.7411586991121097, + "learning_rate": 3.991980973855908e-06, + "loss": 1.0943, + "step": 768 + }, + { + "epoch": 0.057793476627085526, + "grad_norm": 2.272688909553649, + "learning_rate": 3.991937362323183e-06, + "loss": 1.1295, + "step": 769 + }, + { + "epoch": 0.05786863069292049, + "grad_norm": 2.2212199043463383, + "learning_rate": 3.991893632760544e-06, + "loss": 1.0401, + "step": 770 + }, + { + "epoch": 0.057943784758755446, + "grad_norm": 0.7825176635552324, + "learning_rate": 3.991849785170578e-06, + "loss": 0.8489, + "step": 771 + }, + { + "epoch": 0.05801893882459041, + "grad_norm": 1.8858452901603036, + "learning_rate": 3.991805819555885e-06, + "loss": 0.9718, + "step": 772 + }, + { + "epoch": 0.05809409289042537, + "grad_norm": 1.76189162934931, + "learning_rate": 3.991761735919071e-06, + "loss": 0.9857, + "step": 773 + }, + { + "epoch": 0.058169246956260336, + "grad_norm": 1.955346941688041, + "learning_rate": 3.991717534262747e-06, + "loss": 1.094, + "step": 774 + }, + { + "epoch": 0.05824440102209529, + "grad_norm": 1.4119689787193785, + "learning_rate": 3.991673214589532e-06, + "loss": 1.1182, + "step": 775 + }, + { + "epoch": 0.058319555087930255, + "grad_norm": 2.510696986417314, + "learning_rate": 3.991628776902052e-06, + "loss": 1.0205, + "step": 776 + }, + { + "epoch": 0.05839470915376522, + "grad_norm": 1.873848497933274, + "learning_rate": 3.991584221202942e-06, + "loss": 1.0783, + "step": 777 + }, + { + "epoch": 0.05846986321960018, + "grad_norm": 2.2339600931426475, + "learning_rate": 3.991539547494839e-06, + "loss": 1.0972, + "step": 778 + }, + { + "epoch": 0.05854501728543514, + "grad_norm": 1.8332488806375322, + "learning_rate": 3.991494755780392e-06, + "loss": 0.9598, + "step": 779 + }, + { + "epoch": 0.0586201713512701, + "grad_norm": 1.4841158188868682, + "learning_rate": 3.991449846062255e-06, + "loss": 1.1333, + "step": 780 + }, + { + "epoch": 0.058695325417105065, + "grad_norm": 1.6101696682018758, + "learning_rate": 3.991404818343089e-06, + "loss": 1.1102, + "step": 781 + }, + { + "epoch": 0.05877047948294003, + "grad_norm": 1.9581591007832082, + "learning_rate": 3.991359672625562e-06, + "loss": 1.0076, + "step": 782 + }, + { + "epoch": 0.05884563354877499, + "grad_norm": 1.4816983634868122, + "learning_rate": 3.9913144089123485e-06, + "loss": 1.0734, + "step": 783 + }, + { + "epoch": 0.05892078761460995, + "grad_norm": 3.9942543964924075, + "learning_rate": 3.991269027206131e-06, + "loss": 1.02, + "step": 784 + }, + { + "epoch": 0.05899594168044491, + "grad_norm": 1.8270550066304567, + "learning_rate": 3.991223527509599e-06, + "loss": 0.9748, + "step": 785 + }, + { + "epoch": 0.059071095746279875, + "grad_norm": 1.9862274774579225, + "learning_rate": 3.991177909825448e-06, + "loss": 0.9903, + "step": 786 + }, + { + "epoch": 0.05914624981211484, + "grad_norm": 1.6519867351267286, + "learning_rate": 3.991132174156381e-06, + "loss": 1.0609, + "step": 787 + }, + { + "epoch": 0.059221403877949795, + "grad_norm": 1.7434386819745555, + "learning_rate": 3.991086320505108e-06, + "loss": 1.0949, + "step": 788 + }, + { + "epoch": 0.05929655794378476, + "grad_norm": 2.0713554778145666, + "learning_rate": 3.991040348874346e-06, + "loss": 1.0978, + "step": 789 + }, + { + "epoch": 0.05937171200961972, + "grad_norm": 3.0918039694130544, + "learning_rate": 3.99099425926682e-06, + "loss": 1.1164, + "step": 790 + }, + { + "epoch": 0.059446866075454685, + "grad_norm": 4.54644458739249, + "learning_rate": 3.990948051685259e-06, + "loss": 1.0351, + "step": 791 + }, + { + "epoch": 0.05952202014128964, + "grad_norm": 1.582416402769173, + "learning_rate": 3.990901726132403e-06, + "loss": 1.1208, + "step": 792 + }, + { + "epoch": 0.059597174207124605, + "grad_norm": 1.9548511659208483, + "learning_rate": 3.990855282610996e-06, + "loss": 0.9477, + "step": 793 + }, + { + "epoch": 0.05967232827295957, + "grad_norm": 2.28468659620198, + "learning_rate": 3.990808721123789e-06, + "loss": 0.9702, + "step": 794 + }, + { + "epoch": 0.05974748233879453, + "grad_norm": 3.250221244910743, + "learning_rate": 3.990762041673543e-06, + "loss": 1.0517, + "step": 795 + }, + { + "epoch": 0.05982263640462949, + "grad_norm": 2.0963711629300983, + "learning_rate": 3.990715244263023e-06, + "loss": 0.9966, + "step": 796 + }, + { + "epoch": 0.05989779047046445, + "grad_norm": 2.357562853399046, + "learning_rate": 3.9906683288950005e-06, + "loss": 1.0497, + "step": 797 + }, + { + "epoch": 0.059972944536299415, + "grad_norm": 1.3131410304910769, + "learning_rate": 3.990621295572258e-06, + "loss": 1.0636, + "step": 798 + }, + { + "epoch": 0.06004809860213438, + "grad_norm": 1.81091383884827, + "learning_rate": 3.99057414429758e-06, + "loss": 1.1145, + "step": 799 + }, + { + "epoch": 0.060123252667969335, + "grad_norm": 1.7400685926049624, + "learning_rate": 3.9905268750737625e-06, + "loss": 1.0317, + "step": 800 + }, + { + "epoch": 0.0601984067338043, + "grad_norm": 2.5804659301483706, + "learning_rate": 3.990479487903605e-06, + "loss": 0.9973, + "step": 801 + }, + { + "epoch": 0.06027356079963926, + "grad_norm": 1.4298937706070851, + "learning_rate": 3.990431982789917e-06, + "loss": 1.0208, + "step": 802 + }, + { + "epoch": 0.060348714865474225, + "grad_norm": 1.977590783467158, + "learning_rate": 3.9903843597355105e-06, + "loss": 1.1151, + "step": 803 + }, + { + "epoch": 0.06042386893130918, + "grad_norm": 1.9448493498093828, + "learning_rate": 3.99033661874321e-06, + "loss": 1.1663, + "step": 804 + }, + { + "epoch": 0.060499022997144145, + "grad_norm": 1.7044896249586845, + "learning_rate": 3.990288759815843e-06, + "loss": 0.8283, + "step": 805 + }, + { + "epoch": 0.06057417706297911, + "grad_norm": 1.8751043926424162, + "learning_rate": 3.990240782956245e-06, + "loss": 1.1136, + "step": 806 + }, + { + "epoch": 0.06064933112881407, + "grad_norm": 1.787276730231739, + "learning_rate": 3.99019268816726e-06, + "loss": 1.0396, + "step": 807 + }, + { + "epoch": 0.06072448519464903, + "grad_norm": 1.7327480761674607, + "learning_rate": 3.990144475451738e-06, + "loss": 1.0858, + "step": 808 + }, + { + "epoch": 0.06079963926048399, + "grad_norm": 1.4558847943567124, + "learning_rate": 3.990096144812534e-06, + "loss": 1.1287, + "step": 809 + }, + { + "epoch": 0.060874793326318954, + "grad_norm": 1.6359819378870077, + "learning_rate": 3.9900476962525125e-06, + "loss": 1.0769, + "step": 810 + }, + { + "epoch": 0.06094994739215392, + "grad_norm": 1.7278091969892282, + "learning_rate": 3.989999129774546e-06, + "loss": 0.9594, + "step": 811 + }, + { + "epoch": 0.061025101457988874, + "grad_norm": 2.1582785752860585, + "learning_rate": 3.989950445381511e-06, + "loss": 0.9949, + "step": 812 + }, + { + "epoch": 0.06110025552382384, + "grad_norm": 1.4953771722693778, + "learning_rate": 3.98990164307629e-06, + "loss": 1.0309, + "step": 813 + }, + { + "epoch": 0.0611754095896588, + "grad_norm": 2.0827593076170796, + "learning_rate": 3.989852722861778e-06, + "loss": 1.1513, + "step": 814 + }, + { + "epoch": 0.061250563655493764, + "grad_norm": 1.8471694530831004, + "learning_rate": 3.989803684740873e-06, + "loss": 1.0567, + "step": 815 + }, + { + "epoch": 0.06132571772132872, + "grad_norm": 1.9378121647544586, + "learning_rate": 3.9897545287164795e-06, + "loss": 0.9805, + "step": 816 + }, + { + "epoch": 0.061400871787163684, + "grad_norm": 1.473460820034754, + "learning_rate": 3.9897052547915115e-06, + "loss": 1.0755, + "step": 817 + }, + { + "epoch": 0.06147602585299865, + "grad_norm": 1.9649835313102022, + "learning_rate": 3.989655862968887e-06, + "loss": 1.0504, + "step": 818 + }, + { + "epoch": 0.06155117991883361, + "grad_norm": 1.2204870902585545, + "learning_rate": 3.989606353251535e-06, + "loss": 0.9997, + "step": 819 + }, + { + "epoch": 0.06162633398466857, + "grad_norm": 1.33941421074018, + "learning_rate": 3.989556725642388e-06, + "loss": 0.9799, + "step": 820 + }, + { + "epoch": 0.06170148805050353, + "grad_norm": 2.023266759318705, + "learning_rate": 3.989506980144385e-06, + "loss": 1.0231, + "step": 821 + }, + { + "epoch": 0.061776642116338494, + "grad_norm": 1.5002271162390954, + "learning_rate": 3.989457116760477e-06, + "loss": 1.0052, + "step": 822 + }, + { + "epoch": 0.06185179618217346, + "grad_norm": 0.7347081828643911, + "learning_rate": 3.989407135493615e-06, + "loss": 0.9061, + "step": 823 + }, + { + "epoch": 0.061926950248008414, + "grad_norm": 1.5777877598358692, + "learning_rate": 3.9893570363467625e-06, + "loss": 0.8758, + "step": 824 + }, + { + "epoch": 0.06200210431384338, + "grad_norm": 1.551547305675224, + "learning_rate": 3.9893068193228885e-06, + "loss": 1.0588, + "step": 825 + }, + { + "epoch": 0.06207725837967834, + "grad_norm": 2.6228039956429354, + "learning_rate": 3.989256484424968e-06, + "loss": 1.0046, + "step": 826 + }, + { + "epoch": 0.062152412445513304, + "grad_norm": 1.7048966878203344, + "learning_rate": 3.989206031655982e-06, + "loss": 0.9876, + "step": 827 + }, + { + "epoch": 0.06222756651134827, + "grad_norm": 1.78939269904185, + "learning_rate": 3.989155461018923e-06, + "loss": 1.0915, + "step": 828 + }, + { + "epoch": 0.062302720577183224, + "grad_norm": 1.482350223302353, + "learning_rate": 3.989104772516785e-06, + "loss": 1.052, + "step": 829 + }, + { + "epoch": 0.06237787464301819, + "grad_norm": 1.9461573715560776, + "learning_rate": 3.989053966152573e-06, + "loss": 1.0104, + "step": 830 + }, + { + "epoch": 0.06245302870885315, + "grad_norm": 1.8799320204598604, + "learning_rate": 3.9890030419292965e-06, + "loss": 1.0373, + "step": 831 + }, + { + "epoch": 0.0625281827746881, + "grad_norm": 2.529427425594353, + "learning_rate": 3.988951999849974e-06, + "loss": 0.9435, + "step": 832 + }, + { + "epoch": 0.06260333684052308, + "grad_norm": 1.8557767913500143, + "learning_rate": 3.988900839917628e-06, + "loss": 1.0504, + "step": 833 + }, + { + "epoch": 0.06267849090635803, + "grad_norm": 1.481969470323552, + "learning_rate": 3.988849562135293e-06, + "loss": 1.0204, + "step": 834 + }, + { + "epoch": 0.06275364497219299, + "grad_norm": 1.6252706742437966, + "learning_rate": 3.988798166506005e-06, + "loss": 1.0365, + "step": 835 + }, + { + "epoch": 0.06282879903802796, + "grad_norm": 0.8450163063191453, + "learning_rate": 3.98874665303281e-06, + "loss": 0.8773, + "step": 836 + }, + { + "epoch": 0.06290395310386292, + "grad_norm": 1.9229934043354584, + "learning_rate": 3.98869502171876e-06, + "loss": 1.1612, + "step": 837 + }, + { + "epoch": 0.06297910716969789, + "grad_norm": 1.6406797854040998, + "learning_rate": 3.9886432725669146e-06, + "loss": 1.0663, + "step": 838 + }, + { + "epoch": 0.06305426123553284, + "grad_norm": 1.7392321782308713, + "learning_rate": 3.988591405580341e-06, + "loss": 1.0741, + "step": 839 + }, + { + "epoch": 0.0631294153013678, + "grad_norm": 1.3725047445302092, + "learning_rate": 3.988539420762111e-06, + "loss": 1.0356, + "step": 840 + }, + { + "epoch": 0.06320456936720277, + "grad_norm": 1.8096090478348796, + "learning_rate": 3.988487318115306e-06, + "loss": 1.0485, + "step": 841 + }, + { + "epoch": 0.06327972343303773, + "grad_norm": 1.7628081816573218, + "learning_rate": 3.9884350976430136e-06, + "loss": 1.0749, + "step": 842 + }, + { + "epoch": 0.06335487749887268, + "grad_norm": 2.0158738566700696, + "learning_rate": 3.988382759348327e-06, + "loss": 1.0264, + "step": 843 + }, + { + "epoch": 0.06343003156470765, + "grad_norm": 1.655625207482653, + "learning_rate": 3.988330303234347e-06, + "loss": 1.1575, + "step": 844 + }, + { + "epoch": 0.06350518563054261, + "grad_norm": 1.440551090620145, + "learning_rate": 3.988277729304184e-06, + "loss": 1.0618, + "step": 845 + }, + { + "epoch": 0.06358033969637758, + "grad_norm": 2.37026012667064, + "learning_rate": 3.988225037560951e-06, + "loss": 0.9921, + "step": 846 + }, + { + "epoch": 0.06365549376221254, + "grad_norm": 1.839573931498999, + "learning_rate": 3.988172228007771e-06, + "loss": 1.0629, + "step": 847 + }, + { + "epoch": 0.0637306478280475, + "grad_norm": 1.906189776860721, + "learning_rate": 3.9881193006477745e-06, + "loss": 1.0026, + "step": 848 + }, + { + "epoch": 0.06380580189388246, + "grad_norm": 1.5759188651050284, + "learning_rate": 3.9880662554840955e-06, + "loss": 1.0194, + "step": 849 + }, + { + "epoch": 0.06388095595971742, + "grad_norm": 2.0748970373874647, + "learning_rate": 3.9880130925198786e-06, + "loss": 1.1169, + "step": 850 + }, + { + "epoch": 0.06395611002555238, + "grad_norm": 2.6657315839989755, + "learning_rate": 3.987959811758273e-06, + "loss": 0.9808, + "step": 851 + }, + { + "epoch": 0.06403126409138735, + "grad_norm": 1.5937872489615397, + "learning_rate": 3.9879064132024365e-06, + "loss": 0.9743, + "step": 852 + }, + { + "epoch": 0.0641064181572223, + "grad_norm": 1.5135963470320142, + "learning_rate": 3.987852896855532e-06, + "loss": 1.0975, + "step": 853 + }, + { + "epoch": 0.06418157222305727, + "grad_norm": 1.3648724819886449, + "learning_rate": 3.987799262720732e-06, + "loss": 1.0826, + "step": 854 + }, + { + "epoch": 0.06425672628889223, + "grad_norm": 2.8401610815008485, + "learning_rate": 3.987745510801214e-06, + "loss": 1.0387, + "step": 855 + }, + { + "epoch": 0.06433188035472719, + "grad_norm": 2.173470329733295, + "learning_rate": 3.987691641100162e-06, + "loss": 1.0355, + "step": 856 + }, + { + "epoch": 0.06440703442056216, + "grad_norm": 1.691067210129031, + "learning_rate": 3.98763765362077e-06, + "loss": 1.1088, + "step": 857 + }, + { + "epoch": 0.06448218848639711, + "grad_norm": 1.5606629062337327, + "learning_rate": 3.987583548366235e-06, + "loss": 0.9203, + "step": 858 + }, + { + "epoch": 0.06455734255223207, + "grad_norm": 2.3451560317064866, + "learning_rate": 3.987529325339764e-06, + "loss": 0.9419, + "step": 859 + }, + { + "epoch": 0.06463249661806704, + "grad_norm": 1.563014060334892, + "learning_rate": 3.98747498454457e-06, + "loss": 0.95, + "step": 860 + }, + { + "epoch": 0.064707650683902, + "grad_norm": 1.4880000772476176, + "learning_rate": 3.987420525983873e-06, + "loss": 1.1052, + "step": 861 + }, + { + "epoch": 0.06478280474973697, + "grad_norm": 1.3700615021203557, + "learning_rate": 3.9873659496608985e-06, + "loss": 0.9659, + "step": 862 + }, + { + "epoch": 0.06485795881557192, + "grad_norm": 2.1035841963420085, + "learning_rate": 3.9873112555788816e-06, + "loss": 1.0882, + "step": 863 + }, + { + "epoch": 0.06493311288140688, + "grad_norm": 1.8888455047798705, + "learning_rate": 3.987256443741063e-06, + "loss": 0.9819, + "step": 864 + }, + { + "epoch": 0.06500826694724185, + "grad_norm": 1.874195371277666, + "learning_rate": 3.9872015141506905e-06, + "loss": 1.121, + "step": 865 + }, + { + "epoch": 0.0650834210130768, + "grad_norm": 2.0104403980284165, + "learning_rate": 3.987146466811019e-06, + "loss": 1.0643, + "step": 866 + }, + { + "epoch": 0.06515857507891176, + "grad_norm": 2.037579929211619, + "learning_rate": 3.98709130172531e-06, + "loss": 1.0908, + "step": 867 + }, + { + "epoch": 0.06523372914474673, + "grad_norm": 1.7893767101773994, + "learning_rate": 3.987036018896832e-06, + "loss": 0.9586, + "step": 868 + }, + { + "epoch": 0.06530888321058169, + "grad_norm": 2.734874570714625, + "learning_rate": 3.986980618328861e-06, + "loss": 0.9957, + "step": 869 + }, + { + "epoch": 0.06538403727641666, + "grad_norm": 1.6262414298392507, + "learning_rate": 3.98692510002468e-06, + "loss": 0.9921, + "step": 870 + }, + { + "epoch": 0.06545919134225162, + "grad_norm": 2.372135141051388, + "learning_rate": 3.986869463987578e-06, + "loss": 0.9984, + "step": 871 + }, + { + "epoch": 0.06553434540808657, + "grad_norm": 1.9685981934163692, + "learning_rate": 3.9868137102208525e-06, + "loss": 0.9818, + "step": 872 + }, + { + "epoch": 0.06560949947392154, + "grad_norm": 2.0723639922385506, + "learning_rate": 3.9867578387278065e-06, + "loss": 1.1122, + "step": 873 + }, + { + "epoch": 0.0656846535397565, + "grad_norm": 1.5350098673599646, + "learning_rate": 3.986701849511751e-06, + "loss": 1.0319, + "step": 874 + }, + { + "epoch": 0.06575980760559147, + "grad_norm": 1.7299777124325346, + "learning_rate": 3.986645742576002e-06, + "loss": 1.0278, + "step": 875 + }, + { + "epoch": 0.06583496167142643, + "grad_norm": 1.6096576331390633, + "learning_rate": 3.986589517923887e-06, + "loss": 1.0463, + "step": 876 + }, + { + "epoch": 0.06591011573726138, + "grad_norm": 2.308729448541977, + "learning_rate": 3.986533175558735e-06, + "loss": 1.0182, + "step": 877 + }, + { + "epoch": 0.06598526980309635, + "grad_norm": 1.7327033570076615, + "learning_rate": 3.9864767154838856e-06, + "loss": 1.0144, + "step": 878 + }, + { + "epoch": 0.06606042386893131, + "grad_norm": 1.7879009890782056, + "learning_rate": 3.986420137702684e-06, + "loss": 0.9614, + "step": 879 + }, + { + "epoch": 0.06613557793476627, + "grad_norm": 0.7985309150529988, + "learning_rate": 3.9863634422184835e-06, + "loss": 0.8322, + "step": 880 + }, + { + "epoch": 0.06621073200060124, + "grad_norm": 1.8658082728245557, + "learning_rate": 3.986306629034642e-06, + "loss": 1.1231, + "step": 881 + }, + { + "epoch": 0.06628588606643619, + "grad_norm": 1.2453006145527548, + "learning_rate": 3.9862496981545265e-06, + "loss": 0.9615, + "step": 882 + }, + { + "epoch": 0.06636104013227116, + "grad_norm": 1.777600012110376, + "learning_rate": 3.986192649581511e-06, + "loss": 0.9944, + "step": 883 + }, + { + "epoch": 0.06643619419810612, + "grad_norm": 2.007631880061338, + "learning_rate": 3.986135483318975e-06, + "loss": 1.0315, + "step": 884 + }, + { + "epoch": 0.06651134826394108, + "grad_norm": 1.917974778921437, + "learning_rate": 3.986078199370307e-06, + "loss": 0.9739, + "step": 885 + }, + { + "epoch": 0.06658650232977605, + "grad_norm": 2.0106398568925155, + "learning_rate": 3.9860207977388994e-06, + "loss": 0.922, + "step": 886 + }, + { + "epoch": 0.066661656395611, + "grad_norm": 1.9273119679131807, + "learning_rate": 3.985963278428155e-06, + "loss": 1.0825, + "step": 887 + }, + { + "epoch": 0.06673681046144596, + "grad_norm": 0.7684949684449915, + "learning_rate": 3.985905641441482e-06, + "loss": 0.8602, + "step": 888 + }, + { + "epoch": 0.06681196452728093, + "grad_norm": 0.8365190212392364, + "learning_rate": 3.9858478867822945e-06, + "loss": 0.852, + "step": 889 + }, + { + "epoch": 0.06688711859311589, + "grad_norm": 1.469766954646368, + "learning_rate": 3.985790014454016e-06, + "loss": 1.0386, + "step": 890 + }, + { + "epoch": 0.06696227265895086, + "grad_norm": 5.065540860399215, + "learning_rate": 3.985732024460074e-06, + "loss": 1.0985, + "step": 891 + }, + { + "epoch": 0.06703742672478581, + "grad_norm": 1.836602403078415, + "learning_rate": 3.985673916803907e-06, + "loss": 1.0666, + "step": 892 + }, + { + "epoch": 0.06711258079062077, + "grad_norm": 1.6924417686869224, + "learning_rate": 3.9856156914889556e-06, + "loss": 1.1097, + "step": 893 + }, + { + "epoch": 0.06718773485645574, + "grad_norm": 1.6168836225199368, + "learning_rate": 3.985557348518672e-06, + "loss": 1.0, + "step": 894 + }, + { + "epoch": 0.0672628889222907, + "grad_norm": 1.5472324613365156, + "learning_rate": 3.9854988878965125e-06, + "loss": 1.0148, + "step": 895 + }, + { + "epoch": 0.06733804298812565, + "grad_norm": 4.731967225632791, + "learning_rate": 3.98544030962594e-06, + "loss": 1.1051, + "step": 896 + }, + { + "epoch": 0.06741319705396062, + "grad_norm": 1.8925979030587436, + "learning_rate": 3.985381613710427e-06, + "loss": 1.0516, + "step": 897 + }, + { + "epoch": 0.06748835111979558, + "grad_norm": 2.0682255428772067, + "learning_rate": 3.98532280015345e-06, + "loss": 1.0031, + "step": 898 + }, + { + "epoch": 0.06756350518563055, + "grad_norm": 1.905572989890391, + "learning_rate": 3.985263868958496e-06, + "loss": 1.0037, + "step": 899 + }, + { + "epoch": 0.0676386592514655, + "grad_norm": 1.4631689644668866, + "learning_rate": 3.9852048201290545e-06, + "loss": 0.9786, + "step": 900 + }, + { + "epoch": 0.06771381331730046, + "grad_norm": 1.7240781559941167, + "learning_rate": 3.985145653668626e-06, + "loss": 1.0472, + "step": 901 + }, + { + "epoch": 0.06778896738313543, + "grad_norm": 1.5668815215210532, + "learning_rate": 3.985086369580716e-06, + "loss": 1.0327, + "step": 902 + }, + { + "epoch": 0.06786412144897039, + "grad_norm": 1.6420987730438534, + "learning_rate": 3.985026967868837e-06, + "loss": 0.9632, + "step": 903 + }, + { + "epoch": 0.06793927551480534, + "grad_norm": 1.9019131966327052, + "learning_rate": 3.9849674485365094e-06, + "loss": 1.0143, + "step": 904 + }, + { + "epoch": 0.06801442958064031, + "grad_norm": 1.894669360602976, + "learning_rate": 3.98490781158726e-06, + "loss": 1.094, + "step": 905 + }, + { + "epoch": 0.06808958364647527, + "grad_norm": 1.4465271665551787, + "learning_rate": 3.98484805702462e-06, + "loss": 1.0715, + "step": 906 + }, + { + "epoch": 0.06816473771231024, + "grad_norm": 1.399576799256878, + "learning_rate": 3.9847881848521345e-06, + "loss": 0.9923, + "step": 907 + }, + { + "epoch": 0.0682398917781452, + "grad_norm": 1.5375648608909263, + "learning_rate": 3.984728195073347e-06, + "loss": 0.993, + "step": 908 + }, + { + "epoch": 0.06831504584398015, + "grad_norm": 1.9089464596393617, + "learning_rate": 3.984668087691815e-06, + "loss": 1.0715, + "step": 909 + }, + { + "epoch": 0.06839019990981512, + "grad_norm": 1.8294518375192643, + "learning_rate": 3.984607862711099e-06, + "loss": 0.9086, + "step": 910 + }, + { + "epoch": 0.06846535397565008, + "grad_norm": 2.1535845881974844, + "learning_rate": 3.984547520134767e-06, + "loss": 0.9658, + "step": 911 + }, + { + "epoch": 0.06854050804148504, + "grad_norm": 4.777657948452579, + "learning_rate": 3.9844870599663954e-06, + "loss": 0.9794, + "step": 912 + }, + { + "epoch": 0.06861566210732001, + "grad_norm": 1.612232818942595, + "learning_rate": 3.984426482209567e-06, + "loss": 1.0611, + "step": 913 + }, + { + "epoch": 0.06869081617315496, + "grad_norm": 1.697478918582311, + "learning_rate": 3.98436578686787e-06, + "loss": 1.0273, + "step": 914 + }, + { + "epoch": 0.06876597023898993, + "grad_norm": 1.5109711604821707, + "learning_rate": 3.984304973944901e-06, + "loss": 1.0714, + "step": 915 + }, + { + "epoch": 0.06884112430482489, + "grad_norm": 3.065069716440098, + "learning_rate": 3.984244043444264e-06, + "loss": 0.9842, + "step": 916 + }, + { + "epoch": 0.06891627837065985, + "grad_norm": 1.930620346384809, + "learning_rate": 3.98418299536957e-06, + "loss": 1.083, + "step": 917 + }, + { + "epoch": 0.06899143243649482, + "grad_norm": 1.5808863134895528, + "learning_rate": 3.984121829724435e-06, + "loss": 1.0087, + "step": 918 + }, + { + "epoch": 0.06906658650232977, + "grad_norm": 10.073584475598134, + "learning_rate": 3.984060546512484e-06, + "loss": 1.1426, + "step": 919 + }, + { + "epoch": 0.06914174056816474, + "grad_norm": 1.3708143683206742, + "learning_rate": 3.983999145737348e-06, + "loss": 1.0237, + "step": 920 + }, + { + "epoch": 0.0692168946339997, + "grad_norm": 2.1404667259789845, + "learning_rate": 3.983937627402665e-06, + "loss": 0.9778, + "step": 921 + }, + { + "epoch": 0.06929204869983466, + "grad_norm": 1.8327792659780588, + "learning_rate": 3.983875991512082e-06, + "loss": 1.0061, + "step": 922 + }, + { + "epoch": 0.06936720276566963, + "grad_norm": 1.6802465883438036, + "learning_rate": 3.983814238069249e-06, + "loss": 1.0734, + "step": 923 + }, + { + "epoch": 0.06944235683150458, + "grad_norm": 1.5613386797554176, + "learning_rate": 3.983752367077826e-06, + "loss": 0.9938, + "step": 924 + }, + { + "epoch": 0.06951751089733954, + "grad_norm": 1.6364119431079653, + "learning_rate": 3.983690378541478e-06, + "loss": 0.9734, + "step": 925 + }, + { + "epoch": 0.06959266496317451, + "grad_norm": 1.4587192932924193, + "learning_rate": 3.9836282724638805e-06, + "loss": 1.018, + "step": 926 + }, + { + "epoch": 0.06966781902900947, + "grad_norm": 1.6792609094467272, + "learning_rate": 3.983566048848711e-06, + "loss": 1.1045, + "step": 927 + }, + { + "epoch": 0.06974297309484444, + "grad_norm": 1.6869151196791725, + "learning_rate": 3.983503707699658e-06, + "loss": 0.9128, + "step": 928 + }, + { + "epoch": 0.0698181271606794, + "grad_norm": 1.659368853369067, + "learning_rate": 3.983441249020414e-06, + "loss": 1.0923, + "step": 929 + }, + { + "epoch": 0.06989328122651435, + "grad_norm": 1.9181591801784672, + "learning_rate": 3.983378672814682e-06, + "loss": 1.0303, + "step": 930 + }, + { + "epoch": 0.06996843529234932, + "grad_norm": 1.5663726879148776, + "learning_rate": 3.983315979086169e-06, + "loss": 1.0961, + "step": 931 + }, + { + "epoch": 0.07004358935818428, + "grad_norm": 1.5158612568558578, + "learning_rate": 3.9832531678385885e-06, + "loss": 1.0392, + "step": 932 + }, + { + "epoch": 0.07011874342401923, + "grad_norm": 2.3567809945971097, + "learning_rate": 3.983190239075664e-06, + "loss": 1.1237, + "step": 933 + }, + { + "epoch": 0.0701938974898542, + "grad_norm": 2.054237113296628, + "learning_rate": 3.983127192801123e-06, + "loss": 0.9804, + "step": 934 + }, + { + "epoch": 0.07026905155568916, + "grad_norm": 1.729562009482217, + "learning_rate": 3.983064029018703e-06, + "loss": 1.0981, + "step": 935 + }, + { + "epoch": 0.07034420562152413, + "grad_norm": 1.3187922014471571, + "learning_rate": 3.983000747732145e-06, + "loss": 0.9889, + "step": 936 + }, + { + "epoch": 0.07041935968735909, + "grad_norm": 2.416023223544286, + "learning_rate": 3.9829373489452e-06, + "loss": 1.14, + "step": 937 + }, + { + "epoch": 0.07049451375319404, + "grad_norm": 1.7092561167883298, + "learning_rate": 3.982873832661623e-06, + "loss": 1.0596, + "step": 938 + }, + { + "epoch": 0.07056966781902901, + "grad_norm": 1.6556874885445323, + "learning_rate": 3.982810198885179e-06, + "loss": 1.1045, + "step": 939 + }, + { + "epoch": 0.07064482188486397, + "grad_norm": 1.989595589591684, + "learning_rate": 3.982746447619638e-06, + "loss": 1.1131, + "step": 940 + }, + { + "epoch": 0.07071997595069893, + "grad_norm": 1.7864131050575582, + "learning_rate": 3.982682578868777e-06, + "loss": 1.0566, + "step": 941 + }, + { + "epoch": 0.0707951300165339, + "grad_norm": 2.1740590921353062, + "learning_rate": 3.982618592636381e-06, + "loss": 1.1013, + "step": 942 + }, + { + "epoch": 0.07087028408236885, + "grad_norm": 1.8556848018745211, + "learning_rate": 3.982554488926242e-06, + "loss": 1.1172, + "step": 943 + }, + { + "epoch": 0.07094543814820382, + "grad_norm": 1.472099300067228, + "learning_rate": 3.982490267742158e-06, + "loss": 1.0276, + "step": 944 + }, + { + "epoch": 0.07102059221403878, + "grad_norm": 1.3550307915767876, + "learning_rate": 3.9824259290879336e-06, + "loss": 1.0667, + "step": 945 + }, + { + "epoch": 0.07109574627987374, + "grad_norm": 2.045159822481707, + "learning_rate": 3.982361472967382e-06, + "loss": 0.9866, + "step": 946 + }, + { + "epoch": 0.0711709003457087, + "grad_norm": 1.817984799526108, + "learning_rate": 3.982296899384322e-06, + "loss": 1.0301, + "step": 947 + }, + { + "epoch": 0.07124605441154366, + "grad_norm": 1.605055707160845, + "learning_rate": 3.9822322083425805e-06, + "loss": 1.0498, + "step": 948 + }, + { + "epoch": 0.07132120847737862, + "grad_norm": 2.382541128245891, + "learning_rate": 3.982167399845989e-06, + "loss": 1.0041, + "step": 949 + }, + { + "epoch": 0.07139636254321359, + "grad_norm": 1.3532731360560122, + "learning_rate": 3.982102473898391e-06, + "loss": 1.0138, + "step": 950 + }, + { + "epoch": 0.07147151660904855, + "grad_norm": 1.8492642718160595, + "learning_rate": 3.9820374305036295e-06, + "loss": 1.0525, + "step": 951 + }, + { + "epoch": 0.07154667067488352, + "grad_norm": 1.8451641284749556, + "learning_rate": 3.981972269665561e-06, + "loss": 0.9739, + "step": 952 + }, + { + "epoch": 0.07162182474071847, + "grad_norm": 0.7673573261319154, + "learning_rate": 3.981906991388046e-06, + "loss": 0.7838, + "step": 953 + }, + { + "epoch": 0.07169697880655343, + "grad_norm": 1.9032266123036325, + "learning_rate": 3.981841595674952e-06, + "loss": 1.0975, + "step": 954 + }, + { + "epoch": 0.0717721328723884, + "grad_norm": 1.597421195126662, + "learning_rate": 3.981776082530156e-06, + "loss": 1.044, + "step": 955 + }, + { + "epoch": 0.07184728693822336, + "grad_norm": 1.8504647006378925, + "learning_rate": 3.981710451957537e-06, + "loss": 1.0519, + "step": 956 + }, + { + "epoch": 0.07192244100405831, + "grad_norm": 1.5819827978932641, + "learning_rate": 3.981644703960986e-06, + "loss": 1.0915, + "step": 957 + }, + { + "epoch": 0.07199759506989328, + "grad_norm": 1.519196461175952, + "learning_rate": 3.981578838544398e-06, + "loss": 1.1032, + "step": 958 + }, + { + "epoch": 0.07207274913572824, + "grad_norm": 1.9225076393997542, + "learning_rate": 3.981512855711675e-06, + "loss": 0.9816, + "step": 959 + }, + { + "epoch": 0.07214790320156321, + "grad_norm": 2.279636805534804, + "learning_rate": 3.981446755466729e-06, + "loss": 1.0361, + "step": 960 + }, + { + "epoch": 0.07222305726739817, + "grad_norm": 1.8543624631678426, + "learning_rate": 3.981380537813474e-06, + "loss": 1.0843, + "step": 961 + }, + { + "epoch": 0.07229821133323312, + "grad_norm": 2.0197836388016417, + "learning_rate": 3.981314202755835e-06, + "loss": 0.9798, + "step": 962 + }, + { + "epoch": 0.0723733653990681, + "grad_norm": 1.4739999484420623, + "learning_rate": 3.981247750297744e-06, + "loss": 1.0328, + "step": 963 + }, + { + "epoch": 0.07244851946490305, + "grad_norm": 0.763760027468421, + "learning_rate": 3.9811811804431355e-06, + "loss": 0.8613, + "step": 964 + }, + { + "epoch": 0.07252367353073802, + "grad_norm": 2.0575178834826273, + "learning_rate": 3.981114493195956e-06, + "loss": 0.9866, + "step": 965 + }, + { + "epoch": 0.07259882759657298, + "grad_norm": 1.854718701197378, + "learning_rate": 3.981047688560156e-06, + "loss": 1.0038, + "step": 966 + }, + { + "epoch": 0.07267398166240793, + "grad_norm": 2.0079114125152615, + "learning_rate": 3.980980766539696e-06, + "loss": 1.0391, + "step": 967 + }, + { + "epoch": 0.0727491357282429, + "grad_norm": 1.7574902048285672, + "learning_rate": 3.980913727138539e-06, + "loss": 1.0157, + "step": 968 + }, + { + "epoch": 0.07282428979407786, + "grad_norm": 2.2045080294377404, + "learning_rate": 3.980846570360658e-06, + "loss": 0.9507, + "step": 969 + }, + { + "epoch": 0.07289944385991282, + "grad_norm": 1.520508361306701, + "learning_rate": 3.980779296210033e-06, + "loss": 1.0535, + "step": 970 + }, + { + "epoch": 0.07297459792574779, + "grad_norm": 1.6088805689137016, + "learning_rate": 3.98071190469065e-06, + "loss": 1.0292, + "step": 971 + }, + { + "epoch": 0.07304975199158274, + "grad_norm": 1.7794084901274212, + "learning_rate": 3.980644395806502e-06, + "loss": 0.9927, + "step": 972 + }, + { + "epoch": 0.07312490605741771, + "grad_norm": 1.8097676696041225, + "learning_rate": 3.980576769561588e-06, + "loss": 0.9589, + "step": 973 + }, + { + "epoch": 0.07320006012325267, + "grad_norm": 5.1817125875492, + "learning_rate": 3.980509025959918e-06, + "loss": 1.0144, + "step": 974 + }, + { + "epoch": 0.07327521418908763, + "grad_norm": 2.196907957000749, + "learning_rate": 3.980441165005503e-06, + "loss": 1.0747, + "step": 975 + }, + { + "epoch": 0.0733503682549226, + "grad_norm": 1.7999657627664332, + "learning_rate": 3.9803731867023665e-06, + "loss": 1.1237, + "step": 976 + }, + { + "epoch": 0.07342552232075755, + "grad_norm": 2.279661592332384, + "learning_rate": 3.980305091054534e-06, + "loss": 1.0221, + "step": 977 + }, + { + "epoch": 0.07350067638659251, + "grad_norm": 1.6848863211298901, + "learning_rate": 3.980236878066042e-06, + "loss": 1.0547, + "step": 978 + }, + { + "epoch": 0.07357583045242748, + "grad_norm": 1.922357948224124, + "learning_rate": 3.9801685477409336e-06, + "loss": 1.0245, + "step": 979 + }, + { + "epoch": 0.07365098451826244, + "grad_norm": 1.885075898400569, + "learning_rate": 3.980100100083254e-06, + "loss": 0.9309, + "step": 980 + }, + { + "epoch": 0.0737261385840974, + "grad_norm": 1.6674087620700213, + "learning_rate": 3.980031535097063e-06, + "loss": 1.0914, + "step": 981 + }, + { + "epoch": 0.07380129264993236, + "grad_norm": 0.7275615509449085, + "learning_rate": 3.9799628527864205e-06, + "loss": 0.8906, + "step": 982 + }, + { + "epoch": 0.07387644671576732, + "grad_norm": 2.3506663696575933, + "learning_rate": 3.979894053155398e-06, + "loss": 1.0527, + "step": 983 + }, + { + "epoch": 0.07395160078160229, + "grad_norm": 1.5412617572667018, + "learning_rate": 3.979825136208071e-06, + "loss": 0.9946, + "step": 984 + }, + { + "epoch": 0.07402675484743725, + "grad_norm": 1.7477570192209453, + "learning_rate": 3.979756101948523e-06, + "loss": 1.0847, + "step": 985 + }, + { + "epoch": 0.0741019089132722, + "grad_norm": 2.1300732135278855, + "learning_rate": 3.979686950380845e-06, + "loss": 1.0038, + "step": 986 + }, + { + "epoch": 0.07417706297910717, + "grad_norm": 1.970155849778949, + "learning_rate": 3.979617681509135e-06, + "loss": 0.9926, + "step": 987 + }, + { + "epoch": 0.07425221704494213, + "grad_norm": 0.7459951205746, + "learning_rate": 3.979548295337496e-06, + "loss": 0.8231, + "step": 988 + }, + { + "epoch": 0.0743273711107771, + "grad_norm": 1.7149260551148318, + "learning_rate": 3.979478791870041e-06, + "loss": 1.078, + "step": 989 + }, + { + "epoch": 0.07440252517661206, + "grad_norm": 1.5720684046288818, + "learning_rate": 3.9794091711108875e-06, + "loss": 0.975, + "step": 990 + }, + { + "epoch": 0.07447767924244701, + "grad_norm": 1.5505511596023884, + "learning_rate": 3.9793394330641614e-06, + "loss": 1.0118, + "step": 991 + }, + { + "epoch": 0.07455283330828198, + "grad_norm": 0.7548501228469372, + "learning_rate": 3.979269577733994e-06, + "loss": 0.8497, + "step": 992 + }, + { + "epoch": 0.07462798737411694, + "grad_norm": 0.8299283599384694, + "learning_rate": 3.979199605124525e-06, + "loss": 0.9418, + "step": 993 + }, + { + "epoch": 0.0747031414399519, + "grad_norm": 2.4048811747169663, + "learning_rate": 3.979129515239901e-06, + "loss": 0.9307, + "step": 994 + }, + { + "epoch": 0.07477829550578687, + "grad_norm": 1.9937360541833542, + "learning_rate": 3.979059308084274e-06, + "loss": 1.0706, + "step": 995 + }, + { + "epoch": 0.07485344957162182, + "grad_norm": 1.897181154082456, + "learning_rate": 3.9789889836618045e-06, + "loss": 1.0099, + "step": 996 + }, + { + "epoch": 0.07492860363745679, + "grad_norm": 1.679118712176691, + "learning_rate": 3.97891854197666e-06, + "loss": 1.0633, + "step": 997 + }, + { + "epoch": 0.07500375770329175, + "grad_norm": 1.5293812149975405, + "learning_rate": 3.978847983033014e-06, + "loss": 0.9728, + "step": 998 + }, + { + "epoch": 0.0750789117691267, + "grad_norm": 0.6700418437161032, + "learning_rate": 3.978777306835048e-06, + "loss": 0.8583, + "step": 999 + }, + { + "epoch": 0.07515406583496168, + "grad_norm": 1.6627213352263641, + "learning_rate": 3.978706513386949e-06, + "loss": 1.0019, + "step": 1000 + }, + { + "epoch": 0.07522921990079663, + "grad_norm": 3.348530996693241, + "learning_rate": 3.978635602692912e-06, + "loss": 1.0741, + "step": 1001 + }, + { + "epoch": 0.07530437396663159, + "grad_norm": 2.3073987627397585, + "learning_rate": 3.978564574757139e-06, + "loss": 1.0757, + "step": 1002 + }, + { + "epoch": 0.07537952803246656, + "grad_norm": 1.5832541931183557, + "learning_rate": 3.9784934295838385e-06, + "loss": 0.9716, + "step": 1003 + }, + { + "epoch": 0.07545468209830151, + "grad_norm": 1.9127589555340496, + "learning_rate": 3.978422167177226e-06, + "loss": 1.0264, + "step": 1004 + }, + { + "epoch": 0.07552983616413649, + "grad_norm": 1.7812736911202056, + "learning_rate": 3.9783507875415245e-06, + "loss": 1.1806, + "step": 1005 + }, + { + "epoch": 0.07560499022997144, + "grad_norm": 1.3724949732759628, + "learning_rate": 3.9782792906809625e-06, + "loss": 1.0495, + "step": 1006 + }, + { + "epoch": 0.0756801442958064, + "grad_norm": 2.100799081541559, + "learning_rate": 3.978207676599778e-06, + "loss": 1.068, + "step": 1007 + }, + { + "epoch": 0.07575529836164137, + "grad_norm": 1.6172764702985156, + "learning_rate": 3.978135945302213e-06, + "loss": 1.0228, + "step": 1008 + }, + { + "epoch": 0.07583045242747632, + "grad_norm": 1.8155807752360842, + "learning_rate": 3.978064096792519e-06, + "loss": 0.9698, + "step": 1009 + }, + { + "epoch": 0.0759056064933113, + "grad_norm": 1.507178929490619, + "learning_rate": 3.977992131074953e-06, + "loss": 1.0375, + "step": 1010 + }, + { + "epoch": 0.07598076055914625, + "grad_norm": 1.449459498065604, + "learning_rate": 3.9779200481537775e-06, + "loss": 1.0422, + "step": 1011 + }, + { + "epoch": 0.07605591462498121, + "grad_norm": 1.7792444148113347, + "learning_rate": 3.977847848033267e-06, + "loss": 1.0382, + "step": 1012 + }, + { + "epoch": 0.07613106869081618, + "grad_norm": 1.569360245027991, + "learning_rate": 3.977775530717696e-06, + "loss": 1.0282, + "step": 1013 + }, + { + "epoch": 0.07620622275665113, + "grad_norm": 3.1106399186880105, + "learning_rate": 3.977703096211354e-06, + "loss": 1.006, + "step": 1014 + }, + { + "epoch": 0.07628137682248609, + "grad_norm": 1.7513792188785677, + "learning_rate": 3.977630544518529e-06, + "loss": 1.0075, + "step": 1015 + }, + { + "epoch": 0.07635653088832106, + "grad_norm": 2.0543919339829744, + "learning_rate": 3.97755787564352e-06, + "loss": 1.1329, + "step": 1016 + }, + { + "epoch": 0.07643168495415602, + "grad_norm": 1.8762955322433634, + "learning_rate": 3.977485089590636e-06, + "loss": 1.012, + "step": 1017 + }, + { + "epoch": 0.07650683901999099, + "grad_norm": 1.6406056720866715, + "learning_rate": 3.977412186364187e-06, + "loss": 1.0624, + "step": 1018 + }, + { + "epoch": 0.07658199308582594, + "grad_norm": 2.0969649726223443, + "learning_rate": 3.977339165968495e-06, + "loss": 1.1206, + "step": 1019 + }, + { + "epoch": 0.0766571471516609, + "grad_norm": 1.6369055593579485, + "learning_rate": 3.977266028407885e-06, + "loss": 1.018, + "step": 1020 + }, + { + "epoch": 0.07673230121749587, + "grad_norm": 1.8288940377135254, + "learning_rate": 3.977192773686692e-06, + "loss": 1.0408, + "step": 1021 + }, + { + "epoch": 0.07680745528333083, + "grad_norm": 1.2349951429696269, + "learning_rate": 3.977119401809255e-06, + "loss": 0.9921, + "step": 1022 + }, + { + "epoch": 0.07688260934916578, + "grad_norm": 1.3720664027691183, + "learning_rate": 3.977045912779924e-06, + "loss": 1.0215, + "step": 1023 + }, + { + "epoch": 0.07695776341500075, + "grad_norm": 1.2636026779166958, + "learning_rate": 3.9769723066030505e-06, + "loss": 0.9794, + "step": 1024 + }, + { + "epoch": 0.07703291748083571, + "grad_norm": 2.0203177066272695, + "learning_rate": 3.976898583282998e-06, + "loss": 0.9069, + "step": 1025 + }, + { + "epoch": 0.07710807154667068, + "grad_norm": 1.511539576752737, + "learning_rate": 3.976824742824135e-06, + "loss": 1.0089, + "step": 1026 + }, + { + "epoch": 0.07718322561250564, + "grad_norm": 1.7740717354541309, + "learning_rate": 3.976750785230835e-06, + "loss": 1.0395, + "step": 1027 + }, + { + "epoch": 0.0772583796783406, + "grad_norm": 1.2255194049421039, + "learning_rate": 3.976676710507483e-06, + "loss": 1.0119, + "step": 1028 + }, + { + "epoch": 0.07733353374417556, + "grad_norm": 2.6461421956486553, + "learning_rate": 3.976602518658466e-06, + "loss": 1.101, + "step": 1029 + }, + { + "epoch": 0.07740868781001052, + "grad_norm": 1.9209305838613524, + "learning_rate": 3.976528209688181e-06, + "loss": 1.017, + "step": 1030 + }, + { + "epoch": 0.07748384187584548, + "grad_norm": 1.469541782663859, + "learning_rate": 3.976453783601031e-06, + "loss": 1.0274, + "step": 1031 + }, + { + "epoch": 0.07755899594168045, + "grad_norm": 1.8271569191161372, + "learning_rate": 3.976379240401426e-06, + "loss": 1.0198, + "step": 1032 + }, + { + "epoch": 0.0776341500075154, + "grad_norm": 1.671145484983091, + "learning_rate": 3.976304580093782e-06, + "loss": 0.9454, + "step": 1033 + }, + { + "epoch": 0.07770930407335037, + "grad_norm": 1.9152748472421695, + "learning_rate": 3.976229802682524e-06, + "loss": 1.081, + "step": 1034 + }, + { + "epoch": 0.07778445813918533, + "grad_norm": 1.7909136527847893, + "learning_rate": 3.9761549081720845e-06, + "loss": 0.9703, + "step": 1035 + }, + { + "epoch": 0.07785961220502029, + "grad_norm": 1.6726617120816645, + "learning_rate": 3.976079896566898e-06, + "loss": 0.9495, + "step": 1036 + }, + { + "epoch": 0.07793476627085526, + "grad_norm": 1.8234181250267716, + "learning_rate": 3.976004767871411e-06, + "loss": 1.0115, + "step": 1037 + }, + { + "epoch": 0.07800992033669021, + "grad_norm": 1.6501560021837234, + "learning_rate": 3.975929522090075e-06, + "loss": 0.9789, + "step": 1038 + }, + { + "epoch": 0.07808507440252517, + "grad_norm": 2.1938141294490543, + "learning_rate": 3.9758541592273485e-06, + "loss": 1.0611, + "step": 1039 + }, + { + "epoch": 0.07816022846836014, + "grad_norm": 1.9001378109795382, + "learning_rate": 3.975778679287697e-06, + "loss": 1.0416, + "step": 1040 + }, + { + "epoch": 0.0782353825341951, + "grad_norm": 0.7184772388540309, + "learning_rate": 3.975703082275592e-06, + "loss": 0.8472, + "step": 1041 + }, + { + "epoch": 0.07831053660003007, + "grad_norm": 2.2757366915949295, + "learning_rate": 3.975627368195515e-06, + "loss": 0.9483, + "step": 1042 + }, + { + "epoch": 0.07838569066586502, + "grad_norm": 2.3467144338485677, + "learning_rate": 3.9755515370519515e-06, + "loss": 1.0148, + "step": 1043 + }, + { + "epoch": 0.07846084473169998, + "grad_norm": 1.5941898262640168, + "learning_rate": 3.975475588849394e-06, + "loss": 1.0791, + "step": 1044 + }, + { + "epoch": 0.07853599879753495, + "grad_norm": 3.4302149477857355, + "learning_rate": 3.975399523592343e-06, + "loss": 1.071, + "step": 1045 + }, + { + "epoch": 0.0786111528633699, + "grad_norm": 1.4973979474481227, + "learning_rate": 3.975323341285306e-06, + "loss": 1.0555, + "step": 1046 + }, + { + "epoch": 0.07868630692920486, + "grad_norm": 1.6836374370102276, + "learning_rate": 3.975247041932797e-06, + "loss": 1.0343, + "step": 1047 + }, + { + "epoch": 0.07876146099503983, + "grad_norm": 1.801729891594299, + "learning_rate": 3.975170625539338e-06, + "loss": 0.9868, + "step": 1048 + }, + { + "epoch": 0.07883661506087479, + "grad_norm": 1.6146123904002876, + "learning_rate": 3.975094092109455e-06, + "loss": 1.0807, + "step": 1049 + }, + { + "epoch": 0.07891176912670976, + "grad_norm": 1.571849317701087, + "learning_rate": 3.975017441647684e-06, + "loss": 0.9906, + "step": 1050 + }, + { + "epoch": 0.07898692319254472, + "grad_norm": 1.5487653183198766, + "learning_rate": 3.974940674158567e-06, + "loss": 0.9733, + "step": 1051 + }, + { + "epoch": 0.07906207725837967, + "grad_norm": 0.5942310091449803, + "learning_rate": 3.9748637896466526e-06, + "loss": 0.8044, + "step": 1052 + }, + { + "epoch": 0.07913723132421464, + "grad_norm": 1.4732583026998352, + "learning_rate": 3.974786788116496e-06, + "loss": 0.9857, + "step": 1053 + }, + { + "epoch": 0.0792123853900496, + "grad_norm": 1.6117232144654978, + "learning_rate": 3.974709669572661e-06, + "loss": 1.0127, + "step": 1054 + }, + { + "epoch": 0.07928753945588456, + "grad_norm": 1.6516228584512984, + "learning_rate": 3.974632434019716e-06, + "loss": 0.9545, + "step": 1055 + }, + { + "epoch": 0.07936269352171953, + "grad_norm": 1.6096102348184254, + "learning_rate": 3.974555081462237e-06, + "loss": 1.0068, + "step": 1056 + }, + { + "epoch": 0.07943784758755448, + "grad_norm": 18.554731907841596, + "learning_rate": 3.97447761190481e-06, + "loss": 0.9838, + "step": 1057 + }, + { + "epoch": 0.07951300165338945, + "grad_norm": 1.8590664006440933, + "learning_rate": 3.974400025352022e-06, + "loss": 1.1348, + "step": 1058 + }, + { + "epoch": 0.07958815571922441, + "grad_norm": 1.597271562818544, + "learning_rate": 3.974322321808473e-06, + "loss": 1.066, + "step": 1059 + }, + { + "epoch": 0.07966330978505937, + "grad_norm": 2.2598857930114153, + "learning_rate": 3.974244501278766e-06, + "loss": 1.0449, + "step": 1060 + }, + { + "epoch": 0.07973846385089434, + "grad_norm": 1.6879185757908335, + "learning_rate": 3.974166563767513e-06, + "loss": 0.9615, + "step": 1061 + }, + { + "epoch": 0.0798136179167293, + "grad_norm": 1.8247779789525336, + "learning_rate": 3.974088509279331e-06, + "loss": 1.0799, + "step": 1062 + }, + { + "epoch": 0.07988877198256426, + "grad_norm": 1.7136107163203704, + "learning_rate": 3.9740103378188455e-06, + "loss": 0.9645, + "step": 1063 + }, + { + "epoch": 0.07996392604839922, + "grad_norm": 2.256633516722728, + "learning_rate": 3.973932049390688e-06, + "loss": 1.2079, + "step": 1064 + }, + { + "epoch": 0.08003908011423418, + "grad_norm": 1.5470015348834936, + "learning_rate": 3.973853643999499e-06, + "loss": 1.0076, + "step": 1065 + }, + { + "epoch": 0.08011423418006915, + "grad_norm": 1.994789249694263, + "learning_rate": 3.973775121649922e-06, + "loss": 1.0755, + "step": 1066 + }, + { + "epoch": 0.0801893882459041, + "grad_norm": 1.463539214036002, + "learning_rate": 3.973696482346611e-06, + "loss": 0.9488, + "step": 1067 + }, + { + "epoch": 0.08026454231173906, + "grad_norm": 1.7461540331237704, + "learning_rate": 3.973617726094227e-06, + "loss": 1.0696, + "step": 1068 + }, + { + "epoch": 0.08033969637757403, + "grad_norm": 1.4487480777072743, + "learning_rate": 3.973538852897435e-06, + "loss": 1.0702, + "step": 1069 + }, + { + "epoch": 0.08041485044340899, + "grad_norm": 2.017714371193783, + "learning_rate": 3.973459862760908e-06, + "loss": 1.1417, + "step": 1070 + }, + { + "epoch": 0.08049000450924396, + "grad_norm": 2.459041385732083, + "learning_rate": 3.973380755689328e-06, + "loss": 1.0172, + "step": 1071 + }, + { + "epoch": 0.08056515857507891, + "grad_norm": 1.6907772453986183, + "learning_rate": 3.97330153168738e-06, + "loss": 1.0778, + "step": 1072 + }, + { + "epoch": 0.08064031264091387, + "grad_norm": 2.1476382346731557, + "learning_rate": 3.973222190759761e-06, + "loss": 1.0879, + "step": 1073 + }, + { + "epoch": 0.08071546670674884, + "grad_norm": 5.795463287292524, + "learning_rate": 3.973142732911172e-06, + "loss": 1.0887, + "step": 1074 + }, + { + "epoch": 0.0807906207725838, + "grad_norm": 1.6519075002659243, + "learning_rate": 3.9730631581463195e-06, + "loss": 1.0392, + "step": 1075 + }, + { + "epoch": 0.08086577483841875, + "grad_norm": 3.1665228426211924, + "learning_rate": 3.972983466469919e-06, + "loss": 1.0789, + "step": 1076 + }, + { + "epoch": 0.08094092890425372, + "grad_norm": 3.2081194389350056, + "learning_rate": 3.972903657886695e-06, + "loss": 1.041, + "step": 1077 + }, + { + "epoch": 0.08101608297008868, + "grad_norm": 1.9545148253309195, + "learning_rate": 3.972823732401373e-06, + "loss": 1.0824, + "step": 1078 + }, + { + "epoch": 0.08109123703592365, + "grad_norm": 1.6766474692507722, + "learning_rate": 3.972743690018691e-06, + "loss": 1.0356, + "step": 1079 + }, + { + "epoch": 0.0811663911017586, + "grad_norm": 3.1915740362392326, + "learning_rate": 3.9726635307433906e-06, + "loss": 1.1499, + "step": 1080 + }, + { + "epoch": 0.08124154516759356, + "grad_norm": 1.5992451347407068, + "learning_rate": 3.972583254580223e-06, + "loss": 0.986, + "step": 1081 + }, + { + "epoch": 0.08131669923342853, + "grad_norm": 1.3888430805069498, + "learning_rate": 3.972502861533943e-06, + "loss": 1.0568, + "step": 1082 + }, + { + "epoch": 0.08139185329926349, + "grad_norm": 1.9595269224902192, + "learning_rate": 3.9724223516093154e-06, + "loss": 1.0742, + "step": 1083 + }, + { + "epoch": 0.08146700736509845, + "grad_norm": 0.801687035358021, + "learning_rate": 3.972341724811111e-06, + "loss": 0.8536, + "step": 1084 + }, + { + "epoch": 0.08154216143093342, + "grad_norm": 1.5300249142581739, + "learning_rate": 3.972260981144107e-06, + "loss": 1.0866, + "step": 1085 + }, + { + "epoch": 0.08161731549676837, + "grad_norm": 1.9464383831251835, + "learning_rate": 3.972180120613087e-06, + "loss": 1.1012, + "step": 1086 + }, + { + "epoch": 0.08169246956260334, + "grad_norm": 2.4763108375312473, + "learning_rate": 3.972099143222844e-06, + "loss": 1.0831, + "step": 1087 + }, + { + "epoch": 0.0817676236284383, + "grad_norm": 1.468662502847992, + "learning_rate": 3.972018048978174e-06, + "loss": 1.0747, + "step": 1088 + }, + { + "epoch": 0.08184277769427326, + "grad_norm": 1.8235890774131376, + "learning_rate": 3.971936837883884e-06, + "loss": 1.0737, + "step": 1089 + }, + { + "epoch": 0.08191793176010823, + "grad_norm": 1.773836536299322, + "learning_rate": 3.971855509944784e-06, + "loss": 1.0517, + "step": 1090 + }, + { + "epoch": 0.08199308582594318, + "grad_norm": 2.669027763402386, + "learning_rate": 3.971774065165696e-06, + "loss": 0.9564, + "step": 1091 + }, + { + "epoch": 0.08206823989177814, + "grad_norm": 1.3225926230144835, + "learning_rate": 3.971692503551443e-06, + "loss": 1.112, + "step": 1092 + }, + { + "epoch": 0.08214339395761311, + "grad_norm": 2.1108079352383844, + "learning_rate": 3.971610825106859e-06, + "loss": 0.9964, + "step": 1093 + }, + { + "epoch": 0.08221854802344807, + "grad_norm": 1.7058361068476928, + "learning_rate": 3.971529029836785e-06, + "loss": 0.8877, + "step": 1094 + }, + { + "epoch": 0.08229370208928304, + "grad_norm": 1.7191020951723892, + "learning_rate": 3.971447117746065e-06, + "loss": 0.8752, + "step": 1095 + }, + { + "epoch": 0.08236885615511799, + "grad_norm": 1.8697642020614322, + "learning_rate": 3.9713650888395555e-06, + "loss": 0.9825, + "step": 1096 + }, + { + "epoch": 0.08244401022095295, + "grad_norm": 1.4184824115629535, + "learning_rate": 3.971282943122115e-06, + "loss": 1.0117, + "step": 1097 + }, + { + "epoch": 0.08251916428678792, + "grad_norm": 1.8767252931646732, + "learning_rate": 3.971200680598611e-06, + "loss": 0.9685, + "step": 1098 + }, + { + "epoch": 0.08259431835262288, + "grad_norm": 1.4873482089772807, + "learning_rate": 3.971118301273919e-06, + "loss": 0.9709, + "step": 1099 + }, + { + "epoch": 0.08266947241845783, + "grad_norm": 1.5396891446159968, + "learning_rate": 3.97103580515292e-06, + "loss": 0.9991, + "step": 1100 + }, + { + "epoch": 0.0827446264842928, + "grad_norm": 1.6027546237277293, + "learning_rate": 3.970953192240502e-06, + "loss": 1.0855, + "step": 1101 + }, + { + "epoch": 0.08281978055012776, + "grad_norm": 1.5940209700666064, + "learning_rate": 3.970870462541559e-06, + "loss": 1.0382, + "step": 1102 + }, + { + "epoch": 0.08289493461596273, + "grad_norm": 1.8563148898204387, + "learning_rate": 3.970787616060995e-06, + "loss": 1.0176, + "step": 1103 + }, + { + "epoch": 0.08297008868179769, + "grad_norm": 2.1325509047162403, + "learning_rate": 3.970704652803718e-06, + "loss": 1.0039, + "step": 1104 + }, + { + "epoch": 0.08304524274763264, + "grad_norm": 1.392996234376614, + "learning_rate": 3.970621572774644e-06, + "loss": 1.024, + "step": 1105 + }, + { + "epoch": 0.08312039681346761, + "grad_norm": 1.8398768590065977, + "learning_rate": 3.970538375978695e-06, + "loss": 0.9457, + "step": 1106 + }, + { + "epoch": 0.08319555087930257, + "grad_norm": 1.5398506431483365, + "learning_rate": 3.970455062420803e-06, + "loss": 1.0042, + "step": 1107 + }, + { + "epoch": 0.08327070494513754, + "grad_norm": 1.7850794535672903, + "learning_rate": 3.9703716321059026e-06, + "loss": 1.0951, + "step": 1108 + }, + { + "epoch": 0.0833458590109725, + "grad_norm": 0.8478829217017961, + "learning_rate": 3.970288085038938e-06, + "loss": 0.8749, + "step": 1109 + }, + { + "epoch": 0.08342101307680745, + "grad_norm": 1.4187845032651443, + "learning_rate": 3.970204421224859e-06, + "loss": 1.0104, + "step": 1110 + }, + { + "epoch": 0.08349616714264242, + "grad_norm": 1.3834789846717739, + "learning_rate": 3.970120640668623e-06, + "loss": 1.1269, + "step": 1111 + }, + { + "epoch": 0.08357132120847738, + "grad_norm": 1.6222155997969971, + "learning_rate": 3.970036743375197e-06, + "loss": 1.0278, + "step": 1112 + }, + { + "epoch": 0.08364647527431233, + "grad_norm": 1.594932790434383, + "learning_rate": 3.9699527293495485e-06, + "loss": 0.9464, + "step": 1113 + }, + { + "epoch": 0.0837216293401473, + "grad_norm": 1.7562355936891758, + "learning_rate": 3.969868598596658e-06, + "loss": 1.0668, + "step": 1114 + }, + { + "epoch": 0.08379678340598226, + "grad_norm": 1.7130333591960865, + "learning_rate": 3.9697843511215104e-06, + "loss": 1.0299, + "step": 1115 + }, + { + "epoch": 0.08387193747181723, + "grad_norm": 2.3740904220740706, + "learning_rate": 3.969699986929096e-06, + "loss": 0.9982, + "step": 1116 + }, + { + "epoch": 0.08394709153765219, + "grad_norm": 3.78552269515063, + "learning_rate": 3.969615506024416e-06, + "loss": 0.9832, + "step": 1117 + }, + { + "epoch": 0.08402224560348714, + "grad_norm": 1.8103152953089596, + "learning_rate": 3.969530908412474e-06, + "loss": 1.0456, + "step": 1118 + }, + { + "epoch": 0.08409739966932211, + "grad_norm": 3.772442862328883, + "learning_rate": 3.969446194098286e-06, + "loss": 1.0963, + "step": 1119 + }, + { + "epoch": 0.08417255373515707, + "grad_norm": 1.7058553945024546, + "learning_rate": 3.969361363086867e-06, + "loss": 1.083, + "step": 1120 + }, + { + "epoch": 0.08424770780099203, + "grad_norm": 3.0439059928383636, + "learning_rate": 3.969276415383248e-06, + "loss": 1.0498, + "step": 1121 + }, + { + "epoch": 0.084322861866827, + "grad_norm": 1.7195021630773002, + "learning_rate": 3.9691913509924586e-06, + "loss": 1.0895, + "step": 1122 + }, + { + "epoch": 0.08439801593266195, + "grad_norm": 1.8948021134697932, + "learning_rate": 3.969106169919542e-06, + "loss": 1.1148, + "step": 1123 + }, + { + "epoch": 0.08447316999849692, + "grad_norm": 1.81691578927197, + "learning_rate": 3.969020872169545e-06, + "loss": 1.1269, + "step": 1124 + }, + { + "epoch": 0.08454832406433188, + "grad_norm": 1.93170844750672, + "learning_rate": 3.9689354577475206e-06, + "loss": 1.0233, + "step": 1125 + }, + { + "epoch": 0.08462347813016684, + "grad_norm": 11.723521814929043, + "learning_rate": 3.968849926658532e-06, + "loss": 0.9372, + "step": 1126 + }, + { + "epoch": 0.08469863219600181, + "grad_norm": 1.7036484780014567, + "learning_rate": 3.968764278907645e-06, + "loss": 0.9738, + "step": 1127 + }, + { + "epoch": 0.08477378626183676, + "grad_norm": 1.7733294080150632, + "learning_rate": 3.968678514499936e-06, + "loss": 1.0948, + "step": 1128 + }, + { + "epoch": 0.08484894032767172, + "grad_norm": 1.5782080146937008, + "learning_rate": 3.968592633440486e-06, + "loss": 0.9772, + "step": 1129 + }, + { + "epoch": 0.08492409439350669, + "grad_norm": 2.3717444342460157, + "learning_rate": 3.968506635734385e-06, + "loss": 1.0134, + "step": 1130 + }, + { + "epoch": 0.08499924845934165, + "grad_norm": 1.9329886057873595, + "learning_rate": 3.9684205213867275e-06, + "loss": 1.0134, + "step": 1131 + }, + { + "epoch": 0.08507440252517662, + "grad_norm": 1.6229447058368118, + "learning_rate": 3.968334290402616e-06, + "loss": 1.0198, + "step": 1132 + }, + { + "epoch": 0.08514955659101157, + "grad_norm": 1.8124527760634772, + "learning_rate": 3.968247942787161e-06, + "loss": 1.0748, + "step": 1133 + }, + { + "epoch": 0.08522471065684653, + "grad_norm": 1.8372826376307894, + "learning_rate": 3.968161478545479e-06, + "loss": 1.0596, + "step": 1134 + }, + { + "epoch": 0.0852998647226815, + "grad_norm": 2.5647254212407695, + "learning_rate": 3.968074897682692e-06, + "loss": 1.0244, + "step": 1135 + }, + { + "epoch": 0.08537501878851646, + "grad_norm": 1.484324886078395, + "learning_rate": 3.967988200203931e-06, + "loss": 1.0075, + "step": 1136 + }, + { + "epoch": 0.08545017285435141, + "grad_norm": 1.7277879465600512, + "learning_rate": 3.967901386114334e-06, + "loss": 0.952, + "step": 1137 + }, + { + "epoch": 0.08552532692018638, + "grad_norm": 3.309990706185838, + "learning_rate": 3.967814455419044e-06, + "loss": 1.0204, + "step": 1138 + }, + { + "epoch": 0.08560048098602134, + "grad_norm": 3.5101401713495677, + "learning_rate": 3.9677274081232116e-06, + "loss": 0.9648, + "step": 1139 + }, + { + "epoch": 0.08567563505185631, + "grad_norm": 1.727212705778535, + "learning_rate": 3.967640244231996e-06, + "loss": 1.1616, + "step": 1140 + }, + { + "epoch": 0.08575078911769127, + "grad_norm": 1.9921933641534404, + "learning_rate": 3.9675529637505615e-06, + "loss": 1.0702, + "step": 1141 + }, + { + "epoch": 0.08582594318352622, + "grad_norm": 2.133954423655475, + "learning_rate": 3.967465566684079e-06, + "loss": 1.0297, + "step": 1142 + }, + { + "epoch": 0.0859010972493612, + "grad_norm": 1.6787147721651228, + "learning_rate": 3.967378053037728e-06, + "loss": 1.0733, + "step": 1143 + }, + { + "epoch": 0.08597625131519615, + "grad_norm": 1.8849916628949184, + "learning_rate": 3.967290422816693e-06, + "loss": 1.0737, + "step": 1144 + }, + { + "epoch": 0.0860514053810311, + "grad_norm": 1.6704265701838985, + "learning_rate": 3.967202676026169e-06, + "loss": 1.0014, + "step": 1145 + }, + { + "epoch": 0.08612655944686608, + "grad_norm": 2.260034859987645, + "learning_rate": 3.967114812671353e-06, + "loss": 1.0712, + "step": 1146 + }, + { + "epoch": 0.08620171351270103, + "grad_norm": 1.8001042415316533, + "learning_rate": 3.967026832757451e-06, + "loss": 1.1263, + "step": 1147 + }, + { + "epoch": 0.086276867578536, + "grad_norm": 1.4051320188408292, + "learning_rate": 3.966938736289677e-06, + "loss": 1.0215, + "step": 1148 + }, + { + "epoch": 0.08635202164437096, + "grad_norm": 1.650982234940125, + "learning_rate": 3.966850523273251e-06, + "loss": 1.1252, + "step": 1149 + }, + { + "epoch": 0.08642717571020592, + "grad_norm": 1.7550313692468307, + "learning_rate": 3.9667621937134e-06, + "loss": 1.0713, + "step": 1150 + }, + { + "epoch": 0.08650232977604089, + "grad_norm": 1.932639701600322, + "learning_rate": 3.966673747615358e-06, + "loss": 1.0375, + "step": 1151 + }, + { + "epoch": 0.08657748384187584, + "grad_norm": 1.7780027404480832, + "learning_rate": 3.966585184984366e-06, + "loss": 1.0038, + "step": 1152 + }, + { + "epoch": 0.08665263790771081, + "grad_norm": 1.718657681846858, + "learning_rate": 3.96649650582567e-06, + "loss": 0.9863, + "step": 1153 + }, + { + "epoch": 0.08672779197354577, + "grad_norm": 2.4765507550367203, + "learning_rate": 3.966407710144527e-06, + "loss": 1.0176, + "step": 1154 + }, + { + "epoch": 0.08680294603938073, + "grad_norm": 0.8600349666450975, + "learning_rate": 3.966318797946196e-06, + "loss": 0.8402, + "step": 1155 + }, + { + "epoch": 0.0868781001052157, + "grad_norm": 1.5724932203768331, + "learning_rate": 3.966229769235948e-06, + "loss": 1.1191, + "step": 1156 + }, + { + "epoch": 0.08695325417105065, + "grad_norm": 2.5331112647184213, + "learning_rate": 3.966140624019056e-06, + "loss": 0.9953, + "step": 1157 + }, + { + "epoch": 0.08702840823688561, + "grad_norm": 1.8993541874801263, + "learning_rate": 3.966051362300804e-06, + "loss": 0.9843, + "step": 1158 + }, + { + "epoch": 0.08710356230272058, + "grad_norm": 1.9764927787697097, + "learning_rate": 3.9659619840864795e-06, + "loss": 0.9667, + "step": 1159 + }, + { + "epoch": 0.08717871636855554, + "grad_norm": 2.122228947418546, + "learning_rate": 3.96587248938138e-06, + "loss": 1.1028, + "step": 1160 + }, + { + "epoch": 0.0872538704343905, + "grad_norm": 0.6381892442635514, + "learning_rate": 3.965782878190807e-06, + "loss": 0.8145, + "step": 1161 + }, + { + "epoch": 0.08732902450022546, + "grad_norm": 1.7981694741793621, + "learning_rate": 3.965693150520071e-06, + "loss": 0.9536, + "step": 1162 + }, + { + "epoch": 0.08740417856606042, + "grad_norm": 3.1317493033016657, + "learning_rate": 3.965603306374489e-06, + "loss": 0.9736, + "step": 1163 + }, + { + "epoch": 0.08747933263189539, + "grad_norm": 1.2730060150621678, + "learning_rate": 3.9655133457593845e-06, + "loss": 1.0533, + "step": 1164 + }, + { + "epoch": 0.08755448669773035, + "grad_norm": 0.7467641963726775, + "learning_rate": 3.965423268680087e-06, + "loss": 0.8654, + "step": 1165 + }, + { + "epoch": 0.0876296407635653, + "grad_norm": 2.086458633989099, + "learning_rate": 3.965333075141936e-06, + "loss": 1.1056, + "step": 1166 + }, + { + "epoch": 0.08770479482940027, + "grad_norm": 1.851710575154277, + "learning_rate": 3.965242765150274e-06, + "loss": 1.0349, + "step": 1167 + }, + { + "epoch": 0.08777994889523523, + "grad_norm": 2.4076947915567604, + "learning_rate": 3.9651523387104526e-06, + "loss": 1.0006, + "step": 1168 + }, + { + "epoch": 0.0878551029610702, + "grad_norm": 1.8186263104619629, + "learning_rate": 3.96506179582783e-06, + "loss": 1.0145, + "step": 1169 + }, + { + "epoch": 0.08793025702690516, + "grad_norm": 2.0429899896571584, + "learning_rate": 3.964971136507771e-06, + "loss": 1.1343, + "step": 1170 + }, + { + "epoch": 0.08800541109274011, + "grad_norm": 1.9998661234871042, + "learning_rate": 3.964880360755648e-06, + "loss": 1.0655, + "step": 1171 + }, + { + "epoch": 0.08808056515857508, + "grad_norm": 2.195629599262636, + "learning_rate": 3.96478946857684e-06, + "loss": 1.0689, + "step": 1172 + }, + { + "epoch": 0.08815571922441004, + "grad_norm": 3.5358695812166965, + "learning_rate": 3.964698459976732e-06, + "loss": 1.0151, + "step": 1173 + }, + { + "epoch": 0.088230873290245, + "grad_norm": 2.918286246270804, + "learning_rate": 3.9646073349607165e-06, + "loss": 1.1846, + "step": 1174 + }, + { + "epoch": 0.08830602735607997, + "grad_norm": 3.7170832128274967, + "learning_rate": 3.964516093534194e-06, + "loss": 1.0566, + "step": 1175 + }, + { + "epoch": 0.08838118142191492, + "grad_norm": 1.7063247300994469, + "learning_rate": 3.964424735702571e-06, + "loss": 0.9289, + "step": 1176 + }, + { + "epoch": 0.0884563354877499, + "grad_norm": 1.4780485556838059, + "learning_rate": 3.964333261471258e-06, + "loss": 0.9951, + "step": 1177 + }, + { + "epoch": 0.08853148955358485, + "grad_norm": 1.816378226115306, + "learning_rate": 3.964241670845679e-06, + "loss": 0.9953, + "step": 1178 + }, + { + "epoch": 0.0886066436194198, + "grad_norm": 3.0679054045834215, + "learning_rate": 3.9641499638312595e-06, + "loss": 1.0601, + "step": 1179 + }, + { + "epoch": 0.08868179768525478, + "grad_norm": 1.5390983616092813, + "learning_rate": 3.964058140433434e-06, + "loss": 1.0144, + "step": 1180 + }, + { + "epoch": 0.08875695175108973, + "grad_norm": 2.0916274970086515, + "learning_rate": 3.9639662006576405e-06, + "loss": 1.0569, + "step": 1181 + }, + { + "epoch": 0.08883210581692469, + "grad_norm": 1.611769884993019, + "learning_rate": 3.963874144509331e-06, + "loss": 1.0674, + "step": 1182 + }, + { + "epoch": 0.08890725988275966, + "grad_norm": 0.84681477068586, + "learning_rate": 3.963781971993957e-06, + "loss": 0.8649, + "step": 1183 + }, + { + "epoch": 0.08898241394859462, + "grad_norm": 2.14962502691694, + "learning_rate": 3.963689683116982e-06, + "loss": 0.9595, + "step": 1184 + }, + { + "epoch": 0.08905756801442959, + "grad_norm": 1.6328958736470565, + "learning_rate": 3.963597277883874e-06, + "loss": 1.0544, + "step": 1185 + }, + { + "epoch": 0.08913272208026454, + "grad_norm": 1.7222765722551971, + "learning_rate": 3.963504756300107e-06, + "loss": 1.054, + "step": 1186 + }, + { + "epoch": 0.0892078761460995, + "grad_norm": 0.7034972582955472, + "learning_rate": 3.963412118371166e-06, + "loss": 0.8831, + "step": 1187 + }, + { + "epoch": 0.08928303021193447, + "grad_norm": 2.4710150842148324, + "learning_rate": 3.963319364102538e-06, + "loss": 1.0128, + "step": 1188 + }, + { + "epoch": 0.08935818427776943, + "grad_norm": 1.777082486914567, + "learning_rate": 3.96322649349972e-06, + "loss": 1.0458, + "step": 1189 + }, + { + "epoch": 0.08943333834360438, + "grad_norm": 1.9627260357379288, + "learning_rate": 3.963133506568214e-06, + "loss": 1.1253, + "step": 1190 + }, + { + "epoch": 0.08950849240943935, + "grad_norm": 2.9396547571521996, + "learning_rate": 3.96304040331353e-06, + "loss": 1.0092, + "step": 1191 + }, + { + "epoch": 0.08958364647527431, + "grad_norm": 3.3106615254852203, + "learning_rate": 3.9629471837411855e-06, + "loss": 1.1018, + "step": 1192 + }, + { + "epoch": 0.08965880054110928, + "grad_norm": 1.8388948038370376, + "learning_rate": 3.962853847856704e-06, + "loss": 1.1014, + "step": 1193 + }, + { + "epoch": 0.08973395460694424, + "grad_norm": 2.4889886590094648, + "learning_rate": 3.962760395665616e-06, + "loss": 1.0086, + "step": 1194 + }, + { + "epoch": 0.08980910867277919, + "grad_norm": 1.6807097407178746, + "learning_rate": 3.962666827173458e-06, + "loss": 1.0848, + "step": 1195 + }, + { + "epoch": 0.08988426273861416, + "grad_norm": 0.6977853627785631, + "learning_rate": 3.9625731423857745e-06, + "loss": 0.8452, + "step": 1196 + }, + { + "epoch": 0.08995941680444912, + "grad_norm": 6.931583940667073, + "learning_rate": 3.9624793413081185e-06, + "loss": 1.0568, + "step": 1197 + }, + { + "epoch": 0.09003457087028409, + "grad_norm": 1.994340040378088, + "learning_rate": 3.962385423946046e-06, + "loss": 1.0909, + "step": 1198 + }, + { + "epoch": 0.09010972493611905, + "grad_norm": 1.810891457104838, + "learning_rate": 3.962291390305123e-06, + "loss": 1.016, + "step": 1199 + }, + { + "epoch": 0.090184879001954, + "grad_norm": 1.8876679680467436, + "learning_rate": 3.96219724039092e-06, + "loss": 1.0161, + "step": 1200 + }, + { + "epoch": 0.09026003306778897, + "grad_norm": 1.7349352919634826, + "learning_rate": 3.962102974209018e-06, + "loss": 1.0888, + "step": 1201 + }, + { + "epoch": 0.09033518713362393, + "grad_norm": 1.723570704219188, + "learning_rate": 3.962008591765e-06, + "loss": 1.0624, + "step": 1202 + }, + { + "epoch": 0.09041034119945889, + "grad_norm": 2.030428793771988, + "learning_rate": 3.961914093064461e-06, + "loss": 1.0052, + "step": 1203 + }, + { + "epoch": 0.09048549526529386, + "grad_norm": 2.130893612594444, + "learning_rate": 3.961819478112999e-06, + "loss": 1.1097, + "step": 1204 + }, + { + "epoch": 0.09056064933112881, + "grad_norm": 2.6020400150339054, + "learning_rate": 3.961724746916221e-06, + "loss": 1.0194, + "step": 1205 + }, + { + "epoch": 0.09063580339696378, + "grad_norm": 2.4184239564509746, + "learning_rate": 3.961629899479739e-06, + "loss": 1.0872, + "step": 1206 + }, + { + "epoch": 0.09071095746279874, + "grad_norm": 1.4855538756501248, + "learning_rate": 3.961534935809174e-06, + "loss": 1.0634, + "step": 1207 + }, + { + "epoch": 0.0907861115286337, + "grad_norm": 1.486168590911604, + "learning_rate": 3.961439855910154e-06, + "loss": 0.9948, + "step": 1208 + }, + { + "epoch": 0.09086126559446867, + "grad_norm": 2.1111955486388005, + "learning_rate": 3.961344659788311e-06, + "loss": 1.0677, + "step": 1209 + }, + { + "epoch": 0.09093641966030362, + "grad_norm": 3.0509083192667066, + "learning_rate": 3.961249347449286e-06, + "loss": 1.0928, + "step": 1210 + }, + { + "epoch": 0.09101157372613858, + "grad_norm": 1.7923379424191201, + "learning_rate": 3.961153918898727e-06, + "loss": 1.0163, + "step": 1211 + }, + { + "epoch": 0.09108672779197355, + "grad_norm": 1.7622408547858857, + "learning_rate": 3.961058374142289e-06, + "loss": 1.1491, + "step": 1212 + }, + { + "epoch": 0.0911618818578085, + "grad_norm": 1.7633246907032607, + "learning_rate": 3.960962713185633e-06, + "loss": 1.006, + "step": 1213 + }, + { + "epoch": 0.09123703592364348, + "grad_norm": 1.7014246861158486, + "learning_rate": 3.960866936034426e-06, + "loss": 1.0579, + "step": 1214 + }, + { + "epoch": 0.09131218998947843, + "grad_norm": 1.445194218924854, + "learning_rate": 3.960771042694346e-06, + "loss": 1.0553, + "step": 1215 + }, + { + "epoch": 0.09138734405531339, + "grad_norm": 19.085429025154735, + "learning_rate": 3.960675033171072e-06, + "loss": 0.9841, + "step": 1216 + }, + { + "epoch": 0.09146249812114836, + "grad_norm": 1.8972605758012324, + "learning_rate": 3.960578907470295e-06, + "loss": 1.0764, + "step": 1217 + }, + { + "epoch": 0.09153765218698331, + "grad_norm": 2.241371777316925, + "learning_rate": 3.960482665597711e-06, + "loss": 0.9438, + "step": 1218 + }, + { + "epoch": 0.09161280625281827, + "grad_norm": 1.8357093182253248, + "learning_rate": 3.960386307559021e-06, + "loss": 1.0487, + "step": 1219 + }, + { + "epoch": 0.09168796031865324, + "grad_norm": 2.676911899010402, + "learning_rate": 3.960289833359936e-06, + "loss": 1.0105, + "step": 1220 + }, + { + "epoch": 0.0917631143844882, + "grad_norm": 1.3223694347838186, + "learning_rate": 3.9601932430061714e-06, + "loss": 1.0259, + "step": 1221 + }, + { + "epoch": 0.09183826845032317, + "grad_norm": 1.8596027138660474, + "learning_rate": 3.9600965365034515e-06, + "loss": 1.105, + "step": 1222 + }, + { + "epoch": 0.09191342251615812, + "grad_norm": 0.6158692490200987, + "learning_rate": 3.959999713857505e-06, + "loss": 0.7762, + "step": 1223 + }, + { + "epoch": 0.09198857658199308, + "grad_norm": 1.615089788273451, + "learning_rate": 3.959902775074072e-06, + "loss": 1.0987, + "step": 1224 + }, + { + "epoch": 0.09206373064782805, + "grad_norm": 2.5936450247605842, + "learning_rate": 3.9598057201588926e-06, + "loss": 1.0836, + "step": 1225 + }, + { + "epoch": 0.09213888471366301, + "grad_norm": 1.619405064541149, + "learning_rate": 3.959708549117721e-06, + "loss": 1.0194, + "step": 1226 + }, + { + "epoch": 0.09221403877949796, + "grad_norm": 1.8251740202827744, + "learning_rate": 3.959611261956313e-06, + "loss": 1.0482, + "step": 1227 + }, + { + "epoch": 0.09228919284533293, + "grad_norm": 2.083706695161577, + "learning_rate": 3.959513858680434e-06, + "loss": 0.9899, + "step": 1228 + }, + { + "epoch": 0.09236434691116789, + "grad_norm": 1.7691975462068312, + "learning_rate": 3.9594163392958566e-06, + "loss": 1.1058, + "step": 1229 + }, + { + "epoch": 0.09243950097700286, + "grad_norm": 2.8111887100207964, + "learning_rate": 3.959318703808356e-06, + "loss": 0.936, + "step": 1230 + }, + { + "epoch": 0.09251465504283782, + "grad_norm": 1.5845942328646028, + "learning_rate": 3.95922095222372e-06, + "loss": 1.0741, + "step": 1231 + }, + { + "epoch": 0.09258980910867277, + "grad_norm": 1.6626355854474961, + "learning_rate": 3.959123084547741e-06, + "loss": 1.099, + "step": 1232 + }, + { + "epoch": 0.09266496317450774, + "grad_norm": 2.04435504974735, + "learning_rate": 3.959025100786217e-06, + "loss": 1.0863, + "step": 1233 + }, + { + "epoch": 0.0927401172403427, + "grad_norm": 1.9153322329059383, + "learning_rate": 3.958927000944954e-06, + "loss": 1.073, + "step": 1234 + }, + { + "epoch": 0.09281527130617766, + "grad_norm": 1.9070203506552075, + "learning_rate": 3.958828785029765e-06, + "loss": 1.0524, + "step": 1235 + }, + { + "epoch": 0.09289042537201263, + "grad_norm": 3.4514705785596793, + "learning_rate": 3.958730453046469e-06, + "loss": 1.1097, + "step": 1236 + }, + { + "epoch": 0.09296557943784758, + "grad_norm": 1.7354583106324148, + "learning_rate": 3.958632005000895e-06, + "loss": 1.0265, + "step": 1237 + }, + { + "epoch": 0.09304073350368255, + "grad_norm": 1.8455010197506683, + "learning_rate": 3.958533440898873e-06, + "loss": 1.0211, + "step": 1238 + }, + { + "epoch": 0.09311588756951751, + "grad_norm": 1.5350986304887282, + "learning_rate": 3.958434760746245e-06, + "loss": 1.0264, + "step": 1239 + }, + { + "epoch": 0.09319104163535247, + "grad_norm": 0.8316415223223861, + "learning_rate": 3.958335964548859e-06, + "loss": 0.8816, + "step": 1240 + }, + { + "epoch": 0.09326619570118744, + "grad_norm": 1.5732241534212341, + "learning_rate": 3.958237052312568e-06, + "loss": 1.0576, + "step": 1241 + }, + { + "epoch": 0.0933413497670224, + "grad_norm": 2.1251180279434903, + "learning_rate": 3.958138024043232e-06, + "loss": 1.1219, + "step": 1242 + }, + { + "epoch": 0.09341650383285736, + "grad_norm": 0.6865830158194788, + "learning_rate": 3.958038879746721e-06, + "loss": 0.8239, + "step": 1243 + }, + { + "epoch": 0.09349165789869232, + "grad_norm": 2.0923152961474116, + "learning_rate": 3.9579396194289075e-06, + "loss": 0.9196, + "step": 1244 + }, + { + "epoch": 0.09356681196452728, + "grad_norm": 2.4974729163426357, + "learning_rate": 3.957840243095675e-06, + "loss": 0.9892, + "step": 1245 + }, + { + "epoch": 0.09364196603036225, + "grad_norm": 2.1808678497198986, + "learning_rate": 3.95774075075291e-06, + "loss": 0.966, + "step": 1246 + }, + { + "epoch": 0.0937171200961972, + "grad_norm": 1.6577417385520794, + "learning_rate": 3.957641142406509e-06, + "loss": 1.037, + "step": 1247 + }, + { + "epoch": 0.09379227416203216, + "grad_norm": 1.8318999351176186, + "learning_rate": 3.9575414180623746e-06, + "loss": 0.9609, + "step": 1248 + }, + { + "epoch": 0.09386742822786713, + "grad_norm": 2.4806717436623265, + "learning_rate": 3.957441577726415e-06, + "loss": 1.0338, + "step": 1249 + }, + { + "epoch": 0.09394258229370209, + "grad_norm": 1.5726214083333485, + "learning_rate": 3.957341621404547e-06, + "loss": 1.0412, + "step": 1250 + }, + { + "epoch": 0.09401773635953706, + "grad_norm": 2.0963761670081666, + "learning_rate": 3.957241549102692e-06, + "loss": 1.0739, + "step": 1251 + }, + { + "epoch": 0.09409289042537201, + "grad_norm": 1.942060456298405, + "learning_rate": 3.957141360826781e-06, + "loss": 1.0439, + "step": 1252 + }, + { + "epoch": 0.09416804449120697, + "grad_norm": 1.526781764140871, + "learning_rate": 3.95704105658275e-06, + "loss": 0.9302, + "step": 1253 + }, + { + "epoch": 0.09424319855704194, + "grad_norm": 0.9559653749919195, + "learning_rate": 3.9569406363765415e-06, + "loss": 0.878, + "step": 1254 + }, + { + "epoch": 0.0943183526228769, + "grad_norm": 2.1295941002810617, + "learning_rate": 3.956840100214107e-06, + "loss": 0.9799, + "step": 1255 + }, + { + "epoch": 0.09439350668871185, + "grad_norm": 1.7462913179527013, + "learning_rate": 3.956739448101404e-06, + "loss": 1.043, + "step": 1256 + }, + { + "epoch": 0.09446866075454682, + "grad_norm": 1.8293978832444069, + "learning_rate": 3.956638680044396e-06, + "loss": 1.08, + "step": 1257 + }, + { + "epoch": 0.09454381482038178, + "grad_norm": 2.3862567960263688, + "learning_rate": 3.956537796049052e-06, + "loss": 0.9955, + "step": 1258 + }, + { + "epoch": 0.09461896888621675, + "grad_norm": 1.5775455884195138, + "learning_rate": 3.9564367961213536e-06, + "loss": 0.9856, + "step": 1259 + }, + { + "epoch": 0.0946941229520517, + "grad_norm": 1.5201385774936922, + "learning_rate": 3.956335680267282e-06, + "loss": 1.1021, + "step": 1260 + }, + { + "epoch": 0.09476927701788666, + "grad_norm": 1.8338989627924438, + "learning_rate": 3.956234448492831e-06, + "loss": 1.0013, + "step": 1261 + }, + { + "epoch": 0.09484443108372163, + "grad_norm": 1.6576901778126258, + "learning_rate": 3.956133100803996e-06, + "loss": 0.9951, + "step": 1262 + }, + { + "epoch": 0.09491958514955659, + "grad_norm": 3.481310262107106, + "learning_rate": 3.956031637206786e-06, + "loss": 1.0511, + "step": 1263 + }, + { + "epoch": 0.09499473921539155, + "grad_norm": 1.872349391426731, + "learning_rate": 3.955930057707211e-06, + "loss": 0.9468, + "step": 1264 + }, + { + "epoch": 0.09506989328122652, + "grad_norm": 1.4693151157885986, + "learning_rate": 3.95582836231129e-06, + "loss": 1.0316, + "step": 1265 + }, + { + "epoch": 0.09514504734706147, + "grad_norm": 1.9509788012895104, + "learning_rate": 3.9557265510250505e-06, + "loss": 1.0054, + "step": 1266 + }, + { + "epoch": 0.09522020141289644, + "grad_norm": 1.8863868184456607, + "learning_rate": 3.955624623854522e-06, + "loss": 0.975, + "step": 1267 + }, + { + "epoch": 0.0952953554787314, + "grad_norm": 1.6386947175571944, + "learning_rate": 3.9555225808057475e-06, + "loss": 0.9161, + "step": 1268 + }, + { + "epoch": 0.09537050954456636, + "grad_norm": 2.10410945801974, + "learning_rate": 3.9554204218847705e-06, + "loss": 1.0161, + "step": 1269 + }, + { + "epoch": 0.09544566361040133, + "grad_norm": 1.8767446666157477, + "learning_rate": 3.955318147097647e-06, + "loss": 1.0491, + "step": 1270 + }, + { + "epoch": 0.09552081767623628, + "grad_norm": 1.7293521904139075, + "learning_rate": 3.955215756450435e-06, + "loss": 1.0146, + "step": 1271 + }, + { + "epoch": 0.09559597174207124, + "grad_norm": 2.7012064746066766, + "learning_rate": 3.955113249949203e-06, + "loss": 0.9163, + "step": 1272 + }, + { + "epoch": 0.09567112580790621, + "grad_norm": 3.3250701208536566, + "learning_rate": 3.955010627600024e-06, + "loss": 1.0303, + "step": 1273 + }, + { + "epoch": 0.09574627987374117, + "grad_norm": 1.476455361652286, + "learning_rate": 3.954907889408979e-06, + "loss": 1.0998, + "step": 1274 + }, + { + "epoch": 0.09582143393957614, + "grad_norm": 1.8621614238041888, + "learning_rate": 3.954805035382155e-06, + "loss": 1.1093, + "step": 1275 + }, + { + "epoch": 0.0958965880054111, + "grad_norm": 1.7833294693141444, + "learning_rate": 3.954702065525649e-06, + "loss": 0.9856, + "step": 1276 + }, + { + "epoch": 0.09597174207124605, + "grad_norm": 1.694400805899813, + "learning_rate": 3.954598979845559e-06, + "loss": 1.0678, + "step": 1277 + }, + { + "epoch": 0.09604689613708102, + "grad_norm": 2.187200689274666, + "learning_rate": 3.954495778347996e-06, + "loss": 0.9062, + "step": 1278 + }, + { + "epoch": 0.09612205020291598, + "grad_norm": 1.789350328238707, + "learning_rate": 3.954392461039073e-06, + "loss": 1.0907, + "step": 1279 + }, + { + "epoch": 0.09619720426875093, + "grad_norm": 2.0198938866872593, + "learning_rate": 3.954289027924912e-06, + "loss": 0.9733, + "step": 1280 + }, + { + "epoch": 0.0962723583345859, + "grad_norm": 1.6432567015777764, + "learning_rate": 3.954185479011644e-06, + "loss": 1.076, + "step": 1281 + }, + { + "epoch": 0.09634751240042086, + "grad_norm": 0.7786102712073778, + "learning_rate": 3.954081814305403e-06, + "loss": 0.8209, + "step": 1282 + }, + { + "epoch": 0.09642266646625583, + "grad_norm": 1.7806827913642265, + "learning_rate": 3.953978033812332e-06, + "loss": 1.0307, + "step": 1283 + }, + { + "epoch": 0.09649782053209079, + "grad_norm": 2.163068301987929, + "learning_rate": 3.95387413753858e-06, + "loss": 0.9619, + "step": 1284 + }, + { + "epoch": 0.09657297459792574, + "grad_norm": 1.7522320817664485, + "learning_rate": 3.9537701254903034e-06, + "loss": 0.9141, + "step": 1285 + }, + { + "epoch": 0.09664812866376071, + "grad_norm": 2.15576785681813, + "learning_rate": 3.953665997673665e-06, + "loss": 1.0153, + "step": 1286 + }, + { + "epoch": 0.09672328272959567, + "grad_norm": 2.1317726367004903, + "learning_rate": 3.953561754094836e-06, + "loss": 1.0183, + "step": 1287 + }, + { + "epoch": 0.09679843679543064, + "grad_norm": 1.643945295096605, + "learning_rate": 3.953457394759992e-06, + "loss": 1.1083, + "step": 1288 + }, + { + "epoch": 0.0968735908612656, + "grad_norm": 2.2913106352779073, + "learning_rate": 3.953352919675317e-06, + "loss": 1.0863, + "step": 1289 + }, + { + "epoch": 0.09694874492710055, + "grad_norm": 3.062775112951829, + "learning_rate": 3.953248328847001e-06, + "loss": 0.9645, + "step": 1290 + }, + { + "epoch": 0.09702389899293552, + "grad_norm": 1.4904135500084625, + "learning_rate": 3.953143622281243e-06, + "loss": 1.0806, + "step": 1291 + }, + { + "epoch": 0.09709905305877048, + "grad_norm": 1.7924625787593005, + "learning_rate": 3.953038799984246e-06, + "loss": 1.0275, + "step": 1292 + }, + { + "epoch": 0.09717420712460544, + "grad_norm": 1.708654544064466, + "learning_rate": 3.952933861962222e-06, + "loss": 1.1184, + "step": 1293 + }, + { + "epoch": 0.0972493611904404, + "grad_norm": 1.9518344124694469, + "learning_rate": 3.952828808221387e-06, + "loss": 1.0284, + "step": 1294 + }, + { + "epoch": 0.09732451525627536, + "grad_norm": 1.7431961355811636, + "learning_rate": 3.952723638767968e-06, + "loss": 1.1232, + "step": 1295 + }, + { + "epoch": 0.09739966932211033, + "grad_norm": 1.9791804530684172, + "learning_rate": 3.952618353608196e-06, + "loss": 1.0085, + "step": 1296 + }, + { + "epoch": 0.09747482338794529, + "grad_norm": 1.7093236967822887, + "learning_rate": 3.95251295274831e-06, + "loss": 1.0396, + "step": 1297 + }, + { + "epoch": 0.09754997745378025, + "grad_norm": 1.9936644819807132, + "learning_rate": 3.952407436194554e-06, + "loss": 1.0159, + "step": 1298 + }, + { + "epoch": 0.09762513151961522, + "grad_norm": 3.0534137694690195, + "learning_rate": 3.9523018039531816e-06, + "loss": 0.9461, + "step": 1299 + }, + { + "epoch": 0.09770028558545017, + "grad_norm": 2.4942460601674585, + "learning_rate": 3.952196056030451e-06, + "loss": 1.0066, + "step": 1300 + }, + { + "epoch": 0.09777543965128513, + "grad_norm": 0.8127644915337531, + "learning_rate": 3.952090192432629e-06, + "loss": 0.8386, + "step": 1301 + }, + { + "epoch": 0.0978505937171201, + "grad_norm": 2.133646148853949, + "learning_rate": 3.951984213165988e-06, + "loss": 1.0374, + "step": 1302 + }, + { + "epoch": 0.09792574778295506, + "grad_norm": 1.909733512922106, + "learning_rate": 3.951878118236807e-06, + "loss": 0.8682, + "step": 1303 + }, + { + "epoch": 0.09800090184879003, + "grad_norm": 2.1826404544691065, + "learning_rate": 3.951771907651374e-06, + "loss": 0.9847, + "step": 1304 + }, + { + "epoch": 0.09807605591462498, + "grad_norm": 1.684279866265614, + "learning_rate": 3.951665581415982e-06, + "loss": 1.0426, + "step": 1305 + }, + { + "epoch": 0.09815120998045994, + "grad_norm": 1.5504760441853107, + "learning_rate": 3.9515591395369305e-06, + "loss": 1.0246, + "step": 1306 + }, + { + "epoch": 0.09822636404629491, + "grad_norm": 2.7918131965404567, + "learning_rate": 3.9514525820205265e-06, + "loss": 1.0236, + "step": 1307 + }, + { + "epoch": 0.09830151811212987, + "grad_norm": 2.043843477777007, + "learning_rate": 3.951345908873085e-06, + "loss": 0.998, + "step": 1308 + }, + { + "epoch": 0.09837667217796482, + "grad_norm": 3.1854971782987374, + "learning_rate": 3.9512391201009265e-06, + "loss": 1.0093, + "step": 1309 + }, + { + "epoch": 0.09845182624379979, + "grad_norm": 1.9633493978601981, + "learning_rate": 3.9511322157103776e-06, + "loss": 1.0444, + "step": 1310 + }, + { + "epoch": 0.09852698030963475, + "grad_norm": 1.8405759613810149, + "learning_rate": 3.951025195707774e-06, + "loss": 1.0898, + "step": 1311 + }, + { + "epoch": 0.09860213437546972, + "grad_norm": 1.8956204459025299, + "learning_rate": 3.950918060099456e-06, + "loss": 1.1204, + "step": 1312 + }, + { + "epoch": 0.09867728844130468, + "grad_norm": 1.622695976729744, + "learning_rate": 3.950810808891773e-06, + "loss": 0.9725, + "step": 1313 + }, + { + "epoch": 0.09875244250713963, + "grad_norm": 3.541340278096839, + "learning_rate": 3.950703442091079e-06, + "loss": 1.0477, + "step": 1314 + }, + { + "epoch": 0.0988275965729746, + "grad_norm": 2.1299528856394794, + "learning_rate": 3.950595959703736e-06, + "loss": 1.0265, + "step": 1315 + }, + { + "epoch": 0.09890275063880956, + "grad_norm": 1.6718331447602859, + "learning_rate": 3.950488361736114e-06, + "loss": 1.0164, + "step": 1316 + }, + { + "epoch": 0.09897790470464451, + "grad_norm": 1.7151596021762814, + "learning_rate": 3.950380648194587e-06, + "loss": 1.0752, + "step": 1317 + }, + { + "epoch": 0.09905305877047949, + "grad_norm": 1.6356167599999247, + "learning_rate": 3.950272819085538e-06, + "loss": 1.0072, + "step": 1318 + }, + { + "epoch": 0.09912821283631444, + "grad_norm": 3.1697037504859544, + "learning_rate": 3.950164874415357e-06, + "loss": 1.0066, + "step": 1319 + }, + { + "epoch": 0.09920336690214941, + "grad_norm": 2.70308689685934, + "learning_rate": 3.950056814190439e-06, + "loss": 0.9398, + "step": 1320 + }, + { + "epoch": 0.09927852096798437, + "grad_norm": 2.0337776303187525, + "learning_rate": 3.949948638417188e-06, + "loss": 1.0094, + "step": 1321 + }, + { + "epoch": 0.09935367503381932, + "grad_norm": 1.9761527017363394, + "learning_rate": 3.949840347102013e-06, + "loss": 1.0218, + "step": 1322 + }, + { + "epoch": 0.0994288290996543, + "grad_norm": 1.7880149339696554, + "learning_rate": 3.949731940251331e-06, + "loss": 0.9991, + "step": 1323 + }, + { + "epoch": 0.09950398316548925, + "grad_norm": 1.8425647999562902, + "learning_rate": 3.949623417871565e-06, + "loss": 1.0154, + "step": 1324 + }, + { + "epoch": 0.09957913723132421, + "grad_norm": 0.7067863082222359, + "learning_rate": 3.949514779969147e-06, + "loss": 0.8131, + "step": 1325 + }, + { + "epoch": 0.09965429129715918, + "grad_norm": 2.0804796001568246, + "learning_rate": 3.949406026550512e-06, + "loss": 1.1182, + "step": 1326 + }, + { + "epoch": 0.09972944536299413, + "grad_norm": 1.8309680318036519, + "learning_rate": 3.949297157622105e-06, + "loss": 1.0754, + "step": 1327 + }, + { + "epoch": 0.0998045994288291, + "grad_norm": 1.8384707890177459, + "learning_rate": 3.949188173190378e-06, + "loss": 0.989, + "step": 1328 + }, + { + "epoch": 0.09987975349466406, + "grad_norm": 7.876129553787626, + "learning_rate": 3.949079073261788e-06, + "loss": 0.9893, + "step": 1329 + }, + { + "epoch": 0.09995490756049902, + "grad_norm": 2.788650012034124, + "learning_rate": 3.948969857842799e-06, + "loss": 0.9812, + "step": 1330 + }, + { + "epoch": 0.10003006162633399, + "grad_norm": 2.720098665215039, + "learning_rate": 3.948860526939882e-06, + "loss": 1.0277, + "step": 1331 + }, + { + "epoch": 0.10010521569216894, + "grad_norm": 2.202837877975655, + "learning_rate": 3.948751080559517e-06, + "loss": 1.0595, + "step": 1332 + }, + { + "epoch": 0.10018036975800391, + "grad_norm": 7.249545642164223, + "learning_rate": 3.948641518708188e-06, + "loss": 1.0994, + "step": 1333 + }, + { + "epoch": 0.10025552382383887, + "grad_norm": 4.158957381172965, + "learning_rate": 3.9485318413923865e-06, + "loss": 0.9783, + "step": 1334 + }, + { + "epoch": 0.10033067788967383, + "grad_norm": 1.6685219055191898, + "learning_rate": 3.948422048618612e-06, + "loss": 1.0166, + "step": 1335 + }, + { + "epoch": 0.1004058319555088, + "grad_norm": 2.0879653046601026, + "learning_rate": 3.948312140393372e-06, + "loss": 1.0319, + "step": 1336 + }, + { + "epoch": 0.10048098602134375, + "grad_norm": 0.752242748416978, + "learning_rate": 3.948202116723176e-06, + "loss": 0.8168, + "step": 1337 + }, + { + "epoch": 0.10055614008717871, + "grad_norm": 1.3877870125374656, + "learning_rate": 3.948091977614544e-06, + "loss": 0.9849, + "step": 1338 + }, + { + "epoch": 0.10063129415301368, + "grad_norm": 1.6314254783284192, + "learning_rate": 3.947981723074003e-06, + "loss": 0.9718, + "step": 1339 + }, + { + "epoch": 0.10070644821884864, + "grad_norm": 1.9097008659632895, + "learning_rate": 3.947871353108085e-06, + "loss": 0.9526, + "step": 1340 + }, + { + "epoch": 0.10078160228468361, + "grad_norm": 2.1508918198924203, + "learning_rate": 3.947760867723331e-06, + "loss": 1.1198, + "step": 1341 + }, + { + "epoch": 0.10085675635051856, + "grad_norm": 2.1654527515866624, + "learning_rate": 3.9476502669262866e-06, + "loss": 1.0292, + "step": 1342 + }, + { + "epoch": 0.10093191041635352, + "grad_norm": 2.038678129368939, + "learning_rate": 3.947539550723506e-06, + "loss": 1.0049, + "step": 1343 + }, + { + "epoch": 0.10100706448218849, + "grad_norm": 0.7000235502504926, + "learning_rate": 3.94742871912155e-06, + "loss": 0.8391, + "step": 1344 + }, + { + "epoch": 0.10108221854802345, + "grad_norm": 1.644502181931159, + "learning_rate": 3.947317772126985e-06, + "loss": 1.037, + "step": 1345 + }, + { + "epoch": 0.1011573726138584, + "grad_norm": 2.3124017436830115, + "learning_rate": 3.947206709746385e-06, + "loss": 1.1396, + "step": 1346 + }, + { + "epoch": 0.10123252667969337, + "grad_norm": 2.335226812027226, + "learning_rate": 3.947095531986331e-06, + "loss": 1.1121, + "step": 1347 + }, + { + "epoch": 0.10130768074552833, + "grad_norm": 1.970697136407681, + "learning_rate": 3.9469842388534105e-06, + "loss": 1.0467, + "step": 1348 + }, + { + "epoch": 0.1013828348113633, + "grad_norm": 1.9710783738221855, + "learning_rate": 3.946872830354219e-06, + "loss": 1.0112, + "step": 1349 + }, + { + "epoch": 0.10145798887719826, + "grad_norm": 1.596683868496055, + "learning_rate": 3.946761306495357e-06, + "loss": 1.1577, + "step": 1350 + }, + { + "epoch": 0.10153314294303321, + "grad_norm": 2.112788962741031, + "learning_rate": 3.946649667283433e-06, + "loss": 1.025, + "step": 1351 + }, + { + "epoch": 0.10160829700886818, + "grad_norm": 1.8204886551786126, + "learning_rate": 3.946537912725062e-06, + "loss": 1.0021, + "step": 1352 + }, + { + "epoch": 0.10168345107470314, + "grad_norm": 1.713349753844148, + "learning_rate": 3.946426042826865e-06, + "loss": 0.95, + "step": 1353 + }, + { + "epoch": 0.1017586051405381, + "grad_norm": 3.1876548280699213, + "learning_rate": 3.946314057595473e-06, + "loss": 1.0058, + "step": 1354 + }, + { + "epoch": 0.10183375920637307, + "grad_norm": 3.4955838817276774, + "learning_rate": 3.94620195703752e-06, + "loss": 0.9749, + "step": 1355 + }, + { + "epoch": 0.10190891327220802, + "grad_norm": 1.76104060144534, + "learning_rate": 3.946089741159647e-06, + "loss": 0.9045, + "step": 1356 + }, + { + "epoch": 0.101984067338043, + "grad_norm": 0.8132703137280838, + "learning_rate": 3.9459774099685065e-06, + "loss": 0.8759, + "step": 1357 + }, + { + "epoch": 0.10205922140387795, + "grad_norm": 2.034648927215465, + "learning_rate": 3.945864963470752e-06, + "loss": 1.0461, + "step": 1358 + }, + { + "epoch": 0.1021343754697129, + "grad_norm": 7.46211183902684, + "learning_rate": 3.945752401673047e-06, + "loss": 0.9175, + "step": 1359 + }, + { + "epoch": 0.10220952953554788, + "grad_norm": 1.6404992918264714, + "learning_rate": 3.945639724582062e-06, + "loss": 1.1165, + "step": 1360 + }, + { + "epoch": 0.10228468360138283, + "grad_norm": 2.3344123479876067, + "learning_rate": 3.9455269322044725e-06, + "loss": 1.0832, + "step": 1361 + }, + { + "epoch": 0.10235983766721779, + "grad_norm": 1.8830380126633157, + "learning_rate": 3.945414024546963e-06, + "loss": 1.0205, + "step": 1362 + }, + { + "epoch": 0.10243499173305276, + "grad_norm": 1.8934363908098657, + "learning_rate": 3.945301001616222e-06, + "loss": 1.0555, + "step": 1363 + }, + { + "epoch": 0.10251014579888772, + "grad_norm": 3.3180199461297755, + "learning_rate": 3.945187863418949e-06, + "loss": 0.9636, + "step": 1364 + }, + { + "epoch": 0.10258529986472269, + "grad_norm": 1.682866198179746, + "learning_rate": 3.945074609961845e-06, + "loss": 1.0612, + "step": 1365 + }, + { + "epoch": 0.10266045393055764, + "grad_norm": 1.9503064990642773, + "learning_rate": 3.944961241251623e-06, + "loss": 1.0487, + "step": 1366 + }, + { + "epoch": 0.1027356079963926, + "grad_norm": 1.856541116922338, + "learning_rate": 3.944847757295e-06, + "loss": 0.9653, + "step": 1367 + }, + { + "epoch": 0.10281076206222757, + "grad_norm": 2.0411080225422364, + "learning_rate": 3.9447341580987e-06, + "loss": 1.0142, + "step": 1368 + }, + { + "epoch": 0.10288591612806253, + "grad_norm": 2.2170856304571416, + "learning_rate": 3.944620443669453e-06, + "loss": 1.0615, + "step": 1369 + }, + { + "epoch": 0.10296107019389748, + "grad_norm": 1.6515472873358845, + "learning_rate": 3.9445066140139995e-06, + "loss": 0.9391, + "step": 1370 + }, + { + "epoch": 0.10303622425973245, + "grad_norm": 1.8847711908920408, + "learning_rate": 3.944392669139083e-06, + "loss": 1.037, + "step": 1371 + }, + { + "epoch": 0.10311137832556741, + "grad_norm": 1.605169932083285, + "learning_rate": 3.944278609051455e-06, + "loss": 0.9123, + "step": 1372 + }, + { + "epoch": 0.10318653239140238, + "grad_norm": 1.9549587339060728, + "learning_rate": 3.944164433757874e-06, + "loss": 1.0388, + "step": 1373 + }, + { + "epoch": 0.10326168645723734, + "grad_norm": 2.322682823647598, + "learning_rate": 3.944050143265106e-06, + "loss": 0.9495, + "step": 1374 + }, + { + "epoch": 0.1033368405230723, + "grad_norm": 2.358390956574574, + "learning_rate": 3.943935737579923e-06, + "loss": 1.0616, + "step": 1375 + }, + { + "epoch": 0.10341199458890726, + "grad_norm": 2.0992122398761053, + "learning_rate": 3.943821216709103e-06, + "loss": 0.9516, + "step": 1376 + }, + { + "epoch": 0.10348714865474222, + "grad_norm": 2.0027794597946116, + "learning_rate": 3.943706580659433e-06, + "loss": 1.0112, + "step": 1377 + }, + { + "epoch": 0.10356230272057719, + "grad_norm": 1.506935378407261, + "learning_rate": 3.943591829437705e-06, + "loss": 1.1192, + "step": 1378 + }, + { + "epoch": 0.10363745678641215, + "grad_norm": 2.0897839689171613, + "learning_rate": 3.943476963050719e-06, + "loss": 1.0079, + "step": 1379 + }, + { + "epoch": 0.1037126108522471, + "grad_norm": 0.8917316788892004, + "learning_rate": 3.94336198150528e-06, + "loss": 0.9117, + "step": 1380 + }, + { + "epoch": 0.10378776491808207, + "grad_norm": 1.857516730581027, + "learning_rate": 3.9432468848082024e-06, + "loss": 0.9966, + "step": 1381 + }, + { + "epoch": 0.10386291898391703, + "grad_norm": 2.455672186914899, + "learning_rate": 3.9431316729663055e-06, + "loss": 1.0359, + "step": 1382 + }, + { + "epoch": 0.10393807304975199, + "grad_norm": 2.488126022311827, + "learning_rate": 3.943016345986417e-06, + "loss": 0.9484, + "step": 1383 + }, + { + "epoch": 0.10401322711558696, + "grad_norm": 1.5127650540986124, + "learning_rate": 3.942900903875369e-06, + "loss": 1.0545, + "step": 1384 + }, + { + "epoch": 0.10408838118142191, + "grad_norm": 3.354383441878177, + "learning_rate": 3.942785346640003e-06, + "loss": 1.1047, + "step": 1385 + }, + { + "epoch": 0.10416353524725688, + "grad_norm": 3.5152910540178617, + "learning_rate": 3.942669674287166e-06, + "loss": 1.0572, + "step": 1386 + }, + { + "epoch": 0.10423868931309184, + "grad_norm": 1.7569590891687186, + "learning_rate": 3.942553886823711e-06, + "loss": 1.0762, + "step": 1387 + }, + { + "epoch": 0.1043138433789268, + "grad_norm": 2.1054224352440194, + "learning_rate": 3.9424379842565005e-06, + "loss": 1.0513, + "step": 1388 + }, + { + "epoch": 0.10438899744476177, + "grad_norm": 1.9535790487854634, + "learning_rate": 3.9423219665924e-06, + "loss": 1.0743, + "step": 1389 + }, + { + "epoch": 0.10446415151059672, + "grad_norm": 1.81747575672882, + "learning_rate": 3.942205833838287e-06, + "loss": 1.0012, + "step": 1390 + }, + { + "epoch": 0.10453930557643168, + "grad_norm": 2.4366498589446977, + "learning_rate": 3.942089586001039e-06, + "loss": 1.1546, + "step": 1391 + }, + { + "epoch": 0.10461445964226665, + "grad_norm": 1.6840938524565179, + "learning_rate": 3.941973223087548e-06, + "loss": 0.9515, + "step": 1392 + }, + { + "epoch": 0.1046896137081016, + "grad_norm": 1.6402796790599794, + "learning_rate": 3.941856745104707e-06, + "loss": 1.0116, + "step": 1393 + }, + { + "epoch": 0.10476476777393658, + "grad_norm": 2.1230456396974904, + "learning_rate": 3.941740152059418e-06, + "loss": 1.0247, + "step": 1394 + }, + { + "epoch": 0.10483992183977153, + "grad_norm": 1.3294142804352858, + "learning_rate": 3.94162344395859e-06, + "loss": 1.0023, + "step": 1395 + }, + { + "epoch": 0.10491507590560649, + "grad_norm": 2.248753096200403, + "learning_rate": 3.941506620809137e-06, + "loss": 1.055, + "step": 1396 + }, + { + "epoch": 0.10499022997144146, + "grad_norm": 1.8149177605148534, + "learning_rate": 3.941389682617982e-06, + "loss": 0.9779, + "step": 1397 + }, + { + "epoch": 0.10506538403727642, + "grad_norm": 1.9200579729864684, + "learning_rate": 3.9412726293920555e-06, + "loss": 0.9808, + "step": 1398 + }, + { + "epoch": 0.10514053810311137, + "grad_norm": 1.9561598126277089, + "learning_rate": 3.9411554611382914e-06, + "loss": 1.1205, + "step": 1399 + }, + { + "epoch": 0.10521569216894634, + "grad_norm": 2.149652532844062, + "learning_rate": 3.941038177863633e-06, + "loss": 1.0358, + "step": 1400 + }, + { + "epoch": 0.1052908462347813, + "grad_norm": 0.8133505990457489, + "learning_rate": 3.940920779575029e-06, + "loss": 0.8729, + "step": 1401 + }, + { + "epoch": 0.10536600030061627, + "grad_norm": 4.3879475422369, + "learning_rate": 3.940803266279438e-06, + "loss": 1.0888, + "step": 1402 + }, + { + "epoch": 0.10544115436645123, + "grad_norm": 1.8341793001402198, + "learning_rate": 3.940685637983822e-06, + "loss": 1.0237, + "step": 1403 + }, + { + "epoch": 0.10551630843228618, + "grad_norm": 1.7890812902005198, + "learning_rate": 3.940567894695149e-06, + "loss": 1.0787, + "step": 1404 + }, + { + "epoch": 0.10559146249812115, + "grad_norm": 2.178537372975093, + "learning_rate": 3.940450036420397e-06, + "loss": 0.9679, + "step": 1405 + }, + { + "epoch": 0.10566661656395611, + "grad_norm": 0.7493413973950765, + "learning_rate": 3.940332063166551e-06, + "loss": 0.833, + "step": 1406 + }, + { + "epoch": 0.10574177062979107, + "grad_norm": 2.08532511547954, + "learning_rate": 3.9402139749406e-06, + "loss": 1.078, + "step": 1407 + }, + { + "epoch": 0.10581692469562604, + "grad_norm": 1.915829088500327, + "learning_rate": 3.940095771749542e-06, + "loss": 1.0517, + "step": 1408 + }, + { + "epoch": 0.10589207876146099, + "grad_norm": 11.753229225059677, + "learning_rate": 3.939977453600379e-06, + "loss": 0.9967, + "step": 1409 + }, + { + "epoch": 0.10596723282729596, + "grad_norm": 1.7696910390224028, + "learning_rate": 3.939859020500124e-06, + "loss": 1.0102, + "step": 1410 + }, + { + "epoch": 0.10604238689313092, + "grad_norm": 1.897174996247022, + "learning_rate": 3.939740472455794e-06, + "loss": 0.9425, + "step": 1411 + }, + { + "epoch": 0.10611754095896588, + "grad_norm": 1.6893407177877875, + "learning_rate": 3.939621809474413e-06, + "loss": 1.0616, + "step": 1412 + }, + { + "epoch": 0.10619269502480085, + "grad_norm": 1.961141568086377, + "learning_rate": 3.9395030315630124e-06, + "loss": 1.0804, + "step": 1413 + }, + { + "epoch": 0.1062678490906358, + "grad_norm": 1.7666008202259704, + "learning_rate": 3.939384138728631e-06, + "loss": 1.0435, + "step": 1414 + }, + { + "epoch": 0.10634300315647076, + "grad_norm": 1.9216897766673482, + "learning_rate": 3.939265130978312e-06, + "loss": 1.0157, + "step": 1415 + }, + { + "epoch": 0.10641815722230573, + "grad_norm": 1.791346928063045, + "learning_rate": 3.939146008319109e-06, + "loss": 1.0246, + "step": 1416 + }, + { + "epoch": 0.10649331128814069, + "grad_norm": 2.0741174820427086, + "learning_rate": 3.939026770758079e-06, + "loss": 1.0092, + "step": 1417 + }, + { + "epoch": 0.10656846535397566, + "grad_norm": 1.4896578292180507, + "learning_rate": 3.938907418302288e-06, + "loss": 1.0304, + "step": 1418 + }, + { + "epoch": 0.10664361941981061, + "grad_norm": 1.8178072212959702, + "learning_rate": 3.938787950958807e-06, + "loss": 1.0268, + "step": 1419 + }, + { + "epoch": 0.10671877348564557, + "grad_norm": 2.142180385897765, + "learning_rate": 3.938668368734717e-06, + "loss": 0.9848, + "step": 1420 + }, + { + "epoch": 0.10679392755148054, + "grad_norm": 1.7134439581596785, + "learning_rate": 3.938548671637102e-06, + "loss": 1.0059, + "step": 1421 + }, + { + "epoch": 0.1068690816173155, + "grad_norm": 1.9513067708624576, + "learning_rate": 3.938428859673055e-06, + "loss": 1.0558, + "step": 1422 + }, + { + "epoch": 0.10694423568315047, + "grad_norm": 2.1068247885253006, + "learning_rate": 3.9383089328496755e-06, + "loss": 1.0084, + "step": 1423 + }, + { + "epoch": 0.10701938974898542, + "grad_norm": 1.7698177417303402, + "learning_rate": 3.938188891174069e-06, + "loss": 0.9798, + "step": 1424 + }, + { + "epoch": 0.10709454381482038, + "grad_norm": 2.156915423783516, + "learning_rate": 3.9380687346533495e-06, + "loss": 0.9972, + "step": 1425 + }, + { + "epoch": 0.10716969788065535, + "grad_norm": 3.359065267789032, + "learning_rate": 3.9379484632946355e-06, + "loss": 1.0489, + "step": 1426 + }, + { + "epoch": 0.1072448519464903, + "grad_norm": 1.919152603407083, + "learning_rate": 3.937828077105054e-06, + "loss": 0.8877, + "step": 1427 + }, + { + "epoch": 0.10732000601232526, + "grad_norm": 1.85714767790787, + "learning_rate": 3.9377075760917396e-06, + "loss": 1.0833, + "step": 1428 + }, + { + "epoch": 0.10739516007816023, + "grad_norm": 1.602355388597229, + "learning_rate": 3.93758696026183e-06, + "loss": 1.0767, + "step": 1429 + }, + { + "epoch": 0.10747031414399519, + "grad_norm": 2.466900289486594, + "learning_rate": 3.9374662296224746e-06, + "loss": 1.08, + "step": 1430 + }, + { + "epoch": 0.10754546820983016, + "grad_norm": 1.7570052264140714, + "learning_rate": 3.937345384180826e-06, + "loss": 1.027, + "step": 1431 + }, + { + "epoch": 0.10762062227566511, + "grad_norm": 2.6075599433833543, + "learning_rate": 3.937224423944044e-06, + "loss": 1.0519, + "step": 1432 + }, + { + "epoch": 0.10769577634150007, + "grad_norm": 1.6158105267782066, + "learning_rate": 3.937103348919297e-06, + "loss": 0.9453, + "step": 1433 + }, + { + "epoch": 0.10777093040733504, + "grad_norm": 3.018849758608611, + "learning_rate": 3.936982159113759e-06, + "loss": 0.9655, + "step": 1434 + }, + { + "epoch": 0.10784608447317, + "grad_norm": 1.775593526763357, + "learning_rate": 3.936860854534611e-06, + "loss": 0.9608, + "step": 1435 + }, + { + "epoch": 0.10792123853900495, + "grad_norm": 2.536148699922561, + "learning_rate": 3.936739435189041e-06, + "loss": 1.0137, + "step": 1436 + }, + { + "epoch": 0.10799639260483992, + "grad_norm": 1.6783323277101405, + "learning_rate": 3.936617901084243e-06, + "loss": 0.9721, + "step": 1437 + }, + { + "epoch": 0.10807154667067488, + "grad_norm": 2.2789746786521126, + "learning_rate": 3.936496252227417e-06, + "loss": 1.0197, + "step": 1438 + }, + { + "epoch": 0.10814670073650985, + "grad_norm": 2.3097505462840724, + "learning_rate": 3.936374488625775e-06, + "loss": 1.0375, + "step": 1439 + }, + { + "epoch": 0.10822185480234481, + "grad_norm": 1.6193508844556528, + "learning_rate": 3.936252610286528e-06, + "loss": 1.0664, + "step": 1440 + }, + { + "epoch": 0.10829700886817976, + "grad_norm": 1.7779228859861116, + "learning_rate": 3.9361306172169005e-06, + "loss": 0.9514, + "step": 1441 + }, + { + "epoch": 0.10837216293401473, + "grad_norm": 1.929717337806228, + "learning_rate": 3.93600850942412e-06, + "loss": 0.8864, + "step": 1442 + }, + { + "epoch": 0.10844731699984969, + "grad_norm": 1.7326280886156968, + "learning_rate": 3.935886286915421e-06, + "loss": 0.9327, + "step": 1443 + }, + { + "epoch": 0.10852247106568465, + "grad_norm": 1.7284390707358874, + "learning_rate": 3.935763949698047e-06, + "loss": 0.9536, + "step": 1444 + }, + { + "epoch": 0.10859762513151962, + "grad_norm": 2.5253720264384265, + "learning_rate": 3.935641497779247e-06, + "loss": 1.0551, + "step": 1445 + }, + { + "epoch": 0.10867277919735457, + "grad_norm": 6.148165433843197, + "learning_rate": 3.935518931166275e-06, + "loss": 1.0178, + "step": 1446 + }, + { + "epoch": 0.10874793326318954, + "grad_norm": 1.612015419235507, + "learning_rate": 3.935396249866396e-06, + "loss": 0.9901, + "step": 1447 + }, + { + "epoch": 0.1088230873290245, + "grad_norm": 1.7417049680571635, + "learning_rate": 3.935273453886877e-06, + "loss": 1.0045, + "step": 1448 + }, + { + "epoch": 0.10889824139485946, + "grad_norm": 1.8411582575859493, + "learning_rate": 3.935150543234996e-06, + "loss": 0.9541, + "step": 1449 + }, + { + "epoch": 0.10897339546069443, + "grad_norm": 1.9349225035138984, + "learning_rate": 3.935027517918034e-06, + "loss": 1.0166, + "step": 1450 + }, + { + "epoch": 0.10904854952652938, + "grad_norm": 1.8364823851702605, + "learning_rate": 3.9349043779432825e-06, + "loss": 1.0292, + "step": 1451 + }, + { + "epoch": 0.10912370359236434, + "grad_norm": 2.0477999861247698, + "learning_rate": 3.934781123318037e-06, + "loss": 1.054, + "step": 1452 + }, + { + "epoch": 0.10919885765819931, + "grad_norm": 4.462916481393917, + "learning_rate": 3.934657754049602e-06, + "loss": 1.0071, + "step": 1453 + }, + { + "epoch": 0.10927401172403427, + "grad_norm": 4.461406302328673, + "learning_rate": 3.934534270145287e-06, + "loss": 1.0661, + "step": 1454 + }, + { + "epoch": 0.10934916578986924, + "grad_norm": 2.227663216481992, + "learning_rate": 3.934410671612408e-06, + "loss": 1.0604, + "step": 1455 + }, + { + "epoch": 0.1094243198557042, + "grad_norm": 1.9209712935392362, + "learning_rate": 3.934286958458289e-06, + "loss": 1.0633, + "step": 1456 + }, + { + "epoch": 0.10949947392153915, + "grad_norm": 1.7822167968482447, + "learning_rate": 3.934163130690262e-06, + "loss": 1.0314, + "step": 1457 + }, + { + "epoch": 0.10957462798737412, + "grad_norm": 2.0162872882381957, + "learning_rate": 3.9340391883156614e-06, + "loss": 0.95, + "step": 1458 + }, + { + "epoch": 0.10964978205320908, + "grad_norm": 2.0195767012006427, + "learning_rate": 3.933915131341834e-06, + "loss": 1.0553, + "step": 1459 + }, + { + "epoch": 0.10972493611904403, + "grad_norm": 3.8318448424813063, + "learning_rate": 3.93379095977613e-06, + "loss": 0.9467, + "step": 1460 + }, + { + "epoch": 0.109800090184879, + "grad_norm": 1.7803699260099155, + "learning_rate": 3.9336666736259055e-06, + "loss": 1.0461, + "step": 1461 + }, + { + "epoch": 0.10987524425071396, + "grad_norm": 1.7228648779518532, + "learning_rate": 3.933542272898527e-06, + "loss": 0.9949, + "step": 1462 + }, + { + "epoch": 0.10995039831654893, + "grad_norm": 2.478697617554925, + "learning_rate": 3.933417757601365e-06, + "loss": 0.9418, + "step": 1463 + }, + { + "epoch": 0.11002555238238389, + "grad_norm": 1.6806122828946182, + "learning_rate": 3.933293127741796e-06, + "loss": 0.9931, + "step": 1464 + }, + { + "epoch": 0.11010070644821884, + "grad_norm": 2.2302798643820503, + "learning_rate": 3.933168383327207e-06, + "loss": 1.0642, + "step": 1465 + }, + { + "epoch": 0.11017586051405381, + "grad_norm": 2.1407129868293837, + "learning_rate": 3.933043524364989e-06, + "loss": 1.0157, + "step": 1466 + }, + { + "epoch": 0.11025101457988877, + "grad_norm": 1.37841027176938, + "learning_rate": 3.932918550862539e-06, + "loss": 1.0023, + "step": 1467 + }, + { + "epoch": 0.11032616864572373, + "grad_norm": 1.9842193422317371, + "learning_rate": 3.932793462827265e-06, + "loss": 1.0403, + "step": 1468 + }, + { + "epoch": 0.1104013227115587, + "grad_norm": 1.7480305079541036, + "learning_rate": 3.932668260266576e-06, + "loss": 0.9944, + "step": 1469 + }, + { + "epoch": 0.11047647677739365, + "grad_norm": 6.650698643747976, + "learning_rate": 3.932542943187892e-06, + "loss": 1.0972, + "step": 1470 + }, + { + "epoch": 0.11055163084322862, + "grad_norm": 1.8507163130195983, + "learning_rate": 3.932417511598638e-06, + "loss": 1.067, + "step": 1471 + }, + { + "epoch": 0.11062678490906358, + "grad_norm": 2.1207007505973077, + "learning_rate": 3.932291965506247e-06, + "loss": 0.9699, + "step": 1472 + }, + { + "epoch": 0.11070193897489854, + "grad_norm": 2.084760240816251, + "learning_rate": 3.932166304918158e-06, + "loss": 0.9999, + "step": 1473 + }, + { + "epoch": 0.1107770930407335, + "grad_norm": 1.732979152635495, + "learning_rate": 3.9320405298418175e-06, + "loss": 1.0437, + "step": 1474 + }, + { + "epoch": 0.11085224710656846, + "grad_norm": 1.788589414989703, + "learning_rate": 3.931914640284676e-06, + "loss": 1.0056, + "step": 1475 + }, + { + "epoch": 0.11092740117240343, + "grad_norm": 1.7518878699220186, + "learning_rate": 3.931788636254195e-06, + "loss": 1.0545, + "step": 1476 + }, + { + "epoch": 0.11100255523823839, + "grad_norm": 1.6538315155630294, + "learning_rate": 3.931662517757839e-06, + "loss": 0.9664, + "step": 1477 + }, + { + "epoch": 0.11107770930407335, + "grad_norm": 1.8144542418970258, + "learning_rate": 3.931536284803083e-06, + "loss": 0.9763, + "step": 1478 + }, + { + "epoch": 0.11115286336990832, + "grad_norm": 2.593342025108122, + "learning_rate": 3.931409937397406e-06, + "loss": 1.101, + "step": 1479 + }, + { + "epoch": 0.11122801743574327, + "grad_norm": 0.7889992716922468, + "learning_rate": 3.931283475548293e-06, + "loss": 0.9056, + "step": 1480 + }, + { + "epoch": 0.11130317150157823, + "grad_norm": 1.7531608917859745, + "learning_rate": 3.93115689926324e-06, + "loss": 0.9781, + "step": 1481 + }, + { + "epoch": 0.1113783255674132, + "grad_norm": 1.4552786669690299, + "learning_rate": 3.931030208549745e-06, + "loss": 1.0575, + "step": 1482 + }, + { + "epoch": 0.11145347963324816, + "grad_norm": 1.849928895769531, + "learning_rate": 3.930903403415316e-06, + "loss": 1.0217, + "step": 1483 + }, + { + "epoch": 0.11152863369908313, + "grad_norm": 1.965562753343722, + "learning_rate": 3.930776483867467e-06, + "loss": 1.065, + "step": 1484 + }, + { + "epoch": 0.11160378776491808, + "grad_norm": 1.5596579071459864, + "learning_rate": 3.9306494499137175e-06, + "loss": 1.041, + "step": 1485 + }, + { + "epoch": 0.11167894183075304, + "grad_norm": 2.499932097466521, + "learning_rate": 3.930522301561595e-06, + "loss": 1.0114, + "step": 1486 + }, + { + "epoch": 0.11175409589658801, + "grad_norm": 1.5905980375811428, + "learning_rate": 3.930395038818633e-06, + "loss": 1.0202, + "step": 1487 + }, + { + "epoch": 0.11182924996242297, + "grad_norm": 2.195063675323977, + "learning_rate": 3.930267661692374e-06, + "loss": 1.0965, + "step": 1488 + }, + { + "epoch": 0.11190440402825792, + "grad_norm": 3.266210914141452, + "learning_rate": 3.930140170190364e-06, + "loss": 1.0193, + "step": 1489 + }, + { + "epoch": 0.1119795580940929, + "grad_norm": 0.9393370114460452, + "learning_rate": 3.930012564320159e-06, + "loss": 0.9141, + "step": 1490 + }, + { + "epoch": 0.11205471215992785, + "grad_norm": 1.4911395482410394, + "learning_rate": 3.929884844089318e-06, + "loss": 1.0193, + "step": 1491 + }, + { + "epoch": 0.11212986622576282, + "grad_norm": 1.4957078082334376, + "learning_rate": 3.92975700950541e-06, + "loss": 1.0652, + "step": 1492 + }, + { + "epoch": 0.11220502029159778, + "grad_norm": 1.5042506232391126, + "learning_rate": 3.92962906057601e-06, + "loss": 0.9211, + "step": 1493 + }, + { + "epoch": 0.11228017435743273, + "grad_norm": 2.635382329141628, + "learning_rate": 3.929500997308698e-06, + "loss": 1.0409, + "step": 1494 + }, + { + "epoch": 0.1123553284232677, + "grad_norm": 1.3338043851051886, + "learning_rate": 3.929372819711065e-06, + "loss": 1.0025, + "step": 1495 + }, + { + "epoch": 0.11243048248910266, + "grad_norm": 3.8636970566251096, + "learning_rate": 3.929244527790703e-06, + "loss": 0.9864, + "step": 1496 + }, + { + "epoch": 0.11250563655493762, + "grad_norm": 2.025358017720588, + "learning_rate": 3.929116121555216e-06, + "loss": 1.0348, + "step": 1497 + }, + { + "epoch": 0.11258079062077259, + "grad_norm": 1.9575149029862347, + "learning_rate": 3.928987601012212e-06, + "loss": 1.0957, + "step": 1498 + }, + { + "epoch": 0.11265594468660754, + "grad_norm": 1.8887597963999299, + "learning_rate": 3.928858966169306e-06, + "loss": 0.979, + "step": 1499 + }, + { + "epoch": 0.11273109875244251, + "grad_norm": 1.6381650375020198, + "learning_rate": 3.928730217034119e-06, + "loss": 1.0114, + "step": 1500 + }, + { + "epoch": 0.11280625281827747, + "grad_norm": 1.500763063256053, + "learning_rate": 3.928601353614282e-06, + "loss": 0.9391, + "step": 1501 + }, + { + "epoch": 0.11288140688411243, + "grad_norm": 3.8633733262062058, + "learning_rate": 3.92847237591743e-06, + "loss": 1.0018, + "step": 1502 + }, + { + "epoch": 0.1129565609499474, + "grad_norm": 1.4470510388055016, + "learning_rate": 3.928343283951204e-06, + "loss": 1.0291, + "step": 1503 + }, + { + "epoch": 0.11303171501578235, + "grad_norm": 1.9503700052031268, + "learning_rate": 3.928214077723255e-06, + "loss": 1.0215, + "step": 1504 + }, + { + "epoch": 0.11310686908161731, + "grad_norm": 2.028032424765358, + "learning_rate": 3.928084757241239e-06, + "loss": 1.0201, + "step": 1505 + }, + { + "epoch": 0.11318202314745228, + "grad_norm": 4.249841799316048, + "learning_rate": 3.9279553225128165e-06, + "loss": 0.8249, + "step": 1506 + }, + { + "epoch": 0.11325717721328724, + "grad_norm": 2.2299081395135865, + "learning_rate": 3.92782577354566e-06, + "loss": 0.9698, + "step": 1507 + }, + { + "epoch": 0.1133323312791222, + "grad_norm": 1.9783022013236409, + "learning_rate": 3.927696110347443e-06, + "loss": 1.0256, + "step": 1508 + }, + { + "epoch": 0.11340748534495716, + "grad_norm": 2.5203363123015374, + "learning_rate": 3.92756633292585e-06, + "loss": 1.0264, + "step": 1509 + }, + { + "epoch": 0.11348263941079212, + "grad_norm": 19.199527512140843, + "learning_rate": 3.927436441288571e-06, + "loss": 1.0856, + "step": 1510 + }, + { + "epoch": 0.11355779347662709, + "grad_norm": 1.7573071919109486, + "learning_rate": 3.9273064354433025e-06, + "loss": 0.9556, + "step": 1511 + }, + { + "epoch": 0.11363294754246205, + "grad_norm": 1.560176113042334, + "learning_rate": 3.927176315397747e-06, + "loss": 1.0593, + "step": 1512 + }, + { + "epoch": 0.113708101608297, + "grad_norm": 0.6891485087029505, + "learning_rate": 3.927046081159615e-06, + "loss": 0.7971, + "step": 1513 + }, + { + "epoch": 0.11378325567413197, + "grad_norm": 2.9522217676138998, + "learning_rate": 3.926915732736624e-06, + "loss": 0.8889, + "step": 1514 + }, + { + "epoch": 0.11385840973996693, + "grad_norm": 2.623563009852956, + "learning_rate": 3.926785270136497e-06, + "loss": 1.0611, + "step": 1515 + }, + { + "epoch": 0.1139335638058019, + "grad_norm": 2.125249399247694, + "learning_rate": 3.926654693366965e-06, + "loss": 0.9284, + "step": 1516 + }, + { + "epoch": 0.11400871787163686, + "grad_norm": 1.8435095290787953, + "learning_rate": 3.926524002435764e-06, + "loss": 1.0466, + "step": 1517 + }, + { + "epoch": 0.11408387193747181, + "grad_norm": 2.0190135063411145, + "learning_rate": 3.9263931973506395e-06, + "loss": 0.9329, + "step": 1518 + }, + { + "epoch": 0.11415902600330678, + "grad_norm": 2.336336235793844, + "learning_rate": 3.926262278119341e-06, + "loss": 1.0736, + "step": 1519 + }, + { + "epoch": 0.11423418006914174, + "grad_norm": 2.250425298549816, + "learning_rate": 3.9261312447496265e-06, + "loss": 1.0985, + "step": 1520 + }, + { + "epoch": 0.11430933413497671, + "grad_norm": 2.2613573592697023, + "learning_rate": 3.92600009724926e-06, + "loss": 1.0303, + "step": 1521 + }, + { + "epoch": 0.11438448820081167, + "grad_norm": 2.1630336918775135, + "learning_rate": 3.925868835626012e-06, + "loss": 1.073, + "step": 1522 + }, + { + "epoch": 0.11445964226664662, + "grad_norm": 1.693364162064312, + "learning_rate": 3.925737459887662e-06, + "loss": 1.0012, + "step": 1523 + }, + { + "epoch": 0.11453479633248159, + "grad_norm": 1.990587796472921, + "learning_rate": 3.925605970041992e-06, + "loss": 1.0724, + "step": 1524 + }, + { + "epoch": 0.11460995039831655, + "grad_norm": 1.3002775207816637, + "learning_rate": 3.925474366096796e-06, + "loss": 1.007, + "step": 1525 + }, + { + "epoch": 0.1146851044641515, + "grad_norm": 1.5746563097451425, + "learning_rate": 3.92534264805987e-06, + "loss": 1.1212, + "step": 1526 + }, + { + "epoch": 0.11476025852998648, + "grad_norm": 1.5777961463858918, + "learning_rate": 3.92521081593902e-06, + "loss": 1.0753, + "step": 1527 + }, + { + "epoch": 0.11483541259582143, + "grad_norm": 1.5270630561485055, + "learning_rate": 3.925078869742056e-06, + "loss": 0.9918, + "step": 1528 + }, + { + "epoch": 0.1149105666616564, + "grad_norm": 1.5051811066203318, + "learning_rate": 3.924946809476798e-06, + "loss": 0.9554, + "step": 1529 + }, + { + "epoch": 0.11498572072749136, + "grad_norm": 2.4073031214055405, + "learning_rate": 3.924814635151071e-06, + "loss": 1.0314, + "step": 1530 + }, + { + "epoch": 0.11506087479332631, + "grad_norm": 2.1211891785119916, + "learning_rate": 3.924682346772705e-06, + "loss": 1.0918, + "step": 1531 + }, + { + "epoch": 0.11513602885916129, + "grad_norm": 1.7518099553053226, + "learning_rate": 3.92454994434954e-06, + "loss": 1.1059, + "step": 1532 + }, + { + "epoch": 0.11521118292499624, + "grad_norm": 2.1973392999801056, + "learning_rate": 3.9244174278894226e-06, + "loss": 0.9988, + "step": 1533 + }, + { + "epoch": 0.1152863369908312, + "grad_norm": 1.5346866889319182, + "learning_rate": 3.924284797400202e-06, + "loss": 1.04, + "step": 1534 + }, + { + "epoch": 0.11536149105666617, + "grad_norm": 1.762021463780865, + "learning_rate": 3.92415205288974e-06, + "loss": 1.0372, + "step": 1535 + }, + { + "epoch": 0.11543664512250112, + "grad_norm": 2.031233215262683, + "learning_rate": 3.9240191943659e-06, + "loss": 1.0197, + "step": 1536 + }, + { + "epoch": 0.1155117991883361, + "grad_norm": 2.0255750757907727, + "learning_rate": 3.923886221836555e-06, + "loss": 0.991, + "step": 1537 + }, + { + "epoch": 0.11558695325417105, + "grad_norm": 1.366094359304272, + "learning_rate": 3.923753135309584e-06, + "loss": 1.1158, + "step": 1538 + }, + { + "epoch": 0.11566210732000601, + "grad_norm": 1.7932503007553233, + "learning_rate": 3.923619934792873e-06, + "loss": 1.1092, + "step": 1539 + }, + { + "epoch": 0.11573726138584098, + "grad_norm": 1.974354471870977, + "learning_rate": 3.923486620294316e-06, + "loss": 1.0212, + "step": 1540 + }, + { + "epoch": 0.11581241545167593, + "grad_norm": 2.073558292736832, + "learning_rate": 3.923353191821811e-06, + "loss": 1.0006, + "step": 1541 + }, + { + "epoch": 0.11588756951751089, + "grad_norm": 2.0714025542781678, + "learning_rate": 3.923219649383264e-06, + "loss": 1.0196, + "step": 1542 + }, + { + "epoch": 0.11596272358334586, + "grad_norm": 1.8766544672070908, + "learning_rate": 3.923085992986588e-06, + "loss": 1.033, + "step": 1543 + }, + { + "epoch": 0.11603787764918082, + "grad_norm": 2.612857618906171, + "learning_rate": 3.922952222639703e-06, + "loss": 1.0461, + "step": 1544 + }, + { + "epoch": 0.11611303171501579, + "grad_norm": 1.834532959357868, + "learning_rate": 3.922818338350536e-06, + "loss": 1.1124, + "step": 1545 + }, + { + "epoch": 0.11618818578085074, + "grad_norm": 2.098474666127147, + "learning_rate": 3.9226843401270195e-06, + "loss": 1.0155, + "step": 1546 + }, + { + "epoch": 0.1162633398466857, + "grad_norm": 1.4984555240378445, + "learning_rate": 3.922550227977093e-06, + "loss": 1.0466, + "step": 1547 + }, + { + "epoch": 0.11633849391252067, + "grad_norm": 0.7817277973146349, + "learning_rate": 3.9224160019087036e-06, + "loss": 0.8202, + "step": 1548 + }, + { + "epoch": 0.11641364797835563, + "grad_norm": 1.920121051628067, + "learning_rate": 3.922281661929804e-06, + "loss": 0.9532, + "step": 1549 + }, + { + "epoch": 0.11648880204419058, + "grad_norm": 1.7043614230368815, + "learning_rate": 3.922147208048356e-06, + "loss": 0.9816, + "step": 1550 + }, + { + "epoch": 0.11656395611002555, + "grad_norm": 4.255201410531522, + "learning_rate": 3.922012640272325e-06, + "loss": 1.0814, + "step": 1551 + }, + { + "epoch": 0.11663911017586051, + "grad_norm": 1.7879844649235241, + "learning_rate": 3.921877958609685e-06, + "loss": 1.041, + "step": 1552 + }, + { + "epoch": 0.11671426424169548, + "grad_norm": 4.833420500572195, + "learning_rate": 3.9217431630684174e-06, + "loss": 1.1378, + "step": 1553 + }, + { + "epoch": 0.11678941830753044, + "grad_norm": 2.4950805900263657, + "learning_rate": 3.921608253656508e-06, + "loss": 1.0143, + "step": 1554 + }, + { + "epoch": 0.1168645723733654, + "grad_norm": 1.9735216112251048, + "learning_rate": 3.921473230381951e-06, + "loss": 0.9609, + "step": 1555 + }, + { + "epoch": 0.11693972643920036, + "grad_norm": 1.8451634178051473, + "learning_rate": 3.921338093252748e-06, + "loss": 0.9754, + "step": 1556 + }, + { + "epoch": 0.11701488050503532, + "grad_norm": 1.5446185985017942, + "learning_rate": 3.921202842276906e-06, + "loss": 0.999, + "step": 1557 + }, + { + "epoch": 0.11709003457087028, + "grad_norm": 2.1405474794927795, + "learning_rate": 3.921067477462437e-06, + "loss": 0.9512, + "step": 1558 + }, + { + "epoch": 0.11716518863670525, + "grad_norm": 0.815409732049048, + "learning_rate": 3.920931998817365e-06, + "loss": 0.8928, + "step": 1559 + }, + { + "epoch": 0.1172403427025402, + "grad_norm": 1.3889247057852918, + "learning_rate": 3.920796406349717e-06, + "loss": 1.059, + "step": 1560 + }, + { + "epoch": 0.11731549676837517, + "grad_norm": 1.8028881796777168, + "learning_rate": 3.920660700067525e-06, + "loss": 1.0452, + "step": 1561 + }, + { + "epoch": 0.11739065083421013, + "grad_norm": 2.8625985407630274, + "learning_rate": 3.920524879978833e-06, + "loss": 1.0195, + "step": 1562 + }, + { + "epoch": 0.11746580490004509, + "grad_norm": 2.0300197690555497, + "learning_rate": 3.920388946091687e-06, + "loss": 0.9345, + "step": 1563 + }, + { + "epoch": 0.11754095896588006, + "grad_norm": 1.5632545937892048, + "learning_rate": 3.920252898414143e-06, + "loss": 1.0962, + "step": 1564 + }, + { + "epoch": 0.11761611303171501, + "grad_norm": 3.0799293457943886, + "learning_rate": 3.920116736954261e-06, + "loss": 0.9442, + "step": 1565 + }, + { + "epoch": 0.11769126709754998, + "grad_norm": 1.6592547673060212, + "learning_rate": 3.91998046172011e-06, + "loss": 0.9614, + "step": 1566 + }, + { + "epoch": 0.11776642116338494, + "grad_norm": 3.1014130601986416, + "learning_rate": 3.9198440727197645e-06, + "loss": 1.065, + "step": 1567 + }, + { + "epoch": 0.1178415752292199, + "grad_norm": 2.0272821258998865, + "learning_rate": 3.919707569961306e-06, + "loss": 1.0242, + "step": 1568 + }, + { + "epoch": 0.11791672929505487, + "grad_norm": 1.6141140245483818, + "learning_rate": 3.9195709534528235e-06, + "loss": 1.0901, + "step": 1569 + }, + { + "epoch": 0.11799188336088982, + "grad_norm": 2.280611967978404, + "learning_rate": 3.919434223202411e-06, + "loss": 0.9718, + "step": 1570 + }, + { + "epoch": 0.11806703742672478, + "grad_norm": 1.7504192939901426, + "learning_rate": 3.919297379218171e-06, + "loss": 1.1077, + "step": 1571 + }, + { + "epoch": 0.11814219149255975, + "grad_norm": 2.7419843027846693, + "learning_rate": 3.919160421508211e-06, + "loss": 1.0183, + "step": 1572 + }, + { + "epoch": 0.1182173455583947, + "grad_norm": 2.0278506665070473, + "learning_rate": 3.919023350080648e-06, + "loss": 1.075, + "step": 1573 + }, + { + "epoch": 0.11829249962422968, + "grad_norm": 1.7040553920142192, + "learning_rate": 3.918886164943603e-06, + "loss": 1.0759, + "step": 1574 + }, + { + "epoch": 0.11836765369006463, + "grad_norm": 1.8207515465403636, + "learning_rate": 3.918748866105204e-06, + "loss": 1.0874, + "step": 1575 + }, + { + "epoch": 0.11844280775589959, + "grad_norm": 2.536619918867028, + "learning_rate": 3.918611453573589e-06, + "loss": 0.9622, + "step": 1576 + }, + { + "epoch": 0.11851796182173456, + "grad_norm": 1.8036931588990823, + "learning_rate": 3.918473927356896e-06, + "loss": 0.9792, + "step": 1577 + }, + { + "epoch": 0.11859311588756952, + "grad_norm": 1.8505071537690723, + "learning_rate": 3.918336287463279e-06, + "loss": 0.9664, + "step": 1578 + }, + { + "epoch": 0.11866826995340447, + "grad_norm": 2.118013227974394, + "learning_rate": 3.9181985339008895e-06, + "loss": 1.1197, + "step": 1579 + }, + { + "epoch": 0.11874342401923944, + "grad_norm": 2.648676768609706, + "learning_rate": 3.918060666677892e-06, + "loss": 1.0673, + "step": 1580 + }, + { + "epoch": 0.1188185780850744, + "grad_norm": 0.8073935700776395, + "learning_rate": 3.9179226858024555e-06, + "loss": 0.885, + "step": 1581 + }, + { + "epoch": 0.11889373215090937, + "grad_norm": 1.5593052819822046, + "learning_rate": 3.917784591282756e-06, + "loss": 0.9423, + "step": 1582 + }, + { + "epoch": 0.11896888621674433, + "grad_norm": 2.5623404290362517, + "learning_rate": 3.917646383126975e-06, + "loss": 0.9956, + "step": 1583 + }, + { + "epoch": 0.11904404028257928, + "grad_norm": 2.045896456405664, + "learning_rate": 3.917508061343303e-06, + "loss": 1.1131, + "step": 1584 + }, + { + "epoch": 0.11911919434841425, + "grad_norm": 2.4753510312921065, + "learning_rate": 3.917369625939936e-06, + "loss": 1.0729, + "step": 1585 + }, + { + "epoch": 0.11919434841424921, + "grad_norm": 1.7473119799927046, + "learning_rate": 3.917231076925076e-06, + "loss": 1.0212, + "step": 1586 + }, + { + "epoch": 0.11926950248008417, + "grad_norm": 1.8696887130516588, + "learning_rate": 3.917092414306933e-06, + "loss": 1.042, + "step": 1587 + }, + { + "epoch": 0.11934465654591914, + "grad_norm": 2.0034001772874097, + "learning_rate": 3.916953638093725e-06, + "loss": 0.9492, + "step": 1588 + }, + { + "epoch": 0.1194198106117541, + "grad_norm": 6.318449030952001, + "learning_rate": 3.9168147482936715e-06, + "loss": 1.0109, + "step": 1589 + }, + { + "epoch": 0.11949496467758906, + "grad_norm": 2.1080420909856006, + "learning_rate": 3.916675744915005e-06, + "loss": 0.9599, + "step": 1590 + }, + { + "epoch": 0.11957011874342402, + "grad_norm": 1.8162245792886158, + "learning_rate": 3.916536627965961e-06, + "loss": 0.9304, + "step": 1591 + }, + { + "epoch": 0.11964527280925898, + "grad_norm": 1.8633067131301935, + "learning_rate": 3.916397397454783e-06, + "loss": 1.0036, + "step": 1592 + }, + { + "epoch": 0.11972042687509395, + "grad_norm": 1.6207742031396188, + "learning_rate": 3.916258053389721e-06, + "loss": 0.9791, + "step": 1593 + }, + { + "epoch": 0.1197955809409289, + "grad_norm": 3.6308294318628147, + "learning_rate": 3.916118595779031e-06, + "loss": 1.0042, + "step": 1594 + }, + { + "epoch": 0.11987073500676386, + "grad_norm": 1.7580994096365623, + "learning_rate": 3.915979024630977e-06, + "loss": 1.0208, + "step": 1595 + }, + { + "epoch": 0.11994588907259883, + "grad_norm": 2.242936278466572, + "learning_rate": 3.91583933995383e-06, + "loss": 1.0314, + "step": 1596 + }, + { + "epoch": 0.12002104313843379, + "grad_norm": 1.7404761012715142, + "learning_rate": 3.915699541755865e-06, + "loss": 1.0088, + "step": 1597 + }, + { + "epoch": 0.12009619720426876, + "grad_norm": 2.3226843633660064, + "learning_rate": 3.915559630045367e-06, + "loss": 0.981, + "step": 1598 + }, + { + "epoch": 0.12017135127010371, + "grad_norm": 2.232142687116344, + "learning_rate": 3.9154196048306244e-06, + "loss": 1.0765, + "step": 1599 + }, + { + "epoch": 0.12024650533593867, + "grad_norm": 1.827768731247848, + "learning_rate": 3.915279466119937e-06, + "loss": 1.0457, + "step": 1600 + }, + { + "epoch": 0.12032165940177364, + "grad_norm": 1.6400751554899908, + "learning_rate": 3.915139213921606e-06, + "loss": 0.9866, + "step": 1601 + }, + { + "epoch": 0.1203968134676086, + "grad_norm": 1.953320851996757, + "learning_rate": 3.914998848243944e-06, + "loss": 0.9318, + "step": 1602 + }, + { + "epoch": 0.12047196753344355, + "grad_norm": 2.2668513694942978, + "learning_rate": 3.914858369095267e-06, + "loss": 1.0299, + "step": 1603 + }, + { + "epoch": 0.12054712159927852, + "grad_norm": 2.235870607709705, + "learning_rate": 3.914717776483899e-06, + "loss": 1.0066, + "step": 1604 + }, + { + "epoch": 0.12062227566511348, + "grad_norm": 2.4317687709394087, + "learning_rate": 3.9145770704181715e-06, + "loss": 1.0672, + "step": 1605 + }, + { + "epoch": 0.12069742973094845, + "grad_norm": 2.0285790121869307, + "learning_rate": 3.9144362509064194e-06, + "loss": 0.9968, + "step": 1606 + }, + { + "epoch": 0.1207725837967834, + "grad_norm": 1.4447663796138799, + "learning_rate": 3.91429531795699e-06, + "loss": 0.9496, + "step": 1607 + }, + { + "epoch": 0.12084773786261836, + "grad_norm": 2.0639381018901313, + "learning_rate": 3.9141542715782325e-06, + "loss": 0.9242, + "step": 1608 + }, + { + "epoch": 0.12092289192845333, + "grad_norm": 1.7781841596741161, + "learning_rate": 3.9140131117785045e-06, + "loss": 0.9779, + "step": 1609 + }, + { + "epoch": 0.12099804599428829, + "grad_norm": 1.8303257663709103, + "learning_rate": 3.91387183856617e-06, + "loss": 1.0182, + "step": 1610 + }, + { + "epoch": 0.12107320006012326, + "grad_norm": 1.6683494258540195, + "learning_rate": 3.913730451949601e-06, + "loss": 0.9682, + "step": 1611 + }, + { + "epoch": 0.12114835412595822, + "grad_norm": 2.0246399359187763, + "learning_rate": 3.913588951937174e-06, + "loss": 0.9375, + "step": 1612 + }, + { + "epoch": 0.12122350819179317, + "grad_norm": 0.7449704646970067, + "learning_rate": 3.913447338537274e-06, + "loss": 0.946, + "step": 1613 + }, + { + "epoch": 0.12129866225762814, + "grad_norm": 2.044251019489136, + "learning_rate": 3.913305611758292e-06, + "loss": 0.9448, + "step": 1614 + }, + { + "epoch": 0.1213738163234631, + "grad_norm": 1.7229758591114772, + "learning_rate": 3.913163771608627e-06, + "loss": 0.9335, + "step": 1615 + }, + { + "epoch": 0.12144897038929806, + "grad_norm": 1.5149799035882843, + "learning_rate": 3.913021818096682e-06, + "loss": 1.0425, + "step": 1616 + }, + { + "epoch": 0.12152412445513303, + "grad_norm": 1.4170228099077404, + "learning_rate": 3.912879751230868e-06, + "loss": 0.9321, + "step": 1617 + }, + { + "epoch": 0.12159927852096798, + "grad_norm": 2.0123936974545233, + "learning_rate": 3.9127375710196044e-06, + "loss": 1.0297, + "step": 1618 + }, + { + "epoch": 0.12167443258680295, + "grad_norm": 1.9812583177073038, + "learning_rate": 3.912595277471316e-06, + "loss": 0.9247, + "step": 1619 + }, + { + "epoch": 0.12174958665263791, + "grad_norm": 4.952706207958295, + "learning_rate": 3.912452870594433e-06, + "loss": 1.0416, + "step": 1620 + }, + { + "epoch": 0.12182474071847287, + "grad_norm": 1.5705810620878857, + "learning_rate": 3.912310350397394e-06, + "loss": 1.0291, + "step": 1621 + }, + { + "epoch": 0.12189989478430784, + "grad_norm": 1.8614215006828372, + "learning_rate": 3.912167716888644e-06, + "loss": 1.1328, + "step": 1622 + }, + { + "epoch": 0.12197504885014279, + "grad_norm": 1.8224128099471917, + "learning_rate": 3.912024970076636e-06, + "loss": 1.0264, + "step": 1623 + }, + { + "epoch": 0.12205020291597775, + "grad_norm": 1.5353021528103823, + "learning_rate": 3.911882109969825e-06, + "loss": 1.0499, + "step": 1624 + }, + { + "epoch": 0.12212535698181272, + "grad_norm": 2.02971426789779, + "learning_rate": 3.9117391365766785e-06, + "loss": 1.0066, + "step": 1625 + }, + { + "epoch": 0.12220051104764768, + "grad_norm": 1.8120563227129447, + "learning_rate": 3.9115960499056674e-06, + "loss": 1.0321, + "step": 1626 + }, + { + "epoch": 0.12227566511348265, + "grad_norm": 1.823704126211189, + "learning_rate": 3.911452849965271e-06, + "loss": 1.0181, + "step": 1627 + }, + { + "epoch": 0.1223508191793176, + "grad_norm": 5.292729829132548, + "learning_rate": 3.911309536763974e-06, + "loss": 0.9705, + "step": 1628 + }, + { + "epoch": 0.12242597324515256, + "grad_norm": 4.357080436813317, + "learning_rate": 3.911166110310267e-06, + "loss": 1.0174, + "step": 1629 + }, + { + "epoch": 0.12250112731098753, + "grad_norm": 1.7094402293812874, + "learning_rate": 3.91102257061265e-06, + "loss": 1.0963, + "step": 1630 + }, + { + "epoch": 0.12257628137682249, + "grad_norm": 1.7832850158539337, + "learning_rate": 3.9108789176796285e-06, + "loss": 1.0179, + "step": 1631 + }, + { + "epoch": 0.12265143544265744, + "grad_norm": 2.15198515243088, + "learning_rate": 3.910735151519713e-06, + "loss": 1.0614, + "step": 1632 + }, + { + "epoch": 0.12272658950849241, + "grad_norm": 1.8205927031481328, + "learning_rate": 3.910591272141424e-06, + "loss": 0.9567, + "step": 1633 + }, + { + "epoch": 0.12280174357432737, + "grad_norm": 1.8329600113015, + "learning_rate": 3.910447279553285e-06, + "loss": 1.0269, + "step": 1634 + }, + { + "epoch": 0.12287689764016234, + "grad_norm": 1.8845944657953848, + "learning_rate": 3.91030317376383e-06, + "loss": 1.0348, + "step": 1635 + }, + { + "epoch": 0.1229520517059973, + "grad_norm": 0.6901600917012282, + "learning_rate": 3.9101589547815965e-06, + "loss": 0.8229, + "step": 1636 + }, + { + "epoch": 0.12302720577183225, + "grad_norm": 1.6562648268701763, + "learning_rate": 3.91001462261513e-06, + "loss": 1.0513, + "step": 1637 + }, + { + "epoch": 0.12310235983766722, + "grad_norm": 0.7558168699945264, + "learning_rate": 3.909870177272984e-06, + "loss": 0.845, + "step": 1638 + }, + { + "epoch": 0.12317751390350218, + "grad_norm": 7.04051202039151, + "learning_rate": 3.909725618763716e-06, + "loss": 0.918, + "step": 1639 + }, + { + "epoch": 0.12325266796933713, + "grad_norm": 1.6045268096253165, + "learning_rate": 3.909580947095892e-06, + "loss": 0.9708, + "step": 1640 + }, + { + "epoch": 0.1233278220351721, + "grad_norm": 1.4819048757771367, + "learning_rate": 3.909436162278085e-06, + "loss": 1.0244, + "step": 1641 + }, + { + "epoch": 0.12340297610100706, + "grad_norm": 2.4597716461200028, + "learning_rate": 3.9092912643188745e-06, + "loss": 1.0277, + "step": 1642 + }, + { + "epoch": 0.12347813016684203, + "grad_norm": 2.913889371009046, + "learning_rate": 3.909146253226844e-06, + "loss": 1.0809, + "step": 1643 + }, + { + "epoch": 0.12355328423267699, + "grad_norm": 2.0376728335602943, + "learning_rate": 3.909001129010588e-06, + "loss": 1.069, + "step": 1644 + }, + { + "epoch": 0.12362843829851194, + "grad_norm": 1.6767209848259819, + "learning_rate": 3.908855891678706e-06, + "loss": 1.0601, + "step": 1645 + }, + { + "epoch": 0.12370359236434691, + "grad_norm": 2.4920176863739227, + "learning_rate": 3.908710541239802e-06, + "loss": 1.1196, + "step": 1646 + }, + { + "epoch": 0.12377874643018187, + "grad_norm": 2.279494040884507, + "learning_rate": 3.90856507770249e-06, + "loss": 0.9601, + "step": 1647 + }, + { + "epoch": 0.12385390049601683, + "grad_norm": 1.4189968112912417, + "learning_rate": 3.908419501075388e-06, + "loss": 0.9205, + "step": 1648 + }, + { + "epoch": 0.1239290545618518, + "grad_norm": 1.7257506934690512, + "learning_rate": 3.908273811367123e-06, + "loss": 0.9742, + "step": 1649 + }, + { + "epoch": 0.12400420862768675, + "grad_norm": 1.5769271837113028, + "learning_rate": 3.908128008586328e-06, + "loss": 0.9334, + "step": 1650 + }, + { + "epoch": 0.12407936269352172, + "grad_norm": 2.2558038878119113, + "learning_rate": 3.90798209274164e-06, + "loss": 0.8855, + "step": 1651 + }, + { + "epoch": 0.12415451675935668, + "grad_norm": 1.9390916746835742, + "learning_rate": 3.907836063841709e-06, + "loss": 0.9773, + "step": 1652 + }, + { + "epoch": 0.12422967082519164, + "grad_norm": 1.6741836152548772, + "learning_rate": 3.907689921895184e-06, + "loss": 0.9933, + "step": 1653 + }, + { + "epoch": 0.12430482489102661, + "grad_norm": 6.5920711748484315, + "learning_rate": 3.9075436669107265e-06, + "loss": 0.9789, + "step": 1654 + }, + { + "epoch": 0.12437997895686156, + "grad_norm": 5.997675445394742, + "learning_rate": 3.907397298897003e-06, + "loss": 0.9628, + "step": 1655 + }, + { + "epoch": 0.12445513302269653, + "grad_norm": 1.8502117654380354, + "learning_rate": 3.907250817862685e-06, + "loss": 0.9535, + "step": 1656 + }, + { + "epoch": 0.12453028708853149, + "grad_norm": 1.5560349802287334, + "learning_rate": 3.907104223816453e-06, + "loss": 0.9796, + "step": 1657 + }, + { + "epoch": 0.12460544115436645, + "grad_norm": 1.6580741481697547, + "learning_rate": 3.906957516766993e-06, + "loss": 1.0003, + "step": 1658 + }, + { + "epoch": 0.12468059522020142, + "grad_norm": 1.4873058096166976, + "learning_rate": 3.906810696722997e-06, + "loss": 1.0454, + "step": 1659 + }, + { + "epoch": 0.12475574928603637, + "grad_norm": 2.330842755211927, + "learning_rate": 3.906663763693167e-06, + "loss": 0.9602, + "step": 1660 + }, + { + "epoch": 0.12483090335187133, + "grad_norm": 1.5720114555815081, + "learning_rate": 3.906516717686207e-06, + "loss": 1.0715, + "step": 1661 + }, + { + "epoch": 0.1249060574177063, + "grad_norm": 5.890825101251535, + "learning_rate": 3.906369558710831e-06, + "loss": 0.9676, + "step": 1662 + }, + { + "epoch": 0.12498121148354126, + "grad_norm": 1.9640783559664452, + "learning_rate": 3.906222286775759e-06, + "loss": 0.9607, + "step": 1663 + }, + { + "epoch": 0.1250563655493762, + "grad_norm": 1.6792474924454701, + "learning_rate": 3.906074901889717e-06, + "loss": 0.9961, + "step": 1664 + }, + { + "epoch": 0.12513151961521118, + "grad_norm": 1.5826110530895858, + "learning_rate": 3.905927404061439e-06, + "loss": 1.0631, + "step": 1665 + }, + { + "epoch": 0.12520667368104615, + "grad_norm": 2.2439575651851813, + "learning_rate": 3.905779793299662e-06, + "loss": 0.999, + "step": 1666 + }, + { + "epoch": 0.1252818277468811, + "grad_norm": 0.9657344953559841, + "learning_rate": 3.905632069613136e-06, + "loss": 0.8753, + "step": 1667 + }, + { + "epoch": 0.12535698181271607, + "grad_norm": 2.2022158864047032, + "learning_rate": 3.9054842330106125e-06, + "loss": 1.0438, + "step": 1668 + }, + { + "epoch": 0.12543213587855104, + "grad_norm": 1.9124427637249501, + "learning_rate": 3.9053362835008516e-06, + "loss": 1.0499, + "step": 1669 + }, + { + "epoch": 0.12550728994438598, + "grad_norm": 2.2346372650521746, + "learning_rate": 3.9051882210926195e-06, + "loss": 0.8982, + "step": 1670 + }, + { + "epoch": 0.12558244401022095, + "grad_norm": 3.574930155678703, + "learning_rate": 3.90504004579469e-06, + "loss": 1.0022, + "step": 1671 + }, + { + "epoch": 0.12565759807605592, + "grad_norm": 1.7569819543961545, + "learning_rate": 3.904891757615843e-06, + "loss": 1.0118, + "step": 1672 + }, + { + "epoch": 0.12573275214189086, + "grad_norm": 2.2885619778282353, + "learning_rate": 3.904743356564865e-06, + "loss": 1.0073, + "step": 1673 + }, + { + "epoch": 0.12580790620772583, + "grad_norm": 1.7805788214531106, + "learning_rate": 3.90459484265055e-06, + "loss": 1.039, + "step": 1674 + }, + { + "epoch": 0.1258830602735608, + "grad_norm": 1.8271012659472383, + "learning_rate": 3.904446215881697e-06, + "loss": 1.015, + "step": 1675 + }, + { + "epoch": 0.12595821433939577, + "grad_norm": 1.2873472882414452, + "learning_rate": 3.9042974762671125e-06, + "loss": 1.0278, + "step": 1676 + }, + { + "epoch": 0.12603336840523072, + "grad_norm": 1.8683495516701236, + "learning_rate": 3.904148623815611e-06, + "loss": 1.0874, + "step": 1677 + }, + { + "epoch": 0.1261085224710657, + "grad_norm": 1.8177258364063353, + "learning_rate": 3.903999658536012e-06, + "loss": 1.0307, + "step": 1678 + }, + { + "epoch": 0.12618367653690066, + "grad_norm": 1.376417123346926, + "learning_rate": 3.903850580437142e-06, + "loss": 1.1225, + "step": 1679 + }, + { + "epoch": 0.1262588306027356, + "grad_norm": 1.8776337246060735, + "learning_rate": 3.903701389527836e-06, + "loss": 1.0524, + "step": 1680 + }, + { + "epoch": 0.12633398466857057, + "grad_norm": 2.653158026356022, + "learning_rate": 3.903552085816932e-06, + "loss": 1.0063, + "step": 1681 + }, + { + "epoch": 0.12640913873440554, + "grad_norm": 2.4538974396244755, + "learning_rate": 3.903402669313278e-06, + "loss": 1.1229, + "step": 1682 + }, + { + "epoch": 0.12648429280024048, + "grad_norm": 1.847558049179151, + "learning_rate": 3.903253140025726e-06, + "loss": 0.9623, + "step": 1683 + }, + { + "epoch": 0.12655944686607545, + "grad_norm": 1.711617182369511, + "learning_rate": 3.9031034979631385e-06, + "loss": 0.9882, + "step": 1684 + }, + { + "epoch": 0.12663460093191042, + "grad_norm": 1.8815094803716492, + "learning_rate": 3.902953743134381e-06, + "loss": 1.0196, + "step": 1685 + }, + { + "epoch": 0.12670975499774537, + "grad_norm": 2.391441754211848, + "learning_rate": 3.9028038755483275e-06, + "loss": 1.0192, + "step": 1686 + }, + { + "epoch": 0.12678490906358034, + "grad_norm": 1.6558835325933385, + "learning_rate": 3.902653895213858e-06, + "loss": 1.1272, + "step": 1687 + }, + { + "epoch": 0.1268600631294153, + "grad_norm": 2.284617199041184, + "learning_rate": 3.90250380213986e-06, + "loss": 1.0004, + "step": 1688 + }, + { + "epoch": 0.12693521719525025, + "grad_norm": 2.0164460868107543, + "learning_rate": 3.902353596335225e-06, + "loss": 1.0105, + "step": 1689 + }, + { + "epoch": 0.12701037126108522, + "grad_norm": 2.110040969270097, + "learning_rate": 3.902203277808856e-06, + "loss": 1.0483, + "step": 1690 + }, + { + "epoch": 0.1270855253269202, + "grad_norm": 1.8809505135548195, + "learning_rate": 3.902052846569659e-06, + "loss": 1.0805, + "step": 1691 + }, + { + "epoch": 0.12716067939275516, + "grad_norm": 2.083003577520015, + "learning_rate": 3.901902302626547e-06, + "loss": 0.9701, + "step": 1692 + }, + { + "epoch": 0.1272358334585901, + "grad_norm": 1.7575818397925957, + "learning_rate": 3.901751645988441e-06, + "loss": 0.9959, + "step": 1693 + }, + { + "epoch": 0.12731098752442507, + "grad_norm": 3.735370838389807, + "learning_rate": 3.901600876664267e-06, + "loss": 1.0419, + "step": 1694 + }, + { + "epoch": 0.12738614159026004, + "grad_norm": 1.992698633782217, + "learning_rate": 3.9014499946629595e-06, + "loss": 1.0648, + "step": 1695 + }, + { + "epoch": 0.127461295656095, + "grad_norm": 2.336913858454337, + "learning_rate": 3.901298999993459e-06, + "loss": 0.982, + "step": 1696 + }, + { + "epoch": 0.12753644972192996, + "grad_norm": 1.6641943794251848, + "learning_rate": 3.901147892664713e-06, + "loss": 1.0511, + "step": 1697 + }, + { + "epoch": 0.12761160378776493, + "grad_norm": 2.4237117570335918, + "learning_rate": 3.9009966726856725e-06, + "loss": 1.01, + "step": 1698 + }, + { + "epoch": 0.12768675785359987, + "grad_norm": 4.6168954308268075, + "learning_rate": 3.900845340065301e-06, + "loss": 0.9983, + "step": 1699 + }, + { + "epoch": 0.12776191191943484, + "grad_norm": 2.1282949147332917, + "learning_rate": 3.900693894812564e-06, + "loss": 1.0771, + "step": 1700 + }, + { + "epoch": 0.1278370659852698, + "grad_norm": 2.057496098529737, + "learning_rate": 3.900542336936436e-06, + "loss": 0.9968, + "step": 1701 + }, + { + "epoch": 0.12791222005110475, + "grad_norm": 2.9489322346339004, + "learning_rate": 3.900390666445896e-06, + "loss": 1.0376, + "step": 1702 + }, + { + "epoch": 0.12798737411693972, + "grad_norm": 5.2138983737536595, + "learning_rate": 3.900238883349932e-06, + "loss": 1.101, + "step": 1703 + }, + { + "epoch": 0.1280625281827747, + "grad_norm": 1.874595089742924, + "learning_rate": 3.900086987657539e-06, + "loss": 1.0992, + "step": 1704 + }, + { + "epoch": 0.12813768224860966, + "grad_norm": 1.7890155236352712, + "learning_rate": 3.899934979377714e-06, + "loss": 1.0381, + "step": 1705 + }, + { + "epoch": 0.1282128363144446, + "grad_norm": 1.4680427309308361, + "learning_rate": 3.899782858519467e-06, + "loss": 1.0903, + "step": 1706 + }, + { + "epoch": 0.12828799038027958, + "grad_norm": 1.7735150021998556, + "learning_rate": 3.899630625091811e-06, + "loss": 1.0758, + "step": 1707 + }, + { + "epoch": 0.12836314444611455, + "grad_norm": 1.6694523736924465, + "learning_rate": 3.899478279103767e-06, + "loss": 0.9597, + "step": 1708 + }, + { + "epoch": 0.1284382985119495, + "grad_norm": 1.7913750108007227, + "learning_rate": 3.89932582056436e-06, + "loss": 1.0812, + "step": 1709 + }, + { + "epoch": 0.12851345257778446, + "grad_norm": 1.509541760190783, + "learning_rate": 3.899173249482626e-06, + "loss": 1.0452, + "step": 1710 + }, + { + "epoch": 0.12858860664361943, + "grad_norm": 1.6749383544965246, + "learning_rate": 3.899020565867604e-06, + "loss": 1.0781, + "step": 1711 + }, + { + "epoch": 0.12866376070945437, + "grad_norm": 1.6694712962344713, + "learning_rate": 3.898867769728342e-06, + "loss": 0.9945, + "step": 1712 + }, + { + "epoch": 0.12873891477528934, + "grad_norm": 1.8742568768673677, + "learning_rate": 3.8987148610738935e-06, + "loss": 1.0023, + "step": 1713 + }, + { + "epoch": 0.1288140688411243, + "grad_norm": 2.164403564899126, + "learning_rate": 3.898561839913319e-06, + "loss": 1.0268, + "step": 1714 + }, + { + "epoch": 0.12888922290695926, + "grad_norm": 2.5034029688422037, + "learning_rate": 3.898408706255685e-06, + "loss": 1.1103, + "step": 1715 + }, + { + "epoch": 0.12896437697279423, + "grad_norm": 2.279336937141229, + "learning_rate": 3.898255460110066e-06, + "loss": 0.9734, + "step": 1716 + }, + { + "epoch": 0.1290395310386292, + "grad_norm": 1.9825110504689054, + "learning_rate": 3.898102101485542e-06, + "loss": 1.0036, + "step": 1717 + }, + { + "epoch": 0.12911468510446414, + "grad_norm": 2.8614676067219156, + "learning_rate": 3.8979486303912e-06, + "loss": 0.9495, + "step": 1718 + }, + { + "epoch": 0.1291898391702991, + "grad_norm": 2.6872538964011468, + "learning_rate": 3.8977950468361335e-06, + "loss": 1.0774, + "step": 1719 + }, + { + "epoch": 0.12926499323613408, + "grad_norm": 1.6911104221080815, + "learning_rate": 3.897641350829444e-06, + "loss": 0.924, + "step": 1720 + }, + { + "epoch": 0.12934014730196905, + "grad_norm": 2.2679771052622857, + "learning_rate": 3.8974875423802385e-06, + "loss": 1.1826, + "step": 1721 + }, + { + "epoch": 0.129415301367804, + "grad_norm": 2.1405127280092335, + "learning_rate": 3.897333621497629e-06, + "loss": 1.108, + "step": 1722 + }, + { + "epoch": 0.12949045543363896, + "grad_norm": 1.9012685631642714, + "learning_rate": 3.897179588190737e-06, + "loss": 1.0344, + "step": 1723 + }, + { + "epoch": 0.12956560949947393, + "grad_norm": 1.740726148861328, + "learning_rate": 3.89702544246869e-06, + "loss": 0.9575, + "step": 1724 + }, + { + "epoch": 0.12964076356530888, + "grad_norm": 1.743027967595211, + "learning_rate": 3.896871184340622e-06, + "loss": 1.0447, + "step": 1725 + }, + { + "epoch": 0.12971591763114385, + "grad_norm": 2.1519258789287607, + "learning_rate": 3.896716813815672e-06, + "loss": 1.0734, + "step": 1726 + }, + { + "epoch": 0.12979107169697882, + "grad_norm": 2.3668086689895844, + "learning_rate": 3.8965623309029876e-06, + "loss": 1.0839, + "step": 1727 + }, + { + "epoch": 0.12986622576281376, + "grad_norm": 1.6531114170259185, + "learning_rate": 3.896407735611722e-06, + "loss": 1.0337, + "step": 1728 + }, + { + "epoch": 0.12994137982864873, + "grad_norm": 2.0127204253042725, + "learning_rate": 3.896253027951038e-06, + "loss": 1.0136, + "step": 1729 + }, + { + "epoch": 0.1300165338944837, + "grad_norm": 2.954866733289172, + "learning_rate": 3.8960982079301e-06, + "loss": 0.9778, + "step": 1730 + }, + { + "epoch": 0.13009168796031864, + "grad_norm": 1.503711717582844, + "learning_rate": 3.895943275558083e-06, + "loss": 1.0625, + "step": 1731 + }, + { + "epoch": 0.1301668420261536, + "grad_norm": 2.1384418799346974, + "learning_rate": 3.895788230844166e-06, + "loss": 0.9532, + "step": 1732 + }, + { + "epoch": 0.13024199609198858, + "grad_norm": 1.717850883433962, + "learning_rate": 3.895633073797537e-06, + "loss": 0.965, + "step": 1733 + }, + { + "epoch": 0.13031715015782352, + "grad_norm": 1.8042346997314644, + "learning_rate": 3.89547780442739e-06, + "loss": 1.0842, + "step": 1734 + }, + { + "epoch": 0.1303923042236585, + "grad_norm": 1.7495602327647652, + "learning_rate": 3.895322422742924e-06, + "loss": 0.9621, + "step": 1735 + }, + { + "epoch": 0.13046745828949347, + "grad_norm": 2.9192829568539147, + "learning_rate": 3.895166928753348e-06, + "loss": 1.056, + "step": 1736 + }, + { + "epoch": 0.13054261235532844, + "grad_norm": 2.8915365895267136, + "learning_rate": 3.895011322467874e-06, + "loss": 1.0442, + "step": 1737 + }, + { + "epoch": 0.13061776642116338, + "grad_norm": 10.437303735406799, + "learning_rate": 3.894855603895723e-06, + "loss": 0.986, + "step": 1738 + }, + { + "epoch": 0.13069292048699835, + "grad_norm": 2.9008548660450804, + "learning_rate": 3.89469977304612e-06, + "loss": 1.0206, + "step": 1739 + }, + { + "epoch": 0.13076807455283332, + "grad_norm": 2.1146205529255657, + "learning_rate": 3.894543829928302e-06, + "loss": 0.9753, + "step": 1740 + }, + { + "epoch": 0.13084322861866826, + "grad_norm": 2.490035033381712, + "learning_rate": 3.894387774551506e-06, + "loss": 1.0159, + "step": 1741 + }, + { + "epoch": 0.13091838268450323, + "grad_norm": 1.943727207356851, + "learning_rate": 3.894231606924981e-06, + "loss": 1.0186, + "step": 1742 + }, + { + "epoch": 0.1309935367503382, + "grad_norm": 1.8572834978193213, + "learning_rate": 3.89407532705798e-06, + "loss": 0.938, + "step": 1743 + }, + { + "epoch": 0.13106869081617314, + "grad_norm": 0.7763937123883732, + "learning_rate": 3.893918934959762e-06, + "loss": 0.8799, + "step": 1744 + }, + { + "epoch": 0.13114384488200811, + "grad_norm": 5.821004245281392, + "learning_rate": 3.893762430639596e-06, + "loss": 0.9595, + "step": 1745 + }, + { + "epoch": 0.13121899894784309, + "grad_norm": 1.648109584211136, + "learning_rate": 3.893605814106753e-06, + "loss": 1.0353, + "step": 1746 + }, + { + "epoch": 0.13129415301367803, + "grad_norm": 2.9903989381562743, + "learning_rate": 3.893449085370515e-06, + "loss": 1.05, + "step": 1747 + }, + { + "epoch": 0.131369307079513, + "grad_norm": 1.708962541713212, + "learning_rate": 3.893292244440168e-06, + "loss": 1.0753, + "step": 1748 + }, + { + "epoch": 0.13144446114534797, + "grad_norm": 1.8932961124580732, + "learning_rate": 3.893135291325006e-06, + "loss": 1.0784, + "step": 1749 + }, + { + "epoch": 0.13151961521118294, + "grad_norm": 1.8599028278932537, + "learning_rate": 3.892978226034329e-06, + "loss": 1.0291, + "step": 1750 + }, + { + "epoch": 0.13159476927701788, + "grad_norm": 2.062629695629224, + "learning_rate": 3.892821048577443e-06, + "loss": 0.9827, + "step": 1751 + }, + { + "epoch": 0.13166992334285285, + "grad_norm": 2.59978857647983, + "learning_rate": 3.892663758963661e-06, + "loss": 1.0704, + "step": 1752 + }, + { + "epoch": 0.13174507740868782, + "grad_norm": 1.769164596870255, + "learning_rate": 3.892506357202305e-06, + "loss": 0.9173, + "step": 1753 + }, + { + "epoch": 0.13182023147452276, + "grad_norm": 1.9451670764661944, + "learning_rate": 3.8923488433027e-06, + "loss": 1.0249, + "step": 1754 + }, + { + "epoch": 0.13189538554035773, + "grad_norm": 1.655972871447615, + "learning_rate": 3.89219121727418e-06, + "loss": 1.057, + "step": 1755 + }, + { + "epoch": 0.1319705396061927, + "grad_norm": 1.564305673572689, + "learning_rate": 3.892033479126084e-06, + "loss": 1.0059, + "step": 1756 + }, + { + "epoch": 0.13204569367202765, + "grad_norm": 1.6704705301863512, + "learning_rate": 3.89187562886776e-06, + "loss": 1.0631, + "step": 1757 + }, + { + "epoch": 0.13212084773786262, + "grad_norm": 1.5853640266465185, + "learning_rate": 3.89171766650856e-06, + "loss": 1.0627, + "step": 1758 + }, + { + "epoch": 0.1321960018036976, + "grad_norm": 2.0885876454709105, + "learning_rate": 3.891559592057845e-06, + "loss": 1.1199, + "step": 1759 + }, + { + "epoch": 0.13227115586953253, + "grad_norm": 2.314392861656101, + "learning_rate": 3.8914014055249805e-06, + "loss": 0.9749, + "step": 1760 + }, + { + "epoch": 0.1323463099353675, + "grad_norm": 1.6181972134221154, + "learning_rate": 3.89124310691934e-06, + "loss": 0.9303, + "step": 1761 + }, + { + "epoch": 0.13242146400120247, + "grad_norm": 1.9916930538754491, + "learning_rate": 3.891084696250304e-06, + "loss": 1.0804, + "step": 1762 + }, + { + "epoch": 0.1324966180670374, + "grad_norm": 2.7730637849280604, + "learning_rate": 3.890926173527258e-06, + "loss": 1.0245, + "step": 1763 + }, + { + "epoch": 0.13257177213287238, + "grad_norm": 1.6928638406383054, + "learning_rate": 3.8907675387595944e-06, + "loss": 0.9615, + "step": 1764 + }, + { + "epoch": 0.13264692619870735, + "grad_norm": 2.774947322311103, + "learning_rate": 3.890608791956714e-06, + "loss": 0.9711, + "step": 1765 + }, + { + "epoch": 0.13272208026454232, + "grad_norm": 2.0314624381849513, + "learning_rate": 3.890449933128025e-06, + "loss": 1.048, + "step": 1766 + }, + { + "epoch": 0.13279723433037727, + "grad_norm": 1.6057499919988165, + "learning_rate": 3.890290962282937e-06, + "loss": 0.9966, + "step": 1767 + }, + { + "epoch": 0.13287238839621224, + "grad_norm": 1.8960140128201328, + "learning_rate": 3.890131879430871e-06, + "loss": 0.9719, + "step": 1768 + }, + { + "epoch": 0.1329475424620472, + "grad_norm": 2.1500371353136063, + "learning_rate": 3.889972684581253e-06, + "loss": 0.9902, + "step": 1769 + }, + { + "epoch": 0.13302269652788215, + "grad_norm": 3.2729469969712173, + "learning_rate": 3.889813377743517e-06, + "loss": 1.0529, + "step": 1770 + }, + { + "epoch": 0.13309785059371712, + "grad_norm": 1.5775440015263311, + "learning_rate": 3.8896539589271016e-06, + "loss": 1.0699, + "step": 1771 + }, + { + "epoch": 0.1331730046595521, + "grad_norm": 2.619692204573679, + "learning_rate": 3.889494428141453e-06, + "loss": 1.0372, + "step": 1772 + }, + { + "epoch": 0.13324815872538703, + "grad_norm": 1.318537947252227, + "learning_rate": 3.889334785396024e-06, + "loss": 0.9713, + "step": 1773 + }, + { + "epoch": 0.133323312791222, + "grad_norm": 3.370618024125989, + "learning_rate": 3.8891750307002746e-06, + "loss": 1.064, + "step": 1774 + }, + { + "epoch": 0.13339846685705697, + "grad_norm": 1.7114959920066735, + "learning_rate": 3.889015164063671e-06, + "loss": 1.0327, + "step": 1775 + }, + { + "epoch": 0.13347362092289192, + "grad_norm": 2.1956978999240113, + "learning_rate": 3.888855185495685e-06, + "loss": 1.0335, + "step": 1776 + }, + { + "epoch": 0.1335487749887269, + "grad_norm": 2.3130581156500396, + "learning_rate": 3.8886950950057965e-06, + "loss": 0.9077, + "step": 1777 + }, + { + "epoch": 0.13362392905456186, + "grad_norm": 1.792191679865678, + "learning_rate": 3.888534892603491e-06, + "loss": 0.8907, + "step": 1778 + }, + { + "epoch": 0.1336990831203968, + "grad_norm": 1.8466781382198683, + "learning_rate": 3.888374578298261e-06, + "loss": 1.0615, + "step": 1779 + }, + { + "epoch": 0.13377423718623177, + "grad_norm": 1.9998260660832936, + "learning_rate": 3.888214152099607e-06, + "loss": 1.0551, + "step": 1780 + }, + { + "epoch": 0.13384939125206674, + "grad_norm": 1.6871191230734313, + "learning_rate": 3.888053614017034e-06, + "loss": 1.0049, + "step": 1781 + }, + { + "epoch": 0.1339245453179017, + "grad_norm": 1.4735843556945691, + "learning_rate": 3.887892964060054e-06, + "loss": 1.0189, + "step": 1782 + }, + { + "epoch": 0.13399969938373665, + "grad_norm": 2.145013362424973, + "learning_rate": 3.887732202238186e-06, + "loss": 0.9658, + "step": 1783 + }, + { + "epoch": 0.13407485344957162, + "grad_norm": 4.085189138117041, + "learning_rate": 3.887571328560958e-06, + "loss": 0.9908, + "step": 1784 + }, + { + "epoch": 0.1341500075154066, + "grad_norm": 1.9138904104917274, + "learning_rate": 3.8874103430379e-06, + "loss": 1.0157, + "step": 1785 + }, + { + "epoch": 0.13422516158124154, + "grad_norm": 2.3050728168694707, + "learning_rate": 3.887249245678552e-06, + "loss": 1.127, + "step": 1786 + }, + { + "epoch": 0.1343003156470765, + "grad_norm": 1.428783693892503, + "learning_rate": 3.887088036492459e-06, + "loss": 0.9983, + "step": 1787 + }, + { + "epoch": 0.13437546971291148, + "grad_norm": 1.8457850846757988, + "learning_rate": 3.886926715489173e-06, + "loss": 1.0267, + "step": 1788 + }, + { + "epoch": 0.13445062377874642, + "grad_norm": 2.075283545918358, + "learning_rate": 3.8867652826782555e-06, + "loss": 1.0579, + "step": 1789 + }, + { + "epoch": 0.1345257778445814, + "grad_norm": 1.9624711077075851, + "learning_rate": 3.886603738069269e-06, + "loss": 1.0209, + "step": 1790 + }, + { + "epoch": 0.13460093191041636, + "grad_norm": 2.472016599821632, + "learning_rate": 3.886442081671787e-06, + "loss": 1.0079, + "step": 1791 + }, + { + "epoch": 0.1346760859762513, + "grad_norm": 2.9396761686038353, + "learning_rate": 3.886280313495388e-06, + "loss": 0.9486, + "step": 1792 + }, + { + "epoch": 0.13475124004208627, + "grad_norm": 1.9539641751897103, + "learning_rate": 3.886118433549657e-06, + "loss": 1.0563, + "step": 1793 + }, + { + "epoch": 0.13482639410792124, + "grad_norm": 1.3923965485529763, + "learning_rate": 3.8859564418441865e-06, + "loss": 0.9969, + "step": 1794 + }, + { + "epoch": 0.1349015481737562, + "grad_norm": 1.5896790065279771, + "learning_rate": 3.885794338388575e-06, + "loss": 1.0008, + "step": 1795 + }, + { + "epoch": 0.13497670223959116, + "grad_norm": 4.229444723274715, + "learning_rate": 3.8856321231924275e-06, + "loss": 1.0788, + "step": 1796 + }, + { + "epoch": 0.13505185630542613, + "grad_norm": 1.6914832822248114, + "learning_rate": 3.885469796265357e-06, + "loss": 1.0686, + "step": 1797 + }, + { + "epoch": 0.1351270103712611, + "grad_norm": 1.8717264845685329, + "learning_rate": 3.885307357616981e-06, + "loss": 1.0797, + "step": 1798 + }, + { + "epoch": 0.13520216443709604, + "grad_norm": 2.149421161534165, + "learning_rate": 3.8851448072569245e-06, + "loss": 1.0057, + "step": 1799 + }, + { + "epoch": 0.135277318502931, + "grad_norm": 3.343252554995565, + "learning_rate": 3.884982145194819e-06, + "loss": 1.0123, + "step": 1800 + }, + { + "epoch": 0.13535247256876598, + "grad_norm": 1.3132529142773337, + "learning_rate": 3.8848193714403035e-06, + "loss": 1.0721, + "step": 1801 + }, + { + "epoch": 0.13542762663460092, + "grad_norm": 1.3552668088494402, + "learning_rate": 3.884656486003023e-06, + "loss": 1.0205, + "step": 1802 + }, + { + "epoch": 0.1355027807004359, + "grad_norm": 1.982726065312652, + "learning_rate": 3.8844934888926295e-06, + "loss": 1.0213, + "step": 1803 + }, + { + "epoch": 0.13557793476627086, + "grad_norm": 8.297663017250795, + "learning_rate": 3.884330380118779e-06, + "loss": 0.9759, + "step": 1804 + }, + { + "epoch": 0.1356530888321058, + "grad_norm": 1.8912644999289248, + "learning_rate": 3.884167159691139e-06, + "loss": 1.0042, + "step": 1805 + }, + { + "epoch": 0.13572824289794078, + "grad_norm": 5.386050729914262, + "learning_rate": 3.88400382761938e-06, + "loss": 0.9338, + "step": 1806 + }, + { + "epoch": 0.13580339696377575, + "grad_norm": 1.7931329656126853, + "learning_rate": 3.883840383913179e-06, + "loss": 1.0174, + "step": 1807 + }, + { + "epoch": 0.1358785510296107, + "grad_norm": 3.7045021554572752, + "learning_rate": 3.8836768285822225e-06, + "loss": 0.9232, + "step": 1808 + }, + { + "epoch": 0.13595370509544566, + "grad_norm": 1.7412987054173208, + "learning_rate": 3.8835131616362005e-06, + "loss": 1.0445, + "step": 1809 + }, + { + "epoch": 0.13602885916128063, + "grad_norm": 1.8291050167020575, + "learning_rate": 3.883349383084811e-06, + "loss": 1.1091, + "step": 1810 + }, + { + "epoch": 0.1361040132271156, + "grad_norm": 1.3772760107354352, + "learning_rate": 3.883185492937759e-06, + "loss": 1.0813, + "step": 1811 + }, + { + "epoch": 0.13617916729295054, + "grad_norm": 2.17505418556505, + "learning_rate": 3.883021491204755e-06, + "loss": 1.066, + "step": 1812 + }, + { + "epoch": 0.1362543213587855, + "grad_norm": 2.511593449197687, + "learning_rate": 3.8828573778955175e-06, + "loss": 1.1007, + "step": 1813 + }, + { + "epoch": 0.13632947542462048, + "grad_norm": 0.7598400324691899, + "learning_rate": 3.88269315301977e-06, + "loss": 0.8393, + "step": 1814 + }, + { + "epoch": 0.13640462949045543, + "grad_norm": 1.644970398030718, + "learning_rate": 3.882528816587244e-06, + "loss": 0.9759, + "step": 1815 + }, + { + "epoch": 0.1364797835562904, + "grad_norm": 1.7502937070426454, + "learning_rate": 3.882364368607677e-06, + "loss": 0.9708, + "step": 1816 + }, + { + "epoch": 0.13655493762212537, + "grad_norm": 2.0772414085769704, + "learning_rate": 3.882199809090813e-06, + "loss": 0.9782, + "step": 1817 + }, + { + "epoch": 0.1366300916879603, + "grad_norm": 2.1484535910263896, + "learning_rate": 3.8820351380464035e-06, + "loss": 1.1334, + "step": 1818 + }, + { + "epoch": 0.13670524575379528, + "grad_norm": 1.9418239949983287, + "learning_rate": 3.881870355484204e-06, + "loss": 1.0416, + "step": 1819 + }, + { + "epoch": 0.13678039981963025, + "grad_norm": 1.538216140322336, + "learning_rate": 3.88170546141398e-06, + "loss": 1.021, + "step": 1820 + }, + { + "epoch": 0.1368555538854652, + "grad_norm": 4.7179105908895655, + "learning_rate": 3.881540455845503e-06, + "loss": 1.0049, + "step": 1821 + }, + { + "epoch": 0.13693070795130016, + "grad_norm": 1.945378865408334, + "learning_rate": 3.881375338788549e-06, + "loss": 0.9565, + "step": 1822 + }, + { + "epoch": 0.13700586201713513, + "grad_norm": 2.3883904878530733, + "learning_rate": 3.881210110252901e-06, + "loss": 0.9935, + "step": 1823 + }, + { + "epoch": 0.13708101608297008, + "grad_norm": 1.6806202981981677, + "learning_rate": 3.881044770248351e-06, + "loss": 1.0452, + "step": 1824 + }, + { + "epoch": 0.13715617014880505, + "grad_norm": 1.7522926979604638, + "learning_rate": 3.880879318784695e-06, + "loss": 1.0686, + "step": 1825 + }, + { + "epoch": 0.13723132421464002, + "grad_norm": 1.8468633970874024, + "learning_rate": 3.8807137558717375e-06, + "loss": 1.0989, + "step": 1826 + }, + { + "epoch": 0.137306478280475, + "grad_norm": 2.059212694733028, + "learning_rate": 3.880548081519287e-06, + "loss": 0.943, + "step": 1827 + }, + { + "epoch": 0.13738163234630993, + "grad_norm": 2.297284874708129, + "learning_rate": 3.880382295737163e-06, + "loss": 0.914, + "step": 1828 + }, + { + "epoch": 0.1374567864121449, + "grad_norm": 1.9970174246129562, + "learning_rate": 3.880216398535187e-06, + "loss": 0.9837, + "step": 1829 + }, + { + "epoch": 0.13753194047797987, + "grad_norm": 1.6849701603134848, + "learning_rate": 3.8800503899231895e-06, + "loss": 1.0295, + "step": 1830 + }, + { + "epoch": 0.1376070945438148, + "grad_norm": 1.7901273665054585, + "learning_rate": 3.879884269911007e-06, + "loss": 0.8997, + "step": 1831 + }, + { + "epoch": 0.13768224860964978, + "grad_norm": 1.9937053209040076, + "learning_rate": 3.879718038508483e-06, + "loss": 0.9235, + "step": 1832 + }, + { + "epoch": 0.13775740267548475, + "grad_norm": 1.778544060479565, + "learning_rate": 3.8795516957254675e-06, + "loss": 1.0485, + "step": 1833 + }, + { + "epoch": 0.1378325567413197, + "grad_norm": 5.91775989730476, + "learning_rate": 3.8793852415718165e-06, + "loss": 0.9922, + "step": 1834 + }, + { + "epoch": 0.13790771080715467, + "grad_norm": 2.5646460316046613, + "learning_rate": 3.879218676057394e-06, + "loss": 1.0328, + "step": 1835 + }, + { + "epoch": 0.13798286487298964, + "grad_norm": 1.9228890697721093, + "learning_rate": 3.879051999192068e-06, + "loss": 1.0655, + "step": 1836 + }, + { + "epoch": 0.13805801893882458, + "grad_norm": 5.194937097280596, + "learning_rate": 3.8788852109857166e-06, + "loss": 1.0319, + "step": 1837 + }, + { + "epoch": 0.13813317300465955, + "grad_norm": 1.853950987998242, + "learning_rate": 3.878718311448221e-06, + "loss": 1.0423, + "step": 1838 + }, + { + "epoch": 0.13820832707049452, + "grad_norm": 1.6541232723946877, + "learning_rate": 3.878551300589471e-06, + "loss": 0.9604, + "step": 1839 + }, + { + "epoch": 0.1382834811363295, + "grad_norm": 3.601193754636795, + "learning_rate": 3.8783841784193635e-06, + "loss": 1.0453, + "step": 1840 + }, + { + "epoch": 0.13835863520216443, + "grad_norm": 4.355280659125308, + "learning_rate": 3.878216944947801e-06, + "loss": 0.9283, + "step": 1841 + }, + { + "epoch": 0.1384337892679994, + "grad_norm": 1.7379270728540603, + "learning_rate": 3.878049600184692e-06, + "loss": 1.0016, + "step": 1842 + }, + { + "epoch": 0.13850894333383437, + "grad_norm": 1.9128887703878594, + "learning_rate": 3.877882144139952e-06, + "loss": 0.9174, + "step": 1843 + }, + { + "epoch": 0.13858409739966931, + "grad_norm": 3.4602427592587963, + "learning_rate": 3.8777145768235054e-06, + "loss": 0.9947, + "step": 1844 + }, + { + "epoch": 0.13865925146550429, + "grad_norm": 2.7869947445987227, + "learning_rate": 3.877546898245279e-06, + "loss": 1.033, + "step": 1845 + }, + { + "epoch": 0.13873440553133926, + "grad_norm": 2.2711287373412254, + "learning_rate": 3.877379108415209e-06, + "loss": 0.9597, + "step": 1846 + }, + { + "epoch": 0.1388095595971742, + "grad_norm": 2.167101547417954, + "learning_rate": 3.8772112073432385e-06, + "loss": 1.0235, + "step": 1847 + }, + { + "epoch": 0.13888471366300917, + "grad_norm": 2.0600852702371797, + "learning_rate": 3.8770431950393154e-06, + "loss": 1.0345, + "step": 1848 + }, + { + "epoch": 0.13895986772884414, + "grad_norm": 1.5930052868599096, + "learning_rate": 3.876875071513395e-06, + "loss": 1.0205, + "step": 1849 + }, + { + "epoch": 0.13903502179467908, + "grad_norm": 1.8713206748892137, + "learning_rate": 3.87670683677544e-06, + "loss": 0.9268, + "step": 1850 + }, + { + "epoch": 0.13911017586051405, + "grad_norm": 2.0202941044016414, + "learning_rate": 3.876538490835419e-06, + "loss": 0.9982, + "step": 1851 + }, + { + "epoch": 0.13918532992634902, + "grad_norm": 1.6286798772699014, + "learning_rate": 3.876370033703307e-06, + "loss": 1.0927, + "step": 1852 + }, + { + "epoch": 0.13926048399218396, + "grad_norm": 0.8101334586105832, + "learning_rate": 3.876201465389084e-06, + "loss": 0.8637, + "step": 1853 + }, + { + "epoch": 0.13933563805801893, + "grad_norm": 1.8535467533114522, + "learning_rate": 3.87603278590274e-06, + "loss": 1.0893, + "step": 1854 + }, + { + "epoch": 0.1394107921238539, + "grad_norm": 1.5504159953645509, + "learning_rate": 3.8758639952542695e-06, + "loss": 1.0241, + "step": 1855 + }, + { + "epoch": 0.13948594618968888, + "grad_norm": 4.520844818428792, + "learning_rate": 3.875695093453675e-06, + "loss": 1.0031, + "step": 1856 + }, + { + "epoch": 0.13956110025552382, + "grad_norm": 0.6269961428516027, + "learning_rate": 3.875526080510963e-06, + "loss": 0.8092, + "step": 1857 + }, + { + "epoch": 0.1396362543213588, + "grad_norm": 1.8624796872663434, + "learning_rate": 3.8753569564361495e-06, + "loss": 1.1551, + "step": 1858 + }, + { + "epoch": 0.13971140838719376, + "grad_norm": 1.9062973391799034, + "learning_rate": 3.875187721239254e-06, + "loss": 0.9291, + "step": 1859 + }, + { + "epoch": 0.1397865624530287, + "grad_norm": 1.4166027035952509, + "learning_rate": 3.8750183749303066e-06, + "loss": 0.9986, + "step": 1860 + }, + { + "epoch": 0.13986171651886367, + "grad_norm": 2.6338581306699753, + "learning_rate": 3.87484891751934e-06, + "loss": 1.0348, + "step": 1861 + }, + { + "epoch": 0.13993687058469864, + "grad_norm": 3.9544857532853763, + "learning_rate": 3.874679349016396e-06, + "loss": 0.9213, + "step": 1862 + }, + { + "epoch": 0.14001202465053358, + "grad_norm": 1.4650480814520002, + "learning_rate": 3.874509669431521e-06, + "loss": 1.0959, + "step": 1863 + }, + { + "epoch": 0.14008717871636855, + "grad_norm": 2.538883425926751, + "learning_rate": 3.874339878774771e-06, + "loss": 1.0577, + "step": 1864 + }, + { + "epoch": 0.14016233278220352, + "grad_norm": 1.9729880478811597, + "learning_rate": 3.8741699770562065e-06, + "loss": 1.0412, + "step": 1865 + }, + { + "epoch": 0.14023748684803847, + "grad_norm": 1.8382047248991282, + "learning_rate": 3.873999964285893e-06, + "loss": 0.8264, + "step": 1866 + }, + { + "epoch": 0.14031264091387344, + "grad_norm": 2.2301165140616046, + "learning_rate": 3.873829840473906e-06, + "loss": 1.0213, + "step": 1867 + }, + { + "epoch": 0.1403877949797084, + "grad_norm": 2.170205771538799, + "learning_rate": 3.873659605630325e-06, + "loss": 1.0669, + "step": 1868 + }, + { + "epoch": 0.14046294904554335, + "grad_norm": 1.8486176584371283, + "learning_rate": 3.873489259765239e-06, + "loss": 1.0827, + "step": 1869 + }, + { + "epoch": 0.14053810311137832, + "grad_norm": 1.9504391322491992, + "learning_rate": 3.873318802888739e-06, + "loss": 1.0554, + "step": 1870 + }, + { + "epoch": 0.1406132571772133, + "grad_norm": 2.0482755339465117, + "learning_rate": 3.8731482350109276e-06, + "loss": 0.9611, + "step": 1871 + }, + { + "epoch": 0.14068841124304826, + "grad_norm": 5.386250631173452, + "learning_rate": 3.87297755614191e-06, + "loss": 1.0842, + "step": 1872 + }, + { + "epoch": 0.1407635653088832, + "grad_norm": 1.9210321082192579, + "learning_rate": 3.8728067662918e-06, + "loss": 1.0881, + "step": 1873 + }, + { + "epoch": 0.14083871937471817, + "grad_norm": 1.9837112518160664, + "learning_rate": 3.872635865470718e-06, + "loss": 1.0914, + "step": 1874 + }, + { + "epoch": 0.14091387344055314, + "grad_norm": 1.772650523069301, + "learning_rate": 3.8724648536887895e-06, + "loss": 1.0646, + "step": 1875 + }, + { + "epoch": 0.1409890275063881, + "grad_norm": 2.4306637682364727, + "learning_rate": 3.872293730956149e-06, + "loss": 0.9728, + "step": 1876 + }, + { + "epoch": 0.14106418157222306, + "grad_norm": 2.093033853154129, + "learning_rate": 3.872122497282935e-06, + "loss": 0.9886, + "step": 1877 + }, + { + "epoch": 0.14113933563805803, + "grad_norm": 2.5344680298688895, + "learning_rate": 3.871951152679294e-06, + "loss": 1.0348, + "step": 1878 + }, + { + "epoch": 0.14121448970389297, + "grad_norm": 1.6642579872749952, + "learning_rate": 3.871779697155379e-06, + "loss": 1.0709, + "step": 1879 + }, + { + "epoch": 0.14128964376972794, + "grad_norm": 2.170181602227221, + "learning_rate": 3.87160813072135e-06, + "loss": 0.9758, + "step": 1880 + }, + { + "epoch": 0.1413647978355629, + "grad_norm": 1.5490343839253562, + "learning_rate": 3.871436453387372e-06, + "loss": 1.0674, + "step": 1881 + }, + { + "epoch": 0.14143995190139785, + "grad_norm": 1.7421227002601858, + "learning_rate": 3.8712646651636185e-06, + "loss": 1.0479, + "step": 1882 + }, + { + "epoch": 0.14151510596723282, + "grad_norm": 1.5885271248907498, + "learning_rate": 3.8710927660602676e-06, + "loss": 0.9204, + "step": 1883 + }, + { + "epoch": 0.1415902600330678, + "grad_norm": 1.805975714320639, + "learning_rate": 3.870920756087505e-06, + "loss": 1.0228, + "step": 1884 + }, + { + "epoch": 0.14166541409890276, + "grad_norm": 1.8357660091740566, + "learning_rate": 3.870748635255524e-06, + "loss": 0.9981, + "step": 1885 + }, + { + "epoch": 0.1417405681647377, + "grad_norm": 1.662908306032308, + "learning_rate": 3.870576403574523e-06, + "loss": 1.0298, + "step": 1886 + }, + { + "epoch": 0.14181572223057268, + "grad_norm": 3.9267127606763395, + "learning_rate": 3.870404061054706e-06, + "loss": 1.0086, + "step": 1887 + }, + { + "epoch": 0.14189087629640765, + "grad_norm": 1.9480252023198497, + "learning_rate": 3.870231607706287e-06, + "loss": 0.9648, + "step": 1888 + }, + { + "epoch": 0.1419660303622426, + "grad_norm": 2.152159196588044, + "learning_rate": 3.870059043539484e-06, + "loss": 0.9611, + "step": 1889 + }, + { + "epoch": 0.14204118442807756, + "grad_norm": 2.0558945044158645, + "learning_rate": 3.869886368564521e-06, + "loss": 1.2128, + "step": 1890 + }, + { + "epoch": 0.14211633849391253, + "grad_norm": 3.0715377731174023, + "learning_rate": 3.869713582791631e-06, + "loss": 1.0612, + "step": 1891 + }, + { + "epoch": 0.14219149255974747, + "grad_norm": 1.695798672346083, + "learning_rate": 3.869540686231051e-06, + "loss": 1.0231, + "step": 1892 + }, + { + "epoch": 0.14226664662558244, + "grad_norm": 1.947575183930304, + "learning_rate": 3.8693676788930264e-06, + "loss": 1.0038, + "step": 1893 + }, + { + "epoch": 0.1423418006914174, + "grad_norm": 2.105432626842544, + "learning_rate": 3.869194560787808e-06, + "loss": 1.0102, + "step": 1894 + }, + { + "epoch": 0.14241695475725236, + "grad_norm": 1.879967023276195, + "learning_rate": 3.8690213319256555e-06, + "loss": 0.9845, + "step": 1895 + }, + { + "epoch": 0.14249210882308733, + "grad_norm": 0.7457873928760963, + "learning_rate": 3.8688479923168316e-06, + "loss": 0.9271, + "step": 1896 + }, + { + "epoch": 0.1425672628889223, + "grad_norm": 2.2912244774728325, + "learning_rate": 3.868674541971608e-06, + "loss": 0.9509, + "step": 1897 + }, + { + "epoch": 0.14264241695475724, + "grad_norm": 1.668722442513711, + "learning_rate": 3.868500980900262e-06, + "loss": 0.9726, + "step": 1898 + }, + { + "epoch": 0.1427175710205922, + "grad_norm": 2.5292905593852133, + "learning_rate": 3.868327309113079e-06, + "loss": 1.0967, + "step": 1899 + }, + { + "epoch": 0.14279272508642718, + "grad_norm": 2.5460450862179735, + "learning_rate": 3.8681535266203464e-06, + "loss": 1.0182, + "step": 1900 + }, + { + "epoch": 0.14286787915226215, + "grad_norm": 2.7247815805794846, + "learning_rate": 3.867979633432365e-06, + "loss": 1.0028, + "step": 1901 + }, + { + "epoch": 0.1429430332180971, + "grad_norm": 0.7812400054292336, + "learning_rate": 3.867805629559438e-06, + "loss": 0.8412, + "step": 1902 + }, + { + "epoch": 0.14301818728393206, + "grad_norm": 1.9504810595733229, + "learning_rate": 3.867631515011874e-06, + "loss": 1.0483, + "step": 1903 + }, + { + "epoch": 0.14309334134976703, + "grad_norm": 2.0841666144617994, + "learning_rate": 3.8674572897999915e-06, + "loss": 1.0533, + "step": 1904 + }, + { + "epoch": 0.14316849541560198, + "grad_norm": 1.7572095048650394, + "learning_rate": 3.8672829539341136e-06, + "loss": 1.0412, + "step": 1905 + }, + { + "epoch": 0.14324364948143695, + "grad_norm": 1.5645354936230351, + "learning_rate": 3.8671085074245704e-06, + "loss": 1.073, + "step": 1906 + }, + { + "epoch": 0.14331880354727192, + "grad_norm": 1.4428004523475135, + "learning_rate": 3.8669339502816985e-06, + "loss": 0.902, + "step": 1907 + }, + { + "epoch": 0.14339395761310686, + "grad_norm": 3.3691738845731525, + "learning_rate": 3.866759282515841e-06, + "loss": 0.985, + "step": 1908 + }, + { + "epoch": 0.14346911167894183, + "grad_norm": 1.6168515506042243, + "learning_rate": 3.866584504137347e-06, + "loss": 0.9702, + "step": 1909 + }, + { + "epoch": 0.1435442657447768, + "grad_norm": 1.6505608645787144, + "learning_rate": 3.8664096151565755e-06, + "loss": 0.9281, + "step": 1910 + }, + { + "epoch": 0.14361941981061174, + "grad_norm": 2.026969507621331, + "learning_rate": 3.8662346155838855e-06, + "loss": 1.1031, + "step": 1911 + }, + { + "epoch": 0.1436945738764467, + "grad_norm": 1.6058418972629676, + "learning_rate": 3.866059505429649e-06, + "loss": 1.0346, + "step": 1912 + }, + { + "epoch": 0.14376972794228168, + "grad_norm": 2.0540271466623983, + "learning_rate": 3.865884284704241e-06, + "loss": 1.0867, + "step": 1913 + }, + { + "epoch": 0.14384488200811663, + "grad_norm": 1.7023693123062862, + "learning_rate": 3.8657089534180445e-06, + "loss": 1.0112, + "step": 1914 + }, + { + "epoch": 0.1439200360739516, + "grad_norm": 3.3040528465829198, + "learning_rate": 3.865533511581448e-06, + "loss": 1.0167, + "step": 1915 + }, + { + "epoch": 0.14399519013978657, + "grad_norm": 1.923637574339376, + "learning_rate": 3.865357959204847e-06, + "loss": 1.0084, + "step": 1916 + }, + { + "epoch": 0.14407034420562154, + "grad_norm": 1.6667524633576778, + "learning_rate": 3.865182296298644e-06, + "loss": 1.0312, + "step": 1917 + }, + { + "epoch": 0.14414549827145648, + "grad_norm": 1.577372985538678, + "learning_rate": 3.865006522873249e-06, + "loss": 1.1025, + "step": 1918 + }, + { + "epoch": 0.14422065233729145, + "grad_norm": 1.9981173953143403, + "learning_rate": 3.864830638939074e-06, + "loss": 1.0273, + "step": 1919 + }, + { + "epoch": 0.14429580640312642, + "grad_norm": 0.8305722908650295, + "learning_rate": 3.864654644506544e-06, + "loss": 0.9079, + "step": 1920 + }, + { + "epoch": 0.14437096046896136, + "grad_norm": 1.7914016956482732, + "learning_rate": 3.864478539586085e-06, + "loss": 1.0992, + "step": 1921 + }, + { + "epoch": 0.14444611453479633, + "grad_norm": 3.1522091218734936, + "learning_rate": 3.8643023241881344e-06, + "loss": 1.1421, + "step": 1922 + }, + { + "epoch": 0.1445212686006313, + "grad_norm": 1.924529188966, + "learning_rate": 3.864125998323131e-06, + "loss": 1.0333, + "step": 1923 + }, + { + "epoch": 0.14459642266646625, + "grad_norm": 1.7321913435699088, + "learning_rate": 3.863949562001524e-06, + "loss": 0.965, + "step": 1924 + }, + { + "epoch": 0.14467157673230122, + "grad_norm": 1.7253456488134888, + "learning_rate": 3.863773015233769e-06, + "loss": 1.0574, + "step": 1925 + }, + { + "epoch": 0.1447467307981362, + "grad_norm": 1.4286946669282432, + "learning_rate": 3.863596358030326e-06, + "loss": 0.9203, + "step": 1926 + }, + { + "epoch": 0.14482188486397113, + "grad_norm": 1.8131269488740038, + "learning_rate": 3.863419590401661e-06, + "loss": 1.0372, + "step": 1927 + }, + { + "epoch": 0.1448970389298061, + "grad_norm": 1.637619535436613, + "learning_rate": 3.8632427123582505e-06, + "loss": 0.9891, + "step": 1928 + }, + { + "epoch": 0.14497219299564107, + "grad_norm": 1.7871118430865813, + "learning_rate": 3.8630657239105754e-06, + "loss": 1.034, + "step": 1929 + }, + { + "epoch": 0.14504734706147604, + "grad_norm": 2.2173855306675403, + "learning_rate": 3.862888625069121e-06, + "loss": 0.9467, + "step": 1930 + }, + { + "epoch": 0.14512250112731098, + "grad_norm": 1.9807775132602232, + "learning_rate": 3.8627114158443825e-06, + "loss": 1.1192, + "step": 1931 + }, + { + "epoch": 0.14519765519314595, + "grad_norm": 1.8460868638370695, + "learning_rate": 3.862534096246859e-06, + "loss": 1.0585, + "step": 1932 + }, + { + "epoch": 0.14527280925898092, + "grad_norm": 2.2120135489147157, + "learning_rate": 3.862356666287059e-06, + "loss": 0.9642, + "step": 1933 + }, + { + "epoch": 0.14534796332481587, + "grad_norm": 0.8019662416540272, + "learning_rate": 3.862179125975495e-06, + "loss": 0.8389, + "step": 1934 + }, + { + "epoch": 0.14542311739065084, + "grad_norm": 1.6789447650423495, + "learning_rate": 3.862001475322687e-06, + "loss": 1.0931, + "step": 1935 + }, + { + "epoch": 0.1454982714564858, + "grad_norm": 1.741893384252685, + "learning_rate": 3.861823714339162e-06, + "loss": 0.9422, + "step": 1936 + }, + { + "epoch": 0.14557342552232075, + "grad_norm": 19.28617491802742, + "learning_rate": 3.861645843035452e-06, + "loss": 0.9718, + "step": 1937 + }, + { + "epoch": 0.14564857958815572, + "grad_norm": 2.7876238705606395, + "learning_rate": 3.861467861422096e-06, + "loss": 1.061, + "step": 1938 + }, + { + "epoch": 0.1457237336539907, + "grad_norm": 2.3566455337066086, + "learning_rate": 3.861289769509643e-06, + "loss": 0.9561, + "step": 1939 + }, + { + "epoch": 0.14579888771982563, + "grad_norm": 287.26985353496457, + "learning_rate": 3.861111567308643e-06, + "loss": 1.007, + "step": 1940 + }, + { + "epoch": 0.1458740417856606, + "grad_norm": 1.8274022817513322, + "learning_rate": 3.860933254829656e-06, + "loss": 1.1175, + "step": 1941 + }, + { + "epoch": 0.14594919585149557, + "grad_norm": 2.07611950927611, + "learning_rate": 3.860754832083247e-06, + "loss": 0.9302, + "step": 1942 + }, + { + "epoch": 0.14602434991733051, + "grad_norm": 1.5085973722788757, + "learning_rate": 3.86057629907999e-06, + "loss": 0.9298, + "step": 1943 + }, + { + "epoch": 0.14609950398316549, + "grad_norm": 2.3580214299788502, + "learning_rate": 3.8603976558304624e-06, + "loss": 0.9163, + "step": 1944 + }, + { + "epoch": 0.14617465804900046, + "grad_norm": 1.7435695486459886, + "learning_rate": 3.86021890234525e-06, + "loss": 0.9775, + "step": 1945 + }, + { + "epoch": 0.14624981211483543, + "grad_norm": 2.1259536847069467, + "learning_rate": 3.860040038634944e-06, + "loss": 0.9155, + "step": 1946 + }, + { + "epoch": 0.14632496618067037, + "grad_norm": 1.6635667662457814, + "learning_rate": 3.8598610647101426e-06, + "loss": 1.0672, + "step": 1947 + }, + { + "epoch": 0.14640012024650534, + "grad_norm": 1.8052141486278037, + "learning_rate": 3.859681980581452e-06, + "loss": 0.9456, + "step": 1948 + }, + { + "epoch": 0.1464752743123403, + "grad_norm": 2.167466553954687, + "learning_rate": 3.859502786259482e-06, + "loss": 1.0112, + "step": 1949 + }, + { + "epoch": 0.14655042837817525, + "grad_norm": 1.468979594364836, + "learning_rate": 3.8593234817548525e-06, + "loss": 1.015, + "step": 1950 + }, + { + "epoch": 0.14662558244401022, + "grad_norm": 1.7888961114928483, + "learning_rate": 3.859144067078186e-06, + "loss": 0.9961, + "step": 1951 + }, + { + "epoch": 0.1467007365098452, + "grad_norm": 0.7946985479345192, + "learning_rate": 3.858964542240115e-06, + "loss": 0.9081, + "step": 1952 + }, + { + "epoch": 0.14677589057568013, + "grad_norm": 2.3049661758820776, + "learning_rate": 3.8587849072512755e-06, + "loss": 1.0356, + "step": 1953 + }, + { + "epoch": 0.1468510446415151, + "grad_norm": 1.7794520355474504, + "learning_rate": 3.858605162122314e-06, + "loss": 1.0291, + "step": 1954 + }, + { + "epoch": 0.14692619870735008, + "grad_norm": 2.6517907803603125, + "learning_rate": 3.858425306863878e-06, + "loss": 0.8917, + "step": 1955 + }, + { + "epoch": 0.14700135277318502, + "grad_norm": 2.494610412851666, + "learning_rate": 3.858245341486627e-06, + "loss": 1.0379, + "step": 1956 + }, + { + "epoch": 0.14707650683902, + "grad_norm": 4.5611593288568395, + "learning_rate": 3.858065266001224e-06, + "loss": 1.0096, + "step": 1957 + }, + { + "epoch": 0.14715166090485496, + "grad_norm": 2.7477858906809662, + "learning_rate": 3.857885080418339e-06, + "loss": 1.0039, + "step": 1958 + }, + { + "epoch": 0.1472268149706899, + "grad_norm": 2.2666685660672177, + "learning_rate": 3.857704784748648e-06, + "loss": 1.0445, + "step": 1959 + }, + { + "epoch": 0.14730196903652487, + "grad_norm": 2.377010598659357, + "learning_rate": 3.857524379002835e-06, + "loss": 0.9438, + "step": 1960 + }, + { + "epoch": 0.14737712310235984, + "grad_norm": 1.5813423823333055, + "learning_rate": 3.85734386319159e-06, + "loss": 1.0889, + "step": 1961 + }, + { + "epoch": 0.1474522771681948, + "grad_norm": 1.7417259802445357, + "learning_rate": 3.857163237325608e-06, + "loss": 0.9968, + "step": 1962 + }, + { + "epoch": 0.14752743123402975, + "grad_norm": 2.1150710447026135, + "learning_rate": 3.856982501415595e-06, + "loss": 1.0639, + "step": 1963 + }, + { + "epoch": 0.14760258529986472, + "grad_norm": 2.4034389573518, + "learning_rate": 3.8568016554722554e-06, + "loss": 1.0715, + "step": 1964 + }, + { + "epoch": 0.1476777393656997, + "grad_norm": 1.9364435330448297, + "learning_rate": 3.856620699506308e-06, + "loss": 1.0597, + "step": 1965 + }, + { + "epoch": 0.14775289343153464, + "grad_norm": 2.0446413388785034, + "learning_rate": 3.856439633528476e-06, + "loss": 1.0696, + "step": 1966 + }, + { + "epoch": 0.1478280474973696, + "grad_norm": 1.5884859252488657, + "learning_rate": 3.856258457549486e-06, + "loss": 0.9272, + "step": 1967 + }, + { + "epoch": 0.14790320156320458, + "grad_norm": 2.9749443882464726, + "learning_rate": 3.856077171580074e-06, + "loss": 1.0449, + "step": 1968 + }, + { + "epoch": 0.14797835562903952, + "grad_norm": 1.8753051509500227, + "learning_rate": 3.855895775630983e-06, + "loss": 0.9157, + "step": 1969 + }, + { + "epoch": 0.1480535096948745, + "grad_norm": 2.678889920509478, + "learning_rate": 3.85571426971296e-06, + "loss": 1.0926, + "step": 1970 + }, + { + "epoch": 0.14812866376070946, + "grad_norm": 4.306822868522131, + "learning_rate": 3.8555326538367605e-06, + "loss": 1.1039, + "step": 1971 + }, + { + "epoch": 0.1482038178265444, + "grad_norm": 1.6536102130888237, + "learning_rate": 3.855350928013145e-06, + "loss": 1.0805, + "step": 1972 + }, + { + "epoch": 0.14827897189237937, + "grad_norm": 1.8705692073660087, + "learning_rate": 3.855169092252884e-06, + "loss": 1.0291, + "step": 1973 + }, + { + "epoch": 0.14835412595821434, + "grad_norm": 1.7105234420227127, + "learning_rate": 3.85498714656675e-06, + "loss": 0.9935, + "step": 1974 + }, + { + "epoch": 0.14842928002404931, + "grad_norm": 2.008408157516533, + "learning_rate": 3.854805090965525e-06, + "loss": 0.9931, + "step": 1975 + }, + { + "epoch": 0.14850443408988426, + "grad_norm": 1.6793243015531536, + "learning_rate": 3.854622925459994e-06, + "loss": 1.033, + "step": 1976 + }, + { + "epoch": 0.14857958815571923, + "grad_norm": 2.681658139765499, + "learning_rate": 3.854440650060955e-06, + "loss": 0.9975, + "step": 1977 + }, + { + "epoch": 0.1486547422215542, + "grad_norm": 1.5962428624076455, + "learning_rate": 3.854258264779204e-06, + "loss": 1.035, + "step": 1978 + }, + { + "epoch": 0.14872989628738914, + "grad_norm": 1.881894342835547, + "learning_rate": 3.854075769625552e-06, + "loss": 1.0314, + "step": 1979 + }, + { + "epoch": 0.1488050503532241, + "grad_norm": 1.9403050943738023, + "learning_rate": 3.8538931646108105e-06, + "loss": 0.9648, + "step": 1980 + }, + { + "epoch": 0.14888020441905908, + "grad_norm": 2.1834537686432123, + "learning_rate": 3.853710449745801e-06, + "loss": 1.0132, + "step": 1981 + }, + { + "epoch": 0.14895535848489402, + "grad_norm": 1.6497129595223279, + "learning_rate": 3.853527625041347e-06, + "loss": 1.0069, + "step": 1982 + }, + { + "epoch": 0.149030512550729, + "grad_norm": 1.7913935771082965, + "learning_rate": 3.853344690508285e-06, + "loss": 0.9348, + "step": 1983 + }, + { + "epoch": 0.14910566661656396, + "grad_norm": 1.979214423423273, + "learning_rate": 3.853161646157453e-06, + "loss": 1.0573, + "step": 1984 + }, + { + "epoch": 0.1491808206823989, + "grad_norm": 1.743884367876018, + "learning_rate": 3.852978491999697e-06, + "loss": 1.0784, + "step": 1985 + }, + { + "epoch": 0.14925597474823388, + "grad_norm": 2.196901879614186, + "learning_rate": 3.852795228045869e-06, + "loss": 1.0755, + "step": 1986 + }, + { + "epoch": 0.14933112881406885, + "grad_norm": 4.503692383663645, + "learning_rate": 3.85261185430683e-06, + "loss": 1.0279, + "step": 1987 + }, + { + "epoch": 0.1494062828799038, + "grad_norm": 1.7462634659113467, + "learning_rate": 3.8524283707934445e-06, + "loss": 1.1328, + "step": 1988 + }, + { + "epoch": 0.14948143694573876, + "grad_norm": 1.8357405536610654, + "learning_rate": 3.8522447775165845e-06, + "loss": 0.9856, + "step": 1989 + }, + { + "epoch": 0.14955659101157373, + "grad_norm": 1.7091440355106717, + "learning_rate": 3.852061074487129e-06, + "loss": 1.0197, + "step": 1990 + }, + { + "epoch": 0.1496317450774087, + "grad_norm": 1.6495307081589672, + "learning_rate": 3.851877261715961e-06, + "loss": 1.0632, + "step": 1991 + }, + { + "epoch": 0.14970689914324364, + "grad_norm": 3.2459129163773888, + "learning_rate": 3.851693339213976e-06, + "loss": 1.0911, + "step": 1992 + }, + { + "epoch": 0.1497820532090786, + "grad_norm": 1.6868142041348952, + "learning_rate": 3.8515093069920695e-06, + "loss": 1.0111, + "step": 1993 + }, + { + "epoch": 0.14985720727491358, + "grad_norm": 1.556048923355828, + "learning_rate": 3.851325165061147e-06, + "loss": 1.0609, + "step": 1994 + }, + { + "epoch": 0.14993236134074853, + "grad_norm": 1.7390353849341595, + "learning_rate": 3.851140913432118e-06, + "loss": 1.0287, + "step": 1995 + }, + { + "epoch": 0.1500075154065835, + "grad_norm": 1.781772001642065, + "learning_rate": 3.850956552115903e-06, + "loss": 0.9803, + "step": 1996 + }, + { + "epoch": 0.15008266947241847, + "grad_norm": 1.8406959714305682, + "learning_rate": 3.850772081123423e-06, + "loss": 1.0977, + "step": 1997 + }, + { + "epoch": 0.1501578235382534, + "grad_norm": 1.4768129973890405, + "learning_rate": 3.850587500465611e-06, + "loss": 1.0094, + "step": 1998 + }, + { + "epoch": 0.15023297760408838, + "grad_norm": 1.3826824924540575, + "learning_rate": 3.850402810153403e-06, + "loss": 0.9665, + "step": 1999 + }, + { + "epoch": 0.15030813166992335, + "grad_norm": 1.9073969407944948, + "learning_rate": 3.850218010197743e-06, + "loss": 0.9295, + "step": 2000 + }, + { + "epoch": 0.1503832857357583, + "grad_norm": 1.63772537485918, + "learning_rate": 3.850033100609581e-06, + "loss": 1.1044, + "step": 2001 + }, + { + "epoch": 0.15045843980159326, + "grad_norm": 1.7041627714428147, + "learning_rate": 3.8498480813998735e-06, + "loss": 0.9964, + "step": 2002 + }, + { + "epoch": 0.15053359386742823, + "grad_norm": 2.1271173362990656, + "learning_rate": 3.849662952579583e-06, + "loss": 1.0177, + "step": 2003 + }, + { + "epoch": 0.15060874793326318, + "grad_norm": 1.7644292782938584, + "learning_rate": 3.8494777141596805e-06, + "loss": 1.0423, + "step": 2004 + }, + { + "epoch": 0.15068390199909815, + "grad_norm": 0.8338797009998465, + "learning_rate": 3.8492923661511405e-06, + "loss": 0.8841, + "step": 2005 + }, + { + "epoch": 0.15075905606493312, + "grad_norm": 3.371204361145211, + "learning_rate": 3.8491069085649475e-06, + "loss": 1.0755, + "step": 2006 + }, + { + "epoch": 0.1508342101307681, + "grad_norm": 1.6363860818739586, + "learning_rate": 3.848921341412088e-06, + "loss": 1.0404, + "step": 2007 + }, + { + "epoch": 0.15090936419660303, + "grad_norm": 1.7487310849980764, + "learning_rate": 3.848735664703561e-06, + "loss": 1.0787, + "step": 2008 + }, + { + "epoch": 0.150984518262438, + "grad_norm": 1.5047312665775536, + "learning_rate": 3.848549878450365e-06, + "loss": 1.0755, + "step": 2009 + }, + { + "epoch": 0.15105967232827297, + "grad_norm": 1.8589016608400972, + "learning_rate": 3.84836398266351e-06, + "loss": 1.0786, + "step": 2010 + }, + { + "epoch": 0.1511348263941079, + "grad_norm": 2.2274684200766943, + "learning_rate": 3.848177977354012e-06, + "loss": 1.0867, + "step": 2011 + }, + { + "epoch": 0.15120998045994288, + "grad_norm": 0.8384795708894465, + "learning_rate": 3.847991862532892e-06, + "loss": 0.8965, + "step": 2012 + }, + { + "epoch": 0.15128513452577785, + "grad_norm": 1.8026062825861118, + "learning_rate": 3.847805638211177e-06, + "loss": 1.0698, + "step": 2013 + }, + { + "epoch": 0.1513602885916128, + "grad_norm": 1.7774302702584717, + "learning_rate": 3.847619304399902e-06, + "loss": 1.0314, + "step": 2014 + }, + { + "epoch": 0.15143544265744777, + "grad_norm": 2.0729268391677484, + "learning_rate": 3.847432861110109e-06, + "loss": 1.008, + "step": 2015 + }, + { + "epoch": 0.15151059672328274, + "grad_norm": 2.5336897595117747, + "learning_rate": 3.847246308352844e-06, + "loss": 0.955, + "step": 2016 + }, + { + "epoch": 0.15158575078911768, + "grad_norm": 2.0355849738866434, + "learning_rate": 3.847059646139162e-06, + "loss": 1.0226, + "step": 2017 + }, + { + "epoch": 0.15166090485495265, + "grad_norm": 2.9635008332579735, + "learning_rate": 3.846872874480123e-06, + "loss": 1.0444, + "step": 2018 + }, + { + "epoch": 0.15173605892078762, + "grad_norm": 1.6069732818756226, + "learning_rate": 3.8466859933867945e-06, + "loss": 1.0185, + "step": 2019 + }, + { + "epoch": 0.1518112129866226, + "grad_norm": 0.9213233525694313, + "learning_rate": 3.846499002870249e-06, + "loss": 1.0035, + "step": 2020 + }, + { + "epoch": 0.15188636705245753, + "grad_norm": 2.2819906626183517, + "learning_rate": 3.846311902941567e-06, + "loss": 0.9088, + "step": 2021 + }, + { + "epoch": 0.1519615211182925, + "grad_norm": 2.658954544987249, + "learning_rate": 3.846124693611835e-06, + "loss": 0.9821, + "step": 2022 + }, + { + "epoch": 0.15203667518412747, + "grad_norm": 2.420184523743445, + "learning_rate": 3.845937374892145e-06, + "loss": 0.8334, + "step": 2023 + }, + { + "epoch": 0.15211182924996242, + "grad_norm": 2.521325328724765, + "learning_rate": 3.845749946793597e-06, + "loss": 0.8189, + "step": 2024 + }, + { + "epoch": 0.1521869833157974, + "grad_norm": 1.741960108675826, + "learning_rate": 3.845562409327297e-06, + "loss": 1.0419, + "step": 2025 + }, + { + "epoch": 0.15226213738163236, + "grad_norm": 1.6141742885464139, + "learning_rate": 3.8453747625043575e-06, + "loss": 1.0059, + "step": 2026 + }, + { + "epoch": 0.1523372914474673, + "grad_norm": 1.7523026984104915, + "learning_rate": 3.8451870063358966e-06, + "loss": 1.1312, + "step": 2027 + }, + { + "epoch": 0.15241244551330227, + "grad_norm": 2.0480417900888144, + "learning_rate": 3.844999140833039e-06, + "loss": 0.944, + "step": 2028 + }, + { + "epoch": 0.15248759957913724, + "grad_norm": 1.788868123118921, + "learning_rate": 3.844811166006919e-06, + "loss": 1.0314, + "step": 2029 + }, + { + "epoch": 0.15256275364497218, + "grad_norm": 3.1886313431516817, + "learning_rate": 3.844623081868672e-06, + "loss": 1.0326, + "step": 2030 + }, + { + "epoch": 0.15263790771080715, + "grad_norm": 2.060510080018042, + "learning_rate": 3.844434888429444e-06, + "loss": 0.9877, + "step": 2031 + }, + { + "epoch": 0.15271306177664212, + "grad_norm": 1.9570710869120966, + "learning_rate": 3.8442465857003864e-06, + "loss": 1.031, + "step": 2032 + }, + { + "epoch": 0.15278821584247707, + "grad_norm": 1.4934096998176352, + "learning_rate": 3.844058173692657e-06, + "loss": 1.0806, + "step": 2033 + }, + { + "epoch": 0.15286336990831204, + "grad_norm": 1.849622007365125, + "learning_rate": 3.843869652417418e-06, + "loss": 1.0487, + "step": 2034 + }, + { + "epoch": 0.152938523974147, + "grad_norm": 1.456316406962481, + "learning_rate": 3.843681021885842e-06, + "loss": 1.0417, + "step": 2035 + }, + { + "epoch": 0.15301367803998198, + "grad_norm": 16.902426352496335, + "learning_rate": 3.843492282109107e-06, + "loss": 0.8502, + "step": 2036 + }, + { + "epoch": 0.15308883210581692, + "grad_norm": 1.5270279267560949, + "learning_rate": 3.843303433098393e-06, + "loss": 1.0177, + "step": 2037 + }, + { + "epoch": 0.1531639861716519, + "grad_norm": 1.6954347768538038, + "learning_rate": 3.843114474864894e-06, + "loss": 1.1269, + "step": 2038 + }, + { + "epoch": 0.15323914023748686, + "grad_norm": 0.6887941750623499, + "learning_rate": 3.842925407419803e-06, + "loss": 0.8295, + "step": 2039 + }, + { + "epoch": 0.1533142943033218, + "grad_norm": 1.8622738976176207, + "learning_rate": 3.842736230774325e-06, + "loss": 0.9278, + "step": 2040 + }, + { + "epoch": 0.15338944836915677, + "grad_norm": 8.146227528691963, + "learning_rate": 3.842546944939669e-06, + "loss": 0.9424, + "step": 2041 + }, + { + "epoch": 0.15346460243499174, + "grad_norm": 1.2410994268789637, + "learning_rate": 3.842357549927051e-06, + "loss": 1.025, + "step": 2042 + }, + { + "epoch": 0.15353975650082669, + "grad_norm": 1.8867573546572949, + "learning_rate": 3.842168045747693e-06, + "loss": 1.0396, + "step": 2043 + }, + { + "epoch": 0.15361491056666166, + "grad_norm": 1.8694357643977695, + "learning_rate": 3.8419784324128256e-06, + "loss": 1.0442, + "step": 2044 + }, + { + "epoch": 0.15369006463249663, + "grad_norm": 1.7381157614123952, + "learning_rate": 3.841788709933682e-06, + "loss": 0.8597, + "step": 2045 + }, + { + "epoch": 0.15376521869833157, + "grad_norm": 1.7972094017343252, + "learning_rate": 3.841598878321503e-06, + "loss": 1.0635, + "step": 2046 + }, + { + "epoch": 0.15384037276416654, + "grad_norm": 1.753805654652759, + "learning_rate": 3.84140893758754e-06, + "loss": 1.1841, + "step": 2047 + }, + { + "epoch": 0.1539155268300015, + "grad_norm": 3.1053408422459836, + "learning_rate": 3.841218887743046e-06, + "loss": 1.0406, + "step": 2048 + }, + { + "epoch": 0.15399068089583645, + "grad_norm": 1.643113274777302, + "learning_rate": 3.8410287287992825e-06, + "loss": 1.0048, + "step": 2049 + }, + { + "epoch": 0.15406583496167142, + "grad_norm": 1.5139151297734164, + "learning_rate": 3.840838460767517e-06, + "loss": 0.9205, + "step": 2050 + }, + { + "epoch": 0.1541409890275064, + "grad_norm": 1.8324424428604802, + "learning_rate": 3.840648083659024e-06, + "loss": 1.0155, + "step": 2051 + }, + { + "epoch": 0.15421614309334136, + "grad_norm": 2.1463812958643502, + "learning_rate": 3.840457597485083e-06, + "loss": 1.0225, + "step": 2052 + }, + { + "epoch": 0.1542912971591763, + "grad_norm": 0.9411593861422989, + "learning_rate": 3.840267002256983e-06, + "loss": 0.8956, + "step": 2053 + }, + { + "epoch": 0.15436645122501128, + "grad_norm": 2.19864724357744, + "learning_rate": 3.840076297986015e-06, + "loss": 1.0074, + "step": 2054 + }, + { + "epoch": 0.15444160529084625, + "grad_norm": 2.2125854820505486, + "learning_rate": 3.839885484683481e-06, + "loss": 1.0034, + "step": 2055 + }, + { + "epoch": 0.1545167593566812, + "grad_norm": 1.7289711562089998, + "learning_rate": 3.839694562360686e-06, + "loss": 0.994, + "step": 2056 + }, + { + "epoch": 0.15459191342251616, + "grad_norm": 1.7113141903705036, + "learning_rate": 3.839503531028944e-06, + "loss": 0.9721, + "step": 2057 + }, + { + "epoch": 0.15466706748835113, + "grad_norm": 1.9595404888154593, + "learning_rate": 3.839312390699573e-06, + "loss": 1.0382, + "step": 2058 + }, + { + "epoch": 0.15474222155418607, + "grad_norm": 1.463760178658451, + "learning_rate": 3.8391211413839005e-06, + "loss": 1.0417, + "step": 2059 + }, + { + "epoch": 0.15481737562002104, + "grad_norm": 2.009628009440584, + "learning_rate": 3.838929783093258e-06, + "loss": 0.9864, + "step": 2060 + }, + { + "epoch": 0.154892529685856, + "grad_norm": 2.107215963208467, + "learning_rate": 3.838738315838983e-06, + "loss": 1.0417, + "step": 2061 + }, + { + "epoch": 0.15496768375169095, + "grad_norm": 12.89893908019304, + "learning_rate": 3.838546739632423e-06, + "loss": 1.0421, + "step": 2062 + }, + { + "epoch": 0.15504283781752592, + "grad_norm": 2.082274231321674, + "learning_rate": 3.838355054484928e-06, + "loss": 1.0883, + "step": 2063 + }, + { + "epoch": 0.1551179918833609, + "grad_norm": 2.199445533239349, + "learning_rate": 3.838163260407857e-06, + "loss": 1.0073, + "step": 2064 + }, + { + "epoch": 0.15519314594919587, + "grad_norm": 2.1829461790641833, + "learning_rate": 3.837971357412573e-06, + "loss": 0.9306, + "step": 2065 + }, + { + "epoch": 0.1552683000150308, + "grad_norm": 1.9607158398826807, + "learning_rate": 3.837779345510449e-06, + "loss": 1.0056, + "step": 2066 + }, + { + "epoch": 0.15534345408086578, + "grad_norm": 1.6522319927253055, + "learning_rate": 3.837587224712861e-06, + "loss": 0.9581, + "step": 2067 + }, + { + "epoch": 0.15541860814670075, + "grad_norm": 1.4099249399105047, + "learning_rate": 3.837394995031193e-06, + "loss": 0.8579, + "step": 2068 + }, + { + "epoch": 0.1554937622125357, + "grad_norm": 1.8314634960305598, + "learning_rate": 3.837202656476836e-06, + "loss": 1.0655, + "step": 2069 + }, + { + "epoch": 0.15556891627837066, + "grad_norm": 1.7266678583786632, + "learning_rate": 3.837010209061187e-06, + "loss": 0.938, + "step": 2070 + }, + { + "epoch": 0.15564407034420563, + "grad_norm": 2.2047102615403475, + "learning_rate": 3.836817652795648e-06, + "loss": 0.9877, + "step": 2071 + }, + { + "epoch": 0.15571922441004057, + "grad_norm": 2.127659872034742, + "learning_rate": 3.8366249876916294e-06, + "loss": 1.0249, + "step": 2072 + }, + { + "epoch": 0.15579437847587554, + "grad_norm": 2.0556661501622417, + "learning_rate": 3.8364322137605484e-06, + "loss": 0.9632, + "step": 2073 + }, + { + "epoch": 0.15586953254171051, + "grad_norm": 1.8188275350187366, + "learning_rate": 3.836239331013825e-06, + "loss": 1.0702, + "step": 2074 + }, + { + "epoch": 0.15594468660754546, + "grad_norm": 0.6979544306734334, + "learning_rate": 3.836046339462891e-06, + "loss": 0.8511, + "step": 2075 + }, + { + "epoch": 0.15601984067338043, + "grad_norm": 1.637619389848102, + "learning_rate": 3.83585323911918e-06, + "loss": 1.0711, + "step": 2076 + }, + { + "epoch": 0.1560949947392154, + "grad_norm": 2.04978456528207, + "learning_rate": 3.835660029994135e-06, + "loss": 1.0289, + "step": 2077 + }, + { + "epoch": 0.15617014880505034, + "grad_norm": 1.6134837764684464, + "learning_rate": 3.835466712099204e-06, + "loss": 0.9673, + "step": 2078 + }, + { + "epoch": 0.1562453028708853, + "grad_norm": 1.6355497059092634, + "learning_rate": 3.835273285445842e-06, + "loss": 1.0296, + "step": 2079 + }, + { + "epoch": 0.15632045693672028, + "grad_norm": 1.860559807283354, + "learning_rate": 3.83507975004551e-06, + "loss": 1.0189, + "step": 2080 + }, + { + "epoch": 0.15639561100255525, + "grad_norm": 2.3265787404597082, + "learning_rate": 3.8348861059096755e-06, + "loss": 1.0269, + "step": 2081 + }, + { + "epoch": 0.1564707650683902, + "grad_norm": 2.2450758775492132, + "learning_rate": 3.834692353049814e-06, + "loss": 1.0159, + "step": 2082 + }, + { + "epoch": 0.15654591913422516, + "grad_norm": 1.502998613210966, + "learning_rate": 3.834498491477403e-06, + "loss": 1.0599, + "step": 2083 + }, + { + "epoch": 0.15662107320006013, + "grad_norm": 1.7734973241991798, + "learning_rate": 3.834304521203934e-06, + "loss": 1.0526, + "step": 2084 + }, + { + "epoch": 0.15669622726589508, + "grad_norm": 1.7278243757071967, + "learning_rate": 3.834110442240896e-06, + "loss": 1.0318, + "step": 2085 + }, + { + "epoch": 0.15677138133173005, + "grad_norm": 2.316031877650797, + "learning_rate": 3.833916254599792e-06, + "loss": 0.9821, + "step": 2086 + }, + { + "epoch": 0.15684653539756502, + "grad_norm": 1.6640458083770768, + "learning_rate": 3.833721958292128e-06, + "loss": 1.0122, + "step": 2087 + }, + { + "epoch": 0.15692168946339996, + "grad_norm": 1.9531862783355862, + "learning_rate": 3.8335275533294155e-06, + "loss": 1.0242, + "step": 2088 + }, + { + "epoch": 0.15699684352923493, + "grad_norm": 1.8968492936681618, + "learning_rate": 3.833333039723174e-06, + "loss": 0.9728, + "step": 2089 + }, + { + "epoch": 0.1570719975950699, + "grad_norm": 5.856530885260469, + "learning_rate": 3.83313841748493e-06, + "loss": 1.0873, + "step": 2090 + }, + { + "epoch": 0.15714715166090484, + "grad_norm": 1.5832172234107726, + "learning_rate": 3.832943686626215e-06, + "loss": 1.0926, + "step": 2091 + }, + { + "epoch": 0.1572223057267398, + "grad_norm": 1.8587004135074732, + "learning_rate": 3.832748847158568e-06, + "loss": 0.9988, + "step": 2092 + }, + { + "epoch": 0.15729745979257478, + "grad_norm": 2.116654783000277, + "learning_rate": 3.8325538990935346e-06, + "loss": 1.0068, + "step": 2093 + }, + { + "epoch": 0.15737261385840973, + "grad_norm": 17.221870818022538, + "learning_rate": 3.832358842442665e-06, + "loss": 1.0813, + "step": 2094 + }, + { + "epoch": 0.1574477679242447, + "grad_norm": 1.901915265025954, + "learning_rate": 3.832163677217516e-06, + "loss": 1.0119, + "step": 2095 + }, + { + "epoch": 0.15752292199007967, + "grad_norm": 1.6448527042336885, + "learning_rate": 3.831968403429655e-06, + "loss": 0.9607, + "step": 2096 + }, + { + "epoch": 0.15759807605591464, + "grad_norm": 1.9000214926859267, + "learning_rate": 3.83177302109065e-06, + "loss": 1.0073, + "step": 2097 + }, + { + "epoch": 0.15767323012174958, + "grad_norm": 2.451546325046724, + "learning_rate": 3.8315775302120796e-06, + "loss": 1.0395, + "step": 2098 + }, + { + "epoch": 0.15774838418758455, + "grad_norm": 1.896729756911077, + "learning_rate": 3.831381930805526e-06, + "loss": 1.0409, + "step": 2099 + }, + { + "epoch": 0.15782353825341952, + "grad_norm": 1.5579909303765995, + "learning_rate": 3.831186222882582e-06, + "loss": 1.0204, + "step": 2100 + }, + { + "epoch": 0.15789869231925446, + "grad_norm": 1.6730974665796285, + "learning_rate": 3.830990406454841e-06, + "loss": 0.9779, + "step": 2101 + }, + { + "epoch": 0.15797384638508943, + "grad_norm": 1.7473834776879755, + "learning_rate": 3.8307944815339065e-06, + "loss": 1.0068, + "step": 2102 + }, + { + "epoch": 0.1580490004509244, + "grad_norm": 2.002704936957554, + "learning_rate": 3.83059844813139e-06, + "loss": 1.051, + "step": 2103 + }, + { + "epoch": 0.15812415451675935, + "grad_norm": 2.037320383029908, + "learning_rate": 3.830402306258904e-06, + "loss": 0.9827, + "step": 2104 + }, + { + "epoch": 0.15819930858259432, + "grad_norm": 2.216571221651101, + "learning_rate": 3.8302060559280735e-06, + "loss": 0.9806, + "step": 2105 + }, + { + "epoch": 0.1582744626484293, + "grad_norm": 1.7593233524418601, + "learning_rate": 3.830009697150526e-06, + "loss": 0.9412, + "step": 2106 + }, + { + "epoch": 0.15834961671426423, + "grad_norm": 1.589010334190148, + "learning_rate": 3.829813229937896e-06, + "loss": 0.935, + "step": 2107 + }, + { + "epoch": 0.1584247707800992, + "grad_norm": 1.5730950294401067, + "learning_rate": 3.829616654301824e-06, + "loss": 1.0501, + "step": 2108 + }, + { + "epoch": 0.15849992484593417, + "grad_norm": 1.3497672092211845, + "learning_rate": 3.829419970253961e-06, + "loss": 0.9542, + "step": 2109 + }, + { + "epoch": 0.1585750789117691, + "grad_norm": 1.940037757368464, + "learning_rate": 3.829223177805959e-06, + "loss": 1.0864, + "step": 2110 + }, + { + "epoch": 0.15865023297760408, + "grad_norm": 2.1446953382408585, + "learning_rate": 3.8290262769694785e-06, + "loss": 0.9563, + "step": 2111 + }, + { + "epoch": 0.15872538704343905, + "grad_norm": 1.5440381171210777, + "learning_rate": 3.828829267756188e-06, + "loss": 1.0597, + "step": 2112 + }, + { + "epoch": 0.15880054110927402, + "grad_norm": 1.6953506113857755, + "learning_rate": 3.82863215017776e-06, + "loss": 1.0296, + "step": 2113 + }, + { + "epoch": 0.15887569517510897, + "grad_norm": 2.6528700922891986, + "learning_rate": 3.828434924245874e-06, + "loss": 1.0052, + "step": 2114 + }, + { + "epoch": 0.15895084924094394, + "grad_norm": 1.4841194334373156, + "learning_rate": 3.828237589972218e-06, + "loss": 1.0046, + "step": 2115 + }, + { + "epoch": 0.1590260033067789, + "grad_norm": 1.963481635703091, + "learning_rate": 3.828040147368484e-06, + "loss": 1.0331, + "step": 2116 + }, + { + "epoch": 0.15910115737261385, + "grad_norm": 1.4809981635041733, + "learning_rate": 3.827842596446372e-06, + "loss": 1.0063, + "step": 2117 + }, + { + "epoch": 0.15917631143844882, + "grad_norm": 1.5565735384892991, + "learning_rate": 3.827644937217585e-06, + "loss": 0.9939, + "step": 2118 + }, + { + "epoch": 0.1592514655042838, + "grad_norm": 1.609985235869198, + "learning_rate": 3.827447169693839e-06, + "loss": 1.0958, + "step": 2119 + }, + { + "epoch": 0.15932661957011873, + "grad_norm": 1.9600952284898137, + "learning_rate": 3.827249293886849e-06, + "loss": 1.0717, + "step": 2120 + }, + { + "epoch": 0.1594017736359537, + "grad_norm": 2.6157838615931723, + "learning_rate": 3.827051309808342e-06, + "loss": 1.0423, + "step": 2121 + }, + { + "epoch": 0.15947692770178867, + "grad_norm": 1.479538114231473, + "learning_rate": 3.826853217470048e-06, + "loss": 0.9915, + "step": 2122 + }, + { + "epoch": 0.15955208176762362, + "grad_norm": 1.2584640047717504, + "learning_rate": 3.8266550168837065e-06, + "loss": 1.0067, + "step": 2123 + }, + { + "epoch": 0.1596272358334586, + "grad_norm": 1.9486315498298254, + "learning_rate": 3.82645670806106e-06, + "loss": 0.953, + "step": 2124 + }, + { + "epoch": 0.15970238989929356, + "grad_norm": 2.0075911462978113, + "learning_rate": 3.826258291013859e-06, + "loss": 1.0684, + "step": 2125 + }, + { + "epoch": 0.15977754396512853, + "grad_norm": 1.5806643008183976, + "learning_rate": 3.826059765753861e-06, + "loss": 0.9858, + "step": 2126 + }, + { + "epoch": 0.15985269803096347, + "grad_norm": 3.8734765288189785, + "learning_rate": 3.82586113229283e-06, + "loss": 1.0621, + "step": 2127 + }, + { + "epoch": 0.15992785209679844, + "grad_norm": 2.0474822330286218, + "learning_rate": 3.825662390642535e-06, + "loss": 0.9453, + "step": 2128 + }, + { + "epoch": 0.1600030061626334, + "grad_norm": 1.5536434942985147, + "learning_rate": 3.825463540814753e-06, + "loss": 0.9946, + "step": 2129 + }, + { + "epoch": 0.16007816022846835, + "grad_norm": 1.6958687568482713, + "learning_rate": 3.8252645828212655e-06, + "loss": 1.0112, + "step": 2130 + }, + { + "epoch": 0.16015331429430332, + "grad_norm": 2.1004052361548937, + "learning_rate": 3.825065516673862e-06, + "loss": 0.994, + "step": 2131 + }, + { + "epoch": 0.1602284683601383, + "grad_norm": 2.013198338060699, + "learning_rate": 3.824866342384338e-06, + "loss": 1.0801, + "step": 2132 + }, + { + "epoch": 0.16030362242597324, + "grad_norm": 1.7720551333622945, + "learning_rate": 3.824667059964496e-06, + "loss": 0.952, + "step": 2133 + }, + { + "epoch": 0.1603787764918082, + "grad_norm": 1.7600119161202443, + "learning_rate": 3.824467669426143e-06, + "loss": 0.9996, + "step": 2134 + }, + { + "epoch": 0.16045393055764318, + "grad_norm": 1.5624765012881432, + "learning_rate": 3.824268170781094e-06, + "loss": 0.9394, + "step": 2135 + }, + { + "epoch": 0.16052908462347812, + "grad_norm": 0.939930086934684, + "learning_rate": 3.82406856404117e-06, + "loss": 0.8952, + "step": 2136 + }, + { + "epoch": 0.1606042386893131, + "grad_norm": 2.2573337773395665, + "learning_rate": 3.8238688492182e-06, + "loss": 1.0898, + "step": 2137 + }, + { + "epoch": 0.16067939275514806, + "grad_norm": 1.9599625188785972, + "learning_rate": 3.823669026324016e-06, + "loss": 1.029, + "step": 2138 + }, + { + "epoch": 0.160754546820983, + "grad_norm": 2.3581939166301393, + "learning_rate": 3.823469095370459e-06, + "loss": 1.0851, + "step": 2139 + }, + { + "epoch": 0.16082970088681797, + "grad_norm": 1.4797578982209936, + "learning_rate": 3.823269056369376e-06, + "loss": 1.0205, + "step": 2140 + }, + { + "epoch": 0.16090485495265294, + "grad_norm": 1.8789814321342555, + "learning_rate": 3.8230689093326185e-06, + "loss": 1.1048, + "step": 2141 + }, + { + "epoch": 0.1609800090184879, + "grad_norm": 1.6558800050134725, + "learning_rate": 3.822868654272048e-06, + "loss": 0.9507, + "step": 2142 + }, + { + "epoch": 0.16105516308432286, + "grad_norm": 1.9245696986424943, + "learning_rate": 3.822668291199529e-06, + "loss": 1.1099, + "step": 2143 + }, + { + "epoch": 0.16113031715015783, + "grad_norm": 1.7636168546321582, + "learning_rate": 3.822467820126935e-06, + "loss": 0.9498, + "step": 2144 + }, + { + "epoch": 0.1612054712159928, + "grad_norm": 1.77083830739239, + "learning_rate": 3.822267241066143e-06, + "loss": 0.9647, + "step": 2145 + }, + { + "epoch": 0.16128062528182774, + "grad_norm": 2.493049783325125, + "learning_rate": 3.8220665540290395e-06, + "loss": 1.0404, + "step": 2146 + }, + { + "epoch": 0.1613557793476627, + "grad_norm": 1.4506049505620802, + "learning_rate": 3.821865759027515e-06, + "loss": 0.9878, + "step": 2147 + }, + { + "epoch": 0.16143093341349768, + "grad_norm": 1.9661056682871472, + "learning_rate": 3.821664856073469e-06, + "loss": 0.9816, + "step": 2148 + }, + { + "epoch": 0.16150608747933262, + "grad_norm": 2.4373387014377226, + "learning_rate": 3.821463845178803e-06, + "loss": 1.0477, + "step": 2149 + }, + { + "epoch": 0.1615812415451676, + "grad_norm": 2.318673425114357, + "learning_rate": 3.821262726355431e-06, + "loss": 1.0553, + "step": 2150 + }, + { + "epoch": 0.16165639561100256, + "grad_norm": 2.7093642009093104, + "learning_rate": 3.821061499615268e-06, + "loss": 0.961, + "step": 2151 + }, + { + "epoch": 0.1617315496768375, + "grad_norm": 0.831654494686408, + "learning_rate": 3.820860164970237e-06, + "loss": 0.8318, + "step": 2152 + }, + { + "epoch": 0.16180670374267248, + "grad_norm": 2.462011968594425, + "learning_rate": 3.820658722432269e-06, + "loss": 1.0421, + "step": 2153 + }, + { + "epoch": 0.16188185780850745, + "grad_norm": 2.150441514746101, + "learning_rate": 3.820457172013301e-06, + "loss": 1.0431, + "step": 2154 + }, + { + "epoch": 0.1619570118743424, + "grad_norm": 1.5982194201728686, + "learning_rate": 3.820255513725274e-06, + "loss": 1.056, + "step": 2155 + }, + { + "epoch": 0.16203216594017736, + "grad_norm": 1.8022496655681213, + "learning_rate": 3.820053747580137e-06, + "loss": 1.0616, + "step": 2156 + }, + { + "epoch": 0.16210732000601233, + "grad_norm": 2.085771075406864, + "learning_rate": 3.8198518735898465e-06, + "loss": 1.0286, + "step": 2157 + }, + { + "epoch": 0.1621824740718473, + "grad_norm": 1.4878107563754535, + "learning_rate": 3.819649891766364e-06, + "loss": 0.969, + "step": 2158 + }, + { + "epoch": 0.16225762813768224, + "grad_norm": 2.9099827247447685, + "learning_rate": 3.8194478021216566e-06, + "loss": 0.9921, + "step": 2159 + }, + { + "epoch": 0.1623327822035172, + "grad_norm": 1.7010835784526719, + "learning_rate": 3.8192456046677004e-06, + "loss": 1.1711, + "step": 2160 + }, + { + "epoch": 0.16240793626935218, + "grad_norm": 1.784474582866271, + "learning_rate": 3.819043299416476e-06, + "loss": 1.0696, + "step": 2161 + }, + { + "epoch": 0.16248309033518712, + "grad_norm": 1.649882401986248, + "learning_rate": 3.8188408863799706e-06, + "loss": 1.0046, + "step": 2162 + }, + { + "epoch": 0.1625582444010221, + "grad_norm": 2.182610414908396, + "learning_rate": 3.818638365570177e-06, + "loss": 0.9474, + "step": 2163 + }, + { + "epoch": 0.16263339846685707, + "grad_norm": 1.4594815593102068, + "learning_rate": 3.818435736999097e-06, + "loss": 1.0529, + "step": 2164 + }, + { + "epoch": 0.162708552532692, + "grad_norm": 3.290642075888298, + "learning_rate": 3.818233000678736e-06, + "loss": 0.982, + "step": 2165 + }, + { + "epoch": 0.16278370659852698, + "grad_norm": 1.9575937037737339, + "learning_rate": 3.8180301566211075e-06, + "loss": 1.0159, + "step": 2166 + }, + { + "epoch": 0.16285886066436195, + "grad_norm": 2.0971884889675647, + "learning_rate": 3.81782720483823e-06, + "loss": 1.0991, + "step": 2167 + }, + { + "epoch": 0.1629340147301969, + "grad_norm": 2.0918330056768193, + "learning_rate": 3.8176241453421305e-06, + "loss": 1.0614, + "step": 2168 + }, + { + "epoch": 0.16300916879603186, + "grad_norm": 1.6351695592842546, + "learning_rate": 3.81742097814484e-06, + "loss": 0.9711, + "step": 2169 + }, + { + "epoch": 0.16308432286186683, + "grad_norm": 0.8376709621214063, + "learning_rate": 3.817217703258397e-06, + "loss": 0.908, + "step": 2170 + }, + { + "epoch": 0.1631594769277018, + "grad_norm": 0.7442835472806005, + "learning_rate": 3.817014320694846e-06, + "loss": 0.8323, + "step": 2171 + }, + { + "epoch": 0.16323463099353674, + "grad_norm": 2.6777579698142184, + "learning_rate": 3.816810830466239e-06, + "loss": 1.0588, + "step": 2172 + }, + { + "epoch": 0.16330978505937171, + "grad_norm": 1.8553739824688682, + "learning_rate": 3.816607232584633e-06, + "loss": 0.9537, + "step": 2173 + }, + { + "epoch": 0.16338493912520669, + "grad_norm": 1.3785283594794484, + "learning_rate": 3.816403527062093e-06, + "loss": 1.0101, + "step": 2174 + }, + { + "epoch": 0.16346009319104163, + "grad_norm": 2.465686780883192, + "learning_rate": 3.816199713910688e-06, + "loss": 1.0816, + "step": 2175 + }, + { + "epoch": 0.1635352472568766, + "grad_norm": 1.3630063892027235, + "learning_rate": 3.815995793142495e-06, + "loss": 1.0037, + "step": 2176 + }, + { + "epoch": 0.16361040132271157, + "grad_norm": 2.5027082079571374, + "learning_rate": 3.815791764769598e-06, + "loss": 0.9357, + "step": 2177 + }, + { + "epoch": 0.1636855553885465, + "grad_norm": 0.8117540309607374, + "learning_rate": 3.815587628804086e-06, + "loss": 0.8951, + "step": 2178 + }, + { + "epoch": 0.16376070945438148, + "grad_norm": 2.110453351208155, + "learning_rate": 3.815383385258054e-06, + "loss": 1.0153, + "step": 2179 + }, + { + "epoch": 0.16383586352021645, + "grad_norm": 1.9356898804464882, + "learning_rate": 3.8151790341436046e-06, + "loss": 1.0062, + "step": 2180 + }, + { + "epoch": 0.1639110175860514, + "grad_norm": 1.643207805835269, + "learning_rate": 3.814974575472847e-06, + "loss": 1.0608, + "step": 2181 + }, + { + "epoch": 0.16398617165188636, + "grad_norm": 2.2906764463135243, + "learning_rate": 3.814770009257896e-06, + "loss": 1.0105, + "step": 2182 + }, + { + "epoch": 0.16406132571772133, + "grad_norm": 1.6589192887183914, + "learning_rate": 3.814565335510873e-06, + "loss": 1.0541, + "step": 2183 + }, + { + "epoch": 0.16413647978355628, + "grad_norm": 0.6651778335420899, + "learning_rate": 3.814360554243905e-06, + "loss": 0.8298, + "step": 2184 + }, + { + "epoch": 0.16421163384939125, + "grad_norm": 2.5240698810340536, + "learning_rate": 3.814155665469126e-06, + "loss": 0.9389, + "step": 2185 + }, + { + "epoch": 0.16428678791522622, + "grad_norm": 2.922424591109693, + "learning_rate": 3.813950669198678e-06, + "loss": 1.1085, + "step": 2186 + }, + { + "epoch": 0.1643619419810612, + "grad_norm": 0.7659088698428761, + "learning_rate": 3.8137455654447063e-06, + "loss": 0.8553, + "step": 2187 + }, + { + "epoch": 0.16443709604689613, + "grad_norm": 3.2478378146038485, + "learning_rate": 3.8135403542193646e-06, + "loss": 0.9768, + "step": 2188 + }, + { + "epoch": 0.1645122501127311, + "grad_norm": 5.2650669605891345, + "learning_rate": 3.8133350355348125e-06, + "loss": 1.0487, + "step": 2189 + }, + { + "epoch": 0.16458740417856607, + "grad_norm": 1.6729709206315233, + "learning_rate": 3.8131296094032158e-06, + "loss": 0.9556, + "step": 2190 + }, + { + "epoch": 0.164662558244401, + "grad_norm": 1.9841691758952908, + "learning_rate": 3.8129240758367463e-06, + "loss": 1.0244, + "step": 2191 + }, + { + "epoch": 0.16473771231023598, + "grad_norm": 2.1188704779925898, + "learning_rate": 3.8127184348475836e-06, + "loss": 1.0164, + "step": 2192 + }, + { + "epoch": 0.16481286637607095, + "grad_norm": 1.6424116306541598, + "learning_rate": 3.8125126864479123e-06, + "loss": 1.0878, + "step": 2193 + }, + { + "epoch": 0.1648880204419059, + "grad_norm": 1.473636857236414, + "learning_rate": 3.8123068306499236e-06, + "loss": 0.9779, + "step": 2194 + }, + { + "epoch": 0.16496317450774087, + "grad_norm": 2.163943841759038, + "learning_rate": 3.8121008674658154e-06, + "loss": 1.0447, + "step": 2195 + }, + { + "epoch": 0.16503832857357584, + "grad_norm": 1.8976024784967727, + "learning_rate": 3.8118947969077915e-06, + "loss": 1.0475, + "step": 2196 + }, + { + "epoch": 0.16511348263941078, + "grad_norm": 2.0271028879077804, + "learning_rate": 3.8116886189880634e-06, + "loss": 1.0345, + "step": 2197 + }, + { + "epoch": 0.16518863670524575, + "grad_norm": 1.708430713886399, + "learning_rate": 3.811482333718847e-06, + "loss": 0.9272, + "step": 2198 + }, + { + "epoch": 0.16526379077108072, + "grad_norm": 1.8414442871722727, + "learning_rate": 3.811275941112366e-06, + "loss": 1.0424, + "step": 2199 + }, + { + "epoch": 0.16533894483691566, + "grad_norm": 1.8293997078110982, + "learning_rate": 3.811069441180849e-06, + "loss": 1.0723, + "step": 2200 + }, + { + "epoch": 0.16541409890275063, + "grad_norm": 2.6714009539267267, + "learning_rate": 3.810862833936532e-06, + "loss": 1.0288, + "step": 2201 + }, + { + "epoch": 0.1654892529685856, + "grad_norm": 1.906386698839959, + "learning_rate": 3.8106561193916587e-06, + "loss": 0.9597, + "step": 2202 + }, + { + "epoch": 0.16556440703442057, + "grad_norm": 1.9552790246079088, + "learning_rate": 3.810449297558477e-06, + "loss": 0.9429, + "step": 2203 + }, + { + "epoch": 0.16563956110025552, + "grad_norm": 2.305456001042245, + "learning_rate": 3.810242368449241e-06, + "loss": 0.9245, + "step": 2204 + }, + { + "epoch": 0.1657147151660905, + "grad_norm": 2.3038922505747315, + "learning_rate": 3.810035332076214e-06, + "loss": 1.0082, + "step": 2205 + }, + { + "epoch": 0.16578986923192546, + "grad_norm": 1.0469996250046256, + "learning_rate": 3.809828188451662e-06, + "loss": 0.8797, + "step": 2206 + }, + { + "epoch": 0.1658650232977604, + "grad_norm": 1.8000635798139688, + "learning_rate": 3.809620937587859e-06, + "loss": 1.0281, + "step": 2207 + }, + { + "epoch": 0.16594017736359537, + "grad_norm": 2.2224913738757017, + "learning_rate": 3.8094135794970857e-06, + "loss": 1.016, + "step": 2208 + }, + { + "epoch": 0.16601533142943034, + "grad_norm": 3.1702595623140777, + "learning_rate": 3.80920611419163e-06, + "loss": 1.0483, + "step": 2209 + }, + { + "epoch": 0.16609048549526528, + "grad_norm": 1.7349180453949578, + "learning_rate": 3.808998541683784e-06, + "loss": 1.0471, + "step": 2210 + }, + { + "epoch": 0.16616563956110025, + "grad_norm": 1.8499889528099347, + "learning_rate": 3.8087908619858473e-06, + "loss": 1.0188, + "step": 2211 + }, + { + "epoch": 0.16624079362693522, + "grad_norm": 2.3690451204150285, + "learning_rate": 3.8085830751101253e-06, + "loss": 1.0369, + "step": 2212 + }, + { + "epoch": 0.16631594769277017, + "grad_norm": 2.5481461237353944, + "learning_rate": 3.8083751810689306e-06, + "loss": 0.9913, + "step": 2213 + }, + { + "epoch": 0.16639110175860514, + "grad_norm": 1.400117046708577, + "learning_rate": 3.8081671798745817e-06, + "loss": 1.0941, + "step": 2214 + }, + { + "epoch": 0.1664662558244401, + "grad_norm": 1.8933063125808953, + "learning_rate": 3.807959071539404e-06, + "loss": 1.0638, + "step": 2215 + }, + { + "epoch": 0.16654140989027508, + "grad_norm": 5.542792222677962, + "learning_rate": 3.8077508560757275e-06, + "loss": 1.1338, + "step": 2216 + }, + { + "epoch": 0.16661656395611002, + "grad_norm": 0.7731803697820506, + "learning_rate": 3.8075425334958908e-06, + "loss": 0.9155, + "step": 2217 + }, + { + "epoch": 0.166691718021945, + "grad_norm": 1.8495220417484102, + "learning_rate": 3.8073341038122374e-06, + "loss": 0.9659, + "step": 2218 + }, + { + "epoch": 0.16676687208777996, + "grad_norm": 1.687437197611567, + "learning_rate": 3.8071255670371174e-06, + "loss": 0.962, + "step": 2219 + }, + { + "epoch": 0.1668420261536149, + "grad_norm": 2.2958706067539048, + "learning_rate": 3.8069169231828875e-06, + "loss": 1.0255, + "step": 2220 + }, + { + "epoch": 0.16691718021944987, + "grad_norm": 2.2965012849670843, + "learning_rate": 3.8067081722619114e-06, + "loss": 1.0176, + "step": 2221 + }, + { + "epoch": 0.16699233428528484, + "grad_norm": 2.775616544673267, + "learning_rate": 3.8064993142865573e-06, + "loss": 0.9847, + "step": 2222 + }, + { + "epoch": 0.1670674883511198, + "grad_norm": 0.7455556237048128, + "learning_rate": 3.8062903492692014e-06, + "loss": 0.8322, + "step": 2223 + }, + { + "epoch": 0.16714264241695476, + "grad_norm": 1.7443644517766201, + "learning_rate": 3.8060812772222255e-06, + "loss": 0.9398, + "step": 2224 + }, + { + "epoch": 0.16721779648278973, + "grad_norm": 2.42366788397468, + "learning_rate": 3.805872098158018e-06, + "loss": 1.0999, + "step": 2225 + }, + { + "epoch": 0.16729295054862467, + "grad_norm": 2.2092379990204583, + "learning_rate": 3.8056628120889736e-06, + "loss": 1.0651, + "step": 2226 + }, + { + "epoch": 0.16736810461445964, + "grad_norm": 2.0591704340794412, + "learning_rate": 3.805453419027493e-06, + "loss": 1.0412, + "step": 2227 + }, + { + "epoch": 0.1674432586802946, + "grad_norm": 1.4877559505773765, + "learning_rate": 3.805243918985984e-06, + "loss": 1.053, + "step": 2228 + }, + { + "epoch": 0.16751841274612955, + "grad_norm": 2.0280896764186003, + "learning_rate": 3.80503431197686e-06, + "loss": 1.0409, + "step": 2229 + }, + { + "epoch": 0.16759356681196452, + "grad_norm": 2.284085640799849, + "learning_rate": 3.804824598012541e-06, + "loss": 1.0023, + "step": 2230 + }, + { + "epoch": 0.1676687208777995, + "grad_norm": 2.2569387256517257, + "learning_rate": 3.8046147771054536e-06, + "loss": 1.1093, + "step": 2231 + }, + { + "epoch": 0.16774387494363446, + "grad_norm": 3.867006363142612, + "learning_rate": 3.8044048492680297e-06, + "loss": 0.9533, + "step": 2232 + }, + { + "epoch": 0.1678190290094694, + "grad_norm": 2.522841439253031, + "learning_rate": 3.80419481451271e-06, + "loss": 1.1033, + "step": 2233 + }, + { + "epoch": 0.16789418307530438, + "grad_norm": 4.500070359421739, + "learning_rate": 3.8039846728519383e-06, + "loss": 0.9468, + "step": 2234 + }, + { + "epoch": 0.16796933714113935, + "grad_norm": 2.564571380793972, + "learning_rate": 3.803774424298167e-06, + "loss": 1.0155, + "step": 2235 + }, + { + "epoch": 0.1680444912069743, + "grad_norm": 0.7900172585400895, + "learning_rate": 3.8035640688638537e-06, + "loss": 0.8401, + "step": 2236 + }, + { + "epoch": 0.16811964527280926, + "grad_norm": 1.5280702233979622, + "learning_rate": 3.8033536065614625e-06, + "loss": 0.996, + "step": 2237 + }, + { + "epoch": 0.16819479933864423, + "grad_norm": 1.3967297318467764, + "learning_rate": 3.8031430374034653e-06, + "loss": 1.0529, + "step": 2238 + }, + { + "epoch": 0.16826995340447917, + "grad_norm": 1.9368640101940522, + "learning_rate": 3.802932361402338e-06, + "loss": 1.1498, + "step": 2239 + }, + { + "epoch": 0.16834510747031414, + "grad_norm": 2.4073842335533864, + "learning_rate": 3.8027215785705654e-06, + "loss": 0.8969, + "step": 2240 + }, + { + "epoch": 0.1684202615361491, + "grad_norm": 1.9019413391236994, + "learning_rate": 3.8025106889206353e-06, + "loss": 1.008, + "step": 2241 + }, + { + "epoch": 0.16849541560198406, + "grad_norm": 0.7007728652707655, + "learning_rate": 3.802299692465045e-06, + "loss": 0.8613, + "step": 2242 + }, + { + "epoch": 0.16857056966781903, + "grad_norm": 3.6374062647327143, + "learning_rate": 3.802088589216296e-06, + "loss": 0.9349, + "step": 2243 + }, + { + "epoch": 0.168645723733654, + "grad_norm": 1.9370491980044033, + "learning_rate": 3.801877379186898e-06, + "loss": 0.9529, + "step": 2244 + }, + { + "epoch": 0.16872087779948894, + "grad_norm": 3.2729300968340596, + "learning_rate": 3.8016660623893653e-06, + "loss": 1.0522, + "step": 2245 + }, + { + "epoch": 0.1687960318653239, + "grad_norm": 2.154066666643747, + "learning_rate": 3.801454638836219e-06, + "loss": 0.9551, + "step": 2246 + }, + { + "epoch": 0.16887118593115888, + "grad_norm": 1.4461071407413255, + "learning_rate": 3.801243108539987e-06, + "loss": 0.9486, + "step": 2247 + }, + { + "epoch": 0.16894633999699385, + "grad_norm": 3.202713656181078, + "learning_rate": 3.8010314715132037e-06, + "loss": 0.9991, + "step": 2248 + }, + { + "epoch": 0.1690214940628288, + "grad_norm": 1.9800795674953748, + "learning_rate": 3.8008197277684094e-06, + "loss": 1.0273, + "step": 2249 + }, + { + "epoch": 0.16909664812866376, + "grad_norm": 1.7657708301916926, + "learning_rate": 3.80060787731815e-06, + "loss": 1.0426, + "step": 2250 + }, + { + "epoch": 0.16917180219449873, + "grad_norm": 2.9355299816546516, + "learning_rate": 3.8003959201749793e-06, + "loss": 1.0156, + "step": 2251 + }, + { + "epoch": 0.16924695626033368, + "grad_norm": 1.53036637014897, + "learning_rate": 3.800183856351456e-06, + "loss": 1.0752, + "step": 2252 + }, + { + "epoch": 0.16932211032616865, + "grad_norm": 2.0065153808004244, + "learning_rate": 3.7999716858601456e-06, + "loss": 1.0166, + "step": 2253 + }, + { + "epoch": 0.16939726439200362, + "grad_norm": 1.6841090004681298, + "learning_rate": 3.79975940871362e-06, + "loss": 0.9231, + "step": 2254 + }, + { + "epoch": 0.16947241845783856, + "grad_norm": 1.6522970711537874, + "learning_rate": 3.7995470249244582e-06, + "loss": 1.0348, + "step": 2255 + }, + { + "epoch": 0.16954757252367353, + "grad_norm": 2.428277787319675, + "learning_rate": 3.7993345345052445e-06, + "loss": 1.06, + "step": 2256 + }, + { + "epoch": 0.1696227265895085, + "grad_norm": 2.20577821523154, + "learning_rate": 3.799121937468569e-06, + "loss": 1.0591, + "step": 2257 + }, + { + "epoch": 0.16969788065534344, + "grad_norm": 1.9126257038567658, + "learning_rate": 3.7989092338270295e-06, + "loss": 0.9405, + "step": 2258 + }, + { + "epoch": 0.1697730347211784, + "grad_norm": 2.4617611430157984, + "learning_rate": 3.7986964235932293e-06, + "loss": 1.0591, + "step": 2259 + }, + { + "epoch": 0.16984818878701338, + "grad_norm": 1.6345252586133143, + "learning_rate": 3.7984835067797788e-06, + "loss": 1.0045, + "step": 2260 + }, + { + "epoch": 0.16992334285284835, + "grad_norm": 3.604104293254127, + "learning_rate": 3.7982704833992933e-06, + "loss": 1.0394, + "step": 2261 + }, + { + "epoch": 0.1699984969186833, + "grad_norm": 1.7589726667291947, + "learning_rate": 3.7980573534643954e-06, + "loss": 1.063, + "step": 2262 + }, + { + "epoch": 0.17007365098451827, + "grad_norm": 2.2411726406041717, + "learning_rate": 3.7978441169877143e-06, + "loss": 0.9028, + "step": 2263 + }, + { + "epoch": 0.17014880505035324, + "grad_norm": 2.145672381639339, + "learning_rate": 3.7976307739818852e-06, + "loss": 0.971, + "step": 2264 + }, + { + "epoch": 0.17022395911618818, + "grad_norm": 1.5382839499275998, + "learning_rate": 3.7974173244595493e-06, + "loss": 1.0733, + "step": 2265 + }, + { + "epoch": 0.17029911318202315, + "grad_norm": 1.4688719029261232, + "learning_rate": 3.7972037684333534e-06, + "loss": 1.0226, + "step": 2266 + }, + { + "epoch": 0.17037426724785812, + "grad_norm": 1.637522789017607, + "learning_rate": 3.7969901059159524e-06, + "loss": 0.9127, + "step": 2267 + }, + { + "epoch": 0.17044942131369306, + "grad_norm": 2.7689889395710923, + "learning_rate": 3.796776336920007e-06, + "loss": 0.9134, + "step": 2268 + }, + { + "epoch": 0.17052457537952803, + "grad_norm": 2.020702502023633, + "learning_rate": 3.796562461458183e-06, + "loss": 1.0197, + "step": 2269 + }, + { + "epoch": 0.170599729445363, + "grad_norm": 2.7178929172780224, + "learning_rate": 3.7963484795431537e-06, + "loss": 0.9867, + "step": 2270 + }, + { + "epoch": 0.17067488351119794, + "grad_norm": 0.6983618787069192, + "learning_rate": 3.796134391187598e-06, + "loss": 0.8218, + "step": 2271 + }, + { + "epoch": 0.17075003757703291, + "grad_norm": 1.8907707095153827, + "learning_rate": 3.7959201964042024e-06, + "loss": 0.8847, + "step": 2272 + }, + { + "epoch": 0.17082519164286789, + "grad_norm": 1.4997602907016274, + "learning_rate": 3.7957058952056577e-06, + "loss": 0.9423, + "step": 2273 + }, + { + "epoch": 0.17090034570870283, + "grad_norm": 2.2254818083911583, + "learning_rate": 3.7954914876046626e-06, + "loss": 0.977, + "step": 2274 + }, + { + "epoch": 0.1709754997745378, + "grad_norm": 2.1722610459865086, + "learning_rate": 3.795276973613921e-06, + "loss": 0.8297, + "step": 2275 + }, + { + "epoch": 0.17105065384037277, + "grad_norm": 2.1239647306901253, + "learning_rate": 3.795062353246145e-06, + "loss": 1.016, + "step": 2276 + }, + { + "epoch": 0.17112580790620774, + "grad_norm": 2.0219638716491435, + "learning_rate": 3.79484762651405e-06, + "loss": 0.9945, + "step": 2277 + }, + { + "epoch": 0.17120096197204268, + "grad_norm": 2.0321775226092913, + "learning_rate": 3.7946327934303612e-06, + "loss": 1.0234, + "step": 2278 + }, + { + "epoch": 0.17127611603787765, + "grad_norm": 2.3400815610446806, + "learning_rate": 3.7944178540078065e-06, + "loss": 1.0693, + "step": 2279 + }, + { + "epoch": 0.17135127010371262, + "grad_norm": 3.9472351828056693, + "learning_rate": 3.7942028082591227e-06, + "loss": 0.971, + "step": 2280 + }, + { + "epoch": 0.17142642416954756, + "grad_norm": 2.666275671268857, + "learning_rate": 3.7939876561970526e-06, + "loss": 1.1121, + "step": 2281 + }, + { + "epoch": 0.17150157823538253, + "grad_norm": 3.765960425162373, + "learning_rate": 3.7937723978343437e-06, + "loss": 1.0141, + "step": 2282 + }, + { + "epoch": 0.1715767323012175, + "grad_norm": 1.5154152951231274, + "learning_rate": 3.7935570331837514e-06, + "loss": 1.0088, + "step": 2283 + }, + { + "epoch": 0.17165188636705245, + "grad_norm": 3.7542601229049875, + "learning_rate": 3.793341562258037e-06, + "loss": 1.0764, + "step": 2284 + }, + { + "epoch": 0.17172704043288742, + "grad_norm": 1.9108122340868967, + "learning_rate": 3.7931259850699678e-06, + "loss": 1.0432, + "step": 2285 + }, + { + "epoch": 0.1718021944987224, + "grad_norm": 0.8357347973494355, + "learning_rate": 3.7929103016323183e-06, + "loss": 0.8587, + "step": 2286 + }, + { + "epoch": 0.17187734856455733, + "grad_norm": 2.3775839552311004, + "learning_rate": 3.792694511957867e-06, + "loss": 1.0451, + "step": 2287 + }, + { + "epoch": 0.1719525026303923, + "grad_norm": 1.6090351227316582, + "learning_rate": 3.7924786160594016e-06, + "loss": 1.1035, + "step": 2288 + }, + { + "epoch": 0.17202765669622727, + "grad_norm": 3.541960012217729, + "learning_rate": 3.792262613949714e-06, + "loss": 1.048, + "step": 2289 + }, + { + "epoch": 0.1721028107620622, + "grad_norm": 1.8019603915164157, + "learning_rate": 3.792046505641604e-06, + "loss": 1.0283, + "step": 2290 + }, + { + "epoch": 0.17217796482789718, + "grad_norm": 3.8132154700153205, + "learning_rate": 3.7918302911478764e-06, + "loss": 1.03, + "step": 2291 + }, + { + "epoch": 0.17225311889373215, + "grad_norm": 2.240888265681446, + "learning_rate": 3.791613970481342e-06, + "loss": 0.9911, + "step": 2292 + }, + { + "epoch": 0.17232827295956712, + "grad_norm": 2.0561032367495846, + "learning_rate": 3.7913975436548195e-06, + "loss": 0.94, + "step": 2293 + }, + { + "epoch": 0.17240342702540207, + "grad_norm": 1.8684258604093869, + "learning_rate": 3.7911810106811332e-06, + "loss": 1.0367, + "step": 2294 + }, + { + "epoch": 0.17247858109123704, + "grad_norm": 2.6741490195898563, + "learning_rate": 3.7909643715731133e-06, + "loss": 1.0326, + "step": 2295 + }, + { + "epoch": 0.172553735157072, + "grad_norm": 1.86474916228636, + "learning_rate": 3.790747626343596e-06, + "loss": 1.0495, + "step": 2296 + }, + { + "epoch": 0.17262888922290695, + "grad_norm": 1.7680974882270366, + "learning_rate": 3.7905307750054247e-06, + "loss": 1.0788, + "step": 2297 + }, + { + "epoch": 0.17270404328874192, + "grad_norm": 2.0216849849507326, + "learning_rate": 3.790313817571448e-06, + "loss": 1.0792, + "step": 2298 + }, + { + "epoch": 0.1727791973545769, + "grad_norm": 1.8162841755853993, + "learning_rate": 3.790096754054523e-06, + "loss": 0.9524, + "step": 2299 + }, + { + "epoch": 0.17285435142041183, + "grad_norm": 2.042586281680441, + "learning_rate": 3.7898795844675096e-06, + "loss": 1.0247, + "step": 2300 + }, + { + "epoch": 0.1729295054862468, + "grad_norm": 0.8731150760509905, + "learning_rate": 3.789662308823278e-06, + "loss": 0.8794, + "step": 2301 + }, + { + "epoch": 0.17300465955208177, + "grad_norm": 1.9168757725413395, + "learning_rate": 3.7894449271347e-06, + "loss": 1.016, + "step": 2302 + }, + { + "epoch": 0.17307981361791672, + "grad_norm": 4.298088207633783, + "learning_rate": 3.7892274394146592e-06, + "loss": 1.1583, + "step": 2303 + }, + { + "epoch": 0.1731549676837517, + "grad_norm": 1.5566545627022674, + "learning_rate": 3.789009845676041e-06, + "loss": 0.9946, + "step": 2304 + }, + { + "epoch": 0.17323012174958666, + "grad_norm": 2.9592378514059745, + "learning_rate": 3.7887921459317386e-06, + "loss": 1.0398, + "step": 2305 + }, + { + "epoch": 0.17330527581542163, + "grad_norm": 1.9826813927270104, + "learning_rate": 3.7885743401946517e-06, + "loss": 0.914, + "step": 2306 + }, + { + "epoch": 0.17338042988125657, + "grad_norm": 3.0213333417629276, + "learning_rate": 3.7883564284776863e-06, + "loss": 1.06, + "step": 2307 + }, + { + "epoch": 0.17345558394709154, + "grad_norm": 1.8295792884002964, + "learning_rate": 3.7881384107937546e-06, + "loss": 0.9986, + "step": 2308 + }, + { + "epoch": 0.1735307380129265, + "grad_norm": 1.924032783272621, + "learning_rate": 3.7879202871557742e-06, + "loss": 1.0202, + "step": 2309 + }, + { + "epoch": 0.17360589207876145, + "grad_norm": 0.7387442776737532, + "learning_rate": 3.7877020575766714e-06, + "loss": 0.8184, + "step": 2310 + }, + { + "epoch": 0.17368104614459642, + "grad_norm": 2.702109261777815, + "learning_rate": 3.7874837220693756e-06, + "loss": 0.9934, + "step": 2311 + }, + { + "epoch": 0.1737562002104314, + "grad_norm": 0.7589317393837853, + "learning_rate": 3.7872652806468244e-06, + "loss": 0.8606, + "step": 2312 + }, + { + "epoch": 0.17383135427626634, + "grad_norm": 2.48595699587981, + "learning_rate": 3.7870467333219614e-06, + "loss": 0.9871, + "step": 2313 + }, + { + "epoch": 0.1739065083421013, + "grad_norm": 8.297927358666563, + "learning_rate": 3.7868280801077368e-06, + "loss": 1.03, + "step": 2314 + }, + { + "epoch": 0.17398166240793628, + "grad_norm": 1.8433721122778577, + "learning_rate": 3.786609321017106e-06, + "loss": 1.0774, + "step": 2315 + }, + { + "epoch": 0.17405681647377122, + "grad_norm": 2.998573440881328, + "learning_rate": 3.7863904560630315e-06, + "loss": 1.0598, + "step": 2316 + }, + { + "epoch": 0.1741319705396062, + "grad_norm": 0.7137655245801188, + "learning_rate": 3.786171485258482e-06, + "loss": 0.8749, + "step": 2317 + }, + { + "epoch": 0.17420712460544116, + "grad_norm": 2.6144339124884715, + "learning_rate": 3.785952408616432e-06, + "loss": 1.0908, + "step": 2318 + }, + { + "epoch": 0.1742822786712761, + "grad_norm": 1.981795429038677, + "learning_rate": 3.7857332261498635e-06, + "loss": 1.0403, + "step": 2319 + }, + { + "epoch": 0.17435743273711107, + "grad_norm": 1.4851488204521393, + "learning_rate": 3.785513937871763e-06, + "loss": 1.0288, + "step": 2320 + }, + { + "epoch": 0.17443258680294604, + "grad_norm": 5.767611525787979, + "learning_rate": 3.785294543795125e-06, + "loss": 1.0715, + "step": 2321 + }, + { + "epoch": 0.174507740868781, + "grad_norm": 1.5851871613430435, + "learning_rate": 3.7850750439329477e-06, + "loss": 1.0035, + "step": 2322 + }, + { + "epoch": 0.17458289493461596, + "grad_norm": 1.8609971095354283, + "learning_rate": 3.7848554382982398e-06, + "loss": 1.0051, + "step": 2323 + }, + { + "epoch": 0.17465804900045093, + "grad_norm": 1.7416527444870147, + "learning_rate": 3.7846357269040115e-06, + "loss": 1.0313, + "step": 2324 + }, + { + "epoch": 0.1747332030662859, + "grad_norm": 1.5980100356587759, + "learning_rate": 3.784415909763283e-06, + "loss": 0.9463, + "step": 2325 + }, + { + "epoch": 0.17480835713212084, + "grad_norm": 2.3307874164370586, + "learning_rate": 3.784195986889079e-06, + "loss": 1.0721, + "step": 2326 + }, + { + "epoch": 0.1748835111979558, + "grad_norm": 3.0746672907771027, + "learning_rate": 3.7839759582944307e-06, + "loss": 1.093, + "step": 2327 + }, + { + "epoch": 0.17495866526379078, + "grad_norm": 3.7135803448917475, + "learning_rate": 3.783755823992376e-06, + "loss": 0.9232, + "step": 2328 + }, + { + "epoch": 0.17503381932962572, + "grad_norm": 1.5068463802902887, + "learning_rate": 3.783535583995957e-06, + "loss": 1.0516, + "step": 2329 + }, + { + "epoch": 0.1751089733954607, + "grad_norm": 2.0248674325759786, + "learning_rate": 3.783315238318226e-06, + "loss": 1.0397, + "step": 2330 + }, + { + "epoch": 0.17518412746129566, + "grad_norm": 2.1910304326947694, + "learning_rate": 3.7830947869722377e-06, + "loss": 0.9591, + "step": 2331 + }, + { + "epoch": 0.1752592815271306, + "grad_norm": 0.9929618216932424, + "learning_rate": 3.7828742299710558e-06, + "loss": 0.8567, + "step": 2332 + }, + { + "epoch": 0.17533443559296558, + "grad_norm": 2.864739727028143, + "learning_rate": 3.782653567327749e-06, + "loss": 1.1126, + "step": 2333 + }, + { + "epoch": 0.17540958965880055, + "grad_norm": 1.6009686369197733, + "learning_rate": 3.7824327990553914e-06, + "loss": 1.0276, + "step": 2334 + }, + { + "epoch": 0.1754847437246355, + "grad_norm": 2.4654858413317595, + "learning_rate": 3.7822119251670657e-06, + "loss": 0.9927, + "step": 2335 + }, + { + "epoch": 0.17555989779047046, + "grad_norm": 1.524716354006397, + "learning_rate": 3.7819909456758582e-06, + "loss": 1.0082, + "step": 2336 + }, + { + "epoch": 0.17563505185630543, + "grad_norm": 1.4823049466504865, + "learning_rate": 3.7817698605948643e-06, + "loss": 1.1028, + "step": 2337 + }, + { + "epoch": 0.1757102059221404, + "grad_norm": 1.727117388872299, + "learning_rate": 3.7815486699371826e-06, + "loss": 1.0651, + "step": 2338 + }, + { + "epoch": 0.17578535998797534, + "grad_norm": 1.573296136955184, + "learning_rate": 3.7813273737159205e-06, + "loss": 1.0215, + "step": 2339 + }, + { + "epoch": 0.1758605140538103, + "grad_norm": 1.909773400131506, + "learning_rate": 3.78110597194419e-06, + "loss": 1.0524, + "step": 2340 + }, + { + "epoch": 0.17593566811964528, + "grad_norm": 1.7909776190632516, + "learning_rate": 3.780884464635111e-06, + "loss": 0.9949, + "step": 2341 + }, + { + "epoch": 0.17601082218548023, + "grad_norm": 2.0170994769687964, + "learning_rate": 3.7806628518018074e-06, + "loss": 1.0253, + "step": 2342 + }, + { + "epoch": 0.1760859762513152, + "grad_norm": 2.2600698834982733, + "learning_rate": 3.7804411334574116e-06, + "loss": 0.9453, + "step": 2343 + }, + { + "epoch": 0.17616113031715017, + "grad_norm": 2.025707959409847, + "learning_rate": 3.7802193096150606e-06, + "loss": 1.01, + "step": 2344 + }, + { + "epoch": 0.1762362843829851, + "grad_norm": 2.081022837845789, + "learning_rate": 3.7799973802878985e-06, + "loss": 0.9821, + "step": 2345 + }, + { + "epoch": 0.17631143844882008, + "grad_norm": 1.573211650682262, + "learning_rate": 3.779775345489076e-06, + "loss": 0.9665, + "step": 2346 + }, + { + "epoch": 0.17638659251465505, + "grad_norm": 2.3965376993354264, + "learning_rate": 3.779553205231749e-06, + "loss": 1.0024, + "step": 2347 + }, + { + "epoch": 0.17646174658049, + "grad_norm": 1.8131408873752541, + "learning_rate": 3.77933095952908e-06, + "loss": 1.0297, + "step": 2348 + }, + { + "epoch": 0.17653690064632496, + "grad_norm": 2.1083036739791226, + "learning_rate": 3.779108608394238e-06, + "loss": 0.9654, + "step": 2349 + }, + { + "epoch": 0.17661205471215993, + "grad_norm": 3.4617462935371033, + "learning_rate": 3.7788861518403988e-06, + "loss": 0.9436, + "step": 2350 + }, + { + "epoch": 0.1766872087779949, + "grad_norm": 1.668309625818564, + "learning_rate": 3.778663589880743e-06, + "loss": 1.084, + "step": 2351 + }, + { + "epoch": 0.17676236284382985, + "grad_norm": 1.7259813944718843, + "learning_rate": 3.7784409225284585e-06, + "loss": 0.9594, + "step": 2352 + }, + { + "epoch": 0.17683751690966482, + "grad_norm": 2.0762759135062825, + "learning_rate": 3.7782181497967393e-06, + "loss": 0.977, + "step": 2353 + }, + { + "epoch": 0.1769126709754998, + "grad_norm": 2.2577032042980116, + "learning_rate": 3.7779952716987856e-06, + "loss": 0.9665, + "step": 2354 + }, + { + "epoch": 0.17698782504133473, + "grad_norm": 1.8402288402392357, + "learning_rate": 3.7777722882478032e-06, + "loss": 0.9709, + "step": 2355 + }, + { + "epoch": 0.1770629791071697, + "grad_norm": 1.9491377146662194, + "learning_rate": 3.7775491994570057e-06, + "loss": 1.0166, + "step": 2356 + }, + { + "epoch": 0.17713813317300467, + "grad_norm": 1.5295642214732548, + "learning_rate": 3.777326005339611e-06, + "loss": 0.9648, + "step": 2357 + }, + { + "epoch": 0.1772132872388396, + "grad_norm": 0.7048850389521003, + "learning_rate": 3.7771027059088454e-06, + "loss": 0.854, + "step": 2358 + }, + { + "epoch": 0.17728844130467458, + "grad_norm": 2.686603929041143, + "learning_rate": 3.7768793011779383e-06, + "loss": 1.1185, + "step": 2359 + }, + { + "epoch": 0.17736359537050955, + "grad_norm": 1.9020632433239448, + "learning_rate": 3.7766557911601295e-06, + "loss": 1.0281, + "step": 2360 + }, + { + "epoch": 0.1774387494363445, + "grad_norm": 1.804437000674808, + "learning_rate": 3.7764321758686614e-06, + "loss": 1.0541, + "step": 2361 + }, + { + "epoch": 0.17751390350217947, + "grad_norm": 7.346779056573262, + "learning_rate": 3.7762084553167846e-06, + "loss": 1.0168, + "step": 2362 + }, + { + "epoch": 0.17758905756801444, + "grad_norm": 2.8644912057569467, + "learning_rate": 3.7759846295177552e-06, + "loss": 1.0298, + "step": 2363 + }, + { + "epoch": 0.17766421163384938, + "grad_norm": 13.988931367537123, + "learning_rate": 3.775760698484836e-06, + "loss": 1.0488, + "step": 2364 + }, + { + "epoch": 0.17773936569968435, + "grad_norm": 1.6610860218094023, + "learning_rate": 3.7755366622312954e-06, + "loss": 1.0407, + "step": 2365 + }, + { + "epoch": 0.17781451976551932, + "grad_norm": 1.584377780889294, + "learning_rate": 3.7753125207704084e-06, + "loss": 1.0501, + "step": 2366 + }, + { + "epoch": 0.1778896738313543, + "grad_norm": 1.6638419851415505, + "learning_rate": 3.7750882741154566e-06, + "loss": 1.0566, + "step": 2367 + }, + { + "epoch": 0.17796482789718923, + "grad_norm": 1.8164267261950706, + "learning_rate": 3.774863922279727e-06, + "loss": 1.0906, + "step": 2368 + }, + { + "epoch": 0.1780399819630242, + "grad_norm": 1.9696309979490232, + "learning_rate": 3.7746394652765136e-06, + "loss": 0.9003, + "step": 2369 + }, + { + "epoch": 0.17811513602885917, + "grad_norm": 0.8473561542383163, + "learning_rate": 3.774414903119117e-06, + "loss": 0.8891, + "step": 2370 + }, + { + "epoch": 0.17819029009469411, + "grad_norm": 2.4009747830313506, + "learning_rate": 3.7741902358208427e-06, + "loss": 1.0044, + "step": 2371 + }, + { + "epoch": 0.17826544416052909, + "grad_norm": 1.884795098741373, + "learning_rate": 3.7739654633950023e-06, + "loss": 1.0968, + "step": 2372 + }, + { + "epoch": 0.17834059822636406, + "grad_norm": 1.7125444225480515, + "learning_rate": 3.7737405858549156e-06, + "loss": 1.018, + "step": 2373 + }, + { + "epoch": 0.178415752292199, + "grad_norm": 1.63032897730305, + "learning_rate": 3.7735156032139066e-06, + "loss": 1.0749, + "step": 2374 + }, + { + "epoch": 0.17849090635803397, + "grad_norm": 2.5461171900889603, + "learning_rate": 3.773290515485308e-06, + "loss": 0.9016, + "step": 2375 + }, + { + "epoch": 0.17856606042386894, + "grad_norm": 3.0709032262114446, + "learning_rate": 3.773065322682455e-06, + "loss": 1.0052, + "step": 2376 + }, + { + "epoch": 0.17864121448970388, + "grad_norm": 1.5833680333133093, + "learning_rate": 3.772840024818692e-06, + "loss": 0.921, + "step": 2377 + }, + { + "epoch": 0.17871636855553885, + "grad_norm": 2.8030336226897457, + "learning_rate": 3.7726146219073697e-06, + "loss": 1.08, + "step": 2378 + }, + { + "epoch": 0.17879152262137382, + "grad_norm": 1.628098834334566, + "learning_rate": 3.772389113961843e-06, + "loss": 1.034, + "step": 2379 + }, + { + "epoch": 0.17886667668720876, + "grad_norm": 2.618044949034217, + "learning_rate": 3.772163500995474e-06, + "loss": 1.0309, + "step": 2380 + }, + { + "epoch": 0.17894183075304373, + "grad_norm": 1.6735224227656802, + "learning_rate": 3.771937783021632e-06, + "loss": 1.0399, + "step": 2381 + }, + { + "epoch": 0.1790169848188787, + "grad_norm": 2.181570244920046, + "learning_rate": 3.771711960053691e-06, + "loss": 1.0232, + "step": 2382 + }, + { + "epoch": 0.17909213888471368, + "grad_norm": 2.954511367494513, + "learning_rate": 3.7714860321050316e-06, + "loss": 0.903, + "step": 2383 + }, + { + "epoch": 0.17916729295054862, + "grad_norm": 1.9324167079174022, + "learning_rate": 3.771259999189042e-06, + "loss": 1.0053, + "step": 2384 + }, + { + "epoch": 0.1792424470163836, + "grad_norm": 1.5785422198522145, + "learning_rate": 3.7710338613191145e-06, + "loss": 1.0112, + "step": 2385 + }, + { + "epoch": 0.17931760108221856, + "grad_norm": 1.9361017780939487, + "learning_rate": 3.770807618508649e-06, + "loss": 1.0056, + "step": 2386 + }, + { + "epoch": 0.1793927551480535, + "grad_norm": 1.90128969290266, + "learning_rate": 3.770581270771051e-06, + "loss": 1.0526, + "step": 2387 + }, + { + "epoch": 0.17946790921388847, + "grad_norm": 1.8391384060475495, + "learning_rate": 3.770354818119733e-06, + "loss": 1.0896, + "step": 2388 + }, + { + "epoch": 0.17954306327972344, + "grad_norm": 0.777793217592789, + "learning_rate": 3.7701282605681123e-06, + "loss": 0.8539, + "step": 2389 + }, + { + "epoch": 0.17961821734555838, + "grad_norm": 17.551756263953152, + "learning_rate": 3.769901598129615e-06, + "loss": 1.0826, + "step": 2390 + }, + { + "epoch": 0.17969337141139335, + "grad_norm": 2.2968984589708694, + "learning_rate": 3.7696748308176698e-06, + "loss": 1.0457, + "step": 2391 + }, + { + "epoch": 0.17976852547722832, + "grad_norm": 1.968576575012883, + "learning_rate": 3.7694479586457144e-06, + "loss": 1.1158, + "step": 2392 + }, + { + "epoch": 0.17984367954306327, + "grad_norm": 1.6067402577595271, + "learning_rate": 3.7692209816271915e-06, + "loss": 1.0287, + "step": 2393 + }, + { + "epoch": 0.17991883360889824, + "grad_norm": 2.515213547697539, + "learning_rate": 3.7689938997755512e-06, + "loss": 1.0728, + "step": 2394 + }, + { + "epoch": 0.1799939876747332, + "grad_norm": 1.7877534299872773, + "learning_rate": 3.7687667131042487e-06, + "loss": 0.9649, + "step": 2395 + }, + { + "epoch": 0.18006914174056818, + "grad_norm": 3.074382695209561, + "learning_rate": 3.7685394216267444e-06, + "loss": 1.0566, + "step": 2396 + }, + { + "epoch": 0.18014429580640312, + "grad_norm": 2.592804242254622, + "learning_rate": 3.7683120253565076e-06, + "loss": 0.9965, + "step": 2397 + }, + { + "epoch": 0.1802194498722381, + "grad_norm": 2.3444576975170137, + "learning_rate": 3.7680845243070128e-06, + "loss": 1.004, + "step": 2398 + }, + { + "epoch": 0.18029460393807306, + "grad_norm": 3.420204207586769, + "learning_rate": 3.767856918491739e-06, + "loss": 1.0075, + "step": 2399 + }, + { + "epoch": 0.180369758003908, + "grad_norm": 1.5432767536483663, + "learning_rate": 3.767629207924172e-06, + "loss": 1.1136, + "step": 2400 + }, + { + "epoch": 0.18044491206974297, + "grad_norm": 1.89894907145736, + "learning_rate": 3.767401392617807e-06, + "loss": 0.9924, + "step": 2401 + }, + { + "epoch": 0.18052006613557794, + "grad_norm": 1.5410916656569775, + "learning_rate": 3.7671734725861413e-06, + "loss": 0.9964, + "step": 2402 + }, + { + "epoch": 0.1805952202014129, + "grad_norm": 2.888462226693328, + "learning_rate": 3.7669454478426806e-06, + "loss": 1.0152, + "step": 2403 + }, + { + "epoch": 0.18067037426724786, + "grad_norm": 4.031904300553072, + "learning_rate": 3.7667173184009356e-06, + "loss": 0.9587, + "step": 2404 + }, + { + "epoch": 0.18074552833308283, + "grad_norm": 1.9401070067792412, + "learning_rate": 3.7664890842744248e-06, + "loss": 0.9591, + "step": 2405 + }, + { + "epoch": 0.18082068239891777, + "grad_norm": 3.356257353762166, + "learning_rate": 3.7662607454766712e-06, + "loss": 0.9184, + "step": 2406 + }, + { + "epoch": 0.18089583646475274, + "grad_norm": 1.4666716297383577, + "learning_rate": 3.7660323020212047e-06, + "loss": 1.1235, + "step": 2407 + }, + { + "epoch": 0.1809709905305877, + "grad_norm": 12.868604534525709, + "learning_rate": 3.765803753921562e-06, + "loss": 1.0381, + "step": 2408 + }, + { + "epoch": 0.18104614459642265, + "grad_norm": 2.123132895207092, + "learning_rate": 3.7655751011912852e-06, + "loss": 0.9865, + "step": 2409 + }, + { + "epoch": 0.18112129866225762, + "grad_norm": 1.6126703394143234, + "learning_rate": 3.7653463438439225e-06, + "loss": 0.9644, + "step": 2410 + }, + { + "epoch": 0.1811964527280926, + "grad_norm": 2.0542416397041596, + "learning_rate": 3.7651174818930293e-06, + "loss": 1.1213, + "step": 2411 + }, + { + "epoch": 0.18127160679392756, + "grad_norm": 6.236959356755683, + "learning_rate": 3.764888515352166e-06, + "loss": 0.9643, + "step": 2412 + }, + { + "epoch": 0.1813467608597625, + "grad_norm": 1.9708454545705485, + "learning_rate": 3.7646594442349004e-06, + "loss": 1.0468, + "step": 2413 + }, + { + "epoch": 0.18142191492559748, + "grad_norm": 0.9180165379848404, + "learning_rate": 3.764430268554805e-06, + "loss": 0.8808, + "step": 2414 + }, + { + "epoch": 0.18149706899143245, + "grad_norm": 0.8398251553628292, + "learning_rate": 3.7642009883254594e-06, + "loss": 0.8627, + "step": 2415 + }, + { + "epoch": 0.1815722230572674, + "grad_norm": 1.764540752931511, + "learning_rate": 3.7639716035604502e-06, + "loss": 1.0622, + "step": 2416 + }, + { + "epoch": 0.18164737712310236, + "grad_norm": 1.7845881453378045, + "learning_rate": 3.763742114273369e-06, + "loss": 1.016, + "step": 2417 + }, + { + "epoch": 0.18172253118893733, + "grad_norm": 1.5563637596130728, + "learning_rate": 3.763512520477813e-06, + "loss": 1.026, + "step": 2418 + }, + { + "epoch": 0.18179768525477227, + "grad_norm": 1.3869318408844593, + "learning_rate": 3.7632828221873876e-06, + "loss": 0.9905, + "step": 2419 + }, + { + "epoch": 0.18187283932060724, + "grad_norm": 1.8090205502584331, + "learning_rate": 3.763053019415703e-06, + "loss": 0.9828, + "step": 2420 + }, + { + "epoch": 0.1819479933864422, + "grad_norm": 1.7052856179233697, + "learning_rate": 3.7628231121763757e-06, + "loss": 1.0451, + "step": 2421 + }, + { + "epoch": 0.18202314745227716, + "grad_norm": 6.443256952050453, + "learning_rate": 3.7625931004830287e-06, + "loss": 1.1064, + "step": 2422 + }, + { + "epoch": 0.18209830151811213, + "grad_norm": 1.9541495115697047, + "learning_rate": 3.762362984349291e-06, + "loss": 0.9386, + "step": 2423 + }, + { + "epoch": 0.1821734555839471, + "grad_norm": 1.9204441109261468, + "learning_rate": 3.762132763788798e-06, + "loss": 1.0227, + "step": 2424 + }, + { + "epoch": 0.18224860964978204, + "grad_norm": 2.280139600595886, + "learning_rate": 3.7619024388151914e-06, + "loss": 0.9942, + "step": 2425 + }, + { + "epoch": 0.182323763715617, + "grad_norm": 1.954312383211629, + "learning_rate": 3.761672009442118e-06, + "loss": 1.0116, + "step": 2426 + }, + { + "epoch": 0.18239891778145198, + "grad_norm": 2.3163557127510415, + "learning_rate": 3.7614414756832328e-06, + "loss": 0.9974, + "step": 2427 + }, + { + "epoch": 0.18247407184728695, + "grad_norm": 2.130856354009373, + "learning_rate": 3.7612108375521942e-06, + "loss": 1.0243, + "step": 2428 + }, + { + "epoch": 0.1825492259131219, + "grad_norm": 1.5337232740043212, + "learning_rate": 3.76098009506267e-06, + "loss": 0.988, + "step": 2429 + }, + { + "epoch": 0.18262437997895686, + "grad_norm": 2.950700359937576, + "learning_rate": 3.7607492482283315e-06, + "loss": 0.9227, + "step": 2430 + }, + { + "epoch": 0.18269953404479183, + "grad_norm": 1.8421534074859336, + "learning_rate": 3.7605182970628583e-06, + "loss": 0.9445, + "step": 2431 + }, + { + "epoch": 0.18277468811062678, + "grad_norm": 2.1155751976112898, + "learning_rate": 3.7602872415799347e-06, + "loss": 0.9984, + "step": 2432 + }, + { + "epoch": 0.18284984217646175, + "grad_norm": 7.0052636656058995, + "learning_rate": 3.7600560817932506e-06, + "loss": 1.0213, + "step": 2433 + }, + { + "epoch": 0.18292499624229672, + "grad_norm": 1.8224817538296487, + "learning_rate": 3.759824817716504e-06, + "loss": 0.9468, + "step": 2434 + }, + { + "epoch": 0.18300015030813166, + "grad_norm": 2.5010704609776884, + "learning_rate": 3.7595934493633986e-06, + "loss": 1.0102, + "step": 2435 + }, + { + "epoch": 0.18307530437396663, + "grad_norm": 1.782992263811394, + "learning_rate": 3.7593619767476435e-06, + "loss": 1.1201, + "step": 2436 + }, + { + "epoch": 0.1831504584398016, + "grad_norm": 0.9742203664269149, + "learning_rate": 3.759130399882954e-06, + "loss": 0.8269, + "step": 2437 + }, + { + "epoch": 0.18322561250563654, + "grad_norm": 1.9084425027374765, + "learning_rate": 3.758898718783052e-06, + "loss": 0.9381, + "step": 2438 + }, + { + "epoch": 0.1833007665714715, + "grad_norm": 1.8748156774838385, + "learning_rate": 3.758666933461666e-06, + "loss": 1.0764, + "step": 2439 + }, + { + "epoch": 0.18337592063730648, + "grad_norm": 1.5257174914623848, + "learning_rate": 3.7584350439325295e-06, + "loss": 0.9479, + "step": 2440 + }, + { + "epoch": 0.18345107470314145, + "grad_norm": 1.692649402081633, + "learning_rate": 3.7582030502093833e-06, + "loss": 0.8764, + "step": 2441 + }, + { + "epoch": 0.1835262287689764, + "grad_norm": 1.6070360390420713, + "learning_rate": 3.7579709523059736e-06, + "loss": 0.9457, + "step": 2442 + }, + { + "epoch": 0.18360138283481137, + "grad_norm": 1.9639335935984947, + "learning_rate": 3.7577387502360535e-06, + "loss": 0.9953, + "step": 2443 + }, + { + "epoch": 0.18367653690064634, + "grad_norm": 2.5470961024257948, + "learning_rate": 3.757506444013381e-06, + "loss": 1.0629, + "step": 2444 + }, + { + "epoch": 0.18375169096648128, + "grad_norm": 1.4653805168898608, + "learning_rate": 3.7572740336517225e-06, + "loss": 1.0039, + "step": 2445 + }, + { + "epoch": 0.18382684503231625, + "grad_norm": 2.719272607143846, + "learning_rate": 3.757041519164848e-06, + "loss": 1.1087, + "step": 2446 + }, + { + "epoch": 0.18390199909815122, + "grad_norm": 0.8069829329939538, + "learning_rate": 3.7568089005665353e-06, + "loss": 0.8451, + "step": 2447 + }, + { + "epoch": 0.18397715316398616, + "grad_norm": 2.112713016401625, + "learning_rate": 3.7565761778705682e-06, + "loss": 1.0885, + "step": 2448 + }, + { + "epoch": 0.18405230722982113, + "grad_norm": 1.519023349479539, + "learning_rate": 3.756343351090736e-06, + "loss": 1.0041, + "step": 2449 + }, + { + "epoch": 0.1841274612956561, + "grad_norm": 2.7761682955052027, + "learning_rate": 3.756110420240835e-06, + "loss": 1.0022, + "step": 2450 + }, + { + "epoch": 0.18420261536149105, + "grad_norm": 1.6546525268893197, + "learning_rate": 3.755877385334667e-06, + "loss": 1.0455, + "step": 2451 + }, + { + "epoch": 0.18427776942732602, + "grad_norm": 1.4381809280442295, + "learning_rate": 3.7556442463860406e-06, + "loss": 1.0352, + "step": 2452 + }, + { + "epoch": 0.184352923493161, + "grad_norm": 1.9150930938801745, + "learning_rate": 3.7554110034087686e-06, + "loss": 1.0142, + "step": 2453 + }, + { + "epoch": 0.18442807755899593, + "grad_norm": 2.6568438595400803, + "learning_rate": 3.7551776564166736e-06, + "loss": 0.997, + "step": 2454 + }, + { + "epoch": 0.1845032316248309, + "grad_norm": 2.678571374075753, + "learning_rate": 3.7549442054235813e-06, + "loss": 1.0196, + "step": 2455 + }, + { + "epoch": 0.18457838569066587, + "grad_norm": 2.588234886447344, + "learning_rate": 3.754710650443325e-06, + "loss": 0.9053, + "step": 2456 + }, + { + "epoch": 0.18465353975650084, + "grad_norm": 2.471842313953062, + "learning_rate": 3.754476991489743e-06, + "loss": 1.0456, + "step": 2457 + }, + { + "epoch": 0.18472869382233578, + "grad_norm": 1.9502113276680293, + "learning_rate": 3.754243228576681e-06, + "loss": 1.0654, + "step": 2458 + }, + { + "epoch": 0.18480384788817075, + "grad_norm": 1.3545907994728217, + "learning_rate": 3.7540093617179904e-06, + "loss": 1.0767, + "step": 2459 + }, + { + "epoch": 0.18487900195400572, + "grad_norm": 1.716310469082368, + "learning_rate": 3.7537753909275284e-06, + "loss": 1.0515, + "step": 2460 + }, + { + "epoch": 0.18495415601984067, + "grad_norm": 1.7077882253696393, + "learning_rate": 3.7535413162191584e-06, + "loss": 1.0606, + "step": 2461 + }, + { + "epoch": 0.18502931008567564, + "grad_norm": 2.0583344211137913, + "learning_rate": 3.7533071376067514e-06, + "loss": 0.9844, + "step": 2462 + }, + { + "epoch": 0.1851044641515106, + "grad_norm": 1.7670883596118387, + "learning_rate": 3.7530728551041825e-06, + "loss": 1.0099, + "step": 2463 + }, + { + "epoch": 0.18517961821734555, + "grad_norm": 5.95577695555224, + "learning_rate": 3.7528384687253335e-06, + "loss": 1.0171, + "step": 2464 + }, + { + "epoch": 0.18525477228318052, + "grad_norm": 1.9198764861631574, + "learning_rate": 3.752603978484094e-06, + "loss": 0.9781, + "step": 2465 + }, + { + "epoch": 0.1853299263490155, + "grad_norm": 2.9385091793117497, + "learning_rate": 3.752369384394357e-06, + "loss": 1.0445, + "step": 2466 + }, + { + "epoch": 0.18540508041485043, + "grad_norm": 2.096998285572113, + "learning_rate": 3.7521346864700235e-06, + "loss": 1.0463, + "step": 2467 + }, + { + "epoch": 0.1854802344806854, + "grad_norm": 2.0942192121239254, + "learning_rate": 3.751899884725001e-06, + "loss": 0.989, + "step": 2468 + }, + { + "epoch": 0.18555538854652037, + "grad_norm": 0.821986135068848, + "learning_rate": 3.751664979173202e-06, + "loss": 0.8849, + "step": 2469 + }, + { + "epoch": 0.18563054261235531, + "grad_norm": 1.4997815926809572, + "learning_rate": 3.7514299698285447e-06, + "loss": 0.9681, + "step": 2470 + }, + { + "epoch": 0.18570569667819029, + "grad_norm": 0.7206722345942838, + "learning_rate": 3.751194856704955e-06, + "loss": 0.8257, + "step": 2471 + }, + { + "epoch": 0.18578085074402526, + "grad_norm": 1.7568592793057487, + "learning_rate": 3.750959639816365e-06, + "loss": 1.0395, + "step": 2472 + }, + { + "epoch": 0.18585600480986023, + "grad_norm": 3.8366167822764643, + "learning_rate": 3.750724319176711e-06, + "loss": 0.8816, + "step": 2473 + }, + { + "epoch": 0.18593115887569517, + "grad_norm": 0.7997769268281094, + "learning_rate": 3.7504888947999367e-06, + "loss": 0.9107, + "step": 2474 + }, + { + "epoch": 0.18600631294153014, + "grad_norm": 1.6280989807744277, + "learning_rate": 3.7502533666999935e-06, + "loss": 1.0053, + "step": 2475 + }, + { + "epoch": 0.1860814670073651, + "grad_norm": 1.7189497398143474, + "learning_rate": 3.7500177348908354e-06, + "loss": 0.9629, + "step": 2476 + }, + { + "epoch": 0.18615662107320005, + "grad_norm": 1.5745553918024757, + "learning_rate": 3.749781999386425e-06, + "loss": 1.0143, + "step": 2477 + }, + { + "epoch": 0.18623177513903502, + "grad_norm": 2.8884150124557393, + "learning_rate": 3.749546160200731e-06, + "loss": 0.9442, + "step": 2478 + }, + { + "epoch": 0.18630692920487, + "grad_norm": 1.6869135650036895, + "learning_rate": 3.7493102173477277e-06, + "loss": 1.1328, + "step": 2479 + }, + { + "epoch": 0.18638208327070493, + "grad_norm": 2.4897786998839377, + "learning_rate": 3.7490741708413954e-06, + "loss": 1.1279, + "step": 2480 + }, + { + "epoch": 0.1864572373365399, + "grad_norm": 2.0578519291673856, + "learning_rate": 3.748838020695721e-06, + "loss": 0.9912, + "step": 2481 + }, + { + "epoch": 0.18653239140237488, + "grad_norm": 2.332070406228799, + "learning_rate": 3.748601766924697e-06, + "loss": 1.0162, + "step": 2482 + }, + { + "epoch": 0.18660754546820982, + "grad_norm": 1.0157734689031623, + "learning_rate": 3.7483654095423223e-06, + "loss": 0.8194, + "step": 2483 + }, + { + "epoch": 0.1866826995340448, + "grad_norm": 2.1911006177174226, + "learning_rate": 3.7481289485626024e-06, + "loss": 1.0079, + "step": 2484 + }, + { + "epoch": 0.18675785359987976, + "grad_norm": 1.9417152082190379, + "learning_rate": 3.7478923839995477e-06, + "loss": 1.0007, + "step": 2485 + }, + { + "epoch": 0.18683300766571473, + "grad_norm": 1.6655653493767315, + "learning_rate": 3.7476557158671768e-06, + "loss": 0.9902, + "step": 2486 + }, + { + "epoch": 0.18690816173154967, + "grad_norm": 2.4699377751717178, + "learning_rate": 3.747418944179512e-06, + "loss": 1.0411, + "step": 2487 + }, + { + "epoch": 0.18698331579738464, + "grad_norm": 6.617194417794495, + "learning_rate": 3.747182068950584e-06, + "loss": 1.008, + "step": 2488 + }, + { + "epoch": 0.1870584698632196, + "grad_norm": 1.6372689935942883, + "learning_rate": 3.746945090194428e-06, + "loss": 1.0889, + "step": 2489 + }, + { + "epoch": 0.18713362392905455, + "grad_norm": 1.582226927158835, + "learning_rate": 3.7467080079250853e-06, + "loss": 1.0992, + "step": 2490 + }, + { + "epoch": 0.18720877799488952, + "grad_norm": 0.9652433919080207, + "learning_rate": 3.7464708221566052e-06, + "loss": 0.8863, + "step": 2491 + }, + { + "epoch": 0.1872839320607245, + "grad_norm": 1.8372434475914654, + "learning_rate": 3.7462335329030408e-06, + "loss": 1.1562, + "step": 2492 + }, + { + "epoch": 0.18735908612655944, + "grad_norm": 2.147471706750225, + "learning_rate": 3.7459961401784527e-06, + "loss": 1.0568, + "step": 2493 + }, + { + "epoch": 0.1874342401923944, + "grad_norm": 1.837574590090293, + "learning_rate": 3.7457586439969076e-06, + "loss": 1.0478, + "step": 2494 + }, + { + "epoch": 0.18750939425822938, + "grad_norm": 2.215158914415235, + "learning_rate": 3.745521044372478e-06, + "loss": 1.1185, + "step": 2495 + }, + { + "epoch": 0.18758454832406432, + "grad_norm": 2.689774637389733, + "learning_rate": 3.745283341319242e-06, + "loss": 1.0758, + "step": 2496 + }, + { + "epoch": 0.1876597023898993, + "grad_norm": 2.0475628112176065, + "learning_rate": 3.7450455348512854e-06, + "loss": 0.8519, + "step": 2497 + }, + { + "epoch": 0.18773485645573426, + "grad_norm": 2.325878108746373, + "learning_rate": 3.7448076249826987e-06, + "loss": 0.9534, + "step": 2498 + }, + { + "epoch": 0.1878100105215692, + "grad_norm": 1.9325084376181787, + "learning_rate": 3.7445696117275785e-06, + "loss": 0.979, + "step": 2499 + }, + { + "epoch": 0.18788516458740417, + "grad_norm": 2.2288754947437814, + "learning_rate": 3.7443314951000285e-06, + "loss": 0.9404, + "step": 2500 + }, + { + "epoch": 0.18796031865323914, + "grad_norm": 2.8581640462170625, + "learning_rate": 3.744093275114158e-06, + "loss": 1.012, + "step": 2501 + }, + { + "epoch": 0.18803547271907411, + "grad_norm": 0.7935851759502801, + "learning_rate": 3.7438549517840823e-06, + "loss": 0.8821, + "step": 2502 + }, + { + "epoch": 0.18811062678490906, + "grad_norm": 0.9237133704886757, + "learning_rate": 3.743616525123923e-06, + "loss": 0.8996, + "step": 2503 + }, + { + "epoch": 0.18818578085074403, + "grad_norm": 1.4269548158716543, + "learning_rate": 3.743377995147808e-06, + "loss": 0.9783, + "step": 2504 + }, + { + "epoch": 0.188260934916579, + "grad_norm": 2.0907352435549025, + "learning_rate": 3.743139361869871e-06, + "loss": 0.9811, + "step": 2505 + }, + { + "epoch": 0.18833608898241394, + "grad_norm": 2.11567414309788, + "learning_rate": 3.7429006253042524e-06, + "loss": 1.0475, + "step": 2506 + }, + { + "epoch": 0.1884112430482489, + "grad_norm": 2.4944643723112634, + "learning_rate": 3.742661785465097e-06, + "loss": 0.9325, + "step": 2507 + }, + { + "epoch": 0.18848639711408388, + "grad_norm": 2.0136523389192367, + "learning_rate": 3.7424228423665578e-06, + "loss": 1.0269, + "step": 2508 + }, + { + "epoch": 0.18856155117991882, + "grad_norm": 1.8441820446648254, + "learning_rate": 3.7421837960227933e-06, + "loss": 0.9846, + "step": 2509 + }, + { + "epoch": 0.1886367052457538, + "grad_norm": 1.6279881952546675, + "learning_rate": 3.741944646447967e-06, + "loss": 0.9686, + "step": 2510 + }, + { + "epoch": 0.18871185931158876, + "grad_norm": 1.5572850749238845, + "learning_rate": 3.7417053936562503e-06, + "loss": 0.9917, + "step": 2511 + }, + { + "epoch": 0.1887870133774237, + "grad_norm": 1.5104301539728733, + "learning_rate": 3.7414660376618195e-06, + "loss": 0.9138, + "step": 2512 + }, + { + "epoch": 0.18886216744325868, + "grad_norm": 1.55924312668448, + "learning_rate": 3.7412265784788577e-06, + "loss": 1.1101, + "step": 2513 + }, + { + "epoch": 0.18893732150909365, + "grad_norm": 4.002091814489213, + "learning_rate": 3.7409870161215532e-06, + "loss": 0.9373, + "step": 2514 + }, + { + "epoch": 0.1890124755749286, + "grad_norm": 1.5032940299626152, + "learning_rate": 3.740747350604102e-06, + "loss": 1.0729, + "step": 2515 + }, + { + "epoch": 0.18908762964076356, + "grad_norm": 1.7522735812951225, + "learning_rate": 3.7405075819407045e-06, + "loss": 1.0046, + "step": 2516 + }, + { + "epoch": 0.18916278370659853, + "grad_norm": 0.7585528188833983, + "learning_rate": 3.7402677101455672e-06, + "loss": 0.8478, + "step": 2517 + }, + { + "epoch": 0.1892379377724335, + "grad_norm": 9.905195980376561, + "learning_rate": 3.740027735232904e-06, + "loss": 1.0713, + "step": 2518 + }, + { + "epoch": 0.18931309183826844, + "grad_norm": 2.6573429327525186, + "learning_rate": 3.7397876572169355e-06, + "loss": 1.1001, + "step": 2519 + }, + { + "epoch": 0.1893882459041034, + "grad_norm": 2.1570371213927695, + "learning_rate": 3.7395474761118856e-06, + "loss": 0.9147, + "step": 2520 + }, + { + "epoch": 0.18946339996993838, + "grad_norm": 1.6908490185913303, + "learning_rate": 3.7393071919319864e-06, + "loss": 1.0854, + "step": 2521 + }, + { + "epoch": 0.18953855403577333, + "grad_norm": 1.502813958167124, + "learning_rate": 3.739066804691476e-06, + "loss": 1.074, + "step": 2522 + }, + { + "epoch": 0.1896137081016083, + "grad_norm": 2.814242353098371, + "learning_rate": 3.738826314404598e-06, + "loss": 1.0156, + "step": 2523 + }, + { + "epoch": 0.18968886216744327, + "grad_norm": 1.5375943224747521, + "learning_rate": 3.738585721085603e-06, + "loss": 1.1416, + "step": 2524 + }, + { + "epoch": 0.1897640162332782, + "grad_norm": 4.214325796126061, + "learning_rate": 3.738345024748746e-06, + "loss": 0.9887, + "step": 2525 + }, + { + "epoch": 0.18983917029911318, + "grad_norm": 2.042630286106072, + "learning_rate": 3.73810422540829e-06, + "loss": 1.0418, + "step": 2526 + }, + { + "epoch": 0.18991432436494815, + "grad_norm": 2.3112183189387046, + "learning_rate": 3.7378633230785025e-06, + "loss": 1.077, + "step": 2527 + }, + { + "epoch": 0.1899894784307831, + "grad_norm": 1.503613331990403, + "learning_rate": 3.7376223177736587e-06, + "loss": 1.0198, + "step": 2528 + }, + { + "epoch": 0.19006463249661806, + "grad_norm": 1.7064620057466515, + "learning_rate": 3.737381209508039e-06, + "loss": 0.8485, + "step": 2529 + }, + { + "epoch": 0.19013978656245303, + "grad_norm": 3.308271768479157, + "learning_rate": 3.7371399982959294e-06, + "loss": 0.9816, + "step": 2530 + }, + { + "epoch": 0.190214940628288, + "grad_norm": 1.787716688346951, + "learning_rate": 3.736898684151623e-06, + "loss": 1.0499, + "step": 2531 + }, + { + "epoch": 0.19029009469412295, + "grad_norm": 2.088745741414649, + "learning_rate": 3.736657267089419e-06, + "loss": 0.9859, + "step": 2532 + }, + { + "epoch": 0.19036524875995792, + "grad_norm": 1.4524642098198135, + "learning_rate": 3.7364157471236215e-06, + "loss": 0.9961, + "step": 2533 + }, + { + "epoch": 0.1904404028257929, + "grad_norm": 2.173164808003609, + "learning_rate": 3.7361741242685417e-06, + "loss": 1.0634, + "step": 2534 + }, + { + "epoch": 0.19051555689162783, + "grad_norm": 1.9562407398918773, + "learning_rate": 3.7359323985384966e-06, + "loss": 0.9469, + "step": 2535 + }, + { + "epoch": 0.1905907109574628, + "grad_norm": 1.8199798541997636, + "learning_rate": 3.7356905699478096e-06, + "loss": 1.0059, + "step": 2536 + }, + { + "epoch": 0.19066586502329777, + "grad_norm": 2.478341122760174, + "learning_rate": 3.7354486385108103e-06, + "loss": 0.9645, + "step": 2537 + }, + { + "epoch": 0.1907410190891327, + "grad_norm": 2.5874738940704964, + "learning_rate": 3.735206604241834e-06, + "loss": 1.0643, + "step": 2538 + }, + { + "epoch": 0.19081617315496768, + "grad_norm": 3.06491336320461, + "learning_rate": 3.734964467155221e-06, + "loss": 1.0088, + "step": 2539 + }, + { + "epoch": 0.19089132722080265, + "grad_norm": 2.447361291962426, + "learning_rate": 3.73472222726532e-06, + "loss": 0.9237, + "step": 2540 + }, + { + "epoch": 0.1909664812866376, + "grad_norm": 2.313006268319205, + "learning_rate": 3.7344798845864846e-06, + "loss": 0.9371, + "step": 2541 + }, + { + "epoch": 0.19104163535247257, + "grad_norm": 2.4968099268142994, + "learning_rate": 3.734237439133074e-06, + "loss": 1.0578, + "step": 2542 + }, + { + "epoch": 0.19111678941830754, + "grad_norm": 1.900744821722284, + "learning_rate": 3.7339948909194543e-06, + "loss": 1.0484, + "step": 2543 + }, + { + "epoch": 0.19119194348414248, + "grad_norm": 2.1839542534896306, + "learning_rate": 3.7337522399599973e-06, + "loss": 0.9601, + "step": 2544 + }, + { + "epoch": 0.19126709754997745, + "grad_norm": 1.6725516687916795, + "learning_rate": 3.7335094862690814e-06, + "loss": 0.9641, + "step": 2545 + }, + { + "epoch": 0.19134225161581242, + "grad_norm": 1.5276747244940698, + "learning_rate": 3.7332666298610906e-06, + "loss": 1.0492, + "step": 2546 + }, + { + "epoch": 0.1914174056816474, + "grad_norm": 0.7365513066188614, + "learning_rate": 3.733023670750414e-06, + "loss": 0.8291, + "step": 2547 + }, + { + "epoch": 0.19149255974748233, + "grad_norm": 2.0931233849096347, + "learning_rate": 3.7327806089514497e-06, + "loss": 1.0106, + "step": 2548 + }, + { + "epoch": 0.1915677138133173, + "grad_norm": 1.829201080010131, + "learning_rate": 3.7325374444785983e-06, + "loss": 0.9974, + "step": 2549 + }, + { + "epoch": 0.19164286787915227, + "grad_norm": 1.633771769722253, + "learning_rate": 3.7322941773462694e-06, + "loss": 1.0999, + "step": 2550 + }, + { + "epoch": 0.19171802194498722, + "grad_norm": 1.9492457202582096, + "learning_rate": 3.732050807568877e-06, + "loss": 1.0665, + "step": 2551 + }, + { + "epoch": 0.1917931760108222, + "grad_norm": 1.5900443885413194, + "learning_rate": 3.731807335160842e-06, + "loss": 1.0312, + "step": 2552 + }, + { + "epoch": 0.19186833007665716, + "grad_norm": 1.990138058114674, + "learning_rate": 3.7315637601365902e-06, + "loss": 1.0404, + "step": 2553 + }, + { + "epoch": 0.1919434841424921, + "grad_norm": 1.3328570270144018, + "learning_rate": 3.731320082510556e-06, + "loss": 1.0157, + "step": 2554 + }, + { + "epoch": 0.19201863820832707, + "grad_norm": 3.330617195416607, + "learning_rate": 3.7310763022971764e-06, + "loss": 1.0347, + "step": 2555 + }, + { + "epoch": 0.19209379227416204, + "grad_norm": 2.6467003991053675, + "learning_rate": 3.730832419510897e-06, + "loss": 1.115, + "step": 2556 + }, + { + "epoch": 0.19216894633999698, + "grad_norm": 0.810216997420216, + "learning_rate": 3.73058843416617e-06, + "loss": 0.9182, + "step": 2557 + }, + { + "epoch": 0.19224410040583195, + "grad_norm": 2.101723363455948, + "learning_rate": 3.7303443462774505e-06, + "loss": 1.0561, + "step": 2558 + }, + { + "epoch": 0.19231925447166692, + "grad_norm": 1.4155536187303575, + "learning_rate": 3.730100155859203e-06, + "loss": 0.9144, + "step": 2559 + }, + { + "epoch": 0.19239440853750187, + "grad_norm": 2.8099733659561283, + "learning_rate": 3.7298558629258966e-06, + "loss": 1.0258, + "step": 2560 + }, + { + "epoch": 0.19246956260333684, + "grad_norm": 0.6911440853354555, + "learning_rate": 3.729611467492005e-06, + "loss": 0.7957, + "step": 2561 + }, + { + "epoch": 0.1925447166691718, + "grad_norm": 2.410482078550509, + "learning_rate": 3.7293669695720117e-06, + "loss": 0.9934, + "step": 2562 + }, + { + "epoch": 0.19261987073500678, + "grad_norm": 1.8984189130713052, + "learning_rate": 3.7291223691804038e-06, + "loss": 1.0201, + "step": 2563 + }, + { + "epoch": 0.19269502480084172, + "grad_norm": 1.4554068584459297, + "learning_rate": 3.728877666331673e-06, + "loss": 1.0291, + "step": 2564 + }, + { + "epoch": 0.1927701788666767, + "grad_norm": 1.1520435006750005, + "learning_rate": 3.7286328610403207e-06, + "loss": 1.0479, + "step": 2565 + }, + { + "epoch": 0.19284533293251166, + "grad_norm": 2.7287827335758643, + "learning_rate": 3.7283879533208523e-06, + "loss": 0.8676, + "step": 2566 + }, + { + "epoch": 0.1929204869983466, + "grad_norm": 3.8992549673671135, + "learning_rate": 3.7281429431877795e-06, + "loss": 1.0666, + "step": 2567 + }, + { + "epoch": 0.19299564106418157, + "grad_norm": 2.050272444322282, + "learning_rate": 3.727897830655619e-06, + "loss": 0.9934, + "step": 2568 + }, + { + "epoch": 0.19307079513001654, + "grad_norm": 1.743260348356171, + "learning_rate": 3.727652615738896e-06, + "loss": 0.9561, + "step": 2569 + }, + { + "epoch": 0.19314594919585149, + "grad_norm": 1.630656447891402, + "learning_rate": 3.7274072984521395e-06, + "loss": 1.0357, + "step": 2570 + }, + { + "epoch": 0.19322110326168646, + "grad_norm": 1.5789594804566938, + "learning_rate": 3.7271618788098864e-06, + "loss": 0.9939, + "step": 2571 + }, + { + "epoch": 0.19329625732752143, + "grad_norm": 1.7230319696826824, + "learning_rate": 3.7269163568266774e-06, + "loss": 1.0531, + "step": 2572 + }, + { + "epoch": 0.19337141139335637, + "grad_norm": 2.5061792301831227, + "learning_rate": 3.7266707325170623e-06, + "loss": 1.1085, + "step": 2573 + }, + { + "epoch": 0.19344656545919134, + "grad_norm": 1.6870394890758829, + "learning_rate": 3.7264250058955938e-06, + "loss": 0.9645, + "step": 2574 + }, + { + "epoch": 0.1935217195250263, + "grad_norm": 2.330355094570519, + "learning_rate": 3.726179176976833e-06, + "loss": 1.0518, + "step": 2575 + }, + { + "epoch": 0.19359687359086128, + "grad_norm": 1.9347174567810264, + "learning_rate": 3.7259332457753464e-06, + "loss": 1.0035, + "step": 2576 + }, + { + "epoch": 0.19367202765669622, + "grad_norm": 0.6775175071931744, + "learning_rate": 3.725687212305706e-06, + "loss": 0.8329, + "step": 2577 + }, + { + "epoch": 0.1937471817225312, + "grad_norm": 1.98579800263657, + "learning_rate": 3.7254410765824896e-06, + "loss": 1.0665, + "step": 2578 + }, + { + "epoch": 0.19382233578836616, + "grad_norm": 1.9778968016507703, + "learning_rate": 3.7251948386202827e-06, + "loss": 1.0414, + "step": 2579 + }, + { + "epoch": 0.1938974898542011, + "grad_norm": 4.09497890518214, + "learning_rate": 3.724948498433675e-06, + "loss": 0.9803, + "step": 2580 + }, + { + "epoch": 0.19397264392003608, + "grad_norm": 1.802942728915533, + "learning_rate": 3.7247020560372635e-06, + "loss": 1.0642, + "step": 2581 + }, + { + "epoch": 0.19404779798587105, + "grad_norm": 2.004308589528595, + "learning_rate": 3.724455511445651e-06, + "loss": 0.9182, + "step": 2582 + }, + { + "epoch": 0.194122952051706, + "grad_norm": 1.8101414429211706, + "learning_rate": 3.724208864673446e-06, + "loss": 0.9411, + "step": 2583 + }, + { + "epoch": 0.19419810611754096, + "grad_norm": 2.031073107719805, + "learning_rate": 3.7239621157352633e-06, + "loss": 0.99, + "step": 2584 + }, + { + "epoch": 0.19427326018337593, + "grad_norm": 1.950279360102864, + "learning_rate": 3.723715264645724e-06, + "loss": 1.0195, + "step": 2585 + }, + { + "epoch": 0.19434841424921087, + "grad_norm": 2.8145360570438864, + "learning_rate": 3.723468311419455e-06, + "loss": 1.0331, + "step": 2586 + }, + { + "epoch": 0.19442356831504584, + "grad_norm": 1.5512141240941117, + "learning_rate": 3.7232212560710883e-06, + "loss": 1.0101, + "step": 2587 + }, + { + "epoch": 0.1944987223808808, + "grad_norm": 2.72551368281377, + "learning_rate": 3.7229740986152636e-06, + "loss": 0.9447, + "step": 2588 + }, + { + "epoch": 0.19457387644671575, + "grad_norm": 3.0148228976993305, + "learning_rate": 3.722726839066626e-06, + "loss": 0.9141, + "step": 2589 + }, + { + "epoch": 0.19464903051255072, + "grad_norm": 1.4878535419548993, + "learning_rate": 3.722479477439826e-06, + "loss": 1.0849, + "step": 2590 + }, + { + "epoch": 0.1947241845783857, + "grad_norm": 1.743748055733858, + "learning_rate": 3.722232013749522e-06, + "loss": 0.9948, + "step": 2591 + }, + { + "epoch": 0.19479933864422067, + "grad_norm": 1.6567419509105026, + "learning_rate": 3.721984448010376e-06, + "loss": 1.0388, + "step": 2592 + }, + { + "epoch": 0.1948744927100556, + "grad_norm": 1.7103557555689723, + "learning_rate": 3.7217367802370573e-06, + "loss": 1.012, + "step": 2593 + }, + { + "epoch": 0.19494964677589058, + "grad_norm": 1.9700552231653727, + "learning_rate": 3.7214890104442413e-06, + "loss": 1.0048, + "step": 2594 + }, + { + "epoch": 0.19502480084172555, + "grad_norm": 2.433307563523352, + "learning_rate": 3.7212411386466097e-06, + "loss": 1.0844, + "step": 2595 + }, + { + "epoch": 0.1950999549075605, + "grad_norm": 2.8076874991195875, + "learning_rate": 3.72099316485885e-06, + "loss": 0.9608, + "step": 2596 + }, + { + "epoch": 0.19517510897339546, + "grad_norm": 5.59116273855832, + "learning_rate": 3.7207450890956544e-06, + "loss": 1.0493, + "step": 2597 + }, + { + "epoch": 0.19525026303923043, + "grad_norm": 2.1405571697097123, + "learning_rate": 3.720496911371723e-06, + "loss": 0.9562, + "step": 2598 + }, + { + "epoch": 0.19532541710506537, + "grad_norm": 2.1300202700944384, + "learning_rate": 3.720248631701762e-06, + "loss": 0.895, + "step": 2599 + }, + { + "epoch": 0.19540057117090034, + "grad_norm": 2.9618929727962495, + "learning_rate": 3.720000250100482e-06, + "loss": 0.9914, + "step": 2600 + }, + { + "epoch": 0.19547572523673531, + "grad_norm": 2.8687162974918947, + "learning_rate": 3.719751766582601e-06, + "loss": 0.93, + "step": 2601 + }, + { + "epoch": 0.19555087930257026, + "grad_norm": 1.6214621251729444, + "learning_rate": 3.7195031811628422e-06, + "loss": 0.9815, + "step": 2602 + }, + { + "epoch": 0.19562603336840523, + "grad_norm": 2.2357074525903684, + "learning_rate": 3.719254493855936e-06, + "loss": 1.0777, + "step": 2603 + }, + { + "epoch": 0.1957011874342402, + "grad_norm": 0.8896786530741191, + "learning_rate": 3.719005704676617e-06, + "loss": 0.9444, + "step": 2604 + }, + { + "epoch": 0.19577634150007514, + "grad_norm": 1.6130387167780416, + "learning_rate": 3.7187568136396274e-06, + "loss": 1.0548, + "step": 2605 + }, + { + "epoch": 0.1958514955659101, + "grad_norm": 1.7586930845550115, + "learning_rate": 3.7185078207597158e-06, + "loss": 0.9495, + "step": 2606 + }, + { + "epoch": 0.19592664963174508, + "grad_norm": 6.167138708794835, + "learning_rate": 3.7182587260516343e-06, + "loss": 0.9759, + "step": 2607 + }, + { + "epoch": 0.19600180369758005, + "grad_norm": 1.5465850558344407, + "learning_rate": 3.7180095295301443e-06, + "loss": 0.9941, + "step": 2608 + }, + { + "epoch": 0.196076957763415, + "grad_norm": 1.7057705551445352, + "learning_rate": 3.717760231210011e-06, + "loss": 1.0893, + "step": 2609 + }, + { + "epoch": 0.19615211182924996, + "grad_norm": 1.5531986334249226, + "learning_rate": 3.7175108311060057e-06, + "loss": 0.93, + "step": 2610 + }, + { + "epoch": 0.19622726589508493, + "grad_norm": 2.3074799770094034, + "learning_rate": 3.717261329232907e-06, + "loss": 1.0319, + "step": 2611 + }, + { + "epoch": 0.19630241996091988, + "grad_norm": 0.7230375778779344, + "learning_rate": 3.717011725605499e-06, + "loss": 0.7597, + "step": 2612 + }, + { + "epoch": 0.19637757402675485, + "grad_norm": 1.8488889296136346, + "learning_rate": 3.7167620202385715e-06, + "loss": 0.9993, + "step": 2613 + }, + { + "epoch": 0.19645272809258982, + "grad_norm": 1.799364372452585, + "learning_rate": 3.71651221314692e-06, + "loss": 0.905, + "step": 2614 + }, + { + "epoch": 0.19652788215842476, + "grad_norm": 2.1376563879436423, + "learning_rate": 3.7162623043453476e-06, + "loss": 1.1163, + "step": 2615 + }, + { + "epoch": 0.19660303622425973, + "grad_norm": 1.6318502556785068, + "learning_rate": 3.716012293848661e-06, + "loss": 1.0385, + "step": 2616 + }, + { + "epoch": 0.1966781902900947, + "grad_norm": 2.2541586273482475, + "learning_rate": 3.7157621816716747e-06, + "loss": 1.0799, + "step": 2617 + }, + { + "epoch": 0.19675334435592964, + "grad_norm": 1.6657044096548548, + "learning_rate": 3.71551196782921e-06, + "loss": 1.0815, + "step": 2618 + }, + { + "epoch": 0.1968284984217646, + "grad_norm": 2.852111763507111, + "learning_rate": 3.7152616523360913e-06, + "loss": 1.0747, + "step": 2619 + }, + { + "epoch": 0.19690365248759958, + "grad_norm": 1.618224691338352, + "learning_rate": 3.7150112352071514e-06, + "loss": 1.0692, + "step": 2620 + }, + { + "epoch": 0.19697880655343455, + "grad_norm": 0.7057370095092861, + "learning_rate": 3.714760716457229e-06, + "loss": 0.8605, + "step": 2621 + }, + { + "epoch": 0.1970539606192695, + "grad_norm": 2.0360501425813338, + "learning_rate": 3.7145100961011675e-06, + "loss": 0.9808, + "step": 2622 + }, + { + "epoch": 0.19712911468510447, + "grad_norm": 1.7380298218357981, + "learning_rate": 3.714259374153818e-06, + "loss": 0.9872, + "step": 2623 + }, + { + "epoch": 0.19720426875093944, + "grad_norm": 1.434911968973572, + "learning_rate": 3.714008550630036e-06, + "loss": 0.8197, + "step": 2624 + }, + { + "epoch": 0.19727942281677438, + "grad_norm": 1.5831301793052508, + "learning_rate": 3.713757625544684e-06, + "loss": 1.0158, + "step": 2625 + }, + { + "epoch": 0.19735457688260935, + "grad_norm": 1.6518204674323471, + "learning_rate": 3.7135065989126303e-06, + "loss": 0.9556, + "step": 2626 + }, + { + "epoch": 0.19742973094844432, + "grad_norm": 1.8061724741840808, + "learning_rate": 3.7132554707487493e-06, + "loss": 1.0136, + "step": 2627 + }, + { + "epoch": 0.19750488501427926, + "grad_norm": 17.2360987281123, + "learning_rate": 3.713004241067921e-06, + "loss": 0.9439, + "step": 2628 + }, + { + "epoch": 0.19758003908011423, + "grad_norm": 1.6329953762019351, + "learning_rate": 3.712752909885032e-06, + "loss": 0.985, + "step": 2629 + }, + { + "epoch": 0.1976551931459492, + "grad_norm": 1.7336431110329151, + "learning_rate": 3.7125014772149746e-06, + "loss": 1.0747, + "step": 2630 + }, + { + "epoch": 0.19773034721178415, + "grad_norm": 1.7018386462612503, + "learning_rate": 3.712249943072647e-06, + "loss": 1.0117, + "step": 2631 + }, + { + "epoch": 0.19780550127761912, + "grad_norm": 1.4949520525691087, + "learning_rate": 3.7119983074729532e-06, + "loss": 1.0523, + "step": 2632 + }, + { + "epoch": 0.1978806553434541, + "grad_norm": 1.6168319384986316, + "learning_rate": 3.7117465704308045e-06, + "loss": 1.0718, + "step": 2633 + }, + { + "epoch": 0.19795580940928903, + "grad_norm": 1.697637203441497, + "learning_rate": 3.7114947319611164e-06, + "loss": 0.8898, + "step": 2634 + }, + { + "epoch": 0.198030963475124, + "grad_norm": 1.8234874231455491, + "learning_rate": 3.711242792078812e-06, + "loss": 0.9323, + "step": 2635 + }, + { + "epoch": 0.19810611754095897, + "grad_norm": 1.5794169359334564, + "learning_rate": 3.7109907507988192e-06, + "loss": 1.0036, + "step": 2636 + }, + { + "epoch": 0.19818127160679394, + "grad_norm": 1.9651605490987576, + "learning_rate": 3.710738608136073e-06, + "loss": 0.9973, + "step": 2637 + }, + { + "epoch": 0.19825642567262888, + "grad_norm": 1.7022123778913167, + "learning_rate": 3.710486364105513e-06, + "loss": 0.9847, + "step": 2638 + }, + { + "epoch": 0.19833157973846385, + "grad_norm": 2.5353399130854295, + "learning_rate": 3.7102340187220863e-06, + "loss": 1.0143, + "step": 2639 + }, + { + "epoch": 0.19840673380429882, + "grad_norm": 1.8549979485965706, + "learning_rate": 3.7099815720007447e-06, + "loss": 0.9976, + "step": 2640 + }, + { + "epoch": 0.19848188787013377, + "grad_norm": 1.7313045093755055, + "learning_rate": 3.7097290239564478e-06, + "loss": 1.0906, + "step": 2641 + }, + { + "epoch": 0.19855704193596874, + "grad_norm": 2.222112659561492, + "learning_rate": 3.7094763746041584e-06, + "loss": 0.9877, + "step": 2642 + }, + { + "epoch": 0.1986321960018037, + "grad_norm": 2.5680990689143575, + "learning_rate": 3.709223623958848e-06, + "loss": 0.9375, + "step": 2643 + }, + { + "epoch": 0.19870735006763865, + "grad_norm": 1.781181601834063, + "learning_rate": 3.708970772035493e-06, + "loss": 0.9631, + "step": 2644 + }, + { + "epoch": 0.19878250413347362, + "grad_norm": 1.7678747781751918, + "learning_rate": 3.7087178188490754e-06, + "loss": 0.9683, + "step": 2645 + }, + { + "epoch": 0.1988576581993086, + "grad_norm": 2.2685216867622287, + "learning_rate": 3.708464764414584e-06, + "loss": 0.9028, + "step": 2646 + }, + { + "epoch": 0.19893281226514353, + "grad_norm": 2.130278930489102, + "learning_rate": 3.708211608747013e-06, + "loss": 1.039, + "step": 2647 + }, + { + "epoch": 0.1990079663309785, + "grad_norm": 2.996516748680386, + "learning_rate": 3.7079583518613636e-06, + "loss": 1.0146, + "step": 2648 + }, + { + "epoch": 0.19908312039681347, + "grad_norm": 1.6765914416770376, + "learning_rate": 3.707704993772641e-06, + "loss": 1.0353, + "step": 2649 + }, + { + "epoch": 0.19915827446264842, + "grad_norm": 2.126977673433577, + "learning_rate": 3.7074515344958584e-06, + "loss": 1.0198, + "step": 2650 + }, + { + "epoch": 0.1992334285284834, + "grad_norm": 1.6357771684692428, + "learning_rate": 3.7071979740460345e-06, + "loss": 0.9729, + "step": 2651 + }, + { + "epoch": 0.19930858259431836, + "grad_norm": 1.7595457898680715, + "learning_rate": 3.706944312438193e-06, + "loss": 1.0596, + "step": 2652 + }, + { + "epoch": 0.19938373666015333, + "grad_norm": 2.5819590409839672, + "learning_rate": 3.7066905496873646e-06, + "loss": 0.9841, + "step": 2653 + }, + { + "epoch": 0.19945889072598827, + "grad_norm": 1.6603910392337478, + "learning_rate": 3.706436685808586e-06, + "loss": 0.9811, + "step": 2654 + }, + { + "epoch": 0.19953404479182324, + "grad_norm": 2.1470263476378606, + "learning_rate": 3.7061827208168995e-06, + "loss": 1.0331, + "step": 2655 + }, + { + "epoch": 0.1996091988576582, + "grad_norm": 1.9148008829685754, + "learning_rate": 3.705928654727353e-06, + "loss": 0.9535, + "step": 2656 + }, + { + "epoch": 0.19968435292349315, + "grad_norm": 0.8332433691736653, + "learning_rate": 3.7056744875550016e-06, + "loss": 0.9147, + "step": 2657 + }, + { + "epoch": 0.19975950698932812, + "grad_norm": 1.536735285566738, + "learning_rate": 3.7054202193149047e-06, + "loss": 1.0316, + "step": 2658 + }, + { + "epoch": 0.1998346610551631, + "grad_norm": 2.0941371275152663, + "learning_rate": 3.7051658500221297e-06, + "loss": 0.9937, + "step": 2659 + }, + { + "epoch": 0.19990981512099804, + "grad_norm": 1.78967567830272, + "learning_rate": 3.704911379691749e-06, + "loss": 0.9897, + "step": 2660 + }, + { + "epoch": 0.199984969186833, + "grad_norm": 1.8156268016392605, + "learning_rate": 3.70465680833884e-06, + "loss": 1.0403, + "step": 2661 + }, + { + "epoch": 0.20006012325266798, + "grad_norm": 1.7006264654263286, + "learning_rate": 3.704402135978488e-06, + "loss": 1.0035, + "step": 2662 + }, + { + "epoch": 0.20013527731850292, + "grad_norm": 1.619368772918024, + "learning_rate": 3.7041473626257823e-06, + "loss": 0.9747, + "step": 2663 + }, + { + "epoch": 0.2002104313843379, + "grad_norm": 1.5408171899478358, + "learning_rate": 3.7038924882958204e-06, + "loss": 0.9215, + "step": 2664 + }, + { + "epoch": 0.20028558545017286, + "grad_norm": 1.8252867486460078, + "learning_rate": 3.7036375130037037e-06, + "loss": 1.0017, + "step": 2665 + }, + { + "epoch": 0.20036073951600783, + "grad_norm": 1.5967724881049652, + "learning_rate": 3.70338243676454e-06, + "loss": 1.0805, + "step": 2666 + }, + { + "epoch": 0.20043589358184277, + "grad_norm": 3.157179893764655, + "learning_rate": 3.7031272595934453e-06, + "loss": 0.986, + "step": 2667 + }, + { + "epoch": 0.20051104764767774, + "grad_norm": 2.4091426752198446, + "learning_rate": 3.702871981505538e-06, + "loss": 1.0544, + "step": 2668 + }, + { + "epoch": 0.2005862017135127, + "grad_norm": 1.5667596356558366, + "learning_rate": 3.7026166025159454e-06, + "loss": 0.9803, + "step": 2669 + }, + { + "epoch": 0.20066135577934766, + "grad_norm": 1.9172780955205075, + "learning_rate": 3.7023611226397993e-06, + "loss": 0.8935, + "step": 2670 + }, + { + "epoch": 0.20073650984518263, + "grad_norm": 1.8838204499393947, + "learning_rate": 3.702105541892238e-06, + "loss": 1.0206, + "step": 2671 + }, + { + "epoch": 0.2008116639110176, + "grad_norm": 4.611444144625306, + "learning_rate": 3.7018498602884053e-06, + "loss": 0.952, + "step": 2672 + }, + { + "epoch": 0.20088681797685254, + "grad_norm": 2.3932428923425317, + "learning_rate": 3.701594077843452e-06, + "loss": 1.0766, + "step": 2673 + }, + { + "epoch": 0.2009619720426875, + "grad_norm": 1.6760763695417507, + "learning_rate": 3.701338194572533e-06, + "loss": 1.015, + "step": 2674 + }, + { + "epoch": 0.20103712610852248, + "grad_norm": 2.080323855790416, + "learning_rate": 3.7010822104908116e-06, + "loss": 1.0505, + "step": 2675 + }, + { + "epoch": 0.20111228017435742, + "grad_norm": 4.168910655710656, + "learning_rate": 3.7008261256134556e-06, + "loss": 1.0371, + "step": 2676 + }, + { + "epoch": 0.2011874342401924, + "grad_norm": 2.668089387771732, + "learning_rate": 3.7005699399556383e-06, + "loss": 1.0842, + "step": 2677 + }, + { + "epoch": 0.20126258830602736, + "grad_norm": 1.7921369364223148, + "learning_rate": 3.7003136535325405e-06, + "loss": 1.03, + "step": 2678 + }, + { + "epoch": 0.2013377423718623, + "grad_norm": 1.7909334885660968, + "learning_rate": 3.7000572663593475e-06, + "loss": 0.9469, + "step": 2679 + }, + { + "epoch": 0.20141289643769728, + "grad_norm": 0.756539603024332, + "learning_rate": 3.6998007784512515e-06, + "loss": 0.8118, + "step": 2680 + }, + { + "epoch": 0.20148805050353225, + "grad_norm": 1.4859165418813554, + "learning_rate": 3.6995441898234507e-06, + "loss": 0.9714, + "step": 2681 + }, + { + "epoch": 0.20156320456936722, + "grad_norm": 2.1372654300944367, + "learning_rate": 3.6992875004911485e-06, + "loss": 0.9705, + "step": 2682 + }, + { + "epoch": 0.20163835863520216, + "grad_norm": 1.7846788565203109, + "learning_rate": 3.6990307104695547e-06, + "loss": 1.0194, + "step": 2683 + }, + { + "epoch": 0.20171351270103713, + "grad_norm": 1.8046205846757983, + "learning_rate": 3.6987738197738858e-06, + "loss": 1.0035, + "step": 2684 + }, + { + "epoch": 0.2017886667668721, + "grad_norm": 2.1964896633322653, + "learning_rate": 3.698516828419362e-06, + "loss": 1.051, + "step": 2685 + }, + { + "epoch": 0.20186382083270704, + "grad_norm": 1.496108251755353, + "learning_rate": 3.698259736421213e-06, + "loss": 1.0423, + "step": 2686 + }, + { + "epoch": 0.201938974898542, + "grad_norm": 2.3125812413147053, + "learning_rate": 3.698002543794671e-06, + "loss": 0.9918, + "step": 2687 + }, + { + "epoch": 0.20201412896437698, + "grad_norm": 1.277661718520341, + "learning_rate": 3.697745250554977e-06, + "loss": 0.9582, + "step": 2688 + }, + { + "epoch": 0.20208928303021192, + "grad_norm": 1.6317167849274545, + "learning_rate": 3.697487856717375e-06, + "loss": 0.9558, + "step": 2689 + }, + { + "epoch": 0.2021644370960469, + "grad_norm": 1.3931932201085815, + "learning_rate": 3.6972303622971177e-06, + "loss": 1.1123, + "step": 2690 + }, + { + "epoch": 0.20223959116188187, + "grad_norm": 1.9895960689223613, + "learning_rate": 3.6969727673094626e-06, + "loss": 1.0014, + "step": 2691 + }, + { + "epoch": 0.2023147452277168, + "grad_norm": 2.9241979819927764, + "learning_rate": 3.696715071769672e-06, + "loss": 1.0311, + "step": 2692 + }, + { + "epoch": 0.20238989929355178, + "grad_norm": 1.4814817641620013, + "learning_rate": 3.696457275693017e-06, + "loss": 0.9862, + "step": 2693 + }, + { + "epoch": 0.20246505335938675, + "grad_norm": 1.8100124916151425, + "learning_rate": 3.6961993790947722e-06, + "loss": 1.0534, + "step": 2694 + }, + { + "epoch": 0.2025402074252217, + "grad_norm": 1.534270984985391, + "learning_rate": 3.695941381990219e-06, + "loss": 0.9969, + "step": 2695 + }, + { + "epoch": 0.20261536149105666, + "grad_norm": 3.7570268281532644, + "learning_rate": 3.6956832843946445e-06, + "loss": 0.987, + "step": 2696 + }, + { + "epoch": 0.20269051555689163, + "grad_norm": 1.5163294934397242, + "learning_rate": 3.695425086323342e-06, + "loss": 1.0081, + "step": 2697 + }, + { + "epoch": 0.2027656696227266, + "grad_norm": 1.3958544183913053, + "learning_rate": 3.6951667877916113e-06, + "loss": 1.0011, + "step": 2698 + }, + { + "epoch": 0.20284082368856154, + "grad_norm": 1.5634462924292964, + "learning_rate": 3.694908388814757e-06, + "loss": 1.0028, + "step": 2699 + }, + { + "epoch": 0.20291597775439651, + "grad_norm": 1.8792099419775379, + "learning_rate": 3.6946498894080905e-06, + "loss": 0.9681, + "step": 2700 + }, + { + "epoch": 0.20299113182023149, + "grad_norm": 1.3752436422018337, + "learning_rate": 3.694391289586929e-06, + "loss": 1.0812, + "step": 2701 + }, + { + "epoch": 0.20306628588606643, + "grad_norm": 1.6707515328694365, + "learning_rate": 3.6941325893665953e-06, + "loss": 1.0238, + "step": 2702 + }, + { + "epoch": 0.2031414399519014, + "grad_norm": 2.11767381101116, + "learning_rate": 3.693873788762418e-06, + "loss": 0.926, + "step": 2703 + }, + { + "epoch": 0.20321659401773637, + "grad_norm": 2.1132823781911267, + "learning_rate": 3.6936148877897324e-06, + "loss": 1.1019, + "step": 2704 + }, + { + "epoch": 0.2032917480835713, + "grad_norm": 1.7281610329925796, + "learning_rate": 3.6933558864638805e-06, + "loss": 1.0215, + "step": 2705 + }, + { + "epoch": 0.20336690214940628, + "grad_norm": 3.4405117366209725, + "learning_rate": 3.6930967848002065e-06, + "loss": 1.079, + "step": 2706 + }, + { + "epoch": 0.20344205621524125, + "grad_norm": 2.4248529645251997, + "learning_rate": 3.6928375828140658e-06, + "loss": 0.9285, + "step": 2707 + }, + { + "epoch": 0.2035172102810762, + "grad_norm": 1.6468532958943982, + "learning_rate": 3.6925782805208156e-06, + "loss": 0.9473, + "step": 2708 + }, + { + "epoch": 0.20359236434691116, + "grad_norm": 1.6840881188277912, + "learning_rate": 3.692318877935821e-06, + "loss": 1.0202, + "step": 2709 + }, + { + "epoch": 0.20366751841274613, + "grad_norm": 1.9479796115544588, + "learning_rate": 3.692059375074453e-06, + "loss": 1.0239, + "step": 2710 + }, + { + "epoch": 0.2037426724785811, + "grad_norm": 1.7019415427096858, + "learning_rate": 3.6917997719520867e-06, + "loss": 0.9274, + "step": 2711 + }, + { + "epoch": 0.20381782654441605, + "grad_norm": 1.9827935232181748, + "learning_rate": 3.691540068584106e-06, + "loss": 0.9863, + "step": 2712 + }, + { + "epoch": 0.20389298061025102, + "grad_norm": 0.7394163095344476, + "learning_rate": 3.6912802649858995e-06, + "loss": 0.87, + "step": 2713 + }, + { + "epoch": 0.203968134676086, + "grad_norm": 2.6476737302381523, + "learning_rate": 3.6910203611728603e-06, + "loss": 1.0119, + "step": 2714 + }, + { + "epoch": 0.20404328874192093, + "grad_norm": 1.6706519956794488, + "learning_rate": 3.6907603571603895e-06, + "loss": 0.9847, + "step": 2715 + }, + { + "epoch": 0.2041184428077559, + "grad_norm": 1.709546782459028, + "learning_rate": 3.690500252963893e-06, + "loss": 1.0258, + "step": 2716 + }, + { + "epoch": 0.20419359687359087, + "grad_norm": 1.5618593046781977, + "learning_rate": 3.6902400485987835e-06, + "loss": 0.9826, + "step": 2717 + }, + { + "epoch": 0.2042687509394258, + "grad_norm": 1.6168366572224622, + "learning_rate": 3.6899797440804788e-06, + "loss": 1.1086, + "step": 2718 + }, + { + "epoch": 0.20434390500526078, + "grad_norm": 2.1280966242888435, + "learning_rate": 3.689719339424403e-06, + "loss": 0.9851, + "step": 2719 + }, + { + "epoch": 0.20441905907109575, + "grad_norm": 3.1781900614902763, + "learning_rate": 3.689458834645986e-06, + "loss": 0.9231, + "step": 2720 + }, + { + "epoch": 0.2044942131369307, + "grad_norm": 0.838968148778374, + "learning_rate": 3.689198229760663e-06, + "loss": 0.8715, + "step": 2721 + }, + { + "epoch": 0.20456936720276567, + "grad_norm": 2.0887999592895388, + "learning_rate": 3.6889375247838766e-06, + "loss": 0.9541, + "step": 2722 + }, + { + "epoch": 0.20464452126860064, + "grad_norm": 1.7010837186095533, + "learning_rate": 3.6886767197310757e-06, + "loss": 1.0585, + "step": 2723 + }, + { + "epoch": 0.20471967533443558, + "grad_norm": 1.6989203530874082, + "learning_rate": 3.688415814617711e-06, + "loss": 0.972, + "step": 2724 + }, + { + "epoch": 0.20479482940027055, + "grad_norm": 0.7087646227880591, + "learning_rate": 3.688154809459245e-06, + "loss": 0.828, + "step": 2725 + }, + { + "epoch": 0.20486998346610552, + "grad_norm": 1.4559617604537018, + "learning_rate": 3.6878937042711424e-06, + "loss": 1.0632, + "step": 2726 + }, + { + "epoch": 0.2049451375319405, + "grad_norm": 1.6616962784654254, + "learning_rate": 3.687632499068874e-06, + "loss": 0.89, + "step": 2727 + }, + { + "epoch": 0.20502029159777543, + "grad_norm": 1.7343155446902034, + "learning_rate": 3.6873711938679174e-06, + "loss": 1.0406, + "step": 2728 + }, + { + "epoch": 0.2050954456636104, + "grad_norm": 1.576983501679422, + "learning_rate": 3.6871097886837565e-06, + "loss": 0.9371, + "step": 2729 + }, + { + "epoch": 0.20517059972944537, + "grad_norm": 1.5509653438579827, + "learning_rate": 3.68684828353188e-06, + "loss": 0.9795, + "step": 2730 + }, + { + "epoch": 0.20524575379528032, + "grad_norm": 1.6489426259272497, + "learning_rate": 3.6865866784277836e-06, + "loss": 1.113, + "step": 2731 + }, + { + "epoch": 0.2053209078611153, + "grad_norm": 2.4711424425690365, + "learning_rate": 3.6863249733869683e-06, + "loss": 1.0645, + "step": 2732 + }, + { + "epoch": 0.20539606192695026, + "grad_norm": 2.2562665563594093, + "learning_rate": 3.6860631684249403e-06, + "loss": 0.9643, + "step": 2733 + }, + { + "epoch": 0.2054712159927852, + "grad_norm": 1.8531405006004862, + "learning_rate": 3.685801263557214e-06, + "loss": 0.9548, + "step": 2734 + }, + { + "epoch": 0.20554637005862017, + "grad_norm": 1.7467000365284941, + "learning_rate": 3.6855392587993065e-06, + "loss": 1.0408, + "step": 2735 + }, + { + "epoch": 0.20562152412445514, + "grad_norm": 2.323525615527133, + "learning_rate": 3.6852771541667444e-06, + "loss": 0.9682, + "step": 2736 + }, + { + "epoch": 0.20569667819029008, + "grad_norm": 1.5921858797823194, + "learning_rate": 3.6850149496750575e-06, + "loss": 1.088, + "step": 2737 + }, + { + "epoch": 0.20577183225612505, + "grad_norm": 0.7618586093932617, + "learning_rate": 3.684752645339782e-06, + "loss": 0.8781, + "step": 2738 + }, + { + "epoch": 0.20584698632196002, + "grad_norm": 1.6164234955108763, + "learning_rate": 3.6844902411764612e-06, + "loss": 0.9697, + "step": 2739 + }, + { + "epoch": 0.20592214038779497, + "grad_norm": 1.4796932071305842, + "learning_rate": 3.6842277372006434e-06, + "loss": 1.0892, + "step": 2740 + }, + { + "epoch": 0.20599729445362994, + "grad_norm": 1.9391853016984013, + "learning_rate": 3.6839651334278823e-06, + "loss": 1.0731, + "step": 2741 + }, + { + "epoch": 0.2060724485194649, + "grad_norm": 2.8769759975541906, + "learning_rate": 3.6837024298737393e-06, + "loss": 1.0243, + "step": 2742 + }, + { + "epoch": 0.20614760258529988, + "grad_norm": 2.1315653861260766, + "learning_rate": 3.68343962655378e-06, + "loss": 1.0259, + "step": 2743 + }, + { + "epoch": 0.20622275665113482, + "grad_norm": 1.4139201477224366, + "learning_rate": 3.6831767234835763e-06, + "loss": 1.062, + "step": 2744 + }, + { + "epoch": 0.2062979107169698, + "grad_norm": 1.5832644330476238, + "learning_rate": 3.6829137206787065e-06, + "loss": 1.0216, + "step": 2745 + }, + { + "epoch": 0.20637306478280476, + "grad_norm": 1.467494184232297, + "learning_rate": 3.6826506181547543e-06, + "loss": 1.0134, + "step": 2746 + }, + { + "epoch": 0.2064482188486397, + "grad_norm": 1.5256545928872942, + "learning_rate": 3.6823874159273095e-06, + "loss": 0.9392, + "step": 2747 + }, + { + "epoch": 0.20652337291447467, + "grad_norm": 1.977982866384667, + "learning_rate": 3.6821241140119685e-06, + "loss": 0.9746, + "step": 2748 + }, + { + "epoch": 0.20659852698030964, + "grad_norm": 1.8298638705917034, + "learning_rate": 3.6818607124243322e-06, + "loss": 1.0269, + "step": 2749 + }, + { + "epoch": 0.2066736810461446, + "grad_norm": 1.6992789422259171, + "learning_rate": 3.6815972111800082e-06, + "loss": 0.9905, + "step": 2750 + }, + { + "epoch": 0.20674883511197956, + "grad_norm": 2.4307858845401196, + "learning_rate": 3.6813336102946107e-06, + "loss": 0.9592, + "step": 2751 + }, + { + "epoch": 0.20682398917781453, + "grad_norm": 2.2915531708327035, + "learning_rate": 3.681069909783758e-06, + "loss": 0.9665, + "step": 2752 + }, + { + "epoch": 0.20689914324364947, + "grad_norm": 0.7283689131291753, + "learning_rate": 3.6808061096630765e-06, + "loss": 0.8501, + "step": 2753 + }, + { + "epoch": 0.20697429730948444, + "grad_norm": 0.7574257404451068, + "learning_rate": 3.6805422099481965e-06, + "loss": 0.8628, + "step": 2754 + }, + { + "epoch": 0.2070494513753194, + "grad_norm": 1.6784021471924406, + "learning_rate": 3.6802782106547553e-06, + "loss": 1.0081, + "step": 2755 + }, + { + "epoch": 0.20712460544115438, + "grad_norm": 1.9023429980266737, + "learning_rate": 3.6800141117983954e-06, + "loss": 0.9119, + "step": 2756 + }, + { + "epoch": 0.20719975950698932, + "grad_norm": 1.952231912512272, + "learning_rate": 3.679749913394767e-06, + "loss": 0.9426, + "step": 2757 + }, + { + "epoch": 0.2072749135728243, + "grad_norm": 1.5238618699801163, + "learning_rate": 3.6794856154595235e-06, + "loss": 1.0058, + "step": 2758 + }, + { + "epoch": 0.20735006763865926, + "grad_norm": 0.7476245295413574, + "learning_rate": 3.679221218008326e-06, + "loss": 0.8703, + "step": 2759 + }, + { + "epoch": 0.2074252217044942, + "grad_norm": 2.395846679208045, + "learning_rate": 3.6789567210568417e-06, + "loss": 1.0181, + "step": 2760 + }, + { + "epoch": 0.20750037577032918, + "grad_norm": 1.8122735375301802, + "learning_rate": 3.678692124620742e-06, + "loss": 1.0795, + "step": 2761 + }, + { + "epoch": 0.20757552983616415, + "grad_norm": 1.8103024545104722, + "learning_rate": 3.6784274287157066e-06, + "loss": 0.9315, + "step": 2762 + }, + { + "epoch": 0.2076506839019991, + "grad_norm": 2.1988344009101124, + "learning_rate": 3.678162633357418e-06, + "loss": 1.0327, + "step": 2763 + }, + { + "epoch": 0.20772583796783406, + "grad_norm": 2.0496046196729902, + "learning_rate": 3.6778977385615676e-06, + "loss": 1.0102, + "step": 2764 + }, + { + "epoch": 0.20780099203366903, + "grad_norm": 1.5845577458132898, + "learning_rate": 3.677632744343851e-06, + "loss": 1.018, + "step": 2765 + }, + { + "epoch": 0.20787614609950397, + "grad_norm": 1.6172542834659354, + "learning_rate": 3.6773676507199703e-06, + "loss": 0.9385, + "step": 2766 + }, + { + "epoch": 0.20795130016533894, + "grad_norm": 1.4821855962372308, + "learning_rate": 3.6771024577056333e-06, + "loss": 1.0243, + "step": 2767 + }, + { + "epoch": 0.2080264542311739, + "grad_norm": 1.8727121382208667, + "learning_rate": 3.6768371653165537e-06, + "loss": 0.9797, + "step": 2768 + }, + { + "epoch": 0.20810160829700886, + "grad_norm": 2.620203358715069, + "learning_rate": 3.676571773568451e-06, + "loss": 0.9661, + "step": 2769 + }, + { + "epoch": 0.20817676236284383, + "grad_norm": 2.591041529615247, + "learning_rate": 3.67630628247705e-06, + "loss": 1.0431, + "step": 2770 + }, + { + "epoch": 0.2082519164286788, + "grad_norm": 2.926934416072347, + "learning_rate": 3.6760406920580834e-06, + "loss": 1.1713, + "step": 2771 + }, + { + "epoch": 0.20832707049451377, + "grad_norm": 2.237365428777706, + "learning_rate": 3.6757750023272882e-06, + "loss": 1.0322, + "step": 2772 + }, + { + "epoch": 0.2084022245603487, + "grad_norm": 1.5278630069117467, + "learning_rate": 3.6755092133004062e-06, + "loss": 1.0806, + "step": 2773 + }, + { + "epoch": 0.20847737862618368, + "grad_norm": 1.7767883103592002, + "learning_rate": 3.6752433249931876e-06, + "loss": 0.9295, + "step": 2774 + }, + { + "epoch": 0.20855253269201865, + "grad_norm": 2.9783721789228492, + "learning_rate": 3.6749773374213877e-06, + "loss": 0.9498, + "step": 2775 + }, + { + "epoch": 0.2086276867578536, + "grad_norm": 1.4966338853424945, + "learning_rate": 3.674711250600766e-06, + "loss": 0.9532, + "step": 2776 + }, + { + "epoch": 0.20870284082368856, + "grad_norm": 2.5656358093703093, + "learning_rate": 3.6744450645470904e-06, + "loss": 0.9936, + "step": 2777 + }, + { + "epoch": 0.20877799488952353, + "grad_norm": 2.0072575258585945, + "learning_rate": 3.6741787792761324e-06, + "loss": 0.9617, + "step": 2778 + }, + { + "epoch": 0.20885314895535848, + "grad_norm": 2.2263233792582744, + "learning_rate": 3.673912394803671e-06, + "loss": 0.9805, + "step": 2779 + }, + { + "epoch": 0.20892830302119345, + "grad_norm": 1.4516684502737751, + "learning_rate": 3.6736459111454903e-06, + "loss": 1.0132, + "step": 2780 + }, + { + "epoch": 0.20900345708702842, + "grad_norm": 2.222560887226809, + "learning_rate": 3.6733793283173805e-06, + "loss": 1.0212, + "step": 2781 + }, + { + "epoch": 0.20907861115286336, + "grad_norm": 1.3888694740633865, + "learning_rate": 3.673112646335138e-06, + "loss": 1.0844, + "step": 2782 + }, + { + "epoch": 0.20915376521869833, + "grad_norm": 1.5032327308037152, + "learning_rate": 3.672845865214564e-06, + "loss": 1.0411, + "step": 2783 + }, + { + "epoch": 0.2092289192845333, + "grad_norm": 1.6301454363339682, + "learning_rate": 3.6725789849714665e-06, + "loss": 0.9495, + "step": 2784 + }, + { + "epoch": 0.20930407335036824, + "grad_norm": 1.6266219774084898, + "learning_rate": 3.67231200562166e-06, + "loss": 0.975, + "step": 2785 + }, + { + "epoch": 0.2093792274162032, + "grad_norm": 1.9342991014421016, + "learning_rate": 3.6720449271809633e-06, + "loss": 1.0519, + "step": 2786 + }, + { + "epoch": 0.20945438148203818, + "grad_norm": 2.268569716339316, + "learning_rate": 3.6717777496652023e-06, + "loss": 0.9639, + "step": 2787 + }, + { + "epoch": 0.20952953554787315, + "grad_norm": 1.522602576322191, + "learning_rate": 3.6715104730902074e-06, + "loss": 1.0759, + "step": 2788 + }, + { + "epoch": 0.2096046896137081, + "grad_norm": 1.317442081778065, + "learning_rate": 3.671243097471817e-06, + "loss": 0.9872, + "step": 2789 + }, + { + "epoch": 0.20967984367954307, + "grad_norm": 2.057329575072684, + "learning_rate": 3.6709756228258728e-06, + "loss": 1.0815, + "step": 2790 + }, + { + "epoch": 0.20975499774537804, + "grad_norm": 2.1602796412402108, + "learning_rate": 3.6707080491682243e-06, + "loss": 1.0281, + "step": 2791 + }, + { + "epoch": 0.20983015181121298, + "grad_norm": 2.0191506003090796, + "learning_rate": 3.670440376514727e-06, + "loss": 0.9401, + "step": 2792 + }, + { + "epoch": 0.20990530587704795, + "grad_norm": 1.5833062955990524, + "learning_rate": 3.67017260488124e-06, + "loss": 0.9841, + "step": 2793 + }, + { + "epoch": 0.20998045994288292, + "grad_norm": 0.8031029516377463, + "learning_rate": 3.6699047342836313e-06, + "loss": 0.8971, + "step": 2794 + }, + { + "epoch": 0.21005561400871786, + "grad_norm": 1.6865503676781948, + "learning_rate": 3.669636764737772e-06, + "loss": 0.973, + "step": 2795 + }, + { + "epoch": 0.21013076807455283, + "grad_norm": 1.5147062840289636, + "learning_rate": 3.669368696259542e-06, + "loss": 1.101, + "step": 2796 + }, + { + "epoch": 0.2102059221403878, + "grad_norm": 2.5369669086491955, + "learning_rate": 3.669100528864823e-06, + "loss": 1.0651, + "step": 2797 + }, + { + "epoch": 0.21028107620622274, + "grad_norm": 2.0125283991545473, + "learning_rate": 3.6688322625695075e-06, + "loss": 0.9338, + "step": 2798 + }, + { + "epoch": 0.21035623027205771, + "grad_norm": 0.8051778910130978, + "learning_rate": 3.6685638973894896e-06, + "loss": 0.9212, + "step": 2799 + }, + { + "epoch": 0.21043138433789269, + "grad_norm": 1.7602900021221062, + "learning_rate": 3.6682954333406707e-06, + "loss": 1.114, + "step": 2800 + }, + { + "epoch": 0.21050653840372766, + "grad_norm": 1.7859722918361363, + "learning_rate": 3.6680268704389597e-06, + "loss": 0.9906, + "step": 2801 + }, + { + "epoch": 0.2105816924695626, + "grad_norm": 1.3609727261492133, + "learning_rate": 3.6677582087002695e-06, + "loss": 1.0216, + "step": 2802 + }, + { + "epoch": 0.21065684653539757, + "grad_norm": 1.6759637053071759, + "learning_rate": 3.6674894481405184e-06, + "loss": 1.0534, + "step": 2803 + }, + { + "epoch": 0.21073200060123254, + "grad_norm": 1.7772706404520504, + "learning_rate": 3.667220588775633e-06, + "loss": 0.868, + "step": 2804 + }, + { + "epoch": 0.21080715466706748, + "grad_norm": 1.7163112331060788, + "learning_rate": 3.6669516306215433e-06, + "loss": 0.9276, + "step": 2805 + }, + { + "epoch": 0.21088230873290245, + "grad_norm": 1.5688264562205563, + "learning_rate": 3.666682573694186e-06, + "loss": 0.9629, + "step": 2806 + }, + { + "epoch": 0.21095746279873742, + "grad_norm": 2.3936968247207036, + "learning_rate": 3.6664134180095045e-06, + "loss": 1.0315, + "step": 2807 + }, + { + "epoch": 0.21103261686457236, + "grad_norm": 2.507116108162837, + "learning_rate": 3.666144163583446e-06, + "loss": 1.0342, + "step": 2808 + }, + { + "epoch": 0.21110777093040733, + "grad_norm": 1.474688112291523, + "learning_rate": 3.6658748104319667e-06, + "loss": 1.0166, + "step": 2809 + }, + { + "epoch": 0.2111829249962423, + "grad_norm": 1.8994029537552723, + "learning_rate": 3.665605358571026e-06, + "loss": 1.017, + "step": 2810 + }, + { + "epoch": 0.21125807906207725, + "grad_norm": 4.540120669538391, + "learning_rate": 3.6653358080165893e-06, + "loss": 0.9422, + "step": 2811 + }, + { + "epoch": 0.21133323312791222, + "grad_norm": 0.8700860185959572, + "learning_rate": 3.6650661587846283e-06, + "loss": 0.9195, + "step": 2812 + }, + { + "epoch": 0.2114083871937472, + "grad_norm": 1.6506985890553065, + "learning_rate": 3.6647964108911226e-06, + "loss": 1.0855, + "step": 2813 + }, + { + "epoch": 0.21148354125958213, + "grad_norm": 1.4769180919222271, + "learning_rate": 3.6645265643520536e-06, + "loss": 0.9022, + "step": 2814 + }, + { + "epoch": 0.2115586953254171, + "grad_norm": 1.5933841958699708, + "learning_rate": 3.664256619183413e-06, + "loss": 1.0349, + "step": 2815 + }, + { + "epoch": 0.21163384939125207, + "grad_norm": 0.735152948516016, + "learning_rate": 3.6639865754011934e-06, + "loss": 0.8103, + "step": 2816 + }, + { + "epoch": 0.21170900345708704, + "grad_norm": 1.515561131970192, + "learning_rate": 3.663716433021398e-06, + "loss": 0.9918, + "step": 2817 + }, + { + "epoch": 0.21178415752292198, + "grad_norm": 1.6564382769976458, + "learning_rate": 3.6634461920600337e-06, + "loss": 1.0112, + "step": 2818 + }, + { + "epoch": 0.21185931158875695, + "grad_norm": 1.6129405698759964, + "learning_rate": 3.6631758525331124e-06, + "loss": 0.983, + "step": 2819 + }, + { + "epoch": 0.21193446565459192, + "grad_norm": 2.395249624602863, + "learning_rate": 3.662905414456653e-06, + "loss": 0.9658, + "step": 2820 + }, + { + "epoch": 0.21200961972042687, + "grad_norm": 1.7304245919476984, + "learning_rate": 3.66263487784668e-06, + "loss": 1.0312, + "step": 2821 + }, + { + "epoch": 0.21208477378626184, + "grad_norm": 2.5914887832319367, + "learning_rate": 3.6623642427192237e-06, + "loss": 0.9215, + "step": 2822 + }, + { + "epoch": 0.2121599278520968, + "grad_norm": 2.544404124292618, + "learning_rate": 3.6620935090903205e-06, + "loss": 1.0408, + "step": 2823 + }, + { + "epoch": 0.21223508191793175, + "grad_norm": 2.0244743606854483, + "learning_rate": 3.6618226769760127e-06, + "loss": 1.0262, + "step": 2824 + }, + { + "epoch": 0.21231023598376672, + "grad_norm": 1.8376515929302, + "learning_rate": 3.6615517463923477e-06, + "loss": 1.0081, + "step": 2825 + }, + { + "epoch": 0.2123853900496017, + "grad_norm": 1.5069382262602147, + "learning_rate": 3.661280717355379e-06, + "loss": 1.035, + "step": 2826 + }, + { + "epoch": 0.21246054411543663, + "grad_norm": 1.5396553872694538, + "learning_rate": 3.661009589881166e-06, + "loss": 1.011, + "step": 2827 + }, + { + "epoch": 0.2125356981812716, + "grad_norm": 1.7177986286304865, + "learning_rate": 3.660738363985775e-06, + "loss": 0.9666, + "step": 2828 + }, + { + "epoch": 0.21261085224710657, + "grad_norm": 0.7009411913178893, + "learning_rate": 3.660467039685276e-06, + "loss": 0.8089, + "step": 2829 + }, + { + "epoch": 0.21268600631294152, + "grad_norm": 1.5578436325161045, + "learning_rate": 3.660195616995747e-06, + "loss": 1.0662, + "step": 2830 + }, + { + "epoch": 0.2127611603787765, + "grad_norm": 1.628626371714083, + "learning_rate": 3.6599240959332704e-06, + "loss": 1.089, + "step": 2831 + }, + { + "epoch": 0.21283631444461146, + "grad_norm": 1.5217535640197901, + "learning_rate": 3.659652476513934e-06, + "loss": 1.0567, + "step": 2832 + }, + { + "epoch": 0.21291146851044643, + "grad_norm": 1.473118231478239, + "learning_rate": 3.6593807587538343e-06, + "loss": 1.0273, + "step": 2833 + }, + { + "epoch": 0.21298662257628137, + "grad_norm": 0.895002515831933, + "learning_rate": 3.6591089426690695e-06, + "loss": 0.9602, + "step": 2834 + }, + { + "epoch": 0.21306177664211634, + "grad_norm": 3.005949954724805, + "learning_rate": 3.658837028275747e-06, + "loss": 1.043, + "step": 2835 + }, + { + "epoch": 0.2131369307079513, + "grad_norm": 1.597397613109803, + "learning_rate": 3.6585650155899786e-06, + "loss": 1.0428, + "step": 2836 + }, + { + "epoch": 0.21321208477378625, + "grad_norm": 1.659755667434166, + "learning_rate": 3.658292904627882e-06, + "loss": 0.9935, + "step": 2837 + }, + { + "epoch": 0.21328723883962122, + "grad_norm": 2.0496818574739617, + "learning_rate": 3.6580206954055807e-06, + "loss": 1.1096, + "step": 2838 + }, + { + "epoch": 0.2133623929054562, + "grad_norm": 2.2886418812172047, + "learning_rate": 3.657748387939204e-06, + "loss": 0.9769, + "step": 2839 + }, + { + "epoch": 0.21343754697129114, + "grad_norm": 1.3181632793252924, + "learning_rate": 3.657475982244888e-06, + "loss": 0.8632, + "step": 2840 + }, + { + "epoch": 0.2135127010371261, + "grad_norm": 4.41906109568619, + "learning_rate": 3.6572034783387725e-06, + "loss": 0.9164, + "step": 2841 + }, + { + "epoch": 0.21358785510296108, + "grad_norm": 0.8155046805429302, + "learning_rate": 3.6569308762370056e-06, + "loss": 0.8033, + "step": 2842 + }, + { + "epoch": 0.21366300916879602, + "grad_norm": 2.5882298200467813, + "learning_rate": 3.6566581759557387e-06, + "loss": 1.074, + "step": 2843 + }, + { + "epoch": 0.213738163234631, + "grad_norm": 1.549928752738352, + "learning_rate": 3.656385377511132e-06, + "loss": 0.9934, + "step": 2844 + }, + { + "epoch": 0.21381331730046596, + "grad_norm": 1.5657762130840533, + "learning_rate": 3.656112480919348e-06, + "loss": 0.9727, + "step": 2845 + }, + { + "epoch": 0.21388847136630093, + "grad_norm": 1.3870117737559835, + "learning_rate": 3.6558394861965587e-06, + "loss": 0.9553, + "step": 2846 + }, + { + "epoch": 0.21396362543213587, + "grad_norm": 1.7151391680779855, + "learning_rate": 3.6555663933589384e-06, + "loss": 0.9418, + "step": 2847 + }, + { + "epoch": 0.21403877949797084, + "grad_norm": 0.7132955576870779, + "learning_rate": 3.655293202422671e-06, + "loss": 0.8412, + "step": 2848 + }, + { + "epoch": 0.2141139335638058, + "grad_norm": 1.5863170451491444, + "learning_rate": 3.6550199134039414e-06, + "loss": 1.0317, + "step": 2849 + }, + { + "epoch": 0.21418908762964076, + "grad_norm": 4.907763205265213, + "learning_rate": 3.654746526318945e-06, + "loss": 0.9793, + "step": 2850 + }, + { + "epoch": 0.21426424169547573, + "grad_norm": 1.5316920518207038, + "learning_rate": 3.6544730411838805e-06, + "loss": 0.9554, + "step": 2851 + }, + { + "epoch": 0.2143393957613107, + "grad_norm": 1.4645776939373956, + "learning_rate": 3.654199458014953e-06, + "loss": 1.0242, + "step": 2852 + }, + { + "epoch": 0.21441454982714564, + "grad_norm": 1.4693685000400383, + "learning_rate": 3.653925776828373e-06, + "loss": 1.0376, + "step": 2853 + }, + { + "epoch": 0.2144897038929806, + "grad_norm": 1.9601337259761402, + "learning_rate": 3.653651997640358e-06, + "loss": 0.8885, + "step": 2854 + }, + { + "epoch": 0.21456485795881558, + "grad_norm": 0.7122850746444065, + "learning_rate": 3.6533781204671296e-06, + "loss": 0.8673, + "step": 2855 + }, + { + "epoch": 0.21464001202465052, + "grad_norm": 1.6803203011761412, + "learning_rate": 3.6531041453249154e-06, + "loss": 1.0725, + "step": 2856 + }, + { + "epoch": 0.2147151660904855, + "grad_norm": 1.7265705091195742, + "learning_rate": 3.6528300722299515e-06, + "loss": 1.0211, + "step": 2857 + }, + { + "epoch": 0.21479032015632046, + "grad_norm": 1.5469347239052473, + "learning_rate": 3.6525559011984768e-06, + "loss": 0.9489, + "step": 2858 + }, + { + "epoch": 0.2148654742221554, + "grad_norm": 0.8739554778375607, + "learning_rate": 3.652281632246736e-06, + "loss": 0.8552, + "step": 2859 + }, + { + "epoch": 0.21494062828799038, + "grad_norm": 1.4244524054189929, + "learning_rate": 3.6520072653909823e-06, + "loss": 1.0391, + "step": 2860 + }, + { + "epoch": 0.21501578235382535, + "grad_norm": 1.6276392411585348, + "learning_rate": 3.6517328006474717e-06, + "loss": 1.0802, + "step": 2861 + }, + { + "epoch": 0.21509093641966032, + "grad_norm": 2.274002476914801, + "learning_rate": 3.651458238032468e-06, + "loss": 1.047, + "step": 2862 + }, + { + "epoch": 0.21516609048549526, + "grad_norm": 1.3028499431907137, + "learning_rate": 3.65118357756224e-06, + "loss": 1.0569, + "step": 2863 + }, + { + "epoch": 0.21524124455133023, + "grad_norm": 2.845510587547552, + "learning_rate": 3.650908819253062e-06, + "loss": 0.9836, + "step": 2864 + }, + { + "epoch": 0.2153163986171652, + "grad_norm": 1.9186018697114962, + "learning_rate": 3.6506339631212145e-06, + "loss": 1.0923, + "step": 2865 + }, + { + "epoch": 0.21539155268300014, + "grad_norm": 1.9288108780475437, + "learning_rate": 3.650359009182984e-06, + "loss": 0.89, + "step": 2866 + }, + { + "epoch": 0.2154667067488351, + "grad_norm": 3.034984215533517, + "learning_rate": 3.650083957454663e-06, + "loss": 1.0622, + "step": 2867 + }, + { + "epoch": 0.21554186081467008, + "grad_norm": 1.356326796295533, + "learning_rate": 3.6498088079525487e-06, + "loss": 0.9706, + "step": 2868 + }, + { + "epoch": 0.21561701488050503, + "grad_norm": 2.764700869888708, + "learning_rate": 3.649533560692945e-06, + "loss": 1.0208, + "step": 2869 + }, + { + "epoch": 0.21569216894634, + "grad_norm": 1.6308094496708878, + "learning_rate": 3.6492582156921615e-06, + "loss": 1.0258, + "step": 2870 + }, + { + "epoch": 0.21576732301217497, + "grad_norm": 1.6097035489312603, + "learning_rate": 3.648982772966513e-06, + "loss": 1.047, + "step": 2871 + }, + { + "epoch": 0.2158424770780099, + "grad_norm": 1.5631065717157662, + "learning_rate": 3.648707232532321e-06, + "loss": 1.1055, + "step": 2872 + }, + { + "epoch": 0.21591763114384488, + "grad_norm": 1.8753787611824677, + "learning_rate": 3.648431594405912e-06, + "loss": 0.9849, + "step": 2873 + }, + { + "epoch": 0.21599278520967985, + "grad_norm": 1.6035096775925028, + "learning_rate": 3.648155858603619e-06, + "loss": 0.9692, + "step": 2874 + }, + { + "epoch": 0.2160679392755148, + "grad_norm": 1.6777120639796337, + "learning_rate": 3.64788002514178e-06, + "loss": 1.0388, + "step": 2875 + }, + { + "epoch": 0.21614309334134976, + "grad_norm": 1.4978789592271338, + "learning_rate": 3.6476040940367395e-06, + "loss": 1.0043, + "step": 2876 + }, + { + "epoch": 0.21621824740718473, + "grad_norm": 2.079283541934579, + "learning_rate": 3.647328065304847e-06, + "loss": 0.8909, + "step": 2877 + }, + { + "epoch": 0.2162934014730197, + "grad_norm": 1.3326051879503582, + "learning_rate": 3.6470519389624587e-06, + "loss": 1.0255, + "step": 2878 + }, + { + "epoch": 0.21636855553885465, + "grad_norm": 10.840633769071655, + "learning_rate": 3.646775715025936e-06, + "loss": 1.0555, + "step": 2879 + }, + { + "epoch": 0.21644370960468962, + "grad_norm": 1.8316383184449783, + "learning_rate": 3.6464993935116464e-06, + "loss": 1.0012, + "step": 2880 + }, + { + "epoch": 0.2165188636705246, + "grad_norm": 4.804246835737253, + "learning_rate": 3.646222974435963e-06, + "loss": 1.0359, + "step": 2881 + }, + { + "epoch": 0.21659401773635953, + "grad_norm": 0.7731620220920034, + "learning_rate": 3.645946457815264e-06, + "loss": 0.8977, + "step": 2882 + }, + { + "epoch": 0.2166691718021945, + "grad_norm": 2.0012579776335664, + "learning_rate": 3.6456698436659353e-06, + "loss": 0.9607, + "step": 2883 + }, + { + "epoch": 0.21674432586802947, + "grad_norm": 1.3103819059644655, + "learning_rate": 3.645393132004367e-06, + "loss": 0.9764, + "step": 2884 + }, + { + "epoch": 0.2168194799338644, + "grad_norm": 1.8946742053024002, + "learning_rate": 3.6451163228469543e-06, + "loss": 0.8927, + "step": 2885 + }, + { + "epoch": 0.21689463399969938, + "grad_norm": 1.5191063764992654, + "learning_rate": 3.6448394162100994e-06, + "loss": 1.021, + "step": 2886 + }, + { + "epoch": 0.21696978806553435, + "grad_norm": 2.9589452155515237, + "learning_rate": 3.644562412110211e-06, + "loss": 0.9998, + "step": 2887 + }, + { + "epoch": 0.2170449421313693, + "grad_norm": 1.6347560722773837, + "learning_rate": 3.6442853105637024e-06, + "loss": 0.9717, + "step": 2888 + }, + { + "epoch": 0.21712009619720427, + "grad_norm": 1.6608752320640359, + "learning_rate": 3.644008111586993e-06, + "loss": 1.0641, + "step": 2889 + }, + { + "epoch": 0.21719525026303924, + "grad_norm": 1.716104795421602, + "learning_rate": 3.6437308151965074e-06, + "loss": 0.9237, + "step": 2890 + }, + { + "epoch": 0.2172704043288742, + "grad_norm": 2.5096435517187197, + "learning_rate": 3.6434534214086767e-06, + "loss": 1.0455, + "step": 2891 + }, + { + "epoch": 0.21734555839470915, + "grad_norm": 1.8878350598370675, + "learning_rate": 3.643175930239938e-06, + "loss": 1.141, + "step": 2892 + }, + { + "epoch": 0.21742071246054412, + "grad_norm": 1.5001420112778838, + "learning_rate": 3.6428983417067326e-06, + "loss": 0.9609, + "step": 2893 + }, + { + "epoch": 0.2174958665263791, + "grad_norm": 1.5167813952862244, + "learning_rate": 3.64262065582551e-06, + "loss": 0.996, + "step": 2894 + }, + { + "epoch": 0.21757102059221403, + "grad_norm": 1.6841523201979605, + "learning_rate": 3.6423428726127232e-06, + "loss": 0.9697, + "step": 2895 + }, + { + "epoch": 0.217646174658049, + "grad_norm": 2.1916873635981613, + "learning_rate": 3.6420649920848324e-06, + "loss": 1.0232, + "step": 2896 + }, + { + "epoch": 0.21772132872388397, + "grad_norm": 2.03059599691609, + "learning_rate": 3.641787014258303e-06, + "loss": 0.9559, + "step": 2897 + }, + { + "epoch": 0.21779648278971891, + "grad_norm": 2.3830430216361496, + "learning_rate": 3.641508939149606e-06, + "loss": 0.955, + "step": 2898 + }, + { + "epoch": 0.21787163685555389, + "grad_norm": 1.635693795925986, + "learning_rate": 3.6412307667752185e-06, + "loss": 0.9223, + "step": 2899 + }, + { + "epoch": 0.21794679092138886, + "grad_norm": 1.7739824306473855, + "learning_rate": 3.640952497151623e-06, + "loss": 1.0084, + "step": 2900 + }, + { + "epoch": 0.2180219449872238, + "grad_norm": 3.3106875949535612, + "learning_rate": 3.6406741302953093e-06, + "loss": 1.0716, + "step": 2901 + }, + { + "epoch": 0.21809709905305877, + "grad_norm": 1.6173830513043614, + "learning_rate": 3.6403956662227706e-06, + "loss": 1.0167, + "step": 2902 + }, + { + "epoch": 0.21817225311889374, + "grad_norm": 1.5688135384994297, + "learning_rate": 3.640117104950507e-06, + "loss": 0.9252, + "step": 2903 + }, + { + "epoch": 0.21824740718472868, + "grad_norm": 1.7796385318761176, + "learning_rate": 3.639838446495024e-06, + "loss": 1.0394, + "step": 2904 + }, + { + "epoch": 0.21832256125056365, + "grad_norm": 1.7399940538852863, + "learning_rate": 3.6395596908728344e-06, + "loss": 0.9971, + "step": 2905 + }, + { + "epoch": 0.21839771531639862, + "grad_norm": 1.618245686214596, + "learning_rate": 3.639280838100455e-06, + "loss": 0.9717, + "step": 2906 + }, + { + "epoch": 0.2184728693822336, + "grad_norm": 1.555002075267523, + "learning_rate": 3.639001888194408e-06, + "loss": 1.0361, + "step": 2907 + }, + { + "epoch": 0.21854802344806853, + "grad_norm": 1.3844031974728, + "learning_rate": 3.638722841171223e-06, + "loss": 0.9338, + "step": 2908 + }, + { + "epoch": 0.2186231775139035, + "grad_norm": 0.7108380174724745, + "learning_rate": 3.6384436970474353e-06, + "loss": 0.8222, + "step": 2909 + }, + { + "epoch": 0.21869833157973848, + "grad_norm": 1.3351609945910479, + "learning_rate": 3.638164455839584e-06, + "loss": 1.0761, + "step": 2910 + }, + { + "epoch": 0.21877348564557342, + "grad_norm": 1.556803121664793, + "learning_rate": 3.637885117564216e-06, + "loss": 0.9915, + "step": 2911 + }, + { + "epoch": 0.2188486397114084, + "grad_norm": 2.116590239679447, + "learning_rate": 3.6376056822378826e-06, + "loss": 1.108, + "step": 2912 + }, + { + "epoch": 0.21892379377724336, + "grad_norm": 1.9575202011293753, + "learning_rate": 3.6373261498771418e-06, + "loss": 0.9169, + "step": 2913 + }, + { + "epoch": 0.2189989478430783, + "grad_norm": 1.6353519532182592, + "learning_rate": 3.6370465204985567e-06, + "loss": 1.0182, + "step": 2914 + }, + { + "epoch": 0.21907410190891327, + "grad_norm": 1.6450232995189105, + "learning_rate": 3.636766794118697e-06, + "loss": 1.088, + "step": 2915 + }, + { + "epoch": 0.21914925597474824, + "grad_norm": 1.6054801847695737, + "learning_rate": 3.636486970754137e-06, + "loss": 0.9364, + "step": 2916 + }, + { + "epoch": 0.21922441004058318, + "grad_norm": 2.7784871297678606, + "learning_rate": 3.6362070504214577e-06, + "loss": 0.9652, + "step": 2917 + }, + { + "epoch": 0.21929956410641815, + "grad_norm": 2.2882901604069645, + "learning_rate": 3.6359270331372447e-06, + "loss": 1.0333, + "step": 2918 + }, + { + "epoch": 0.21937471817225312, + "grad_norm": 2.0459433029463083, + "learning_rate": 3.6356469189180907e-06, + "loss": 0.8791, + "step": 2919 + }, + { + "epoch": 0.21944987223808807, + "grad_norm": 2.570850721781664, + "learning_rate": 3.6353667077805934e-06, + "loss": 1.0393, + "step": 2920 + }, + { + "epoch": 0.21952502630392304, + "grad_norm": 1.3801058989126305, + "learning_rate": 3.635086399741357e-06, + "loss": 0.9941, + "step": 2921 + }, + { + "epoch": 0.219600180369758, + "grad_norm": 1.6963716353993818, + "learning_rate": 3.6348059948169894e-06, + "loss": 1.0071, + "step": 2922 + }, + { + "epoch": 0.21967533443559298, + "grad_norm": 1.7578082275338702, + "learning_rate": 3.6345254930241075e-06, + "loss": 1.0201, + "step": 2923 + }, + { + "epoch": 0.21975048850142792, + "grad_norm": 1.639631424634415, + "learning_rate": 3.63424489437933e-06, + "loss": 1.0333, + "step": 2924 + }, + { + "epoch": 0.2198256425672629, + "grad_norm": 1.6128196517765072, + "learning_rate": 3.6339641988992853e-06, + "loss": 1.0392, + "step": 2925 + }, + { + "epoch": 0.21990079663309786, + "grad_norm": 1.6461353588231653, + "learning_rate": 3.633683406600605e-06, + "loss": 0.9905, + "step": 2926 + }, + { + "epoch": 0.2199759506989328, + "grad_norm": 1.5976457292354926, + "learning_rate": 3.633402517499927e-06, + "loss": 0.9773, + "step": 2927 + }, + { + "epoch": 0.22005110476476777, + "grad_norm": 2.2759471556524034, + "learning_rate": 3.633121531613895e-06, + "loss": 1.0648, + "step": 2928 + }, + { + "epoch": 0.22012625883060274, + "grad_norm": 1.9544424267767422, + "learning_rate": 3.6328404489591585e-06, + "loss": 1.0894, + "step": 2929 + }, + { + "epoch": 0.2202014128964377, + "grad_norm": 1.4241413038899737, + "learning_rate": 3.6325592695523727e-06, + "loss": 1.0911, + "step": 2930 + }, + { + "epoch": 0.22027656696227266, + "grad_norm": 1.5043085526236237, + "learning_rate": 3.6322779934101995e-06, + "loss": 1.0707, + "step": 2931 + }, + { + "epoch": 0.22035172102810763, + "grad_norm": 1.7850795203482321, + "learning_rate": 3.6319966205493044e-06, + "loss": 1.1135, + "step": 2932 + }, + { + "epoch": 0.22042687509394257, + "grad_norm": 1.6830968986627257, + "learning_rate": 3.63171515098636e-06, + "loss": 1.0425, + "step": 2933 + }, + { + "epoch": 0.22050202915977754, + "grad_norm": 1.917025767424306, + "learning_rate": 3.6314335847380443e-06, + "loss": 0.9814, + "step": 2934 + }, + { + "epoch": 0.2205771832256125, + "grad_norm": 1.573358418997441, + "learning_rate": 3.631151921821042e-06, + "loss": 1.0243, + "step": 2935 + }, + { + "epoch": 0.22065233729144745, + "grad_norm": 2.039183748228518, + "learning_rate": 3.630870162252042e-06, + "loss": 1.0201, + "step": 2936 + }, + { + "epoch": 0.22072749135728242, + "grad_norm": 0.8172511788934728, + "learning_rate": 3.63058830604774e-06, + "loss": 0.8934, + "step": 2937 + }, + { + "epoch": 0.2208026454231174, + "grad_norm": 1.6894469156316527, + "learning_rate": 3.6303063532248367e-06, + "loss": 1.0288, + "step": 2938 + }, + { + "epoch": 0.22087779948895236, + "grad_norm": 2.045078100377956, + "learning_rate": 3.6300243038000397e-06, + "loss": 0.9798, + "step": 2939 + }, + { + "epoch": 0.2209529535547873, + "grad_norm": 2.011953988256341, + "learning_rate": 3.6297421577900608e-06, + "loss": 0.9632, + "step": 2940 + }, + { + "epoch": 0.22102810762062228, + "grad_norm": 1.6103247645492373, + "learning_rate": 3.629459915211618e-06, + "loss": 0.9966, + "step": 2941 + }, + { + "epoch": 0.22110326168645725, + "grad_norm": 1.9359247510488709, + "learning_rate": 3.6291775760814358e-06, + "loss": 1.0137, + "step": 2942 + }, + { + "epoch": 0.2211784157522922, + "grad_norm": 3.2737007126333335, + "learning_rate": 3.6288951404162433e-06, + "loss": 1.1605, + "step": 2943 + }, + { + "epoch": 0.22125356981812716, + "grad_norm": 2.032128833508787, + "learning_rate": 3.6286126082327764e-06, + "loss": 1.0161, + "step": 2944 + }, + { + "epoch": 0.22132872388396213, + "grad_norm": 1.4432776070483044, + "learning_rate": 3.6283299795477767e-06, + "loss": 1.0271, + "step": 2945 + }, + { + "epoch": 0.22140387794979707, + "grad_norm": 4.377638974600734, + "learning_rate": 3.62804725437799e-06, + "loss": 1.0203, + "step": 2946 + }, + { + "epoch": 0.22147903201563204, + "grad_norm": 2.286276726835026, + "learning_rate": 3.6277644327401687e-06, + "loss": 1.0568, + "step": 2947 + }, + { + "epoch": 0.221554186081467, + "grad_norm": 1.4346870598182675, + "learning_rate": 3.627481514651073e-06, + "loss": 0.9958, + "step": 2948 + }, + { + "epoch": 0.22162934014730196, + "grad_norm": 1.438439352534475, + "learning_rate": 3.6271985001274647e-06, + "loss": 0.9025, + "step": 2949 + }, + { + "epoch": 0.22170449421313693, + "grad_norm": 1.7908422288397103, + "learning_rate": 3.6269153891861137e-06, + "loss": 1.0288, + "step": 2950 + }, + { + "epoch": 0.2217796482789719, + "grad_norm": 2.3979714403655006, + "learning_rate": 3.6266321818437967e-06, + "loss": 0.9762, + "step": 2951 + }, + { + "epoch": 0.22185480234480687, + "grad_norm": 1.8732195984376598, + "learning_rate": 3.6263488781172946e-06, + "loss": 0.92, + "step": 2952 + }, + { + "epoch": 0.2219299564106418, + "grad_norm": 1.7074673093512074, + "learning_rate": 3.6260654780233927e-06, + "loss": 1.0763, + "step": 2953 + }, + { + "epoch": 0.22200511047647678, + "grad_norm": 1.8023276484993416, + "learning_rate": 3.6257819815788854e-06, + "loss": 0.9787, + "step": 2954 + }, + { + "epoch": 0.22208026454231175, + "grad_norm": 0.748776033168629, + "learning_rate": 3.6254983888005697e-06, + "loss": 0.9126, + "step": 2955 + }, + { + "epoch": 0.2221554186081467, + "grad_norm": 1.6416234928915125, + "learning_rate": 3.6252146997052507e-06, + "loss": 1.0551, + "step": 2956 + }, + { + "epoch": 0.22223057267398166, + "grad_norm": 1.8461908251164452, + "learning_rate": 3.624930914309736e-06, + "loss": 1.0427, + "step": 2957 + }, + { + "epoch": 0.22230572673981663, + "grad_norm": 1.770883005978901, + "learning_rate": 3.624647032630844e-06, + "loss": 1.05, + "step": 2958 + }, + { + "epoch": 0.22238088080565158, + "grad_norm": 1.555892553988509, + "learning_rate": 3.6243630546853932e-06, + "loss": 0.9099, + "step": 2959 + }, + { + "epoch": 0.22245603487148655, + "grad_norm": 2.304441794536264, + "learning_rate": 3.6240789804902116e-06, + "loss": 1.0451, + "step": 2960 + }, + { + "epoch": 0.22253118893732152, + "grad_norm": 1.6879702725165684, + "learning_rate": 3.623794810062131e-06, + "loss": 1.1132, + "step": 2961 + }, + { + "epoch": 0.22260634300315646, + "grad_norm": 1.6404923884974718, + "learning_rate": 3.62351054341799e-06, + "loss": 0.9327, + "step": 2962 + }, + { + "epoch": 0.22268149706899143, + "grad_norm": 0.8054875960860138, + "learning_rate": 3.623226180574633e-06, + "loss": 0.8056, + "step": 2963 + }, + { + "epoch": 0.2227566511348264, + "grad_norm": 1.502206530481191, + "learning_rate": 3.6229417215489084e-06, + "loss": 1.0412, + "step": 2964 + }, + { + "epoch": 0.22283180520066134, + "grad_norm": 1.7281519965327463, + "learning_rate": 3.6226571663576727e-06, + "loss": 0.9873, + "step": 2965 + }, + { + "epoch": 0.2229069592664963, + "grad_norm": 1.8166932238232942, + "learning_rate": 3.6223725150177858e-06, + "loss": 0.9051, + "step": 2966 + }, + { + "epoch": 0.22298211333233128, + "grad_norm": 1.600098922770898, + "learning_rate": 3.622087767546116e-06, + "loss": 1.0803, + "step": 2967 + }, + { + "epoch": 0.22305726739816625, + "grad_norm": 1.7664150394725484, + "learning_rate": 3.6218029239595332e-06, + "loss": 0.9026, + "step": 2968 + }, + { + "epoch": 0.2231324214640012, + "grad_norm": 1.4998272160515802, + "learning_rate": 3.6215179842749172e-06, + "loss": 1.0719, + "step": 2969 + }, + { + "epoch": 0.22320757552983617, + "grad_norm": 2.151676801583286, + "learning_rate": 3.6212329485091518e-06, + "loss": 0.9619, + "step": 2970 + }, + { + "epoch": 0.22328272959567114, + "grad_norm": 1.7593542500927253, + "learning_rate": 3.620947816679126e-06, + "loss": 1.1373, + "step": 2971 + }, + { + "epoch": 0.22335788366150608, + "grad_norm": 3.755728097923521, + "learning_rate": 3.6206625888017355e-06, + "loss": 0.9935, + "step": 2972 + }, + { + "epoch": 0.22343303772734105, + "grad_norm": 1.7885520240100181, + "learning_rate": 3.62037726489388e-06, + "loss": 0.9333, + "step": 2973 + }, + { + "epoch": 0.22350819179317602, + "grad_norm": 1.4128460789271662, + "learning_rate": 3.620091844972467e-06, + "loss": 1.0283, + "step": 2974 + }, + { + "epoch": 0.22358334585901096, + "grad_norm": 1.622874704127787, + "learning_rate": 3.619806329054408e-06, + "loss": 1.0598, + "step": 2975 + }, + { + "epoch": 0.22365849992484593, + "grad_norm": 1.6512477838355533, + "learning_rate": 3.619520717156622e-06, + "loss": 1.028, + "step": 2976 + }, + { + "epoch": 0.2237336539906809, + "grad_norm": 1.5624179055582046, + "learning_rate": 3.6192350092960315e-06, + "loss": 1.0785, + "step": 2977 + }, + { + "epoch": 0.22380880805651585, + "grad_norm": 1.4898808411180375, + "learning_rate": 3.6189492054895667e-06, + "loss": 0.9912, + "step": 2978 + }, + { + "epoch": 0.22388396212235082, + "grad_norm": 1.4074448489670603, + "learning_rate": 3.6186633057541617e-06, + "loss": 0.9986, + "step": 2979 + }, + { + "epoch": 0.2239591161881858, + "grad_norm": 1.7030566840512964, + "learning_rate": 3.6183773101067575e-06, + "loss": 1.0895, + "step": 2980 + }, + { + "epoch": 0.22403427025402073, + "grad_norm": 0.7525741192760483, + "learning_rate": 3.618091218564301e-06, + "loss": 0.8674, + "step": 2981 + }, + { + "epoch": 0.2241094243198557, + "grad_norm": 1.7455211955182028, + "learning_rate": 3.6178050311437432e-06, + "loss": 0.9977, + "step": 2982 + }, + { + "epoch": 0.22418457838569067, + "grad_norm": 1.5707964554698364, + "learning_rate": 3.6175187478620424e-06, + "loss": 1.0776, + "step": 2983 + }, + { + "epoch": 0.22425973245152564, + "grad_norm": 1.4377419641277145, + "learning_rate": 3.617232368736162e-06, + "loss": 0.9853, + "step": 2984 + }, + { + "epoch": 0.22433488651736058, + "grad_norm": 1.602517043572069, + "learning_rate": 3.616945893783071e-06, + "loss": 0.9448, + "step": 2985 + }, + { + "epoch": 0.22441004058319555, + "grad_norm": 1.740107230905276, + "learning_rate": 3.616659323019744e-06, + "loss": 1.0056, + "step": 2986 + }, + { + "epoch": 0.22448519464903052, + "grad_norm": 2.382056744902921, + "learning_rate": 3.616372656463161e-06, + "loss": 1.0399, + "step": 2987 + }, + { + "epoch": 0.22456034871486547, + "grad_norm": 1.5741258314553526, + "learning_rate": 3.6160858941303095e-06, + "loss": 1.0913, + "step": 2988 + }, + { + "epoch": 0.22463550278070044, + "grad_norm": 74.01250950008384, + "learning_rate": 3.61579903603818e-06, + "loss": 1.0092, + "step": 2989 + }, + { + "epoch": 0.2247106568465354, + "grad_norm": 1.9902027490740422, + "learning_rate": 3.6155120822037707e-06, + "loss": 1.0131, + "step": 2990 + }, + { + "epoch": 0.22478581091237035, + "grad_norm": 2.4219780684658643, + "learning_rate": 3.6152250326440833e-06, + "loss": 1.0545, + "step": 2991 + }, + { + "epoch": 0.22486096497820532, + "grad_norm": 1.4736038518629564, + "learning_rate": 3.614937887376128e-06, + "loss": 0.8862, + "step": 2992 + }, + { + "epoch": 0.2249361190440403, + "grad_norm": 1.6783961100172307, + "learning_rate": 3.61465064641692e-06, + "loss": 0.987, + "step": 2993 + }, + { + "epoch": 0.22501127310987523, + "grad_norm": 1.852452897001962, + "learning_rate": 3.614363309783477e-06, + "loss": 1.0269, + "step": 2994 + }, + { + "epoch": 0.2250864271757102, + "grad_norm": 1.7620459546646614, + "learning_rate": 3.6140758774928265e-06, + "loss": 1.0218, + "step": 2995 + }, + { + "epoch": 0.22516158124154517, + "grad_norm": 1.1098775128493898, + "learning_rate": 3.613788349561999e-06, + "loss": 0.953, + "step": 2996 + }, + { + "epoch": 0.22523673530738014, + "grad_norm": 1.7683956721418774, + "learning_rate": 3.6135007260080334e-06, + "loss": 0.9939, + "step": 2997 + }, + { + "epoch": 0.22531188937321509, + "grad_norm": 2.008257271155984, + "learning_rate": 3.61321300684797e-06, + "loss": 1.0202, + "step": 2998 + }, + { + "epoch": 0.22538704343905006, + "grad_norm": 2.1325196005668405, + "learning_rate": 3.6129251920988594e-06, + "loss": 1.0521, + "step": 2999 + }, + { + "epoch": 0.22546219750488503, + "grad_norm": 1.8289354924109245, + "learning_rate": 3.612637281777755e-06, + "loss": 1.0529, + "step": 3000 + }, + { + "epoch": 0.22553735157071997, + "grad_norm": 2.9492991051242696, + "learning_rate": 3.6123492759017155e-06, + "loss": 0.9704, + "step": 3001 + }, + { + "epoch": 0.22561250563655494, + "grad_norm": 2.0432599007924273, + "learning_rate": 3.6120611744878076e-06, + "loss": 1.0814, + "step": 3002 + }, + { + "epoch": 0.2256876597023899, + "grad_norm": 2.002585885138502, + "learning_rate": 3.6117729775531028e-06, + "loss": 1.0179, + "step": 3003 + }, + { + "epoch": 0.22576281376822485, + "grad_norm": 1.830646569101178, + "learning_rate": 3.6114846851146767e-06, + "loss": 0.8992, + "step": 3004 + }, + { + "epoch": 0.22583796783405982, + "grad_norm": 1.6538377144893357, + "learning_rate": 3.611196297189612e-06, + "loss": 1.1149, + "step": 3005 + }, + { + "epoch": 0.2259131218998948, + "grad_norm": 1.9341313395081026, + "learning_rate": 3.6109078137949975e-06, + "loss": 0.9409, + "step": 3006 + }, + { + "epoch": 0.22598827596572973, + "grad_norm": 1.8760020439486562, + "learning_rate": 3.6106192349479263e-06, + "loss": 1.1266, + "step": 3007 + }, + { + "epoch": 0.2260634300315647, + "grad_norm": 1.8739687308814117, + "learning_rate": 3.610330560665498e-06, + "loss": 1.0433, + "step": 3008 + }, + { + "epoch": 0.22613858409739968, + "grad_norm": 1.6317838504132844, + "learning_rate": 3.6100417909648175e-06, + "loss": 0.8843, + "step": 3009 + }, + { + "epoch": 0.22621373816323462, + "grad_norm": 1.6474161464207868, + "learning_rate": 3.6097529258629952e-06, + "loss": 1.0478, + "step": 3010 + }, + { + "epoch": 0.2262888922290696, + "grad_norm": 3.9286491584519325, + "learning_rate": 3.6094639653771486e-06, + "loss": 0.9477, + "step": 3011 + }, + { + "epoch": 0.22636404629490456, + "grad_norm": 2.0765734164924803, + "learning_rate": 3.6091749095243986e-06, + "loss": 1.0004, + "step": 3012 + }, + { + "epoch": 0.22643920036073953, + "grad_norm": 1.9832436047086288, + "learning_rate": 3.6088857583218735e-06, + "loss": 1.0661, + "step": 3013 + }, + { + "epoch": 0.22651435442657447, + "grad_norm": 1.5648138842348989, + "learning_rate": 3.6085965117867066e-06, + "loss": 0.949, + "step": 3014 + }, + { + "epoch": 0.22658950849240944, + "grad_norm": 1.9153514648098895, + "learning_rate": 3.608307169936036e-06, + "loss": 0.9568, + "step": 3015 + }, + { + "epoch": 0.2266646625582444, + "grad_norm": 2.197681058817544, + "learning_rate": 3.608017732787007e-06, + "loss": 0.9261, + "step": 3016 + }, + { + "epoch": 0.22673981662407935, + "grad_norm": 4.713239066455105, + "learning_rate": 3.60772820035677e-06, + "loss": 0.9777, + "step": 3017 + }, + { + "epoch": 0.22681497068991432, + "grad_norm": 2.0602225242720316, + "learning_rate": 3.607438572662481e-06, + "loss": 1.0513, + "step": 3018 + }, + { + "epoch": 0.2268901247557493, + "grad_norm": 1.5860367166086238, + "learning_rate": 3.6071488497213017e-06, + "loss": 0.9465, + "step": 3019 + }, + { + "epoch": 0.22696527882158424, + "grad_norm": 1.681478707452379, + "learning_rate": 3.6068590315503976e-06, + "loss": 0.9153, + "step": 3020 + }, + { + "epoch": 0.2270404328874192, + "grad_norm": 1.9889992127759784, + "learning_rate": 3.606569118166944e-06, + "loss": 0.9931, + "step": 3021 + }, + { + "epoch": 0.22711558695325418, + "grad_norm": 1.8075808496187176, + "learning_rate": 3.6062791095881174e-06, + "loss": 1.0245, + "step": 3022 + }, + { + "epoch": 0.22719074101908912, + "grad_norm": 1.3380811569536055, + "learning_rate": 3.6059890058311025e-06, + "loss": 1.0064, + "step": 3023 + }, + { + "epoch": 0.2272658950849241, + "grad_norm": 1.3279779240571516, + "learning_rate": 3.6056988069130903e-06, + "loss": 0.9242, + "step": 3024 + }, + { + "epoch": 0.22734104915075906, + "grad_norm": 9.059711737688984, + "learning_rate": 3.6054085128512747e-06, + "loss": 0.9856, + "step": 3025 + }, + { + "epoch": 0.227416203216594, + "grad_norm": 1.5398878798603792, + "learning_rate": 3.605118123662857e-06, + "loss": 0.9525, + "step": 3026 + }, + { + "epoch": 0.22749135728242897, + "grad_norm": 1.4442568661270836, + "learning_rate": 3.6048276393650434e-06, + "loss": 1.0129, + "step": 3027 + }, + { + "epoch": 0.22756651134826394, + "grad_norm": 1.7929174212253882, + "learning_rate": 3.6045370599750482e-06, + "loss": 0.9955, + "step": 3028 + }, + { + "epoch": 0.22764166541409891, + "grad_norm": 2.173587371024859, + "learning_rate": 3.6042463855100876e-06, + "loss": 0.9684, + "step": 3029 + }, + { + "epoch": 0.22771681947993386, + "grad_norm": 2.096600883292875, + "learning_rate": 3.603955615987385e-06, + "loss": 0.9746, + "step": 3030 + }, + { + "epoch": 0.22779197354576883, + "grad_norm": 4.763682685642714, + "learning_rate": 3.603664751424171e-06, + "loss": 1.0027, + "step": 3031 + }, + { + "epoch": 0.2278671276116038, + "grad_norm": 1.6627186824984908, + "learning_rate": 3.603373791837679e-06, + "loss": 1.0592, + "step": 3032 + }, + { + "epoch": 0.22794228167743874, + "grad_norm": 1.5294928297290664, + "learning_rate": 3.6030827372451506e-06, + "loss": 0.8963, + "step": 3033 + }, + { + "epoch": 0.2280174357432737, + "grad_norm": 1.626122380496975, + "learning_rate": 3.602791587663831e-06, + "loss": 0.9104, + "step": 3034 + }, + { + "epoch": 0.22809258980910868, + "grad_norm": 1.4375131855235546, + "learning_rate": 3.6025003431109722e-06, + "loss": 1.0114, + "step": 3035 + }, + { + "epoch": 0.22816774387494362, + "grad_norm": 1.6320181196459749, + "learning_rate": 3.6022090036038326e-06, + "loss": 1.1181, + "step": 3036 + }, + { + "epoch": 0.2282428979407786, + "grad_norm": 1.3553534027182987, + "learning_rate": 3.601917569159673e-06, + "loss": 1.1101, + "step": 3037 + }, + { + "epoch": 0.22831805200661356, + "grad_norm": 2.1965728074408717, + "learning_rate": 3.6016260397957642e-06, + "loss": 1.0332, + "step": 3038 + }, + { + "epoch": 0.2283932060724485, + "grad_norm": 0.6727652307499431, + "learning_rate": 3.6013344155293792e-06, + "loss": 0.8098, + "step": 3039 + }, + { + "epoch": 0.22846836013828348, + "grad_norm": 2.025014373351526, + "learning_rate": 3.6010426963777985e-06, + "loss": 0.9813, + "step": 3040 + }, + { + "epoch": 0.22854351420411845, + "grad_norm": 1.4747297427332822, + "learning_rate": 3.600750882358307e-06, + "loss": 1.0938, + "step": 3041 + }, + { + "epoch": 0.22861866826995342, + "grad_norm": 1.736744679451762, + "learning_rate": 3.6004589734881953e-06, + "loss": 1.0112, + "step": 3042 + }, + { + "epoch": 0.22869382233578836, + "grad_norm": 1.6592766799103713, + "learning_rate": 3.600166969784762e-06, + "loss": 1.0265, + "step": 3043 + }, + { + "epoch": 0.22876897640162333, + "grad_norm": 1.714968179346385, + "learning_rate": 3.5998748712653077e-06, + "loss": 1.0065, + "step": 3044 + }, + { + "epoch": 0.2288441304674583, + "grad_norm": 2.7406067381090544, + "learning_rate": 3.5995826779471408e-06, + "loss": 1.0254, + "step": 3045 + }, + { + "epoch": 0.22891928453329324, + "grad_norm": 2.282632696132179, + "learning_rate": 3.5992903898475757e-06, + "loss": 0.9754, + "step": 3046 + }, + { + "epoch": 0.2289944385991282, + "grad_norm": 1.3356733005128236, + "learning_rate": 3.5989980069839304e-06, + "loss": 0.9305, + "step": 3047 + }, + { + "epoch": 0.22906959266496318, + "grad_norm": 4.945894083594271, + "learning_rate": 3.5987055293735305e-06, + "loss": 1.0565, + "step": 3048 + }, + { + "epoch": 0.22914474673079813, + "grad_norm": 1.7090835629698224, + "learning_rate": 3.5984129570337056e-06, + "loss": 1.0338, + "step": 3049 + }, + { + "epoch": 0.2292199007966331, + "grad_norm": 1.8561943482556167, + "learning_rate": 3.598120289981793e-06, + "loss": 1.0256, + "step": 3050 + }, + { + "epoch": 0.22929505486246807, + "grad_norm": 1.8955805124356961, + "learning_rate": 3.597827528235133e-06, + "loss": 1.0453, + "step": 3051 + }, + { + "epoch": 0.229370208928303, + "grad_norm": 2.38343787819867, + "learning_rate": 3.597534671811074e-06, + "loss": 0.9508, + "step": 3052 + }, + { + "epoch": 0.22944536299413798, + "grad_norm": 1.9300151346038208, + "learning_rate": 3.5972417207269675e-06, + "loss": 1.0588, + "step": 3053 + }, + { + "epoch": 0.22952051705997295, + "grad_norm": 2.54795524295561, + "learning_rate": 3.596948675000173e-06, + "loss": 1.0519, + "step": 3054 + }, + { + "epoch": 0.2295956711258079, + "grad_norm": 3.614636174382802, + "learning_rate": 3.596655534648055e-06, + "loss": 0.9653, + "step": 3055 + }, + { + "epoch": 0.22967082519164286, + "grad_norm": 2.090091298495681, + "learning_rate": 3.596362299687982e-06, + "loss": 1.0367, + "step": 3056 + }, + { + "epoch": 0.22974597925747783, + "grad_norm": 2.417080175166135, + "learning_rate": 3.59606897013733e-06, + "loss": 0.9547, + "step": 3057 + }, + { + "epoch": 0.2298211333233128, + "grad_norm": 1.5029458682730341, + "learning_rate": 3.59577554601348e-06, + "loss": 0.874, + "step": 3058 + }, + { + "epoch": 0.22989628738914775, + "grad_norm": 1.8112281414684865, + "learning_rate": 3.595482027333818e-06, + "loss": 0.9263, + "step": 3059 + }, + { + "epoch": 0.22997144145498272, + "grad_norm": 1.460329306579439, + "learning_rate": 3.5951884141157365e-06, + "loss": 0.9073, + "step": 3060 + }, + { + "epoch": 0.2300465955208177, + "grad_norm": 1.414132638108626, + "learning_rate": 3.5948947063766334e-06, + "loss": 1.0422, + "step": 3061 + }, + { + "epoch": 0.23012174958665263, + "grad_norm": 3.846459155069271, + "learning_rate": 3.5946009041339114e-06, + "loss": 1.0555, + "step": 3062 + }, + { + "epoch": 0.2301969036524876, + "grad_norm": 3.447376005226458, + "learning_rate": 3.5943070074049797e-06, + "loss": 0.9462, + "step": 3063 + }, + { + "epoch": 0.23027205771832257, + "grad_norm": 1.8324177218463114, + "learning_rate": 3.5940130162072525e-06, + "loss": 1.0372, + "step": 3064 + }, + { + "epoch": 0.2303472117841575, + "grad_norm": 1.6518130340673458, + "learning_rate": 3.593718930558151e-06, + "loss": 1.0171, + "step": 3065 + }, + { + "epoch": 0.23042236584999248, + "grad_norm": 1.9146027717253873, + "learning_rate": 3.5934247504750995e-06, + "loss": 0.9399, + "step": 3066 + }, + { + "epoch": 0.23049751991582745, + "grad_norm": 1.8847150252275169, + "learning_rate": 3.59313047597553e-06, + "loss": 0.9205, + "step": 3067 + }, + { + "epoch": 0.2305726739816624, + "grad_norm": 2.079524207350104, + "learning_rate": 3.5928361070768788e-06, + "loss": 0.924, + "step": 3068 + }, + { + "epoch": 0.23064782804749737, + "grad_norm": 1.637080914981124, + "learning_rate": 3.59254164379659e-06, + "loss": 1.0589, + "step": 3069 + }, + { + "epoch": 0.23072298211333234, + "grad_norm": 0.8066576939732719, + "learning_rate": 3.5922470861521098e-06, + "loss": 0.8417, + "step": 3070 + }, + { + "epoch": 0.23079813617916728, + "grad_norm": 2.2117579794543922, + "learning_rate": 3.5919524341608923e-06, + "loss": 1.0335, + "step": 3071 + }, + { + "epoch": 0.23087329024500225, + "grad_norm": 0.7586097456309332, + "learning_rate": 3.5916576878403975e-06, + "loss": 0.8548, + "step": 3072 + }, + { + "epoch": 0.23094844431083722, + "grad_norm": 3.067436599138422, + "learning_rate": 3.59136284720809e-06, + "loss": 0.9882, + "step": 3073 + }, + { + "epoch": 0.2310235983766722, + "grad_norm": 2.0245756387521827, + "learning_rate": 3.59106791228144e-06, + "loss": 1.0652, + "step": 3074 + }, + { + "epoch": 0.23109875244250713, + "grad_norm": 0.7199191660009904, + "learning_rate": 3.5907728830779236e-06, + "loss": 0.8853, + "step": 3075 + }, + { + "epoch": 0.2311739065083421, + "grad_norm": 1.5811964304900368, + "learning_rate": 3.5904777596150222e-06, + "loss": 1.013, + "step": 3076 + }, + { + "epoch": 0.23124906057417707, + "grad_norm": 1.669586040791975, + "learning_rate": 3.5901825419102238e-06, + "loss": 0.9691, + "step": 3077 + }, + { + "epoch": 0.23132421464001202, + "grad_norm": 1.8629515414157858, + "learning_rate": 3.58988722998102e-06, + "loss": 0.9613, + "step": 3078 + }, + { + "epoch": 0.231399368705847, + "grad_norm": 1.8697182331877957, + "learning_rate": 3.58959182384491e-06, + "loss": 0.9754, + "step": 3079 + }, + { + "epoch": 0.23147452277168196, + "grad_norm": 1.4820439556825962, + "learning_rate": 3.5892963235193968e-06, + "loss": 1.007, + "step": 3080 + }, + { + "epoch": 0.2315496768375169, + "grad_norm": 2.8670880458951795, + "learning_rate": 3.589000729021991e-06, + "loss": 0.8725, + "step": 3081 + }, + { + "epoch": 0.23162483090335187, + "grad_norm": 1.6883383893498705, + "learning_rate": 3.5887050403702073e-06, + "loss": 0.9828, + "step": 3082 + }, + { + "epoch": 0.23169998496918684, + "grad_norm": 1.9214963462411694, + "learning_rate": 3.588409257581567e-06, + "loss": 0.9375, + "step": 3083 + }, + { + "epoch": 0.23177513903502178, + "grad_norm": 1.9622533874093673, + "learning_rate": 3.5881133806735956e-06, + "loss": 1.0766, + "step": 3084 + }, + { + "epoch": 0.23185029310085675, + "grad_norm": 1.4617736382019075, + "learning_rate": 3.587817409663824e-06, + "loss": 1.0272, + "step": 3085 + }, + { + "epoch": 0.23192544716669172, + "grad_norm": 0.6291773194133311, + "learning_rate": 3.5875213445697917e-06, + "loss": 0.8415, + "step": 3086 + }, + { + "epoch": 0.2320006012325267, + "grad_norm": 2.4818969459687352, + "learning_rate": 3.587225185409041e-06, + "loss": 0.9947, + "step": 3087 + }, + { + "epoch": 0.23207575529836164, + "grad_norm": 1.829138645950998, + "learning_rate": 3.5869289321991195e-06, + "loss": 0.9709, + "step": 3088 + }, + { + "epoch": 0.2321509093641966, + "grad_norm": 1.7116415587047509, + "learning_rate": 3.586632584957582e-06, + "loss": 0.9564, + "step": 3089 + }, + { + "epoch": 0.23222606343003158, + "grad_norm": 1.4524511600234336, + "learning_rate": 3.5863361437019885e-06, + "loss": 1.0468, + "step": 3090 + }, + { + "epoch": 0.23230121749586652, + "grad_norm": 1.3349834850447035, + "learning_rate": 3.5860396084499043e-06, + "loss": 1.0603, + "step": 3091 + }, + { + "epoch": 0.2323763715617015, + "grad_norm": 1.7861862049496524, + "learning_rate": 3.5857429792188996e-06, + "loss": 1.1211, + "step": 3092 + }, + { + "epoch": 0.23245152562753646, + "grad_norm": 0.7668726350705866, + "learning_rate": 3.585446256026551e-06, + "loss": 0.8289, + "step": 3093 + }, + { + "epoch": 0.2325266796933714, + "grad_norm": 1.6069310714808525, + "learning_rate": 3.5851494388904406e-06, + "loss": 0.8865, + "step": 3094 + }, + { + "epoch": 0.23260183375920637, + "grad_norm": 1.4962091227136671, + "learning_rate": 3.5848525278281564e-06, + "loss": 0.9598, + "step": 3095 + }, + { + "epoch": 0.23267698782504134, + "grad_norm": 2.107726859013528, + "learning_rate": 3.5845555228572907e-06, + "loss": 0.8926, + "step": 3096 + }, + { + "epoch": 0.23275214189087629, + "grad_norm": 1.7566193324642114, + "learning_rate": 3.5842584239954426e-06, + "loss": 1.0281, + "step": 3097 + }, + { + "epoch": 0.23282729595671126, + "grad_norm": 1.7586770877216071, + "learning_rate": 3.5839612312602166e-06, + "loss": 0.9663, + "step": 3098 + }, + { + "epoch": 0.23290245002254623, + "grad_norm": 2.1065111852965073, + "learning_rate": 3.5836639446692223e-06, + "loss": 1.0323, + "step": 3099 + }, + { + "epoch": 0.23297760408838117, + "grad_norm": 0.8657478548310306, + "learning_rate": 3.5833665642400747e-06, + "loss": 0.849, + "step": 3100 + }, + { + "epoch": 0.23305275815421614, + "grad_norm": 1.6002929866281197, + "learning_rate": 3.5830690899903954e-06, + "loss": 0.9325, + "step": 3101 + }, + { + "epoch": 0.2331279122200511, + "grad_norm": 1.8078301879756042, + "learning_rate": 3.582771521937811e-06, + "loss": 1.0037, + "step": 3102 + }, + { + "epoch": 0.23320306628588608, + "grad_norm": 1.5906913428720117, + "learning_rate": 3.582473860099952e-06, + "loss": 0.9924, + "step": 3103 + }, + { + "epoch": 0.23327822035172102, + "grad_norm": 1.8888940374765475, + "learning_rate": 3.582176104494458e-06, + "loss": 0.9945, + "step": 3104 + }, + { + "epoch": 0.233353374417556, + "grad_norm": 4.167175846777271, + "learning_rate": 3.581878255138971e-06, + "loss": 0.9956, + "step": 3105 + }, + { + "epoch": 0.23342852848339096, + "grad_norm": 2.0349000267947, + "learning_rate": 3.5815803120511395e-06, + "loss": 0.9703, + "step": 3106 + }, + { + "epoch": 0.2335036825492259, + "grad_norm": 2.0897243002150843, + "learning_rate": 3.5812822752486187e-06, + "loss": 1.0305, + "step": 3107 + }, + { + "epoch": 0.23357883661506088, + "grad_norm": 1.6572098739583996, + "learning_rate": 3.5809841447490674e-06, + "loss": 1.0501, + "step": 3108 + }, + { + "epoch": 0.23365399068089585, + "grad_norm": 1.6620073836308333, + "learning_rate": 3.5806859205701523e-06, + "loss": 1.0759, + "step": 3109 + }, + { + "epoch": 0.2337291447467308, + "grad_norm": 1.8191986605524384, + "learning_rate": 3.5803876027295433e-06, + "loss": 1.022, + "step": 3110 + }, + { + "epoch": 0.23380429881256576, + "grad_norm": 0.7301091823840092, + "learning_rate": 3.580089191244917e-06, + "loss": 0.865, + "step": 3111 + }, + { + "epoch": 0.23387945287840073, + "grad_norm": 4.487592121336937, + "learning_rate": 3.5797906861339556e-06, + "loss": 1.0938, + "step": 3112 + }, + { + "epoch": 0.23395460694423567, + "grad_norm": 1.7340255935214528, + "learning_rate": 3.579492087414347e-06, + "loss": 1.0716, + "step": 3113 + }, + { + "epoch": 0.23402976101007064, + "grad_norm": 2.8249171489618123, + "learning_rate": 3.5791933951037834e-06, + "loss": 0.9836, + "step": 3114 + }, + { + "epoch": 0.2341049150759056, + "grad_norm": 2.116474327099317, + "learning_rate": 3.5788946092199643e-06, + "loss": 0.9961, + "step": 3115 + }, + { + "epoch": 0.23418006914174055, + "grad_norm": 1.9234913177770256, + "learning_rate": 3.578595729780593e-06, + "loss": 0.9462, + "step": 3116 + }, + { + "epoch": 0.23425522320757552, + "grad_norm": 2.086257167138736, + "learning_rate": 3.5782967568033805e-06, + "loss": 0.9961, + "step": 3117 + }, + { + "epoch": 0.2343303772734105, + "grad_norm": 1.4427336909777477, + "learning_rate": 3.5779976903060412e-06, + "loss": 0.9734, + "step": 3118 + }, + { + "epoch": 0.23440553133924547, + "grad_norm": 1.8779465728651303, + "learning_rate": 3.5776985303062965e-06, + "loss": 1.0257, + "step": 3119 + }, + { + "epoch": 0.2344806854050804, + "grad_norm": 2.123891204576391, + "learning_rate": 3.5773992768218724e-06, + "loss": 1.0747, + "step": 3120 + }, + { + "epoch": 0.23455583947091538, + "grad_norm": 2.921672467234878, + "learning_rate": 3.577099929870501e-06, + "loss": 1.0207, + "step": 3121 + }, + { + "epoch": 0.23463099353675035, + "grad_norm": 1.3806624427232241, + "learning_rate": 3.5768004894699192e-06, + "loss": 0.8652, + "step": 3122 + }, + { + "epoch": 0.2347061476025853, + "grad_norm": 1.8321097871583838, + "learning_rate": 3.57650095563787e-06, + "loss": 0.9796, + "step": 3123 + }, + { + "epoch": 0.23478130166842026, + "grad_norm": 0.714048766957785, + "learning_rate": 3.5762013283921033e-06, + "loss": 0.8218, + "step": 3124 + }, + { + "epoch": 0.23485645573425523, + "grad_norm": 2.3743138576988594, + "learning_rate": 3.5759016077503716e-06, + "loss": 1.0565, + "step": 3125 + }, + { + "epoch": 0.23493160980009017, + "grad_norm": 2.224418491282414, + "learning_rate": 3.5756017937304356e-06, + "loss": 0.97, + "step": 3126 + }, + { + "epoch": 0.23500676386592514, + "grad_norm": 2.275979210689912, + "learning_rate": 3.57530188635006e-06, + "loss": 1.0518, + "step": 3127 + }, + { + "epoch": 0.23508191793176011, + "grad_norm": 1.4935244816962818, + "learning_rate": 3.5750018856270153e-06, + "loss": 1.0388, + "step": 3128 + }, + { + "epoch": 0.23515707199759506, + "grad_norm": 2.0856943738886464, + "learning_rate": 3.5747017915790776e-06, + "loss": 1.0783, + "step": 3129 + }, + { + "epoch": 0.23523222606343003, + "grad_norm": 1.6980447155307108, + "learning_rate": 3.5744016042240287e-06, + "loss": 0.9992, + "step": 3130 + }, + { + "epoch": 0.235307380129265, + "grad_norm": 1.7690633193005563, + "learning_rate": 3.574101323579656e-06, + "loss": 1.0435, + "step": 3131 + }, + { + "epoch": 0.23538253419509997, + "grad_norm": 2.21522424514276, + "learning_rate": 3.5738009496637523e-06, + "loss": 1.0282, + "step": 3132 + }, + { + "epoch": 0.2354576882609349, + "grad_norm": 5.0286376995203295, + "learning_rate": 3.573500482494116e-06, + "loss": 1.0291, + "step": 3133 + }, + { + "epoch": 0.23553284232676988, + "grad_norm": 2.3538807945761535, + "learning_rate": 3.573199922088551e-06, + "loss": 0.999, + "step": 3134 + }, + { + "epoch": 0.23560799639260485, + "grad_norm": 1.5857578420118872, + "learning_rate": 3.5728992684648657e-06, + "loss": 0.938, + "step": 3135 + }, + { + "epoch": 0.2356831504584398, + "grad_norm": 1.8892313336697946, + "learning_rate": 3.572598521640876e-06, + "loss": 0.9941, + "step": 3136 + }, + { + "epoch": 0.23575830452427476, + "grad_norm": 2.005909058309354, + "learning_rate": 3.572297681634402e-06, + "loss": 1.0236, + "step": 3137 + }, + { + "epoch": 0.23583345859010973, + "grad_norm": 1.5511416537727847, + "learning_rate": 3.57199674846327e-06, + "loss": 1.0301, + "step": 3138 + }, + { + "epoch": 0.23590861265594468, + "grad_norm": 2.5772520494755415, + "learning_rate": 3.5716957221453106e-06, + "loss": 0.9714, + "step": 3139 + }, + { + "epoch": 0.23598376672177965, + "grad_norm": 2.7477289705954644, + "learning_rate": 3.571394602698362e-06, + "loss": 1.0134, + "step": 3140 + }, + { + "epoch": 0.23605892078761462, + "grad_norm": 1.5617606893521376, + "learning_rate": 3.5710933901402652e-06, + "loss": 1.0357, + "step": 3141 + }, + { + "epoch": 0.23613407485344956, + "grad_norm": 2.3292402630019313, + "learning_rate": 3.570792084488869e-06, + "loss": 1.0032, + "step": 3142 + }, + { + "epoch": 0.23620922891928453, + "grad_norm": 2.7437248593788315, + "learning_rate": 3.5704906857620268e-06, + "loss": 1.0207, + "step": 3143 + }, + { + "epoch": 0.2362843829851195, + "grad_norm": 1.384373704896756, + "learning_rate": 3.5701891939775974e-06, + "loss": 1.0225, + "step": 3144 + }, + { + "epoch": 0.23635953705095444, + "grad_norm": 1.3901409635489064, + "learning_rate": 3.5698876091534465e-06, + "loss": 0.9742, + "step": 3145 + }, + { + "epoch": 0.2364346911167894, + "grad_norm": 1.6854709152840077, + "learning_rate": 3.5695859313074425e-06, + "loss": 1.0245, + "step": 3146 + }, + { + "epoch": 0.23650984518262438, + "grad_norm": 1.861513270709666, + "learning_rate": 3.5692841604574617e-06, + "loss": 0.9759, + "step": 3147 + }, + { + "epoch": 0.23658499924845935, + "grad_norm": 1.9712036827901398, + "learning_rate": 3.568982296621386e-06, + "loss": 0.9978, + "step": 3148 + }, + { + "epoch": 0.2366601533142943, + "grad_norm": 2.1557576819929647, + "learning_rate": 3.5686803398171007e-06, + "loss": 1.0149, + "step": 3149 + }, + { + "epoch": 0.23673530738012927, + "grad_norm": 1.7269473790433632, + "learning_rate": 3.5683782900624986e-06, + "loss": 1.0556, + "step": 3150 + }, + { + "epoch": 0.23681046144596424, + "grad_norm": 1.6024665327936665, + "learning_rate": 3.5680761473754767e-06, + "loss": 0.9712, + "step": 3151 + }, + { + "epoch": 0.23688561551179918, + "grad_norm": 1.5596467478921145, + "learning_rate": 3.5677739117739385e-06, + "loss": 1.0125, + "step": 3152 + }, + { + "epoch": 0.23696076957763415, + "grad_norm": 2.077760363260458, + "learning_rate": 3.5674715832757927e-06, + "loss": 1.0294, + "step": 3153 + }, + { + "epoch": 0.23703592364346912, + "grad_norm": 2.4125434515800372, + "learning_rate": 3.5671691618989533e-06, + "loss": 1.0409, + "step": 3154 + }, + { + "epoch": 0.23711107770930406, + "grad_norm": 1.8870554874269527, + "learning_rate": 3.56686664766134e-06, + "loss": 1.055, + "step": 3155 + }, + { + "epoch": 0.23718623177513903, + "grad_norm": 2.1967744676749623, + "learning_rate": 3.5665640405808785e-06, + "loss": 0.9821, + "step": 3156 + }, + { + "epoch": 0.237261385840974, + "grad_norm": 1.585569968646298, + "learning_rate": 3.566261340675498e-06, + "loss": 1.0366, + "step": 3157 + }, + { + "epoch": 0.23733653990680895, + "grad_norm": 1.622681136989618, + "learning_rate": 3.5659585479631357e-06, + "loss": 1.0006, + "step": 3158 + }, + { + "epoch": 0.23741169397264392, + "grad_norm": 1.6902375798010985, + "learning_rate": 3.565655662461733e-06, + "loss": 1.1107, + "step": 3159 + }, + { + "epoch": 0.2374868480384789, + "grad_norm": 1.8530209748531854, + "learning_rate": 3.565352684189237e-06, + "loss": 0.9773, + "step": 3160 + }, + { + "epoch": 0.23756200210431383, + "grad_norm": 2.0122681099115476, + "learning_rate": 3.5650496131636006e-06, + "loss": 1.0341, + "step": 3161 + }, + { + "epoch": 0.2376371561701488, + "grad_norm": 1.2199526013093822, + "learning_rate": 3.564746449402781e-06, + "loss": 0.9484, + "step": 3162 + }, + { + "epoch": 0.23771231023598377, + "grad_norm": 1.9434965919089853, + "learning_rate": 3.5644431929247432e-06, + "loss": 1.0403, + "step": 3163 + }, + { + "epoch": 0.23778746430181874, + "grad_norm": 1.5511430371192056, + "learning_rate": 3.5641398437474546e-06, + "loss": 1.0926, + "step": 3164 + }, + { + "epoch": 0.23786261836765368, + "grad_norm": 2.315544186725071, + "learning_rate": 3.563836401888892e-06, + "loss": 0.9013, + "step": 3165 + }, + { + "epoch": 0.23793777243348865, + "grad_norm": 1.8438621745495896, + "learning_rate": 3.5635328673670335e-06, + "loss": 1.0043, + "step": 3166 + }, + { + "epoch": 0.23801292649932362, + "grad_norm": 1.5922927177785422, + "learning_rate": 3.5632292401998657e-06, + "loss": 1.0148, + "step": 3167 + }, + { + "epoch": 0.23808808056515857, + "grad_norm": 0.6092235425871632, + "learning_rate": 3.562925520405379e-06, + "loss": 0.8311, + "step": 3168 + }, + { + "epoch": 0.23816323463099354, + "grad_norm": 2.097883443054764, + "learning_rate": 3.562621708001571e-06, + "loss": 1.0985, + "step": 3169 + }, + { + "epoch": 0.2382383886968285, + "grad_norm": 1.5159303809765312, + "learning_rate": 3.5623178030064426e-06, + "loss": 0.9846, + "step": 3170 + }, + { + "epoch": 0.23831354276266345, + "grad_norm": 1.4942719445457684, + "learning_rate": 3.562013805438002e-06, + "loss": 1.0745, + "step": 3171 + }, + { + "epoch": 0.23838869682849842, + "grad_norm": 1.8091639367634569, + "learning_rate": 3.5617097153142623e-06, + "loss": 1.0613, + "step": 3172 + }, + { + "epoch": 0.2384638508943334, + "grad_norm": 1.6517994662857534, + "learning_rate": 3.5614055326532416e-06, + "loss": 0.9396, + "step": 3173 + }, + { + "epoch": 0.23853900496016833, + "grad_norm": 2.1071567881310695, + "learning_rate": 3.561101257472964e-06, + "loss": 0.9986, + "step": 3174 + }, + { + "epoch": 0.2386141590260033, + "grad_norm": 4.3865656974635705, + "learning_rate": 3.560796889791459e-06, + "loss": 1.0958, + "step": 3175 + }, + { + "epoch": 0.23868931309183827, + "grad_norm": 1.8283258311495336, + "learning_rate": 3.5604924296267616e-06, + "loss": 1.039, + "step": 3176 + }, + { + "epoch": 0.23876446715767324, + "grad_norm": 2.065376818546168, + "learning_rate": 3.5601878769969123e-06, + "loss": 1.101, + "step": 3177 + }, + { + "epoch": 0.2388396212235082, + "grad_norm": 2.0895448274826673, + "learning_rate": 3.559883231919957e-06, + "loss": 0.959, + "step": 3178 + }, + { + "epoch": 0.23891477528934316, + "grad_norm": 1.8036929606237346, + "learning_rate": 3.559578494413947e-06, + "loss": 1.0538, + "step": 3179 + }, + { + "epoch": 0.23898992935517813, + "grad_norm": 0.7986277956302401, + "learning_rate": 3.559273664496939e-06, + "loss": 0.8496, + "step": 3180 + }, + { + "epoch": 0.23906508342101307, + "grad_norm": 1.503454997742472, + "learning_rate": 3.5589687421869957e-06, + "loss": 0.8437, + "step": 3181 + }, + { + "epoch": 0.23914023748684804, + "grad_norm": 1.8045948880034959, + "learning_rate": 3.558663727502185e-06, + "loss": 0.9759, + "step": 3182 + }, + { + "epoch": 0.239215391552683, + "grad_norm": 1.6574922617275676, + "learning_rate": 3.5583586204605796e-06, + "loss": 1.0176, + "step": 3183 + }, + { + "epoch": 0.23929054561851795, + "grad_norm": 2.8491568673445005, + "learning_rate": 3.5580534210802587e-06, + "loss": 1.0073, + "step": 3184 + }, + { + "epoch": 0.23936569968435292, + "grad_norm": 1.7084866741800564, + "learning_rate": 3.5577481293793063e-06, + "loss": 1.0333, + "step": 3185 + }, + { + "epoch": 0.2394408537501879, + "grad_norm": 1.624744541922689, + "learning_rate": 3.5574427453758124e-06, + "loss": 0.9336, + "step": 3186 + }, + { + "epoch": 0.23951600781602284, + "grad_norm": 2.644342187521466, + "learning_rate": 3.557137269087872e-06, + "loss": 1.0293, + "step": 3187 + }, + { + "epoch": 0.2395911618818578, + "grad_norm": 1.5800898944053328, + "learning_rate": 3.5568317005335852e-06, + "loss": 1.0725, + "step": 3188 + }, + { + "epoch": 0.23966631594769278, + "grad_norm": 1.8100533249940178, + "learning_rate": 3.556526039731059e-06, + "loss": 0.9214, + "step": 3189 + }, + { + "epoch": 0.23974147001352772, + "grad_norm": 1.7619065144492114, + "learning_rate": 3.5562202866984045e-06, + "loss": 0.9692, + "step": 3190 + }, + { + "epoch": 0.2398166240793627, + "grad_norm": 1.6536920335778804, + "learning_rate": 3.555914441453739e-06, + "loss": 1.0195, + "step": 3191 + }, + { + "epoch": 0.23989177814519766, + "grad_norm": 2.543163942501533, + "learning_rate": 3.555608504015185e-06, + "loss": 0.9865, + "step": 3192 + }, + { + "epoch": 0.23996693221103263, + "grad_norm": 1.7575088577806977, + "learning_rate": 3.5553024744008697e-06, + "loss": 1.0006, + "step": 3193 + }, + { + "epoch": 0.24004208627686757, + "grad_norm": 2.0212603901045787, + "learning_rate": 3.5549963526289276e-06, + "loss": 0.9372, + "step": 3194 + }, + { + "epoch": 0.24011724034270254, + "grad_norm": 1.7924648399575738, + "learning_rate": 3.5546901387174975e-06, + "loss": 1.1093, + "step": 3195 + }, + { + "epoch": 0.2401923944085375, + "grad_norm": 1.9329056565479, + "learning_rate": 3.554383832684723e-06, + "loss": 1.0667, + "step": 3196 + }, + { + "epoch": 0.24026754847437246, + "grad_norm": 1.5619850072934898, + "learning_rate": 3.554077434548754e-06, + "loss": 0.9854, + "step": 3197 + }, + { + "epoch": 0.24034270254020743, + "grad_norm": 1.9759942244279294, + "learning_rate": 3.5537709443277465e-06, + "loss": 0.8965, + "step": 3198 + }, + { + "epoch": 0.2404178566060424, + "grad_norm": 1.5749896034018684, + "learning_rate": 3.55346436203986e-06, + "loss": 1.0724, + "step": 3199 + }, + { + "epoch": 0.24049301067187734, + "grad_norm": 1.8489563704645904, + "learning_rate": 3.5531576877032627e-06, + "loss": 1.057, + "step": 3200 + }, + { + "epoch": 0.2405681647377123, + "grad_norm": 2.786170977930414, + "learning_rate": 3.552850921336124e-06, + "loss": 0.9113, + "step": 3201 + }, + { + "epoch": 0.24064331880354728, + "grad_norm": 2.0900196607477195, + "learning_rate": 3.5525440629566223e-06, + "loss": 1.0069, + "step": 3202 + }, + { + "epoch": 0.24071847286938222, + "grad_norm": 8.157482770222956, + "learning_rate": 3.5522371125829395e-06, + "loss": 1.0979, + "step": 3203 + }, + { + "epoch": 0.2407936269352172, + "grad_norm": 0.8256323746056587, + "learning_rate": 3.551930070233264e-06, + "loss": 0.8994, + "step": 3204 + }, + { + "epoch": 0.24086878100105216, + "grad_norm": 1.6777784987876865, + "learning_rate": 3.551622935925789e-06, + "loss": 0.9675, + "step": 3205 + }, + { + "epoch": 0.2409439350668871, + "grad_norm": 4.5245290437378065, + "learning_rate": 3.5513157096787143e-06, + "loss": 0.9316, + "step": 3206 + }, + { + "epoch": 0.24101908913272208, + "grad_norm": 1.2789520031779145, + "learning_rate": 3.551008391510242e-06, + "loss": 0.9609, + "step": 3207 + }, + { + "epoch": 0.24109424319855705, + "grad_norm": 4.711288514985634, + "learning_rate": 3.5507009814385846e-06, + "loss": 1.0715, + "step": 3208 + }, + { + "epoch": 0.24116939726439202, + "grad_norm": 1.669473081368874, + "learning_rate": 3.550393479481955e-06, + "loss": 1.0437, + "step": 3209 + }, + { + "epoch": 0.24124455133022696, + "grad_norm": 1.6468663977266218, + "learning_rate": 3.550085885658576e-06, + "loss": 0.9822, + "step": 3210 + }, + { + "epoch": 0.24131970539606193, + "grad_norm": 1.791384992710432, + "learning_rate": 3.5497781999866715e-06, + "loss": 0.9971, + "step": 3211 + }, + { + "epoch": 0.2413948594618969, + "grad_norm": 1.4898173897486104, + "learning_rate": 3.5494704224844746e-06, + "loss": 0.8941, + "step": 3212 + }, + { + "epoch": 0.24147001352773184, + "grad_norm": 2.83091556013781, + "learning_rate": 3.549162553170222e-06, + "loss": 0.9272, + "step": 3213 + }, + { + "epoch": 0.2415451675935668, + "grad_norm": 1.6117608616366634, + "learning_rate": 3.548854592062156e-06, + "loss": 0.961, + "step": 3214 + }, + { + "epoch": 0.24162032165940178, + "grad_norm": 1.2831994090033714, + "learning_rate": 3.548546539178524e-06, + "loss": 1.0385, + "step": 3215 + }, + { + "epoch": 0.24169547572523672, + "grad_norm": 1.5597096514098545, + "learning_rate": 3.548238394537581e-06, + "loss": 1.0436, + "step": 3216 + }, + { + "epoch": 0.2417706297910717, + "grad_norm": 3.7722198859613143, + "learning_rate": 3.5479301581575827e-06, + "loss": 1.0254, + "step": 3217 + }, + { + "epoch": 0.24184578385690667, + "grad_norm": 1.6949933714160321, + "learning_rate": 3.547621830056796e-06, + "loss": 1.0774, + "step": 3218 + }, + { + "epoch": 0.2419209379227416, + "grad_norm": 1.4947535635138012, + "learning_rate": 3.5473134102534895e-06, + "loss": 0.9937, + "step": 3219 + }, + { + "epoch": 0.24199609198857658, + "grad_norm": 2.1287226886531667, + "learning_rate": 3.5470048987659387e-06, + "loss": 0.9537, + "step": 3220 + }, + { + "epoch": 0.24207124605441155, + "grad_norm": 1.636574240338832, + "learning_rate": 3.5466962956124235e-06, + "loss": 0.977, + "step": 3221 + }, + { + "epoch": 0.24214640012024652, + "grad_norm": 2.3874602469285353, + "learning_rate": 3.54638760081123e-06, + "loss": 1.0095, + "step": 3222 + }, + { + "epoch": 0.24222155418608146, + "grad_norm": 1.7168288157402742, + "learning_rate": 3.5460788143806505e-06, + "loss": 1.0429, + "step": 3223 + }, + { + "epoch": 0.24229670825191643, + "grad_norm": 2.7009730069316844, + "learning_rate": 3.54576993633898e-06, + "loss": 1.0182, + "step": 3224 + }, + { + "epoch": 0.2423718623177514, + "grad_norm": 1.583412979866649, + "learning_rate": 3.545460966704522e-06, + "loss": 0.9983, + "step": 3225 + }, + { + "epoch": 0.24244701638358634, + "grad_norm": 1.5300458727922215, + "learning_rate": 3.5451519054955836e-06, + "loss": 1.006, + "step": 3226 + }, + { + "epoch": 0.24252217044942131, + "grad_norm": 2.055091615105826, + "learning_rate": 3.544842752730478e-06, + "loss": 1.0591, + "step": 3227 + }, + { + "epoch": 0.24259732451525629, + "grad_norm": 1.5563535724782842, + "learning_rate": 3.5445335084275235e-06, + "loss": 0.8987, + "step": 3228 + }, + { + "epoch": 0.24267247858109123, + "grad_norm": 1.8497992380855615, + "learning_rate": 3.5442241726050444e-06, + "loss": 0.9926, + "step": 3229 + }, + { + "epoch": 0.2427476326469262, + "grad_norm": 2.0047842262582174, + "learning_rate": 3.5439147452813696e-06, + "loss": 1.011, + "step": 3230 + }, + { + "epoch": 0.24282278671276117, + "grad_norm": 1.7754787847035094, + "learning_rate": 3.5436052264748348e-06, + "loss": 0.9171, + "step": 3231 + }, + { + "epoch": 0.2428979407785961, + "grad_norm": 1.9744162988645337, + "learning_rate": 3.543295616203779e-06, + "loss": 1.0464, + "step": 3232 + }, + { + "epoch": 0.24297309484443108, + "grad_norm": 1.5650374598788281, + "learning_rate": 3.5429859144865486e-06, + "loss": 0.994, + "step": 3233 + }, + { + "epoch": 0.24304824891026605, + "grad_norm": 3.0894966557477392, + "learning_rate": 3.542676121341494e-06, + "loss": 0.8425, + "step": 3234 + }, + { + "epoch": 0.243123402976101, + "grad_norm": 1.5189430647847262, + "learning_rate": 3.5423662367869716e-06, + "loss": 0.95, + "step": 3235 + }, + { + "epoch": 0.24319855704193596, + "grad_norm": 2.4113483664786064, + "learning_rate": 3.542056260841344e-06, + "loss": 0.9319, + "step": 3236 + }, + { + "epoch": 0.24327371110777093, + "grad_norm": 1.6044874324716516, + "learning_rate": 3.5417461935229777e-06, + "loss": 0.945, + "step": 3237 + }, + { + "epoch": 0.2433488651736059, + "grad_norm": 1.4267792853502654, + "learning_rate": 3.5414360348502463e-06, + "loss": 0.8534, + "step": 3238 + }, + { + "epoch": 0.24342401923944085, + "grad_norm": 1.7291999798365496, + "learning_rate": 3.5411257848415266e-06, + "loss": 0.9513, + "step": 3239 + }, + { + "epoch": 0.24349917330527582, + "grad_norm": 1.748211423492204, + "learning_rate": 3.5408154435152034e-06, + "loss": 1.0189, + "step": 3240 + }, + { + "epoch": 0.2435743273711108, + "grad_norm": 2.180288252540535, + "learning_rate": 3.5405050108896645e-06, + "loss": 0.8636, + "step": 3241 + }, + { + "epoch": 0.24364948143694573, + "grad_norm": 2.2323278906288153, + "learning_rate": 3.5401944869833046e-06, + "loss": 0.9479, + "step": 3242 + }, + { + "epoch": 0.2437246355027807, + "grad_norm": 1.840795898275601, + "learning_rate": 3.539883871814524e-06, + "loss": 1.0537, + "step": 3243 + }, + { + "epoch": 0.24379978956861567, + "grad_norm": 2.319057240656883, + "learning_rate": 3.5395731654017277e-06, + "loss": 1.0098, + "step": 3244 + }, + { + "epoch": 0.2438749436344506, + "grad_norm": 16.20827067672792, + "learning_rate": 3.539262367763325e-06, + "loss": 0.9554, + "step": 3245 + }, + { + "epoch": 0.24395009770028558, + "grad_norm": 1.3350830915216811, + "learning_rate": 3.5389514789177334e-06, + "loss": 0.9713, + "step": 3246 + }, + { + "epoch": 0.24402525176612055, + "grad_norm": 1.7064720652071266, + "learning_rate": 3.5386404988833732e-06, + "loss": 1.0318, + "step": 3247 + }, + { + "epoch": 0.2441004058319555, + "grad_norm": 12.823114529448436, + "learning_rate": 3.538329427678672e-06, + "loss": 0.9483, + "step": 3248 + }, + { + "epoch": 0.24417555989779047, + "grad_norm": 0.8947590034097963, + "learning_rate": 3.5380182653220613e-06, + "loss": 0.8584, + "step": 3249 + }, + { + "epoch": 0.24425071396362544, + "grad_norm": 1.7899150566106738, + "learning_rate": 3.5377070118319788e-06, + "loss": 0.9824, + "step": 3250 + }, + { + "epoch": 0.24432586802946038, + "grad_norm": 2.481296671469271, + "learning_rate": 3.5373956672268683e-06, + "loss": 1.0206, + "step": 3251 + }, + { + "epoch": 0.24440102209529535, + "grad_norm": 1.7946597832570887, + "learning_rate": 3.5370842315251766e-06, + "loss": 1.0569, + "step": 3252 + }, + { + "epoch": 0.24447617616113032, + "grad_norm": 2.151948037438395, + "learning_rate": 3.5367727047453583e-06, + "loss": 1.0704, + "step": 3253 + }, + { + "epoch": 0.2445513302269653, + "grad_norm": 1.847508624189635, + "learning_rate": 3.536461086905873e-06, + "loss": 1.1121, + "step": 3254 + }, + { + "epoch": 0.24462648429280023, + "grad_norm": 2.1612249286669027, + "learning_rate": 3.536149378025185e-06, + "loss": 0.9801, + "step": 3255 + }, + { + "epoch": 0.2447016383586352, + "grad_norm": 0.7370939446205338, + "learning_rate": 3.5358375781217634e-06, + "loss": 0.8739, + "step": 3256 + }, + { + "epoch": 0.24477679242447017, + "grad_norm": 1.4424266149937521, + "learning_rate": 3.5355256872140846e-06, + "loss": 0.9848, + "step": 3257 + }, + { + "epoch": 0.24485194649030512, + "grad_norm": 4.214187528560981, + "learning_rate": 3.535213705320629e-06, + "loss": 0.8999, + "step": 3258 + }, + { + "epoch": 0.2449271005561401, + "grad_norm": 1.6750759249887537, + "learning_rate": 3.534901632459882e-06, + "loss": 0.949, + "step": 3259 + }, + { + "epoch": 0.24500225462197506, + "grad_norm": 3.1344163551530886, + "learning_rate": 3.5345894686503366e-06, + "loss": 0.9221, + "step": 3260 + }, + { + "epoch": 0.24507740868781, + "grad_norm": 1.7787250223771838, + "learning_rate": 3.5342772139104884e-06, + "loss": 1.0481, + "step": 3261 + }, + { + "epoch": 0.24515256275364497, + "grad_norm": 1.5983496469348661, + "learning_rate": 3.5339648682588397e-06, + "loss": 0.913, + "step": 3262 + }, + { + "epoch": 0.24522771681947994, + "grad_norm": 2.284400541107963, + "learning_rate": 3.5336524317138993e-06, + "loss": 0.9636, + "step": 3263 + }, + { + "epoch": 0.24530287088531488, + "grad_norm": 1.9626297658793543, + "learning_rate": 3.5333399042941797e-06, + "loss": 1.0004, + "step": 3264 + }, + { + "epoch": 0.24537802495114985, + "grad_norm": 1.6786698918320446, + "learning_rate": 3.5330272860181985e-06, + "loss": 0.9645, + "step": 3265 + }, + { + "epoch": 0.24545317901698482, + "grad_norm": 1.4146707817692519, + "learning_rate": 3.532714576904481e-06, + "loss": 0.9861, + "step": 3266 + }, + { + "epoch": 0.2455283330828198, + "grad_norm": 0.7309701596025867, + "learning_rate": 3.5324017769715548e-06, + "loss": 0.8925, + "step": 3267 + }, + { + "epoch": 0.24560348714865474, + "grad_norm": 1.690285890877209, + "learning_rate": 3.532088886237956e-06, + "loss": 0.8783, + "step": 3268 + }, + { + "epoch": 0.2456786412144897, + "grad_norm": 2.2311788253758738, + "learning_rate": 3.5317759047222235e-06, + "loss": 0.9977, + "step": 3269 + }, + { + "epoch": 0.24575379528032468, + "grad_norm": 1.4239395579412069, + "learning_rate": 3.531462832442903e-06, + "loss": 1.0345, + "step": 3270 + }, + { + "epoch": 0.24582894934615962, + "grad_norm": 3.246212145761285, + "learning_rate": 3.531149669418546e-06, + "loss": 0.9661, + "step": 3271 + }, + { + "epoch": 0.2459041034119946, + "grad_norm": 2.295184804585663, + "learning_rate": 3.530836415667708e-06, + "loss": 0.8819, + "step": 3272 + }, + { + "epoch": 0.24597925747782956, + "grad_norm": 2.5224678373223965, + "learning_rate": 3.53052307120895e-06, + "loss": 1.0947, + "step": 3273 + }, + { + "epoch": 0.2460544115436645, + "grad_norm": 2.166763107892198, + "learning_rate": 3.5302096360608385e-06, + "loss": 1.0083, + "step": 3274 + }, + { + "epoch": 0.24612956560949947, + "grad_norm": 2.2263647158933884, + "learning_rate": 3.5298961102419477e-06, + "loss": 0.9401, + "step": 3275 + }, + { + "epoch": 0.24620471967533444, + "grad_norm": 1.5308672368187701, + "learning_rate": 3.5295824937708537e-06, + "loss": 1.0649, + "step": 3276 + }, + { + "epoch": 0.2462798737411694, + "grad_norm": 1.9756439849124765, + "learning_rate": 3.5292687866661396e-06, + "loss": 1.1082, + "step": 3277 + }, + { + "epoch": 0.24635502780700436, + "grad_norm": 4.681669551042262, + "learning_rate": 3.528954988946394e-06, + "loss": 0.9151, + "step": 3278 + }, + { + "epoch": 0.24643018187283933, + "grad_norm": 1.6681472003898754, + "learning_rate": 3.5286411006302107e-06, + "loss": 1.0005, + "step": 3279 + }, + { + "epoch": 0.24650533593867427, + "grad_norm": 1.6176617060727467, + "learning_rate": 3.528327121736188e-06, + "loss": 0.969, + "step": 3280 + }, + { + "epoch": 0.24658049000450924, + "grad_norm": 1.6167503171430577, + "learning_rate": 3.5280130522829317e-06, + "loss": 0.9153, + "step": 3281 + }, + { + "epoch": 0.2466556440703442, + "grad_norm": 1.8016405020880382, + "learning_rate": 3.5276988922890503e-06, + "loss": 0.9584, + "step": 3282 + }, + { + "epoch": 0.24673079813617918, + "grad_norm": 2.3419916296271017, + "learning_rate": 3.52738464177316e-06, + "loss": 1.0355, + "step": 3283 + }, + { + "epoch": 0.24680595220201412, + "grad_norm": 1.7894226111587765, + "learning_rate": 3.527070300753881e-06, + "loss": 1.0385, + "step": 3284 + }, + { + "epoch": 0.2468811062678491, + "grad_norm": 1.8213631540727049, + "learning_rate": 3.526755869249839e-06, + "loss": 0.9622, + "step": 3285 + }, + { + "epoch": 0.24695626033368406, + "grad_norm": 1.803238191803677, + "learning_rate": 3.5264413472796653e-06, + "loss": 0.9087, + "step": 3286 + }, + { + "epoch": 0.247031414399519, + "grad_norm": 13.542714114638185, + "learning_rate": 3.5261267348619964e-06, + "loss": 1.0311, + "step": 3287 + }, + { + "epoch": 0.24710656846535398, + "grad_norm": 2.7650733871792292, + "learning_rate": 3.5258120320154755e-06, + "loss": 0.9719, + "step": 3288 + }, + { + "epoch": 0.24718172253118895, + "grad_norm": 2.09932183714249, + "learning_rate": 3.5254972387587483e-06, + "loss": 0.9144, + "step": 3289 + }, + { + "epoch": 0.2472568765970239, + "grad_norm": 4.3522328488804485, + "learning_rate": 3.525182355110468e-06, + "loss": 0.9271, + "step": 3290 + }, + { + "epoch": 0.24733203066285886, + "grad_norm": 2.7558322961938146, + "learning_rate": 3.524867381089293e-06, + "loss": 0.9623, + "step": 3291 + }, + { + "epoch": 0.24740718472869383, + "grad_norm": 1.7572874514249013, + "learning_rate": 3.524552316713887e-06, + "loss": 1.0007, + "step": 3292 + }, + { + "epoch": 0.24748233879452877, + "grad_norm": 1.5980236125438068, + "learning_rate": 3.5242371620029176e-06, + "loss": 1.0016, + "step": 3293 + }, + { + "epoch": 0.24755749286036374, + "grad_norm": 2.0516657783677847, + "learning_rate": 3.5239219169750604e-06, + "loss": 0.9927, + "step": 3294 + }, + { + "epoch": 0.2476326469261987, + "grad_norm": 0.6650494814153224, + "learning_rate": 3.5236065816489938e-06, + "loss": 0.7684, + "step": 3295 + }, + { + "epoch": 0.24770780099203366, + "grad_norm": 0.8081506445666594, + "learning_rate": 3.5232911560434023e-06, + "loss": 0.8313, + "step": 3296 + }, + { + "epoch": 0.24778295505786863, + "grad_norm": 2.285939139563076, + "learning_rate": 3.5229756401769775e-06, + "loss": 1.0674, + "step": 3297 + }, + { + "epoch": 0.2478581091237036, + "grad_norm": 1.942693088120547, + "learning_rate": 3.522660034068414e-06, + "loss": 0.9318, + "step": 3298 + }, + { + "epoch": 0.24793326318953857, + "grad_norm": 1.742567422949827, + "learning_rate": 3.5223443377364133e-06, + "loss": 0.9855, + "step": 3299 + }, + { + "epoch": 0.2480084172553735, + "grad_norm": 2.607688770135982, + "learning_rate": 3.5220285511996802e-06, + "loss": 0.9534, + "step": 3300 + }, + { + "epoch": 0.24808357132120848, + "grad_norm": 2.253159953149154, + "learning_rate": 3.521712674476928e-06, + "loss": 1.0245, + "step": 3301 + }, + { + "epoch": 0.24815872538704345, + "grad_norm": 0.7392499915733505, + "learning_rate": 3.521396707586872e-06, + "loss": 0.8003, + "step": 3302 + }, + { + "epoch": 0.2482338794528784, + "grad_norm": 0.7940868894276089, + "learning_rate": 3.521080650548236e-06, + "loss": 0.8273, + "step": 3303 + }, + { + "epoch": 0.24830903351871336, + "grad_norm": 1.5800871783964985, + "learning_rate": 3.5207645033797464e-06, + "loss": 1.0158, + "step": 3304 + }, + { + "epoch": 0.24838418758454833, + "grad_norm": 1.5868707175570769, + "learning_rate": 3.5204482661001373e-06, + "loss": 0.9527, + "step": 3305 + }, + { + "epoch": 0.24845934165038328, + "grad_norm": 3.454417267387737, + "learning_rate": 3.5201319387281455e-06, + "loss": 0.9585, + "step": 3306 + }, + { + "epoch": 0.24853449571621825, + "grad_norm": 1.5057138652237356, + "learning_rate": 3.519815521282515e-06, + "loss": 1.0415, + "step": 3307 + }, + { + "epoch": 0.24860964978205322, + "grad_norm": 1.5138437874907023, + "learning_rate": 3.519499013781996e-06, + "loss": 1.0478, + "step": 3308 + }, + { + "epoch": 0.24868480384788816, + "grad_norm": 1.5381386399700916, + "learning_rate": 3.5191824162453417e-06, + "loss": 1.0589, + "step": 3309 + }, + { + "epoch": 0.24875995791372313, + "grad_norm": 4.1785757136992485, + "learning_rate": 3.5188657286913115e-06, + "loss": 1.0103, + "step": 3310 + }, + { + "epoch": 0.2488351119795581, + "grad_norm": 3.1930695288471633, + "learning_rate": 3.5185489511386712e-06, + "loss": 1.0064, + "step": 3311 + }, + { + "epoch": 0.24891026604539307, + "grad_norm": 6.184980708025093, + "learning_rate": 3.5182320836061906e-06, + "loss": 1.0032, + "step": 3312 + }, + { + "epoch": 0.248985420111228, + "grad_norm": 1.6779814816840966, + "learning_rate": 3.517915126112645e-06, + "loss": 0.9408, + "step": 3313 + }, + { + "epoch": 0.24906057417706298, + "grad_norm": 2.1488580465530616, + "learning_rate": 3.517598078676816e-06, + "loss": 0.9568, + "step": 3314 + }, + { + "epoch": 0.24913572824289795, + "grad_norm": 1.810866441887757, + "learning_rate": 3.517280941317489e-06, + "loss": 1.0195, + "step": 3315 + }, + { + "epoch": 0.2492108823087329, + "grad_norm": 3.753354289017262, + "learning_rate": 3.5169637140534573e-06, + "loss": 1.0706, + "step": 3316 + }, + { + "epoch": 0.24928603637456787, + "grad_norm": 1.8284129380613976, + "learning_rate": 3.5166463969035157e-06, + "loss": 0.8607, + "step": 3317 + }, + { + "epoch": 0.24936119044040284, + "grad_norm": 1.717998270888785, + "learning_rate": 3.5163289898864675e-06, + "loss": 0.9025, + "step": 3318 + }, + { + "epoch": 0.24943634450623778, + "grad_norm": 4.561756439236216, + "learning_rate": 3.5160114930211203e-06, + "loss": 0.8111, + "step": 3319 + }, + { + "epoch": 0.24951149857207275, + "grad_norm": 1.4605625099718527, + "learning_rate": 3.5156939063262875e-06, + "loss": 0.954, + "step": 3320 + }, + { + "epoch": 0.24958665263790772, + "grad_norm": 1.742525281791135, + "learning_rate": 3.515376229820787e-06, + "loss": 0.981, + "step": 3321 + }, + { + "epoch": 0.24966180670374266, + "grad_norm": 5.048711956170785, + "learning_rate": 3.5150584635234416e-06, + "loss": 0.9398, + "step": 3322 + }, + { + "epoch": 0.24973696076957763, + "grad_norm": 1.9632439290638601, + "learning_rate": 3.5147406074530805e-06, + "loss": 0.9662, + "step": 3323 + }, + { + "epoch": 0.2498121148354126, + "grad_norm": 0.703638842698805, + "learning_rate": 3.5144226616285384e-06, + "loss": 0.8764, + "step": 3324 + }, + { + "epoch": 0.24988726890124754, + "grad_norm": 2.571550992670387, + "learning_rate": 3.5141046260686537e-06, + "loss": 0.984, + "step": 3325 + }, + { + "epoch": 0.24996242296708251, + "grad_norm": 3.9207207550415615, + "learning_rate": 3.5137865007922726e-06, + "loss": 1.0179, + "step": 3326 + }, + { + "epoch": 0.25003757703291746, + "grad_norm": 1.4242301132744881, + "learning_rate": 3.5134682858182448e-06, + "loss": 0.9866, + "step": 3327 + }, + { + "epoch": 0.2501127310987524, + "grad_norm": 0.7489864733153196, + "learning_rate": 3.5131499811654253e-06, + "loss": 0.8493, + "step": 3328 + }, + { + "epoch": 0.2501878851645874, + "grad_norm": 1.50554110358615, + "learning_rate": 3.5128315868526755e-06, + "loss": 0.94, + "step": 3329 + }, + { + "epoch": 0.25026303923042237, + "grad_norm": 1.5545674953439481, + "learning_rate": 3.512513102898861e-06, + "loss": 0.9503, + "step": 3330 + }, + { + "epoch": 0.25033819329625734, + "grad_norm": 3.28538311006047, + "learning_rate": 3.512194529322853e-06, + "loss": 0.9147, + "step": 3331 + }, + { + "epoch": 0.2504133473620923, + "grad_norm": 1.9698011226282515, + "learning_rate": 3.511875866143529e-06, + "loss": 1.0106, + "step": 3332 + }, + { + "epoch": 0.2504885014279272, + "grad_norm": 1.2188483467524935, + "learning_rate": 3.511557113379771e-06, + "loss": 1.0078, + "step": 3333 + }, + { + "epoch": 0.2505636554937622, + "grad_norm": 2.0216610448915833, + "learning_rate": 3.511238271050465e-06, + "loss": 0.9904, + "step": 3334 + }, + { + "epoch": 0.25063880955959716, + "grad_norm": 4.0779213105411705, + "learning_rate": 3.510919339174505e-06, + "loss": 1.0758, + "step": 3335 + }, + { + "epoch": 0.25071396362543213, + "grad_norm": 1.7611574348724461, + "learning_rate": 3.5106003177707882e-06, + "loss": 0.9562, + "step": 3336 + }, + { + "epoch": 0.2507891176912671, + "grad_norm": 1.6392888032021555, + "learning_rate": 3.5102812068582183e-06, + "loss": 0.9875, + "step": 3337 + }, + { + "epoch": 0.2508642717571021, + "grad_norm": 1.8028095597514118, + "learning_rate": 3.509962006455704e-06, + "loss": 1.0636, + "step": 3338 + }, + { + "epoch": 0.25093942582293705, + "grad_norm": 1.8415004778946054, + "learning_rate": 3.5096427165821583e-06, + "loss": 1.0856, + "step": 3339 + }, + { + "epoch": 0.25101457988877196, + "grad_norm": 1.3357857065799132, + "learning_rate": 3.509323337256501e-06, + "loss": 0.9501, + "step": 3340 + }, + { + "epoch": 0.25108973395460693, + "grad_norm": 0.7911537709108969, + "learning_rate": 3.5090038684976563e-06, + "loss": 0.883, + "step": 3341 + }, + { + "epoch": 0.2511648880204419, + "grad_norm": 2.169373093507765, + "learning_rate": 3.5086843103245542e-06, + "loss": 0.9738, + "step": 3342 + }, + { + "epoch": 0.25124004208627687, + "grad_norm": 1.6608742272139867, + "learning_rate": 3.508364662756129e-06, + "loss": 1.1192, + "step": 3343 + }, + { + "epoch": 0.25131519615211184, + "grad_norm": 2.1760436743883287, + "learning_rate": 3.5080449258113224e-06, + "loss": 0.9826, + "step": 3344 + }, + { + "epoch": 0.2513903502179468, + "grad_norm": 1.682838571500577, + "learning_rate": 3.5077250995090786e-06, + "loss": 1.0194, + "step": 3345 + }, + { + "epoch": 0.2514655042837817, + "grad_norm": 1.6759395924700395, + "learning_rate": 3.5074051838683497e-06, + "loss": 1.0535, + "step": 3346 + }, + { + "epoch": 0.2515406583496167, + "grad_norm": 1.7222002251905446, + "learning_rate": 3.507085178908091e-06, + "loss": 1.0933, + "step": 3347 + }, + { + "epoch": 0.25161581241545167, + "grad_norm": 1.663291788493358, + "learning_rate": 3.506765084647265e-06, + "loss": 1.0423, + "step": 3348 + }, + { + "epoch": 0.25169096648128664, + "grad_norm": 1.728046797626438, + "learning_rate": 3.506444901104837e-06, + "loss": 1.0218, + "step": 3349 + }, + { + "epoch": 0.2517661205471216, + "grad_norm": 1.6508574598623598, + "learning_rate": 3.506124628299781e-06, + "loss": 0.8303, + "step": 3350 + }, + { + "epoch": 0.2518412746129566, + "grad_norm": 6.070376912564466, + "learning_rate": 3.505804266251073e-06, + "loss": 1.0322, + "step": 3351 + }, + { + "epoch": 0.25191642867879155, + "grad_norm": 0.7090673056616575, + "learning_rate": 3.505483814977696e-06, + "loss": 0.8064, + "step": 3352 + }, + { + "epoch": 0.25199158274462646, + "grad_norm": 1.7056135143258142, + "learning_rate": 3.5051632744986384e-06, + "loss": 1.0292, + "step": 3353 + }, + { + "epoch": 0.25206673681046143, + "grad_norm": 1.881110470689, + "learning_rate": 3.5048426448328926e-06, + "loss": 1.0321, + "step": 3354 + }, + { + "epoch": 0.2521418908762964, + "grad_norm": 1.6780042864184368, + "learning_rate": 3.504521925999458e-06, + "loss": 1.0115, + "step": 3355 + }, + { + "epoch": 0.2522170449421314, + "grad_norm": 0.6392426579605388, + "learning_rate": 3.5042011180173386e-06, + "loss": 0.7667, + "step": 3356 + }, + { + "epoch": 0.25229219900796634, + "grad_norm": 3.5000464572548027, + "learning_rate": 3.5038802209055424e-06, + "loss": 0.9909, + "step": 3357 + }, + { + "epoch": 0.2523673530738013, + "grad_norm": 1.8159203976817246, + "learning_rate": 3.5035592346830846e-06, + "loss": 1.0463, + "step": 3358 + }, + { + "epoch": 0.25244250713963623, + "grad_norm": 0.6720323378397357, + "learning_rate": 3.5032381593689843e-06, + "loss": 0.8804, + "step": 3359 + }, + { + "epoch": 0.2525176612054712, + "grad_norm": 1.7031380189170555, + "learning_rate": 3.502916994982267e-06, + "loss": 0.8779, + "step": 3360 + }, + { + "epoch": 0.25259281527130617, + "grad_norm": 2.0730278202587864, + "learning_rate": 3.502595741541963e-06, + "loss": 1.0073, + "step": 3361 + }, + { + "epoch": 0.25266796933714114, + "grad_norm": 3.386134254502346, + "learning_rate": 3.502274399067107e-06, + "loss": 0.9879, + "step": 3362 + }, + { + "epoch": 0.2527431234029761, + "grad_norm": 2.0087184420695943, + "learning_rate": 3.5019529675767403e-06, + "loss": 0.9212, + "step": 3363 + }, + { + "epoch": 0.2528182774688111, + "grad_norm": 1.7209463478028568, + "learning_rate": 3.501631447089909e-06, + "loss": 1.0248, + "step": 3364 + }, + { + "epoch": 0.25289343153464605, + "grad_norm": 1.8176069331555709, + "learning_rate": 3.5013098376256645e-06, + "loss": 1.0836, + "step": 3365 + }, + { + "epoch": 0.25296858560048097, + "grad_norm": 1.3758015463792483, + "learning_rate": 3.5009881392030633e-06, + "loss": 1.0183, + "step": 3366 + }, + { + "epoch": 0.25304373966631594, + "grad_norm": 2.1454710299603605, + "learning_rate": 3.5006663518411666e-06, + "loss": 1.0346, + "step": 3367 + }, + { + "epoch": 0.2531188937321509, + "grad_norm": 2.1098233346969106, + "learning_rate": 3.500344475559043e-06, + "loss": 1.0458, + "step": 3368 + }, + { + "epoch": 0.2531940477979859, + "grad_norm": 1.5938419240733115, + "learning_rate": 3.5000225103757634e-06, + "loss": 1.0584, + "step": 3369 + }, + { + "epoch": 0.25326920186382085, + "grad_norm": 1.747541062468819, + "learning_rate": 3.499700456310406e-06, + "loss": 1.0186, + "step": 3370 + }, + { + "epoch": 0.2533443559296558, + "grad_norm": 1.8541400850369414, + "learning_rate": 3.499378313382054e-06, + "loss": 1.0047, + "step": 3371 + }, + { + "epoch": 0.25341950999549073, + "grad_norm": 0.8348889414660892, + "learning_rate": 3.4990560816097954e-06, + "loss": 0.9159, + "step": 3372 + }, + { + "epoch": 0.2534946640613257, + "grad_norm": 1.6350771152614063, + "learning_rate": 3.4987337610127237e-06, + "loss": 1.0351, + "step": 3373 + }, + { + "epoch": 0.2535698181271607, + "grad_norm": 1.683732312512403, + "learning_rate": 3.498411351609938e-06, + "loss": 1.0638, + "step": 3374 + }, + { + "epoch": 0.25364497219299564, + "grad_norm": 1.5561932503644458, + "learning_rate": 3.4980888534205414e-06, + "loss": 1.033, + "step": 3375 + }, + { + "epoch": 0.2537201262588306, + "grad_norm": 2.247426999002354, + "learning_rate": 3.4977662664636443e-06, + "loss": 0.9533, + "step": 3376 + }, + { + "epoch": 0.2537952803246656, + "grad_norm": 1.8305562472627548, + "learning_rate": 3.4974435907583597e-06, + "loss": 0.9045, + "step": 3377 + }, + { + "epoch": 0.2538704343905005, + "grad_norm": 1.611437022835349, + "learning_rate": 3.497120826323809e-06, + "loss": 1.0186, + "step": 3378 + }, + { + "epoch": 0.25394558845633547, + "grad_norm": 1.892646780432032, + "learning_rate": 3.496797973179116e-06, + "loss": 0.9848, + "step": 3379 + }, + { + "epoch": 0.25402074252217044, + "grad_norm": 1.6145281464364056, + "learning_rate": 3.4964750313434114e-06, + "loss": 1.0616, + "step": 3380 + }, + { + "epoch": 0.2540958965880054, + "grad_norm": 0.6685153219639826, + "learning_rate": 3.496152000835831e-06, + "loss": 0.7496, + "step": 3381 + }, + { + "epoch": 0.2541710506538404, + "grad_norm": 1.4839497810462579, + "learning_rate": 3.495828881675516e-06, + "loss": 0.9745, + "step": 3382 + }, + { + "epoch": 0.25424620471967535, + "grad_norm": 1.5877015023492715, + "learning_rate": 3.4955056738816113e-06, + "loss": 0.9498, + "step": 3383 + }, + { + "epoch": 0.2543213587855103, + "grad_norm": 0.7259605128844095, + "learning_rate": 3.4951823774732686e-06, + "loss": 0.8374, + "step": 3384 + }, + { + "epoch": 0.25439651285134524, + "grad_norm": 1.6500704663575168, + "learning_rate": 3.4948589924696447e-06, + "loss": 0.9784, + "step": 3385 + }, + { + "epoch": 0.2544716669171802, + "grad_norm": 1.8810324584293905, + "learning_rate": 3.4945355188899013e-06, + "loss": 1.0969, + "step": 3386 + }, + { + "epoch": 0.2545468209830152, + "grad_norm": 1.9431068141278878, + "learning_rate": 3.494211956753206e-06, + "loss": 0.9135, + "step": 3387 + }, + { + "epoch": 0.25462197504885015, + "grad_norm": 1.667351995073648, + "learning_rate": 3.49388830607873e-06, + "loss": 1.0348, + "step": 3388 + }, + { + "epoch": 0.2546971291146851, + "grad_norm": 2.3096396824941463, + "learning_rate": 3.493564566885651e-06, + "loss": 1.0198, + "step": 3389 + }, + { + "epoch": 0.2547722831805201, + "grad_norm": 1.5003980267458097, + "learning_rate": 3.4932407391931527e-06, + "loss": 0.9889, + "step": 3390 + }, + { + "epoch": 0.254847437246355, + "grad_norm": 2.0617878002125893, + "learning_rate": 3.4929168230204226e-06, + "loss": 1.0258, + "step": 3391 + }, + { + "epoch": 0.25492259131219, + "grad_norm": 1.6598970098604027, + "learning_rate": 3.4925928183866534e-06, + "loss": 1.0351, + "step": 3392 + }, + { + "epoch": 0.25499774537802494, + "grad_norm": 1.6221447582489295, + "learning_rate": 3.492268725311045e-06, + "loss": 1.0262, + "step": 3393 + }, + { + "epoch": 0.2550728994438599, + "grad_norm": 2.405632038775142, + "learning_rate": 3.4919445438128e-06, + "loss": 1.0734, + "step": 3394 + }, + { + "epoch": 0.2551480535096949, + "grad_norm": 1.3710947933356072, + "learning_rate": 3.491620273911128e-06, + "loss": 0.8662, + "step": 3395 + }, + { + "epoch": 0.25522320757552985, + "grad_norm": 1.3964384058075368, + "learning_rate": 3.491295915625243e-06, + "loss": 1.0759, + "step": 3396 + }, + { + "epoch": 0.2552983616413648, + "grad_norm": 1.963260687869529, + "learning_rate": 3.490971468974364e-06, + "loss": 1.0987, + "step": 3397 + }, + { + "epoch": 0.25537351570719974, + "grad_norm": 1.6222291210686874, + "learning_rate": 3.490646933977716e-06, + "loss": 1.0573, + "step": 3398 + }, + { + "epoch": 0.2554486697730347, + "grad_norm": 1.5672457531098527, + "learning_rate": 3.49032231065453e-06, + "loss": 0.9226, + "step": 3399 + }, + { + "epoch": 0.2555238238388697, + "grad_norm": 2.21937219196852, + "learning_rate": 3.48999759902404e-06, + "loss": 0.977, + "step": 3400 + }, + { + "epoch": 0.25559897790470465, + "grad_norm": 1.4530212201211778, + "learning_rate": 3.4896727991054856e-06, + "loss": 1.0102, + "step": 3401 + }, + { + "epoch": 0.2556741319705396, + "grad_norm": 4.126083000661127, + "learning_rate": 3.4893479109181144e-06, + "loss": 0.8525, + "step": 3402 + }, + { + "epoch": 0.2557492860363746, + "grad_norm": 1.83403546441919, + "learning_rate": 3.489022934481176e-06, + "loss": 0.9813, + "step": 3403 + }, + { + "epoch": 0.2558244401022095, + "grad_norm": 1.842150560159119, + "learning_rate": 3.4886978698139275e-06, + "loss": 1.036, + "step": 3404 + }, + { + "epoch": 0.2558995941680445, + "grad_norm": 1.930651342941472, + "learning_rate": 3.4883727169356293e-06, + "loss": 1.0109, + "step": 3405 + }, + { + "epoch": 0.25597474823387945, + "grad_norm": 8.946780208467894, + "learning_rate": 3.4880474758655485e-06, + "loss": 1.0616, + "step": 3406 + }, + { + "epoch": 0.2560499022997144, + "grad_norm": 0.7958597934445069, + "learning_rate": 3.487722146622956e-06, + "loss": 0.9297, + "step": 3407 + }, + { + "epoch": 0.2561250563655494, + "grad_norm": 1.8111148671668382, + "learning_rate": 3.48739672922713e-06, + "loss": 1.0713, + "step": 3408 + }, + { + "epoch": 0.25620021043138436, + "grad_norm": 1.6264158096640886, + "learning_rate": 3.4870712236973524e-06, + "loss": 1.0806, + "step": 3409 + }, + { + "epoch": 0.2562753644972193, + "grad_norm": 1.6383867527249225, + "learning_rate": 3.4867456300529096e-06, + "loss": 1.031, + "step": 3410 + }, + { + "epoch": 0.25635051856305424, + "grad_norm": 3.158123253618577, + "learning_rate": 3.4864199483130957e-06, + "loss": 0.8683, + "step": 3411 + }, + { + "epoch": 0.2564256726288892, + "grad_norm": 1.6049526155316258, + "learning_rate": 3.4860941784972077e-06, + "loss": 1.0194, + "step": 3412 + }, + { + "epoch": 0.2565008266947242, + "grad_norm": 2.2087309377438165, + "learning_rate": 3.485768320624549e-06, + "loss": 1.0826, + "step": 3413 + }, + { + "epoch": 0.25657598076055915, + "grad_norm": 4.319546236649373, + "learning_rate": 3.485442374714428e-06, + "loss": 0.9874, + "step": 3414 + }, + { + "epoch": 0.2566511348263941, + "grad_norm": 1.7542318538897697, + "learning_rate": 3.485116340786158e-06, + "loss": 1.0623, + "step": 3415 + }, + { + "epoch": 0.2567262888922291, + "grad_norm": 2.227329055272028, + "learning_rate": 3.4847902188590582e-06, + "loss": 0.97, + "step": 3416 + }, + { + "epoch": 0.256801442958064, + "grad_norm": 1.6099126713922334, + "learning_rate": 3.484464008952452e-06, + "loss": 0.9776, + "step": 3417 + }, + { + "epoch": 0.256876597023899, + "grad_norm": 14.816781628351874, + "learning_rate": 3.484137711085669e-06, + "loss": 1.0425, + "step": 3418 + }, + { + "epoch": 0.25695175108973395, + "grad_norm": 1.5504405226286446, + "learning_rate": 3.4838113252780435e-06, + "loss": 1.1198, + "step": 3419 + }, + { + "epoch": 0.2570269051555689, + "grad_norm": 1.4703225989810749, + "learning_rate": 3.4834848515489154e-06, + "loss": 0.9563, + "step": 3420 + }, + { + "epoch": 0.2571020592214039, + "grad_norm": 1.5325011766438532, + "learning_rate": 3.4831582899176286e-06, + "loss": 1.0201, + "step": 3421 + }, + { + "epoch": 0.25717721328723886, + "grad_norm": 1.8814978224036931, + "learning_rate": 3.4828316404035345e-06, + "loss": 1.0607, + "step": 3422 + }, + { + "epoch": 0.2572523673530738, + "grad_norm": 1.9306767202299093, + "learning_rate": 3.4825049030259868e-06, + "loss": 1.0041, + "step": 3423 + }, + { + "epoch": 0.25732752141890874, + "grad_norm": 4.1521873408777115, + "learning_rate": 3.482178077804347e-06, + "loss": 0.9327, + "step": 3424 + }, + { + "epoch": 0.2574026754847437, + "grad_norm": 2.681535622915823, + "learning_rate": 3.48185116475798e-06, + "loss": 1.0188, + "step": 3425 + }, + { + "epoch": 0.2574778295505787, + "grad_norm": 2.122630761997037, + "learning_rate": 3.481524163906258e-06, + "loss": 0.9647, + "step": 3426 + }, + { + "epoch": 0.25755298361641366, + "grad_norm": 1.4418558386168978, + "learning_rate": 3.4811970752685555e-06, + "loss": 0.9474, + "step": 3427 + }, + { + "epoch": 0.2576281376822486, + "grad_norm": 0.7293406460605794, + "learning_rate": 3.4808698988642547e-06, + "loss": 0.8339, + "step": 3428 + }, + { + "epoch": 0.2577032917480836, + "grad_norm": 1.6793349494885923, + "learning_rate": 3.4805426347127416e-06, + "loss": 1.0443, + "step": 3429 + }, + { + "epoch": 0.2577784458139185, + "grad_norm": 1.5602982839879909, + "learning_rate": 3.4802152828334083e-06, + "loss": 0.9983, + "step": 3430 + }, + { + "epoch": 0.2578535998797535, + "grad_norm": 1.6120880369563273, + "learning_rate": 3.479887843245651e-06, + "loss": 1.0408, + "step": 3431 + }, + { + "epoch": 0.25792875394558845, + "grad_norm": 2.0894514910052733, + "learning_rate": 3.4795603159688725e-06, + "loss": 1.0169, + "step": 3432 + }, + { + "epoch": 0.2580039080114234, + "grad_norm": 1.8168616595597, + "learning_rate": 3.4792327010224794e-06, + "loss": 0.7988, + "step": 3433 + }, + { + "epoch": 0.2580790620772584, + "grad_norm": 1.723693189563101, + "learning_rate": 3.478904998425884e-06, + "loss": 0.8807, + "step": 3434 + }, + { + "epoch": 0.25815421614309336, + "grad_norm": 2.422216772759672, + "learning_rate": 3.478577208198505e-06, + "loss": 0.9894, + "step": 3435 + }, + { + "epoch": 0.2582293702089283, + "grad_norm": 1.7244459616981027, + "learning_rate": 3.478249330359764e-06, + "loss": 1.0346, + "step": 3436 + }, + { + "epoch": 0.25830452427476325, + "grad_norm": 1.8525480713572464, + "learning_rate": 3.4779213649290907e-06, + "loss": 0.9474, + "step": 3437 + }, + { + "epoch": 0.2583796783405982, + "grad_norm": 2.477223208889372, + "learning_rate": 3.4775933119259162e-06, + "loss": 0.9875, + "step": 3438 + }, + { + "epoch": 0.2584548324064332, + "grad_norm": 1.519260882568553, + "learning_rate": 3.47726517136968e-06, + "loss": 0.9686, + "step": 3439 + }, + { + "epoch": 0.25852998647226816, + "grad_norm": 0.719108906574286, + "learning_rate": 3.4769369432798258e-06, + "loss": 0.8554, + "step": 3440 + }, + { + "epoch": 0.25860514053810313, + "grad_norm": 2.1319473244491935, + "learning_rate": 3.4766086276758014e-06, + "loss": 1.0094, + "step": 3441 + }, + { + "epoch": 0.2586802946039381, + "grad_norm": 1.3341222108196853, + "learning_rate": 3.4762802245770627e-06, + "loss": 0.8692, + "step": 3442 + }, + { + "epoch": 0.258755448669773, + "grad_norm": 0.7078406406223311, + "learning_rate": 3.4759517340030674e-06, + "loss": 0.8328, + "step": 3443 + }, + { + "epoch": 0.258830602735608, + "grad_norm": 1.580551321802054, + "learning_rate": 3.475623155973279e-06, + "loss": 0.8776, + "step": 3444 + }, + { + "epoch": 0.25890575680144295, + "grad_norm": 1.8287014826650674, + "learning_rate": 3.4752944905071687e-06, + "loss": 0.9236, + "step": 3445 + }, + { + "epoch": 0.2589809108672779, + "grad_norm": 1.820723785957548, + "learning_rate": 3.474965737624211e-06, + "loss": 0.8616, + "step": 3446 + }, + { + "epoch": 0.2590560649331129, + "grad_norm": 2.0834931757325186, + "learning_rate": 3.474636897343885e-06, + "loss": 0.8828, + "step": 3447 + }, + { + "epoch": 0.25913121899894787, + "grad_norm": 1.325410469821777, + "learning_rate": 3.474307969685676e-06, + "loss": 0.9506, + "step": 3448 + }, + { + "epoch": 0.2592063730647828, + "grad_norm": 1.2533132987162765, + "learning_rate": 3.473978954669074e-06, + "loss": 1.0099, + "step": 3449 + }, + { + "epoch": 0.25928152713061775, + "grad_norm": 0.7091980499184827, + "learning_rate": 3.473649852313575e-06, + "loss": 0.8905, + "step": 3450 + }, + { + "epoch": 0.2593566811964527, + "grad_norm": 1.980863513013173, + "learning_rate": 3.4733206626386794e-06, + "loss": 1.1009, + "step": 3451 + }, + { + "epoch": 0.2594318352622877, + "grad_norm": 1.3952028169485404, + "learning_rate": 3.472991385663893e-06, + "loss": 0.9889, + "step": 3452 + }, + { + "epoch": 0.25950698932812266, + "grad_norm": 1.6219020804494593, + "learning_rate": 3.4726620214087264e-06, + "loss": 1.0658, + "step": 3453 + }, + { + "epoch": 0.25958214339395763, + "grad_norm": 2.157773902984311, + "learning_rate": 3.4723325698926953e-06, + "loss": 0.8354, + "step": 3454 + }, + { + "epoch": 0.2596572974597926, + "grad_norm": 1.5180013059225985, + "learning_rate": 3.4720030311353216e-06, + "loss": 1.0014, + "step": 3455 + }, + { + "epoch": 0.2597324515256275, + "grad_norm": 1.7628270488342193, + "learning_rate": 3.4716734051561324e-06, + "loss": 1.0284, + "step": 3456 + }, + { + "epoch": 0.2598076055914625, + "grad_norm": 2.2034474671197506, + "learning_rate": 3.471343691974658e-06, + "loss": 0.9941, + "step": 3457 + }, + { + "epoch": 0.25988275965729746, + "grad_norm": 1.4090601499617847, + "learning_rate": 3.471013891610436e-06, + "loss": 1.1123, + "step": 3458 + }, + { + "epoch": 0.2599579137231324, + "grad_norm": 1.9396703316107122, + "learning_rate": 3.4706840040830076e-06, + "loss": 1.0457, + "step": 3459 + }, + { + "epoch": 0.2600330677889674, + "grad_norm": 2.005076997757756, + "learning_rate": 3.4703540294119204e-06, + "loss": 1.0697, + "step": 3460 + }, + { + "epoch": 0.26010822185480237, + "grad_norm": 1.9811991712145243, + "learning_rate": 3.4700239676167264e-06, + "loss": 0.9357, + "step": 3461 + }, + { + "epoch": 0.2601833759206373, + "grad_norm": 1.7954171942852264, + "learning_rate": 3.4696938187169836e-06, + "loss": 1.052, + "step": 3462 + }, + { + "epoch": 0.26025852998647225, + "grad_norm": 2.2716742425551093, + "learning_rate": 3.469363582732254e-06, + "loss": 0.9901, + "step": 3463 + }, + { + "epoch": 0.2603336840523072, + "grad_norm": 1.9142346363126144, + "learning_rate": 3.4690332596821065e-06, + "loss": 0.9312, + "step": 3464 + }, + { + "epoch": 0.2604088381181422, + "grad_norm": 2.0124740219808923, + "learning_rate": 3.468702849586112e-06, + "loss": 0.9722, + "step": 3465 + }, + { + "epoch": 0.26048399218397716, + "grad_norm": 0.7802200394598119, + "learning_rate": 3.4683723524638494e-06, + "loss": 0.7813, + "step": 3466 + }, + { + "epoch": 0.26055914624981213, + "grad_norm": 1.5200787418448727, + "learning_rate": 3.4680417683349024e-06, + "loss": 1.0829, + "step": 3467 + }, + { + "epoch": 0.26063430031564705, + "grad_norm": 0.8138473050586534, + "learning_rate": 3.46771109721886e-06, + "loss": 0.937, + "step": 3468 + }, + { + "epoch": 0.260709454381482, + "grad_norm": 3.3276279418560883, + "learning_rate": 3.467380339135314e-06, + "loss": 0.9948, + "step": 3469 + }, + { + "epoch": 0.260784608447317, + "grad_norm": 2.017593014450746, + "learning_rate": 3.4670494941038642e-06, + "loss": 0.9832, + "step": 3470 + }, + { + "epoch": 0.26085976251315196, + "grad_norm": 0.6927324711506835, + "learning_rate": 3.466718562144114e-06, + "loss": 0.8416, + "step": 3471 + }, + { + "epoch": 0.26093491657898693, + "grad_norm": 0.6795928713817007, + "learning_rate": 3.4663875432756726e-06, + "loss": 0.845, + "step": 3472 + }, + { + "epoch": 0.2610100706448219, + "grad_norm": 1.7068632204647343, + "learning_rate": 3.466056437518154e-06, + "loss": 1.024, + "step": 3473 + }, + { + "epoch": 0.26108522471065687, + "grad_norm": 1.6412508542827815, + "learning_rate": 3.465725244891178e-06, + "loss": 1.0364, + "step": 3474 + }, + { + "epoch": 0.2611603787764918, + "grad_norm": 1.5431592606379378, + "learning_rate": 3.465393965414368e-06, + "loss": 1.0062, + "step": 3475 + }, + { + "epoch": 0.26123553284232676, + "grad_norm": 1.7089166420729343, + "learning_rate": 3.4650625991073543e-06, + "loss": 0.9804, + "step": 3476 + }, + { + "epoch": 0.2613106869081617, + "grad_norm": 1.60792320470905, + "learning_rate": 3.464731145989772e-06, + "loss": 0.967, + "step": 3477 + }, + { + "epoch": 0.2613858409739967, + "grad_norm": 7.72971791017934, + "learning_rate": 3.46439960608126e-06, + "loss": 1.0962, + "step": 3478 + }, + { + "epoch": 0.26146099503983167, + "grad_norm": 2.7489786852485203, + "learning_rate": 3.464067979401464e-06, + "loss": 1.0219, + "step": 3479 + }, + { + "epoch": 0.26153614910566664, + "grad_norm": 6.566727502737939, + "learning_rate": 3.4637362659700337e-06, + "loss": 0.983, + "step": 3480 + }, + { + "epoch": 0.26161130317150155, + "grad_norm": 2.2893997308518474, + "learning_rate": 3.463404465806625e-06, + "loss": 0.9747, + "step": 3481 + }, + { + "epoch": 0.2616864572373365, + "grad_norm": 1.8934966418304764, + "learning_rate": 3.4630725789308974e-06, + "loss": 1.0405, + "step": 3482 + }, + { + "epoch": 0.2617616113031715, + "grad_norm": 2.3915199150581774, + "learning_rate": 3.4627406053625175e-06, + "loss": 1.0225, + "step": 3483 + }, + { + "epoch": 0.26183676536900646, + "grad_norm": 2.047718835112218, + "learning_rate": 3.462408545121155e-06, + "loss": 0.9574, + "step": 3484 + }, + { + "epoch": 0.26191191943484143, + "grad_norm": 1.7255753677122165, + "learning_rate": 3.462076398226487e-06, + "loss": 0.9648, + "step": 3485 + }, + { + "epoch": 0.2619870735006764, + "grad_norm": 1.8703505407813266, + "learning_rate": 3.4617441646981935e-06, + "loss": 1.0424, + "step": 3486 + }, + { + "epoch": 0.2620622275665114, + "grad_norm": 1.5253539722058915, + "learning_rate": 3.461411844555961e-06, + "loss": 1.0551, + "step": 3487 + }, + { + "epoch": 0.2621373816323463, + "grad_norm": 1.724748858538691, + "learning_rate": 3.46107943781948e-06, + "loss": 1.024, + "step": 3488 + }, + { + "epoch": 0.26221253569818126, + "grad_norm": 8.95936226073824, + "learning_rate": 3.460746944508448e-06, + "loss": 0.9505, + "step": 3489 + }, + { + "epoch": 0.26228768976401623, + "grad_norm": 1.5824382493926483, + "learning_rate": 3.4604143646425655e-06, + "loss": 0.9468, + "step": 3490 + }, + { + "epoch": 0.2623628438298512, + "grad_norm": 2.451279352878988, + "learning_rate": 3.46008169824154e-06, + "loss": 1.0116, + "step": 3491 + }, + { + "epoch": 0.26243799789568617, + "grad_norm": 1.8327568260381086, + "learning_rate": 3.4597489453250824e-06, + "loss": 1.0257, + "step": 3492 + }, + { + "epoch": 0.26251315196152114, + "grad_norm": 2.1224676638814577, + "learning_rate": 3.4594161059129102e-06, + "loss": 0.8755, + "step": 3493 + }, + { + "epoch": 0.26258830602735606, + "grad_norm": 1.4993381629607112, + "learning_rate": 3.4590831800247457e-06, + "loss": 0.9419, + "step": 3494 + }, + { + "epoch": 0.262663460093191, + "grad_norm": 2.4471150052786768, + "learning_rate": 3.458750167680315e-06, + "loss": 0.8524, + "step": 3495 + }, + { + "epoch": 0.262738614159026, + "grad_norm": 1.3152131240075065, + "learning_rate": 3.458417068899351e-06, + "loss": 0.9772, + "step": 3496 + }, + { + "epoch": 0.26281376822486097, + "grad_norm": 1.4816014126517563, + "learning_rate": 3.4580838837015915e-06, + "loss": 0.9327, + "step": 3497 + }, + { + "epoch": 0.26288892229069594, + "grad_norm": 2.1940274533177258, + "learning_rate": 3.4577506121067784e-06, + "loss": 0.9358, + "step": 3498 + }, + { + "epoch": 0.2629640763565309, + "grad_norm": 1.8818959898143925, + "learning_rate": 3.457417254134659e-06, + "loss": 0.9355, + "step": 3499 + }, + { + "epoch": 0.2630392304223659, + "grad_norm": 2.9237371212666265, + "learning_rate": 3.457083809804986e-06, + "loss": 0.9476, + "step": 3500 + }, + { + "epoch": 0.2631143844882008, + "grad_norm": 1.7485539047314111, + "learning_rate": 3.456750279137519e-06, + "loss": 0.9998, + "step": 3501 + }, + { + "epoch": 0.26318953855403576, + "grad_norm": 2.39574975130545, + "learning_rate": 3.4564166621520193e-06, + "loss": 0.8916, + "step": 3502 + }, + { + "epoch": 0.26326469261987073, + "grad_norm": 1.8845202034190895, + "learning_rate": 3.456082958868255e-06, + "loss": 1.0332, + "step": 3503 + }, + { + "epoch": 0.2633398466857057, + "grad_norm": 2.0203551155900343, + "learning_rate": 3.455749169306e-06, + "loss": 0.902, + "step": 3504 + }, + { + "epoch": 0.2634150007515407, + "grad_norm": 1.6697044900311406, + "learning_rate": 3.455415293485032e-06, + "loss": 1.0581, + "step": 3505 + }, + { + "epoch": 0.26349015481737564, + "grad_norm": 1.5873454942382939, + "learning_rate": 3.455081331425135e-06, + "loss": 0.9232, + "step": 3506 + }, + { + "epoch": 0.26356530888321056, + "grad_norm": 4.934706078523917, + "learning_rate": 3.4547472831460973e-06, + "loss": 0.9221, + "step": 3507 + }, + { + "epoch": 0.26364046294904553, + "grad_norm": 1.6906702144873385, + "learning_rate": 3.4544131486677124e-06, + "loss": 1.0852, + "step": 3508 + }, + { + "epoch": 0.2637156170148805, + "grad_norm": 4.32659157811757, + "learning_rate": 3.454078928009779e-06, + "loss": 1.0303, + "step": 3509 + }, + { + "epoch": 0.26379077108071547, + "grad_norm": 1.6508778231057892, + "learning_rate": 3.4537446211921008e-06, + "loss": 1.0068, + "step": 3510 + }, + { + "epoch": 0.26386592514655044, + "grad_norm": 1.7147306435373062, + "learning_rate": 3.4534102282344876e-06, + "loss": 1.069, + "step": 3511 + }, + { + "epoch": 0.2639410792123854, + "grad_norm": 1.6507292811982521, + "learning_rate": 3.453075749156753e-06, + "loss": 1.053, + "step": 3512 + }, + { + "epoch": 0.2640162332782203, + "grad_norm": 4.183578704892415, + "learning_rate": 3.4527411839787152e-06, + "loss": 1.0054, + "step": 3513 + }, + { + "epoch": 0.2640913873440553, + "grad_norm": 1.3143656958061603, + "learning_rate": 3.4524065327202e-06, + "loss": 0.9153, + "step": 3514 + }, + { + "epoch": 0.26416654140989027, + "grad_norm": 1.6733449018060773, + "learning_rate": 3.4520717954010356e-06, + "loss": 0.96, + "step": 3515 + }, + { + "epoch": 0.26424169547572524, + "grad_norm": 1.6183542658941348, + "learning_rate": 3.4517369720410576e-06, + "loss": 1.0226, + "step": 3516 + }, + { + "epoch": 0.2643168495415602, + "grad_norm": 4.086084541898873, + "learning_rate": 3.4514020626601044e-06, + "loss": 1.0811, + "step": 3517 + }, + { + "epoch": 0.2643920036073952, + "grad_norm": 1.8873034216975149, + "learning_rate": 3.451067067278021e-06, + "loss": 0.9875, + "step": 3518 + }, + { + "epoch": 0.26446715767323015, + "grad_norm": 2.0361728581943295, + "learning_rate": 3.4507319859146585e-06, + "loss": 1.0733, + "step": 3519 + }, + { + "epoch": 0.26454231173906506, + "grad_norm": 7.5941536250596, + "learning_rate": 3.4503968185898696e-06, + "loss": 0.9719, + "step": 3520 + }, + { + "epoch": 0.26461746580490003, + "grad_norm": 1.6287921527083322, + "learning_rate": 3.450061565323516e-06, + "loss": 0.9768, + "step": 3521 + }, + { + "epoch": 0.264692619870735, + "grad_norm": 1.8488649443202783, + "learning_rate": 3.449726226135461e-06, + "loss": 1.0715, + "step": 3522 + }, + { + "epoch": 0.26476777393657, + "grad_norm": 3.0567602749645046, + "learning_rate": 3.4493908010455762e-06, + "loss": 0.9905, + "step": 3523 + }, + { + "epoch": 0.26484292800240494, + "grad_norm": 6.062145419418908, + "learning_rate": 3.4490552900737363e-06, + "loss": 1.0255, + "step": 3524 + }, + { + "epoch": 0.2649180820682399, + "grad_norm": 2.4327799751395904, + "learning_rate": 3.448719693239822e-06, + "loss": 1.0398, + "step": 3525 + }, + { + "epoch": 0.2649932361340748, + "grad_norm": 2.1540668880097527, + "learning_rate": 3.448384010563718e-06, + "loss": 0.9017, + "step": 3526 + }, + { + "epoch": 0.2650683901999098, + "grad_norm": 1.3999050261435542, + "learning_rate": 3.4480482420653153e-06, + "loss": 0.9669, + "step": 3527 + }, + { + "epoch": 0.26514354426574477, + "grad_norm": 1.3951779530277217, + "learning_rate": 3.4477123877645093e-06, + "loss": 1.0079, + "step": 3528 + }, + { + "epoch": 0.26521869833157974, + "grad_norm": 2.354368341860307, + "learning_rate": 3.4473764476812004e-06, + "loss": 1.086, + "step": 3529 + }, + { + "epoch": 0.2652938523974147, + "grad_norm": 1.7078283618075223, + "learning_rate": 3.447040421835295e-06, + "loss": 0.9457, + "step": 3530 + }, + { + "epoch": 0.2653690064632497, + "grad_norm": 1.7166525787713693, + "learning_rate": 3.446704310246703e-06, + "loss": 0.8822, + "step": 3531 + }, + { + "epoch": 0.26544416052908465, + "grad_norm": 4.457023119104108, + "learning_rate": 3.4463681129353413e-06, + "loss": 1.0269, + "step": 3532 + }, + { + "epoch": 0.26551931459491956, + "grad_norm": 2.1032642154047707, + "learning_rate": 3.4460318299211304e-06, + "loss": 1.0028, + "step": 3533 + }, + { + "epoch": 0.26559446866075453, + "grad_norm": 2.705618177137895, + "learning_rate": 3.4456954612239964e-06, + "loss": 1.011, + "step": 3534 + }, + { + "epoch": 0.2656696227265895, + "grad_norm": 1.6727227174184238, + "learning_rate": 3.44535900686387e-06, + "loss": 0.9387, + "step": 3535 + }, + { + "epoch": 0.2657447767924245, + "grad_norm": 2.7019947312965105, + "learning_rate": 3.4450224668606884e-06, + "loss": 1.0567, + "step": 3536 + }, + { + "epoch": 0.26581993085825945, + "grad_norm": 2.732084612042984, + "learning_rate": 3.444685841234392e-06, + "loss": 1.0524, + "step": 3537 + }, + { + "epoch": 0.2658950849240944, + "grad_norm": 2.1932680649404213, + "learning_rate": 3.444349130004927e-06, + "loss": 1.0515, + "step": 3538 + }, + { + "epoch": 0.26597023898992933, + "grad_norm": 2.7049419492743447, + "learning_rate": 3.4440123331922457e-06, + "loss": 1.0466, + "step": 3539 + }, + { + "epoch": 0.2660453930557643, + "grad_norm": 1.7881217382108612, + "learning_rate": 3.443675450816304e-06, + "loss": 0.9735, + "step": 3540 + }, + { + "epoch": 0.26612054712159927, + "grad_norm": 1.8815983066468502, + "learning_rate": 3.4433384828970636e-06, + "loss": 0.9777, + "step": 3541 + }, + { + "epoch": 0.26619570118743424, + "grad_norm": 1.7216388353509071, + "learning_rate": 3.443001429454491e-06, + "loss": 1.0204, + "step": 3542 + }, + { + "epoch": 0.2662708552532692, + "grad_norm": 2.1214955548220313, + "learning_rate": 3.4426642905085585e-06, + "loss": 0.9637, + "step": 3543 + }, + { + "epoch": 0.2663460093191042, + "grad_norm": 2.2792794064307573, + "learning_rate": 3.4423270660792422e-06, + "loss": 0.9754, + "step": 3544 + }, + { + "epoch": 0.26642116338493915, + "grad_norm": 1.970192577515134, + "learning_rate": 3.4419897561865242e-06, + "loss": 1.0313, + "step": 3545 + }, + { + "epoch": 0.26649631745077407, + "grad_norm": 2.1036514054221693, + "learning_rate": 3.4416523608503914e-06, + "loss": 0.9639, + "step": 3546 + }, + { + "epoch": 0.26657147151660904, + "grad_norm": 1.6561973131544736, + "learning_rate": 3.4413148800908364e-06, + "loss": 1.0906, + "step": 3547 + }, + { + "epoch": 0.266646625582444, + "grad_norm": 2.9516263519698875, + "learning_rate": 3.4409773139278546e-06, + "loss": 1.0594, + "step": 3548 + }, + { + "epoch": 0.266721779648279, + "grad_norm": 2.844473945775957, + "learning_rate": 3.44063966238145e-06, + "loss": 1.0309, + "step": 3549 + }, + { + "epoch": 0.26679693371411395, + "grad_norm": 1.8402062968000938, + "learning_rate": 3.440301925471628e-06, + "loss": 1.015, + "step": 3550 + }, + { + "epoch": 0.2668720877799489, + "grad_norm": 1.5509280656554147, + "learning_rate": 3.439964103218402e-06, + "loss": 0.9195, + "step": 3551 + }, + { + "epoch": 0.26694724184578383, + "grad_norm": 1.7071121860058702, + "learning_rate": 3.439626195641789e-06, + "loss": 1.0453, + "step": 3552 + }, + { + "epoch": 0.2670223959116188, + "grad_norm": 2.0327129084522677, + "learning_rate": 3.4392882027618113e-06, + "loss": 0.9744, + "step": 3553 + }, + { + "epoch": 0.2670975499774538, + "grad_norm": 3.460171927000781, + "learning_rate": 3.438950124598496e-06, + "loss": 0.798, + "step": 3554 + }, + { + "epoch": 0.26717270404328874, + "grad_norm": 1.4486424207529462, + "learning_rate": 3.438611961171875e-06, + "loss": 0.921, + "step": 3555 + }, + { + "epoch": 0.2672478581091237, + "grad_norm": 2.3716929651099345, + "learning_rate": 3.4382737125019874e-06, + "loss": 1.0689, + "step": 3556 + }, + { + "epoch": 0.2673230121749587, + "grad_norm": 2.2719032380291924, + "learning_rate": 3.4379353786088748e-06, + "loss": 1.1205, + "step": 3557 + }, + { + "epoch": 0.2673981662407936, + "grad_norm": 2.0895364981131634, + "learning_rate": 3.437596959512585e-06, + "loss": 1.0369, + "step": 3558 + }, + { + "epoch": 0.26747332030662857, + "grad_norm": 1.9378603323299122, + "learning_rate": 3.4372584552331694e-06, + "loss": 0.9839, + "step": 3559 + }, + { + "epoch": 0.26754847437246354, + "grad_norm": 1.6052540730720417, + "learning_rate": 3.4369198657906875e-06, + "loss": 0.9689, + "step": 3560 + }, + { + "epoch": 0.2676236284382985, + "grad_norm": 1.724024430435982, + "learning_rate": 3.4365811912052013e-06, + "loss": 1.0201, + "step": 3561 + }, + { + "epoch": 0.2676987825041335, + "grad_norm": 1.7773821816114528, + "learning_rate": 3.4362424314967777e-06, + "loss": 1.0069, + "step": 3562 + }, + { + "epoch": 0.26777393656996845, + "grad_norm": 0.7965277869052623, + "learning_rate": 3.4359035866854907e-06, + "loss": 0.8854, + "step": 3563 + }, + { + "epoch": 0.2678490906358034, + "grad_norm": 1.7616996679520196, + "learning_rate": 3.435564656791418e-06, + "loss": 0.9644, + "step": 3564 + }, + { + "epoch": 0.26792424470163834, + "grad_norm": 2.0755927928159115, + "learning_rate": 3.435225641834642e-06, + "loss": 1.0991, + "step": 3565 + }, + { + "epoch": 0.2679993987674733, + "grad_norm": 1.4974337719520174, + "learning_rate": 3.434886541835251e-06, + "loss": 1.004, + "step": 3566 + }, + { + "epoch": 0.2680745528333083, + "grad_norm": 2.238782789883722, + "learning_rate": 3.434547356813338e-06, + "loss": 1.0247, + "step": 3567 + }, + { + "epoch": 0.26814970689914325, + "grad_norm": 1.6193795206234316, + "learning_rate": 3.4342080867890006e-06, + "loss": 0.9877, + "step": 3568 + }, + { + "epoch": 0.2682248609649782, + "grad_norm": 2.1721187976152154, + "learning_rate": 3.4338687317823425e-06, + "loss": 1.0479, + "step": 3569 + }, + { + "epoch": 0.2683000150308132, + "grad_norm": 1.980244279589034, + "learning_rate": 3.4335292918134713e-06, + "loss": 1.0134, + "step": 3570 + }, + { + "epoch": 0.2683751690966481, + "grad_norm": 1.7799678002348938, + "learning_rate": 3.4331897669024996e-06, + "loss": 1.0691, + "step": 3571 + }, + { + "epoch": 0.2684503231624831, + "grad_norm": 1.8502839257189492, + "learning_rate": 3.432850157069546e-06, + "loss": 1.078, + "step": 3572 + }, + { + "epoch": 0.26852547722831804, + "grad_norm": 1.7289505406018804, + "learning_rate": 3.4325104623347345e-06, + "loss": 1.0432, + "step": 3573 + }, + { + "epoch": 0.268600631294153, + "grad_norm": 1.4419866287680259, + "learning_rate": 3.432170682718193e-06, + "loss": 0.9787, + "step": 3574 + }, + { + "epoch": 0.268675785359988, + "grad_norm": 1.6586484996152375, + "learning_rate": 3.431830818240054e-06, + "loss": 0.9051, + "step": 3575 + }, + { + "epoch": 0.26875093942582295, + "grad_norm": 1.6476924713190577, + "learning_rate": 3.431490868920456e-06, + "loss": 0.9867, + "step": 3576 + }, + { + "epoch": 0.2688260934916579, + "grad_norm": 1.6947569046566378, + "learning_rate": 3.4311508347795427e-06, + "loss": 1.0194, + "step": 3577 + }, + { + "epoch": 0.26890124755749284, + "grad_norm": 3.9886813957796345, + "learning_rate": 3.430810715837462e-06, + "loss": 0.9464, + "step": 3578 + }, + { + "epoch": 0.2689764016233278, + "grad_norm": 1.6686001371034056, + "learning_rate": 3.4304705121143674e-06, + "loss": 1.0059, + "step": 3579 + }, + { + "epoch": 0.2690515556891628, + "grad_norm": 3.422716847161729, + "learning_rate": 3.4301302236304174e-06, + "loss": 1.0024, + "step": 3580 + }, + { + "epoch": 0.26912670975499775, + "grad_norm": 1.9944571338215653, + "learning_rate": 3.4297898504057754e-06, + "loss": 0.9904, + "step": 3581 + }, + { + "epoch": 0.2692018638208327, + "grad_norm": 1.4760166350050135, + "learning_rate": 3.4294493924606095e-06, + "loss": 1.0645, + "step": 3582 + }, + { + "epoch": 0.2692770178866677, + "grad_norm": 2.8315387165993466, + "learning_rate": 3.429108849815094e-06, + "loss": 1.1052, + "step": 3583 + }, + { + "epoch": 0.2693521719525026, + "grad_norm": 1.649902488254655, + "learning_rate": 3.428768222489406e-06, + "loss": 0.9223, + "step": 3584 + }, + { + "epoch": 0.2694273260183376, + "grad_norm": 1.862700237800844, + "learning_rate": 3.4284275105037298e-06, + "loss": 1.0173, + "step": 3585 + }, + { + "epoch": 0.26950248008417255, + "grad_norm": 1.8107243752326094, + "learning_rate": 3.4280867138782544e-06, + "loss": 0.9932, + "step": 3586 + }, + { + "epoch": 0.2695776341500075, + "grad_norm": 1.6617283457204137, + "learning_rate": 3.427745832633172e-06, + "loss": 1.0232, + "step": 3587 + }, + { + "epoch": 0.2696527882158425, + "grad_norm": 1.5929607606306746, + "learning_rate": 3.4274048667886826e-06, + "loss": 0.9915, + "step": 3588 + }, + { + "epoch": 0.26972794228167746, + "grad_norm": 1.6150556878329725, + "learning_rate": 3.4270638163649884e-06, + "loss": 1.0638, + "step": 3589 + }, + { + "epoch": 0.2698030963475124, + "grad_norm": 2.711614079105186, + "learning_rate": 3.4267226813822983e-06, + "loss": 0.9683, + "step": 3590 + }, + { + "epoch": 0.26987825041334734, + "grad_norm": 1.722039975330347, + "learning_rate": 3.426381461860826e-06, + "loss": 1.033, + "step": 3591 + }, + { + "epoch": 0.2699534044791823, + "grad_norm": 1.8882705909134598, + "learning_rate": 3.4260401578207904e-06, + "loss": 1.0023, + "step": 3592 + }, + { + "epoch": 0.2700285585450173, + "grad_norm": 1.586496414392112, + "learning_rate": 3.425698769282415e-06, + "loss": 1.0187, + "step": 3593 + }, + { + "epoch": 0.27010371261085225, + "grad_norm": 1.776167864152127, + "learning_rate": 3.4253572962659276e-06, + "loss": 1.0163, + "step": 3594 + }, + { + "epoch": 0.2701788666766872, + "grad_norm": 2.416773585442678, + "learning_rate": 3.425015738791563e-06, + "loss": 0.9286, + "step": 3595 + }, + { + "epoch": 0.2702540207425222, + "grad_norm": 1.9122598060918037, + "learning_rate": 3.424674096879559e-06, + "loss": 0.9498, + "step": 3596 + }, + { + "epoch": 0.2703291748083571, + "grad_norm": 2.21824780005745, + "learning_rate": 3.424332370550159e-06, + "loss": 1.0106, + "step": 3597 + }, + { + "epoch": 0.2704043288741921, + "grad_norm": 1.5726405863543256, + "learning_rate": 3.4239905598236115e-06, + "loss": 1.0124, + "step": 3598 + }, + { + "epoch": 0.27047948294002705, + "grad_norm": 2.5127219752095393, + "learning_rate": 3.423648664720171e-06, + "loss": 1.0422, + "step": 3599 + }, + { + "epoch": 0.270554637005862, + "grad_norm": 1.640032924810066, + "learning_rate": 3.4233066852600958e-06, + "loss": 1.0729, + "step": 3600 + }, + { + "epoch": 0.270629791071697, + "grad_norm": 1.408326713651791, + "learning_rate": 3.422964621463649e-06, + "loss": 0.9997, + "step": 3601 + }, + { + "epoch": 0.27070494513753196, + "grad_norm": 3.545170983843727, + "learning_rate": 3.4226224733511e-06, + "loss": 1.063, + "step": 3602 + }, + { + "epoch": 0.2707800992033669, + "grad_norm": 1.9089563887956504, + "learning_rate": 3.4222802409427216e-06, + "loss": 0.9404, + "step": 3603 + }, + { + "epoch": 0.27085525326920185, + "grad_norm": 1.5196328411121218, + "learning_rate": 3.421937924258792e-06, + "loss": 1.0492, + "step": 3604 + }, + { + "epoch": 0.2709304073350368, + "grad_norm": 1.7110907081045394, + "learning_rate": 3.421595523319596e-06, + "loss": 1.0901, + "step": 3605 + }, + { + "epoch": 0.2710055614008718, + "grad_norm": 3.19500701115515, + "learning_rate": 3.421253038145421e-06, + "loss": 1.0352, + "step": 3606 + }, + { + "epoch": 0.27108071546670676, + "grad_norm": 1.6295835701417112, + "learning_rate": 3.420910468756562e-06, + "loss": 0.9145, + "step": 3607 + }, + { + "epoch": 0.2711558695325417, + "grad_norm": 2.3277879509391215, + "learning_rate": 3.4205678151733162e-06, + "loss": 0.9655, + "step": 3608 + }, + { + "epoch": 0.2712310235983767, + "grad_norm": 2.0453349136278463, + "learning_rate": 3.420225077415988e-06, + "loss": 0.9945, + "step": 3609 + }, + { + "epoch": 0.2713061776642116, + "grad_norm": 2.031056204134146, + "learning_rate": 3.4198822555048856e-06, + "loss": 1.1081, + "step": 3610 + }, + { + "epoch": 0.2713813317300466, + "grad_norm": 1.602800960398825, + "learning_rate": 3.419539349460322e-06, + "loss": 1.0929, + "step": 3611 + }, + { + "epoch": 0.27145648579588155, + "grad_norm": 3.4015083557501784, + "learning_rate": 3.4191963593026163e-06, + "loss": 1.0791, + "step": 3612 + }, + { + "epoch": 0.2715316398617165, + "grad_norm": 2.622881897647152, + "learning_rate": 3.4188532850520924e-06, + "loss": 1.0285, + "step": 3613 + }, + { + "epoch": 0.2716067939275515, + "grad_norm": 2.4207312067214963, + "learning_rate": 3.4185101267290773e-06, + "loss": 1.0055, + "step": 3614 + }, + { + "epoch": 0.27168194799338646, + "grad_norm": 3.010430957447878, + "learning_rate": 3.418166884353906e-06, + "loss": 1.0078, + "step": 3615 + }, + { + "epoch": 0.2717571020592214, + "grad_norm": 1.7526881824627158, + "learning_rate": 3.4178235579469154e-06, + "loss": 1.0276, + "step": 3616 + }, + { + "epoch": 0.27183225612505635, + "grad_norm": 1.6190522725441785, + "learning_rate": 3.417480147528451e-06, + "loss": 1.0142, + "step": 3617 + }, + { + "epoch": 0.2719074101908913, + "grad_norm": 2.1785417505605777, + "learning_rate": 3.4171366531188596e-06, + "loss": 1.0239, + "step": 3618 + }, + { + "epoch": 0.2719825642567263, + "grad_norm": 2.079198574313137, + "learning_rate": 3.4167930747384947e-06, + "loss": 0.9904, + "step": 3619 + }, + { + "epoch": 0.27205771832256126, + "grad_norm": 1.8683963198541238, + "learning_rate": 3.416449412407715e-06, + "loss": 1.0036, + "step": 3620 + }, + { + "epoch": 0.27213287238839623, + "grad_norm": 2.157841523546381, + "learning_rate": 3.4161056661468834e-06, + "loss": 1.139, + "step": 3621 + }, + { + "epoch": 0.2722080264542312, + "grad_norm": 2.3532362142880636, + "learning_rate": 3.4157618359763687e-06, + "loss": 0.9519, + "step": 3622 + }, + { + "epoch": 0.2722831805200661, + "grad_norm": 1.5452135970477021, + "learning_rate": 3.4154179219165435e-06, + "loss": 1.0311, + "step": 3623 + }, + { + "epoch": 0.2723583345859011, + "grad_norm": 2.038433109356498, + "learning_rate": 3.415073923987787e-06, + "loss": 0.9844, + "step": 3624 + }, + { + "epoch": 0.27243348865173606, + "grad_norm": 2.6870148686635407, + "learning_rate": 3.4147298422104815e-06, + "loss": 0.9819, + "step": 3625 + }, + { + "epoch": 0.272508642717571, + "grad_norm": 2.8398165485071045, + "learning_rate": 3.4143856766050157e-06, + "loss": 0.934, + "step": 3626 + }, + { + "epoch": 0.272583796783406, + "grad_norm": 34.216670170811845, + "learning_rate": 3.4140414271917825e-06, + "loss": 1.0676, + "step": 3627 + }, + { + "epoch": 0.27265895084924097, + "grad_norm": 3.4996993071953613, + "learning_rate": 3.4136970939911793e-06, + "loss": 1.0325, + "step": 3628 + }, + { + "epoch": 0.2727341049150759, + "grad_norm": 0.9259417109115281, + "learning_rate": 3.413352677023611e-06, + "loss": 0.8396, + "step": 3629 + }, + { + "epoch": 0.27280925898091085, + "grad_norm": 1.6277270542801314, + "learning_rate": 3.4130081763094836e-06, + "loss": 0.9812, + "step": 3630 + }, + { + "epoch": 0.2728844130467458, + "grad_norm": 1.7087277988199892, + "learning_rate": 3.4126635918692114e-06, + "loss": 1.0061, + "step": 3631 + }, + { + "epoch": 0.2729595671125808, + "grad_norm": 2.102160173819498, + "learning_rate": 3.412318923723212e-06, + "loss": 0.9968, + "step": 3632 + }, + { + "epoch": 0.27303472117841576, + "grad_norm": 2.1954226211676575, + "learning_rate": 3.411974171891908e-06, + "loss": 1.0437, + "step": 3633 + }, + { + "epoch": 0.27310987524425073, + "grad_norm": 1.536958602238563, + "learning_rate": 3.4116293363957276e-06, + "loss": 1.0037, + "step": 3634 + }, + { + "epoch": 0.2731850293100857, + "grad_norm": 2.2990688304909894, + "learning_rate": 3.4112844172551034e-06, + "loss": 0.9027, + "step": 3635 + }, + { + "epoch": 0.2732601833759206, + "grad_norm": 1.5879863414561364, + "learning_rate": 3.410939414490474e-06, + "loss": 0.9757, + "step": 3636 + }, + { + "epoch": 0.2733353374417556, + "grad_norm": 2.024113486941881, + "learning_rate": 3.4105943281222804e-06, + "loss": 1.0725, + "step": 3637 + }, + { + "epoch": 0.27341049150759056, + "grad_norm": 1.6457978578760084, + "learning_rate": 3.4102491581709717e-06, + "loss": 1.0228, + "step": 3638 + }, + { + "epoch": 0.27348564557342553, + "grad_norm": 2.2190984532922493, + "learning_rate": 3.4099039046570006e-06, + "loss": 1.1434, + "step": 3639 + }, + { + "epoch": 0.2735607996392605, + "grad_norm": 2.1765044566765446, + "learning_rate": 3.4095585676008234e-06, + "loss": 0.9199, + "step": 3640 + }, + { + "epoch": 0.27363595370509547, + "grad_norm": 0.7321653603541323, + "learning_rate": 3.4092131470229045e-06, + "loss": 0.8697, + "step": 3641 + }, + { + "epoch": 0.2737111077709304, + "grad_norm": 2.2446771815255473, + "learning_rate": 3.40886764294371e-06, + "loss": 1.0503, + "step": 3642 + }, + { + "epoch": 0.27378626183676535, + "grad_norm": 0.656900764417385, + "learning_rate": 3.4085220553837133e-06, + "loss": 0.8497, + "step": 3643 + }, + { + "epoch": 0.2738614159026003, + "grad_norm": 3.181691730760899, + "learning_rate": 3.40817638436339e-06, + "loss": 0.9597, + "step": 3644 + }, + { + "epoch": 0.2739365699684353, + "grad_norm": 2.2670365770348235, + "learning_rate": 3.407830629903224e-06, + "loss": 1.0255, + "step": 3645 + }, + { + "epoch": 0.27401172403427027, + "grad_norm": 1.4743089382480015, + "learning_rate": 3.4074847920237032e-06, + "loss": 1.0042, + "step": 3646 + }, + { + "epoch": 0.27408687810010524, + "grad_norm": 2.6688879616162526, + "learning_rate": 3.407138870745318e-06, + "loss": 1.0137, + "step": 3647 + }, + { + "epoch": 0.27416203216594015, + "grad_norm": 3.097020895838492, + "learning_rate": 3.4067928660885665e-06, + "loss": 1.0349, + "step": 3648 + }, + { + "epoch": 0.2742371862317751, + "grad_norm": 2.062861208641867, + "learning_rate": 3.406446778073951e-06, + "loss": 0.9934, + "step": 3649 + }, + { + "epoch": 0.2743123402976101, + "grad_norm": 1.5483382269238246, + "learning_rate": 3.4061006067219776e-06, + "loss": 0.9723, + "step": 3650 + }, + { + "epoch": 0.27438749436344506, + "grad_norm": 1.8276827024463336, + "learning_rate": 3.40575435205316e-06, + "loss": 1.0019, + "step": 3651 + }, + { + "epoch": 0.27446264842928003, + "grad_norm": 1.7147681843090568, + "learning_rate": 3.405408014088013e-06, + "loss": 1.0546, + "step": 3652 + }, + { + "epoch": 0.274537802495115, + "grad_norm": 1.5230814982429262, + "learning_rate": 3.40506159284706e-06, + "loss": 1.0298, + "step": 3653 + }, + { + "epoch": 0.27461295656095, + "grad_norm": 2.11598886716254, + "learning_rate": 3.4047150883508274e-06, + "loss": 1.0169, + "step": 3654 + }, + { + "epoch": 0.2746881106267849, + "grad_norm": 4.13726993163156, + "learning_rate": 3.4043685006198465e-06, + "loss": 0.9279, + "step": 3655 + }, + { + "epoch": 0.27476326469261986, + "grad_norm": 2.4804774006369823, + "learning_rate": 3.4040218296746544e-06, + "loss": 1.017, + "step": 3656 + }, + { + "epoch": 0.2748384187584548, + "grad_norm": 2.287836260652016, + "learning_rate": 3.403675075535793e-06, + "loss": 1.0216, + "step": 3657 + }, + { + "epoch": 0.2749135728242898, + "grad_norm": 2.4697837114038395, + "learning_rate": 3.403328238223808e-06, + "loss": 0.913, + "step": 3658 + }, + { + "epoch": 0.27498872689012477, + "grad_norm": 1.8003707715804356, + "learning_rate": 3.4029813177592504e-06, + "loss": 0.9738, + "step": 3659 + }, + { + "epoch": 0.27506388095595974, + "grad_norm": 1.4757714145508345, + "learning_rate": 3.402634314162678e-06, + "loss": 0.971, + "step": 3660 + }, + { + "epoch": 0.27513903502179465, + "grad_norm": 2.3244494411839467, + "learning_rate": 3.4022872274546517e-06, + "loss": 0.9893, + "step": 3661 + }, + { + "epoch": 0.2752141890876296, + "grad_norm": 1.8332114901657641, + "learning_rate": 3.4019400576557377e-06, + "loss": 1.0958, + "step": 3662 + }, + { + "epoch": 0.2752893431534646, + "grad_norm": 1.8067551035309495, + "learning_rate": 3.4015928047865056e-06, + "loss": 1.0663, + "step": 3663 + }, + { + "epoch": 0.27536449721929956, + "grad_norm": 1.6810011583831446, + "learning_rate": 3.401245468867534e-06, + "loss": 0.9979, + "step": 3664 + }, + { + "epoch": 0.27543965128513453, + "grad_norm": 1.421870137300665, + "learning_rate": 3.4008980499194025e-06, + "loss": 1.0454, + "step": 3665 + }, + { + "epoch": 0.2755148053509695, + "grad_norm": 1.7541685863976686, + "learning_rate": 3.4005505479626965e-06, + "loss": 0.9476, + "step": 3666 + }, + { + "epoch": 0.2755899594168045, + "grad_norm": 1.718280381288072, + "learning_rate": 3.4002029630180074e-06, + "loss": 0.9705, + "step": 3667 + }, + { + "epoch": 0.2756651134826394, + "grad_norm": 2.601258131213405, + "learning_rate": 3.399855295105932e-06, + "loss": 1.0545, + "step": 3668 + }, + { + "epoch": 0.27574026754847436, + "grad_norm": 5.4408750035112, + "learning_rate": 3.3995075442470694e-06, + "loss": 0.9824, + "step": 3669 + }, + { + "epoch": 0.27581542161430933, + "grad_norm": 1.9226872658271996, + "learning_rate": 3.3991597104620253e-06, + "loss": 0.9916, + "step": 3670 + }, + { + "epoch": 0.2758905756801443, + "grad_norm": 1.5231233713096706, + "learning_rate": 3.3988117937714114e-06, + "loss": 1.0714, + "step": 3671 + }, + { + "epoch": 0.27596572974597927, + "grad_norm": 3.3367828798841246, + "learning_rate": 3.398463794195842e-06, + "loss": 0.9774, + "step": 3672 + }, + { + "epoch": 0.27604088381181424, + "grad_norm": 0.6790781468547453, + "learning_rate": 3.3981157117559376e-06, + "loss": 0.8503, + "step": 3673 + }, + { + "epoch": 0.27611603787764916, + "grad_norm": 1.8121280946317675, + "learning_rate": 3.397767546472323e-06, + "loss": 0.9689, + "step": 3674 + }, + { + "epoch": 0.2761911919434841, + "grad_norm": 0.7167919688875442, + "learning_rate": 3.39741929836563e-06, + "loss": 0.7949, + "step": 3675 + }, + { + "epoch": 0.2762663460093191, + "grad_norm": 2.563511811513607, + "learning_rate": 3.3970709674564918e-06, + "loss": 1.0388, + "step": 3676 + }, + { + "epoch": 0.27634150007515407, + "grad_norm": 1.777887550415379, + "learning_rate": 3.3967225537655492e-06, + "loss": 1.0378, + "step": 3677 + }, + { + "epoch": 0.27641665414098904, + "grad_norm": 2.58015842587117, + "learning_rate": 3.396374057313447e-06, + "loss": 0.983, + "step": 3678 + }, + { + "epoch": 0.276491808206824, + "grad_norm": 1.9953442265580383, + "learning_rate": 3.396025478120835e-06, + "loss": 0.9573, + "step": 3679 + }, + { + "epoch": 0.276566962272659, + "grad_norm": 1.5095648983742185, + "learning_rate": 3.395676816208367e-06, + "loss": 0.9925, + "step": 3680 + }, + { + "epoch": 0.2766421163384939, + "grad_norm": 5.081571284010246, + "learning_rate": 3.3953280715967036e-06, + "loss": 1.0245, + "step": 3681 + }, + { + "epoch": 0.27671727040432886, + "grad_norm": 1.98512175919236, + "learning_rate": 3.394979244306509e-06, + "loss": 0.9049, + "step": 3682 + }, + { + "epoch": 0.27679242447016383, + "grad_norm": 1.6194668982933762, + "learning_rate": 3.3946303343584523e-06, + "loss": 0.9205, + "step": 3683 + }, + { + "epoch": 0.2768675785359988, + "grad_norm": 2.2713219936383746, + "learning_rate": 3.3942813417732083e-06, + "loss": 1.0262, + "step": 3684 + }, + { + "epoch": 0.2769427326018338, + "grad_norm": 1.4577986509068024, + "learning_rate": 3.3939322665714548e-06, + "loss": 1.029, + "step": 3685 + }, + { + "epoch": 0.27701788666766874, + "grad_norm": 1.7917669918744392, + "learning_rate": 3.3935831087738774e-06, + "loss": 0.9814, + "step": 3686 + }, + { + "epoch": 0.27709304073350366, + "grad_norm": 1.6498285002156743, + "learning_rate": 3.3932338684011646e-06, + "loss": 0.9885, + "step": 3687 + }, + { + "epoch": 0.27716819479933863, + "grad_norm": 1.6421567759738505, + "learning_rate": 3.3928845454740097e-06, + "loss": 1.0768, + "step": 3688 + }, + { + "epoch": 0.2772433488651736, + "grad_norm": 1.7134312958425002, + "learning_rate": 3.3925351400131118e-06, + "loss": 1.0747, + "step": 3689 + }, + { + "epoch": 0.27731850293100857, + "grad_norm": 2.036249317522678, + "learning_rate": 3.392185652039175e-06, + "loss": 0.9692, + "step": 3690 + }, + { + "epoch": 0.27739365699684354, + "grad_norm": 2.2556936492155417, + "learning_rate": 3.3918360815729066e-06, + "loss": 1.0572, + "step": 3691 + }, + { + "epoch": 0.2774688110626785, + "grad_norm": 1.7453741515223864, + "learning_rate": 3.391486428635021e-06, + "loss": 1.0323, + "step": 3692 + }, + { + "epoch": 0.2775439651285134, + "grad_norm": 3.036483651127075, + "learning_rate": 3.391136693246236e-06, + "loss": 0.9022, + "step": 3693 + }, + { + "epoch": 0.2776191191943484, + "grad_norm": 1.7236418034910685, + "learning_rate": 3.390786875427275e-06, + "loss": 1.0682, + "step": 3694 + }, + { + "epoch": 0.27769427326018337, + "grad_norm": 1.806626372409997, + "learning_rate": 3.3904369751988657e-06, + "loss": 1.0359, + "step": 3695 + }, + { + "epoch": 0.27776942732601834, + "grad_norm": 1.6298589687534695, + "learning_rate": 3.3900869925817416e-06, + "loss": 1.018, + "step": 3696 + }, + { + "epoch": 0.2778445813918533, + "grad_norm": 2.428624745593116, + "learning_rate": 3.3897369275966404e-06, + "loss": 0.942, + "step": 3697 + }, + { + "epoch": 0.2779197354576883, + "grad_norm": 1.8291136847652893, + "learning_rate": 3.389386780264304e-06, + "loss": 0.9878, + "step": 3698 + }, + { + "epoch": 0.27799488952352325, + "grad_norm": 2.7573370591248527, + "learning_rate": 3.389036550605481e-06, + "loss": 1.0275, + "step": 3699 + }, + { + "epoch": 0.27807004358935816, + "grad_norm": 1.750229071202258, + "learning_rate": 3.3886862386409233e-06, + "loss": 0.9951, + "step": 3700 + }, + { + "epoch": 0.27814519765519313, + "grad_norm": 8.672888894135536, + "learning_rate": 3.3883358443913883e-06, + "loss": 1.0645, + "step": 3701 + }, + { + "epoch": 0.2782203517210281, + "grad_norm": 1.571951016053601, + "learning_rate": 3.387985367877638e-06, + "loss": 1.0176, + "step": 3702 + }, + { + "epoch": 0.2782955057868631, + "grad_norm": 1.6079233529863852, + "learning_rate": 3.38763480912044e-06, + "loss": 0.9546, + "step": 3703 + }, + { + "epoch": 0.27837065985269804, + "grad_norm": 1.9252322886630249, + "learning_rate": 3.3872841681405654e-06, + "loss": 1.0439, + "step": 3704 + }, + { + "epoch": 0.278445813918533, + "grad_norm": 4.601931597636072, + "learning_rate": 3.3869334449587925e-06, + "loss": 1.008, + "step": 3705 + }, + { + "epoch": 0.27852096798436793, + "grad_norm": 1.6280386464361192, + "learning_rate": 3.3865826395959018e-06, + "loss": 0.945, + "step": 3706 + }, + { + "epoch": 0.2785961220502029, + "grad_norm": 7.6072676793583485, + "learning_rate": 3.38623175207268e-06, + "loss": 1.0402, + "step": 3707 + }, + { + "epoch": 0.27867127611603787, + "grad_norm": 2.075767499241421, + "learning_rate": 3.3858807824099182e-06, + "loss": 1.0623, + "step": 3708 + }, + { + "epoch": 0.27874643018187284, + "grad_norm": 1.3695693276786598, + "learning_rate": 3.385529730628414e-06, + "loss": 0.9647, + "step": 3709 + }, + { + "epoch": 0.2788215842477078, + "grad_norm": 1.8412362112032685, + "learning_rate": 3.385178596748967e-06, + "loss": 1.1065, + "step": 3710 + }, + { + "epoch": 0.2788967383135428, + "grad_norm": 1.6273724037464392, + "learning_rate": 3.3848273807923836e-06, + "loss": 1.0165, + "step": 3711 + }, + { + "epoch": 0.27897189237937775, + "grad_norm": 3.7823313278238198, + "learning_rate": 3.384476082779476e-06, + "loss": 1.0313, + "step": 3712 + }, + { + "epoch": 0.27904704644521267, + "grad_norm": 1.7086474974890693, + "learning_rate": 3.3841247027310584e-06, + "loss": 1.064, + "step": 3713 + }, + { + "epoch": 0.27912220051104764, + "grad_norm": 2.00943746751474, + "learning_rate": 3.3837732406679524e-06, + "loss": 1.0034, + "step": 3714 + }, + { + "epoch": 0.2791973545768826, + "grad_norm": 1.600489335930002, + "learning_rate": 3.3834216966109827e-06, + "loss": 0.9856, + "step": 3715 + }, + { + "epoch": 0.2792725086427176, + "grad_norm": 0.6909479680642149, + "learning_rate": 3.3830700705809802e-06, + "loss": 0.8467, + "step": 3716 + }, + { + "epoch": 0.27934766270855255, + "grad_norm": 1.950246902876078, + "learning_rate": 3.38271836259878e-06, + "loss": 1.0705, + "step": 3717 + }, + { + "epoch": 0.2794228167743875, + "grad_norm": 1.4792459314447348, + "learning_rate": 3.382366572685222e-06, + "loss": 0.9518, + "step": 3718 + }, + { + "epoch": 0.27949797084022243, + "grad_norm": 2.81271827698445, + "learning_rate": 3.3820147008611512e-06, + "loss": 1.0434, + "step": 3719 + }, + { + "epoch": 0.2795731249060574, + "grad_norm": 1.7171151448800301, + "learning_rate": 3.3816627471474166e-06, + "loss": 1.0017, + "step": 3720 + }, + { + "epoch": 0.2796482789718924, + "grad_norm": 6.208816646051574, + "learning_rate": 3.381310711564874e-06, + "loss": 0.9187, + "step": 3721 + }, + { + "epoch": 0.27972343303772734, + "grad_norm": 1.7079286638769002, + "learning_rate": 3.380958594134382e-06, + "loss": 1.085, + "step": 3722 + }, + { + "epoch": 0.2797985871035623, + "grad_norm": 2.161111410097931, + "learning_rate": 3.380606394876806e-06, + "loss": 0.9642, + "step": 3723 + }, + { + "epoch": 0.2798737411693973, + "grad_norm": 1.4482981593196114, + "learning_rate": 3.380254113813014e-06, + "loss": 1.0441, + "step": 3724 + }, + { + "epoch": 0.27994889523523225, + "grad_norm": 1.9106162673805536, + "learning_rate": 3.3799017509638805e-06, + "loss": 1.0649, + "step": 3725 + }, + { + "epoch": 0.28002404930106717, + "grad_norm": 1.6394468891351504, + "learning_rate": 3.3795493063502836e-06, + "loss": 0.8628, + "step": 3726 + }, + { + "epoch": 0.28009920336690214, + "grad_norm": 2.430508882589143, + "learning_rate": 3.3791967799931085e-06, + "loss": 1.0017, + "step": 3727 + }, + { + "epoch": 0.2801743574327371, + "grad_norm": 5.26541055757704, + "learning_rate": 3.3788441719132425e-06, + "loss": 0.9573, + "step": 3728 + }, + { + "epoch": 0.2802495114985721, + "grad_norm": 2.0401526566028467, + "learning_rate": 3.37849148213158e-06, + "loss": 1.0091, + "step": 3729 + }, + { + "epoch": 0.28032466556440705, + "grad_norm": 1.6764005925906789, + "learning_rate": 3.3781387106690175e-06, + "loss": 1.0086, + "step": 3730 + }, + { + "epoch": 0.280399819630242, + "grad_norm": 1.9913676650633145, + "learning_rate": 3.37778585754646e-06, + "loss": 1.0184, + "step": 3731 + }, + { + "epoch": 0.28047497369607693, + "grad_norm": 2.6840486765567833, + "learning_rate": 3.3774329227848144e-06, + "loss": 0.8827, + "step": 3732 + }, + { + "epoch": 0.2805501277619119, + "grad_norm": 1.5553366681153256, + "learning_rate": 3.3770799064049927e-06, + "loss": 1.0933, + "step": 3733 + }, + { + "epoch": 0.2806252818277469, + "grad_norm": 1.721584202741625, + "learning_rate": 3.3767268084279143e-06, + "loss": 1.0033, + "step": 3734 + }, + { + "epoch": 0.28070043589358185, + "grad_norm": 2.394936755944074, + "learning_rate": 3.376373628874501e-06, + "loss": 0.918, + "step": 3735 + }, + { + "epoch": 0.2807755899594168, + "grad_norm": 1.8319085248822815, + "learning_rate": 3.3760203677656786e-06, + "loss": 0.9829, + "step": 3736 + }, + { + "epoch": 0.2808507440252518, + "grad_norm": 1.5273218958235077, + "learning_rate": 3.3756670251223813e-06, + "loss": 0.9771, + "step": 3737 + }, + { + "epoch": 0.2809258980910867, + "grad_norm": 2.5234728358956864, + "learning_rate": 3.375313600965544e-06, + "loss": 1.0091, + "step": 3738 + }, + { + "epoch": 0.28100105215692167, + "grad_norm": 1.6053043476214819, + "learning_rate": 3.3749600953161102e-06, + "loss": 1.04, + "step": 3739 + }, + { + "epoch": 0.28107620622275664, + "grad_norm": 1.5239404875031355, + "learning_rate": 3.3746065081950253e-06, + "loss": 1.0698, + "step": 3740 + }, + { + "epoch": 0.2811513602885916, + "grad_norm": 1.9437153705811703, + "learning_rate": 3.374252839623241e-06, + "loss": 1.0233, + "step": 3741 + }, + { + "epoch": 0.2812265143544266, + "grad_norm": 1.6103930910399074, + "learning_rate": 3.373899089621714e-06, + "loss": 1.0631, + "step": 3742 + }, + { + "epoch": 0.28130166842026155, + "grad_norm": 1.5272154302913585, + "learning_rate": 3.3735452582114046e-06, + "loss": 0.9284, + "step": 3743 + }, + { + "epoch": 0.2813768224860965, + "grad_norm": 0.7304986141852277, + "learning_rate": 3.373191345413279e-06, + "loss": 0.8316, + "step": 3744 + }, + { + "epoch": 0.28145197655193144, + "grad_norm": 2.1952270273966525, + "learning_rate": 3.3728373512483083e-06, + "loss": 1.0553, + "step": 3745 + }, + { + "epoch": 0.2815271306177664, + "grad_norm": 2.444918957324709, + "learning_rate": 3.3724832757374674e-06, + "loss": 1.011, + "step": 3746 + }, + { + "epoch": 0.2816022846836014, + "grad_norm": 1.6668389310821308, + "learning_rate": 3.3721291189017363e-06, + "loss": 1.0732, + "step": 3747 + }, + { + "epoch": 0.28167743874943635, + "grad_norm": 1.501739764591935, + "learning_rate": 3.371774880762101e-06, + "loss": 1.061, + "step": 3748 + }, + { + "epoch": 0.2817525928152713, + "grad_norm": 1.605000894136269, + "learning_rate": 3.3714205613395513e-06, + "loss": 1.0554, + "step": 3749 + }, + { + "epoch": 0.2818277468811063, + "grad_norm": 1.5224808256698903, + "learning_rate": 3.371066160655082e-06, + "loss": 0.9762, + "step": 3750 + }, + { + "epoch": 0.2819029009469412, + "grad_norm": 2.0698588324062315, + "learning_rate": 3.3707116787296918e-06, + "loss": 0.9947, + "step": 3751 + }, + { + "epoch": 0.2819780550127762, + "grad_norm": 1.7880584698019983, + "learning_rate": 3.3703571155843866e-06, + "loss": 1.0929, + "step": 3752 + }, + { + "epoch": 0.28205320907861114, + "grad_norm": 1.9979142399502343, + "learning_rate": 3.370002471240174e-06, + "loss": 0.9394, + "step": 3753 + }, + { + "epoch": 0.2821283631444461, + "grad_norm": 1.4399270839244058, + "learning_rate": 3.36964774571807e-06, + "loss": 0.9955, + "step": 3754 + }, + { + "epoch": 0.2822035172102811, + "grad_norm": 2.7647327772701495, + "learning_rate": 3.3692929390390914e-06, + "loss": 0.9758, + "step": 3755 + }, + { + "epoch": 0.28227867127611606, + "grad_norm": 1.69076525940158, + "learning_rate": 3.3689380512242627e-06, + "loss": 0.9851, + "step": 3756 + }, + { + "epoch": 0.282353825341951, + "grad_norm": 1.6262704577907496, + "learning_rate": 3.3685830822946134e-06, + "loss": 0.8576, + "step": 3757 + }, + { + "epoch": 0.28242897940778594, + "grad_norm": 1.5361637759862135, + "learning_rate": 3.3682280322711753e-06, + "loss": 1.0772, + "step": 3758 + }, + { + "epoch": 0.2825041334736209, + "grad_norm": 1.8772117920788656, + "learning_rate": 3.367872901174987e-06, + "loss": 1.0095, + "step": 3759 + }, + { + "epoch": 0.2825792875394559, + "grad_norm": 1.4354007153485213, + "learning_rate": 3.367517689027091e-06, + "loss": 0.9204, + "step": 3760 + }, + { + "epoch": 0.28265444160529085, + "grad_norm": 1.623018083190775, + "learning_rate": 3.3671623958485354e-06, + "loss": 0.9962, + "step": 3761 + }, + { + "epoch": 0.2827295956711258, + "grad_norm": 2.041034782356265, + "learning_rate": 3.3668070216603736e-06, + "loss": 1.0901, + "step": 3762 + }, + { + "epoch": 0.2828047497369608, + "grad_norm": 1.7351553596174831, + "learning_rate": 3.366451566483661e-06, + "loss": 0.9366, + "step": 3763 + }, + { + "epoch": 0.2828799038027957, + "grad_norm": 1.59922426313187, + "learning_rate": 3.366096030339461e-06, + "loss": 1.04, + "step": 3764 + }, + { + "epoch": 0.2829550578686307, + "grad_norm": 1.647638570257468, + "learning_rate": 3.3657404132488403e-06, + "loss": 1.0135, + "step": 3765 + }, + { + "epoch": 0.28303021193446565, + "grad_norm": 1.6361821644755514, + "learning_rate": 3.3653847152328694e-06, + "loss": 1.0689, + "step": 3766 + }, + { + "epoch": 0.2831053660003006, + "grad_norm": 2.4274304077670283, + "learning_rate": 3.3650289363126266e-06, + "loss": 1.0459, + "step": 3767 + }, + { + "epoch": 0.2831805200661356, + "grad_norm": 1.567379846302684, + "learning_rate": 3.3646730765091916e-06, + "loss": 0.987, + "step": 3768 + }, + { + "epoch": 0.28325567413197056, + "grad_norm": 0.679764250006522, + "learning_rate": 3.3643171358436513e-06, + "loss": 0.8566, + "step": 3769 + }, + { + "epoch": 0.28333082819780553, + "grad_norm": 1.4833207251337959, + "learning_rate": 3.3639611143370967e-06, + "loss": 0.9058, + "step": 3770 + }, + { + "epoch": 0.28340598226364044, + "grad_norm": 1.4943511615097838, + "learning_rate": 3.3636050120106233e-06, + "loss": 1.0671, + "step": 3771 + }, + { + "epoch": 0.2834811363294754, + "grad_norm": 2.167964571012301, + "learning_rate": 3.363248828885331e-06, + "loss": 0.9766, + "step": 3772 + }, + { + "epoch": 0.2835562903953104, + "grad_norm": 2.0556728770594868, + "learning_rate": 3.362892564982325e-06, + "loss": 0.9418, + "step": 3773 + }, + { + "epoch": 0.28363144446114535, + "grad_norm": 1.8507666185089717, + "learning_rate": 3.3625362203227167e-06, + "loss": 1.0086, + "step": 3774 + }, + { + "epoch": 0.2837065985269803, + "grad_norm": 1.2609783161274388, + "learning_rate": 3.3621797949276188e-06, + "loss": 1.0016, + "step": 3775 + }, + { + "epoch": 0.2837817525928153, + "grad_norm": 1.6025930671030644, + "learning_rate": 3.3618232888181524e-06, + "loss": 0.9979, + "step": 3776 + }, + { + "epoch": 0.2838569066586502, + "grad_norm": 2.7429568523778034, + "learning_rate": 3.3614667020154415e-06, + "loss": 1.1041, + "step": 3777 + }, + { + "epoch": 0.2839320607244852, + "grad_norm": 1.6606225649091149, + "learning_rate": 3.3611100345406146e-06, + "loss": 0.9378, + "step": 3778 + }, + { + "epoch": 0.28400721479032015, + "grad_norm": 1.7720296371771493, + "learning_rate": 3.3607532864148063e-06, + "loss": 1.0093, + "step": 3779 + }, + { + "epoch": 0.2840823688561551, + "grad_norm": 2.1159650926504976, + "learning_rate": 3.3603964576591553e-06, + "loss": 0.9018, + "step": 3780 + }, + { + "epoch": 0.2841575229219901, + "grad_norm": 2.4247338925680206, + "learning_rate": 3.360039548294805e-06, + "loss": 1.0815, + "step": 3781 + }, + { + "epoch": 0.28423267698782506, + "grad_norm": 1.696229466965496, + "learning_rate": 3.3596825583429033e-06, + "loss": 0.9474, + "step": 3782 + }, + { + "epoch": 0.28430783105366, + "grad_norm": 2.25216157714356, + "learning_rate": 3.3593254878246035e-06, + "loss": 0.9914, + "step": 3783 + }, + { + "epoch": 0.28438298511949495, + "grad_norm": 1.7224482888707755, + "learning_rate": 3.358968336761063e-06, + "loss": 0.9975, + "step": 3784 + }, + { + "epoch": 0.2844581391853299, + "grad_norm": 2.1646271667839705, + "learning_rate": 3.3586111051734455e-06, + "loss": 1.0009, + "step": 3785 + }, + { + "epoch": 0.2845332932511649, + "grad_norm": 1.511690913296478, + "learning_rate": 3.358253793082917e-06, + "loss": 0.9935, + "step": 3786 + }, + { + "epoch": 0.28460844731699986, + "grad_norm": 5.325520077233545, + "learning_rate": 3.3578964005106496e-06, + "loss": 0.9497, + "step": 3787 + }, + { + "epoch": 0.2846836013828348, + "grad_norm": 1.5830564591441882, + "learning_rate": 3.3575389274778214e-06, + "loss": 1.0642, + "step": 3788 + }, + { + "epoch": 0.2847587554486698, + "grad_norm": 2.0910391267666872, + "learning_rate": 3.3571813740056135e-06, + "loss": 1.0725, + "step": 3789 + }, + { + "epoch": 0.2848339095145047, + "grad_norm": 2.0000319478344335, + "learning_rate": 3.356823740115212e-06, + "loss": 0.9867, + "step": 3790 + }, + { + "epoch": 0.2849090635803397, + "grad_norm": 1.8876998593452743, + "learning_rate": 3.3564660258278085e-06, + "loss": 0.9286, + "step": 3791 + }, + { + "epoch": 0.28498421764617465, + "grad_norm": 1.6364434131287762, + "learning_rate": 3.3561082311645982e-06, + "loss": 1.0622, + "step": 3792 + }, + { + "epoch": 0.2850593717120096, + "grad_norm": 2.404776481186764, + "learning_rate": 3.3557503561467832e-06, + "loss": 1.0142, + "step": 3793 + }, + { + "epoch": 0.2851345257778446, + "grad_norm": 1.6879848207631942, + "learning_rate": 3.3553924007955673e-06, + "loss": 0.8955, + "step": 3794 + }, + { + "epoch": 0.28520967984367956, + "grad_norm": 1.8051444792052114, + "learning_rate": 3.355034365132162e-06, + "loss": 0.8917, + "step": 3795 + }, + { + "epoch": 0.2852848339095145, + "grad_norm": 1.3898609076690465, + "learning_rate": 3.354676249177781e-06, + "loss": 1.0656, + "step": 3796 + }, + { + "epoch": 0.28535998797534945, + "grad_norm": 1.7539014242738804, + "learning_rate": 3.354318052953646e-06, + "loss": 0.9934, + "step": 3797 + }, + { + "epoch": 0.2854351420411844, + "grad_norm": 1.7592065327348276, + "learning_rate": 3.3539597764809794e-06, + "loss": 0.9082, + "step": 3798 + }, + { + "epoch": 0.2855102961070194, + "grad_norm": 2.6473312543164234, + "learning_rate": 3.3536014197810115e-06, + "loss": 0.9956, + "step": 3799 + }, + { + "epoch": 0.28558545017285436, + "grad_norm": 1.7358954011634795, + "learning_rate": 3.3532429828749768e-06, + "loss": 0.9336, + "step": 3800 + }, + { + "epoch": 0.28566060423868933, + "grad_norm": 1.6504101272423919, + "learning_rate": 3.3528844657841128e-06, + "loss": 1.0402, + "step": 3801 + }, + { + "epoch": 0.2857357583045243, + "grad_norm": 1.6958019762943888, + "learning_rate": 3.352525868529664e-06, + "loss": 1.0033, + "step": 3802 + }, + { + "epoch": 0.2858109123703592, + "grad_norm": 1.5165252371580018, + "learning_rate": 3.352167191132878e-06, + "loss": 1.0272, + "step": 3803 + }, + { + "epoch": 0.2858860664361942, + "grad_norm": 1.8305095542693801, + "learning_rate": 3.3518084336150084e-06, + "loss": 1.0626, + "step": 3804 + }, + { + "epoch": 0.28596122050202916, + "grad_norm": 2.4701896043352556, + "learning_rate": 3.3514495959973125e-06, + "loss": 0.9613, + "step": 3805 + }, + { + "epoch": 0.2860363745678641, + "grad_norm": 1.7807837428160609, + "learning_rate": 3.3510906783010536e-06, + "loss": 1.0023, + "step": 3806 + }, + { + "epoch": 0.2861115286336991, + "grad_norm": 4.335656179288769, + "learning_rate": 3.3507316805474976e-06, + "loss": 0.8959, + "step": 3807 + }, + { + "epoch": 0.28618668269953407, + "grad_norm": 1.4180644189597242, + "learning_rate": 3.3503726027579175e-06, + "loss": 0.9985, + "step": 3808 + }, + { + "epoch": 0.286261836765369, + "grad_norm": 1.4017426629089607, + "learning_rate": 3.3500134449535894e-06, + "loss": 0.9866, + "step": 3809 + }, + { + "epoch": 0.28633699083120395, + "grad_norm": 2.7457955037020936, + "learning_rate": 3.3496542071557955e-06, + "loss": 0.8859, + "step": 3810 + }, + { + "epoch": 0.2864121448970389, + "grad_norm": 1.437246383187346, + "learning_rate": 3.3492948893858217e-06, + "loss": 0.9656, + "step": 3811 + }, + { + "epoch": 0.2864872989628739, + "grad_norm": 1.9829184525651364, + "learning_rate": 3.3489354916649593e-06, + "loss": 0.9288, + "step": 3812 + }, + { + "epoch": 0.28656245302870886, + "grad_norm": 2.5526234165584816, + "learning_rate": 3.348576014014503e-06, + "loss": 1.0204, + "step": 3813 + }, + { + "epoch": 0.28663760709454383, + "grad_norm": 5.577064049960193, + "learning_rate": 3.3482164564557537e-06, + "loss": 0.9921, + "step": 3814 + }, + { + "epoch": 0.2867127611603788, + "grad_norm": 1.7378611539201183, + "learning_rate": 3.3478568190100173e-06, + "loss": 1.0101, + "step": 3815 + }, + { + "epoch": 0.2867879152262137, + "grad_norm": 11.598440367985972, + "learning_rate": 3.3474971016986024e-06, + "loss": 0.9088, + "step": 3816 + }, + { + "epoch": 0.2868630692920487, + "grad_norm": 1.4467815459618774, + "learning_rate": 3.3471373045428248e-06, + "loss": 1.0776, + "step": 3817 + }, + { + "epoch": 0.28693822335788366, + "grad_norm": 2.2772882628183493, + "learning_rate": 3.346777427564003e-06, + "loss": 0.9164, + "step": 3818 + }, + { + "epoch": 0.28701337742371863, + "grad_norm": 1.8242354790453486, + "learning_rate": 3.3464174707834618e-06, + "loss": 1.0632, + "step": 3819 + }, + { + "epoch": 0.2870885314895536, + "grad_norm": 1.677970540985024, + "learning_rate": 3.34605743422253e-06, + "loss": 0.9836, + "step": 3820 + }, + { + "epoch": 0.28716368555538857, + "grad_norm": 2.216451609607558, + "learning_rate": 3.34569731790254e-06, + "loss": 0.9164, + "step": 3821 + }, + { + "epoch": 0.2872388396212235, + "grad_norm": 1.8752697432719188, + "learning_rate": 3.3453371218448318e-06, + "loss": 1.0545, + "step": 3822 + }, + { + "epoch": 0.28731399368705846, + "grad_norm": 1.8784370708956917, + "learning_rate": 3.3449768460707465e-06, + "loss": 0.9913, + "step": 3823 + }, + { + "epoch": 0.2873891477528934, + "grad_norm": 0.8183428050255087, + "learning_rate": 3.344616490601633e-06, + "loss": 0.8252, + "step": 3824 + }, + { + "epoch": 0.2874643018187284, + "grad_norm": 1.734406479558419, + "learning_rate": 3.3442560554588444e-06, + "loss": 1.0112, + "step": 3825 + }, + { + "epoch": 0.28753945588456337, + "grad_norm": 1.8088890147749357, + "learning_rate": 3.3438955406637365e-06, + "loss": 1.006, + "step": 3826 + }, + { + "epoch": 0.28761460995039834, + "grad_norm": 2.559201979232552, + "learning_rate": 3.3435349462376713e-06, + "loss": 0.9584, + "step": 3827 + }, + { + "epoch": 0.28768976401623325, + "grad_norm": 1.8625501561932036, + "learning_rate": 3.343174272202017e-06, + "loss": 0.9729, + "step": 3828 + }, + { + "epoch": 0.2877649180820682, + "grad_norm": 2.376090652605282, + "learning_rate": 3.3428135185781425e-06, + "loss": 0.9568, + "step": 3829 + }, + { + "epoch": 0.2878400721479032, + "grad_norm": 1.6779153390971824, + "learning_rate": 3.3424526853874252e-06, + "loss": 1.0363, + "step": 3830 + }, + { + "epoch": 0.28791522621373816, + "grad_norm": 1.7877586977830715, + "learning_rate": 3.342091772651246e-06, + "loss": 0.8685, + "step": 3831 + }, + { + "epoch": 0.28799038027957313, + "grad_norm": 1.7846642281510048, + "learning_rate": 3.34173078039099e-06, + "loss": 1.0127, + "step": 3832 + }, + { + "epoch": 0.2880655343454081, + "grad_norm": 1.6447683421682227, + "learning_rate": 3.341369708628047e-06, + "loss": 1.1322, + "step": 3833 + }, + { + "epoch": 0.2881406884112431, + "grad_norm": 0.7101421308635154, + "learning_rate": 3.341008557383813e-06, + "loss": 0.8093, + "step": 3834 + }, + { + "epoch": 0.288215842477078, + "grad_norm": 0.8075901382141747, + "learning_rate": 3.3406473266796865e-06, + "loss": 0.8197, + "step": 3835 + }, + { + "epoch": 0.28829099654291296, + "grad_norm": 1.7195464629661976, + "learning_rate": 3.3402860165370724e-06, + "loss": 1.0062, + "step": 3836 + }, + { + "epoch": 0.28836615060874793, + "grad_norm": 1.6508817946310876, + "learning_rate": 3.3399246269773796e-06, + "loss": 1.0362, + "step": 3837 + }, + { + "epoch": 0.2884413046745829, + "grad_norm": 2.0461747231033427, + "learning_rate": 3.3395631580220213e-06, + "loss": 0.9522, + "step": 3838 + }, + { + "epoch": 0.28851645874041787, + "grad_norm": 1.5166587843358266, + "learning_rate": 3.3392016096924168e-06, + "loss": 1.0, + "step": 3839 + }, + { + "epoch": 0.28859161280625284, + "grad_norm": 2.119466083338932, + "learning_rate": 3.3388399820099887e-06, + "loss": 1.0097, + "step": 3840 + }, + { + "epoch": 0.28866676687208775, + "grad_norm": 1.7832318707893204, + "learning_rate": 3.3384782749961646e-06, + "loss": 1.0334, + "step": 3841 + }, + { + "epoch": 0.2887419209379227, + "grad_norm": 2.0641664650641927, + "learning_rate": 3.3381164886723777e-06, + "loss": 0.98, + "step": 3842 + }, + { + "epoch": 0.2888170750037577, + "grad_norm": 1.9425437250702529, + "learning_rate": 3.337754623060065e-06, + "loss": 1.014, + "step": 3843 + }, + { + "epoch": 0.28889222906959267, + "grad_norm": 2.360946435109871, + "learning_rate": 3.337392678180668e-06, + "loss": 1.0667, + "step": 3844 + }, + { + "epoch": 0.28896738313542764, + "grad_norm": 1.6513029387172602, + "learning_rate": 3.3370306540556336e-06, + "loss": 1.0718, + "step": 3845 + }, + { + "epoch": 0.2890425372012626, + "grad_norm": 1.4647534151833406, + "learning_rate": 3.336668550706413e-06, + "loss": 1.0367, + "step": 3846 + }, + { + "epoch": 0.2891176912670976, + "grad_norm": 2.303634765193626, + "learning_rate": 3.3363063681544628e-06, + "loss": 0.9851, + "step": 3847 + }, + { + "epoch": 0.2891928453329325, + "grad_norm": 2.253880227928748, + "learning_rate": 3.335944106421243e-06, + "loss": 1.0013, + "step": 3848 + }, + { + "epoch": 0.28926799939876746, + "grad_norm": 1.8570599930480278, + "learning_rate": 3.3355817655282188e-06, + "loss": 1.0049, + "step": 3849 + }, + { + "epoch": 0.28934315346460243, + "grad_norm": 6.23978688242914, + "learning_rate": 3.3352193454968607e-06, + "loss": 1.0006, + "step": 3850 + }, + { + "epoch": 0.2894183075304374, + "grad_norm": 1.4660083261334094, + "learning_rate": 3.334856846348644e-06, + "loss": 0.9833, + "step": 3851 + }, + { + "epoch": 0.2894934615962724, + "grad_norm": 1.7423700485044193, + "learning_rate": 3.3344942681050477e-06, + "loss": 0.8087, + "step": 3852 + }, + { + "epoch": 0.28956861566210734, + "grad_norm": 2.3231421277120643, + "learning_rate": 3.3341316107875552e-06, + "loss": 1.0152, + "step": 3853 + }, + { + "epoch": 0.28964376972794226, + "grad_norm": 1.5695555271608541, + "learning_rate": 3.3337688744176564e-06, + "loss": 1.0394, + "step": 3854 + }, + { + "epoch": 0.2897189237937772, + "grad_norm": 2.075967457451926, + "learning_rate": 3.3334060590168447e-06, + "loss": 0.9859, + "step": 3855 + }, + { + "epoch": 0.2897940778596122, + "grad_norm": 1.7537663665247065, + "learning_rate": 3.333043164606618e-06, + "loss": 1.0712, + "step": 3856 + }, + { + "epoch": 0.28986923192544717, + "grad_norm": 1.8445814165453376, + "learning_rate": 3.332680191208479e-06, + "loss": 0.925, + "step": 3857 + }, + { + "epoch": 0.28994438599128214, + "grad_norm": 2.6956717873644913, + "learning_rate": 3.3323171388439353e-06, + "loss": 1.0629, + "step": 3858 + }, + { + "epoch": 0.2900195400571171, + "grad_norm": 1.5467657956194962, + "learning_rate": 3.3319540075344996e-06, + "loss": 0.9069, + "step": 3859 + }, + { + "epoch": 0.2900946941229521, + "grad_norm": 2.448497319175432, + "learning_rate": 3.331590797301689e-06, + "loss": 0.9316, + "step": 3860 + }, + { + "epoch": 0.290169848188787, + "grad_norm": 2.024661014582072, + "learning_rate": 3.331227508167024e-06, + "loss": 0.9956, + "step": 3861 + }, + { + "epoch": 0.29024500225462196, + "grad_norm": 1.8751349718469421, + "learning_rate": 3.330864140152032e-06, + "loss": 1.1142, + "step": 3862 + }, + { + "epoch": 0.29032015632045693, + "grad_norm": 1.4406451772075344, + "learning_rate": 3.3305006932782435e-06, + "loss": 1.0496, + "step": 3863 + }, + { + "epoch": 0.2903953103862919, + "grad_norm": 2.2482916386355214, + "learning_rate": 3.3301371675671935e-06, + "loss": 0.9331, + "step": 3864 + }, + { + "epoch": 0.2904704644521269, + "grad_norm": 0.7402575647752823, + "learning_rate": 3.329773563040423e-06, + "loss": 0.8431, + "step": 3865 + }, + { + "epoch": 0.29054561851796185, + "grad_norm": 1.8559078299351834, + "learning_rate": 3.3294098797194776e-06, + "loss": 0.9601, + "step": 3866 + }, + { + "epoch": 0.29062077258379676, + "grad_norm": 1.6559299663625013, + "learning_rate": 3.3290461176259054e-06, + "loss": 0.9783, + "step": 3867 + }, + { + "epoch": 0.29069592664963173, + "grad_norm": 0.7588111353821797, + "learning_rate": 3.3286822767812618e-06, + "loss": 0.8517, + "step": 3868 + }, + { + "epoch": 0.2907710807154667, + "grad_norm": 2.05228251411964, + "learning_rate": 3.3283183572071054e-06, + "loss": 0.9957, + "step": 3869 + }, + { + "epoch": 0.29084623478130167, + "grad_norm": 2.1650103204494404, + "learning_rate": 3.3279543589249998e-06, + "loss": 1.0549, + "step": 3870 + }, + { + "epoch": 0.29092138884713664, + "grad_norm": 5.291110612784771, + "learning_rate": 3.3275902819565127e-06, + "loss": 0.9525, + "step": 3871 + }, + { + "epoch": 0.2909965429129716, + "grad_norm": 2.530582198602823, + "learning_rate": 3.3272261263232195e-06, + "loss": 0.9101, + "step": 3872 + }, + { + "epoch": 0.2910716969788065, + "grad_norm": 2.42067152085479, + "learning_rate": 3.326861892046694e-06, + "loss": 0.9705, + "step": 3873 + }, + { + "epoch": 0.2911468510446415, + "grad_norm": 1.7285583145990777, + "learning_rate": 3.3264975791485218e-06, + "loss": 1.0272, + "step": 3874 + }, + { + "epoch": 0.29122200511047647, + "grad_norm": 2.176111055920114, + "learning_rate": 3.3261331876502884e-06, + "loss": 0.9421, + "step": 3875 + }, + { + "epoch": 0.29129715917631144, + "grad_norm": 2.448495079585548, + "learning_rate": 3.325768717573585e-06, + "loss": 1.0344, + "step": 3876 + }, + { + "epoch": 0.2913723132421464, + "grad_norm": 1.8125904981448082, + "learning_rate": 3.325404168940009e-06, + "loss": 1.0144, + "step": 3877 + }, + { + "epoch": 0.2914474673079814, + "grad_norm": 1.4962714266537536, + "learning_rate": 3.3250395417711605e-06, + "loss": 1.0534, + "step": 3878 + }, + { + "epoch": 0.29152262137381635, + "grad_norm": 2.131151607624999, + "learning_rate": 3.3246748360886453e-06, + "loss": 1.0656, + "step": 3879 + }, + { + "epoch": 0.29159777543965126, + "grad_norm": 1.4349305782738522, + "learning_rate": 3.324310051914073e-06, + "loss": 0.9974, + "step": 3880 + }, + { + "epoch": 0.29167292950548623, + "grad_norm": 1.3536774949992814, + "learning_rate": 3.323945189269059e-06, + "loss": 1.0013, + "step": 3881 + }, + { + "epoch": 0.2917480835713212, + "grad_norm": 2.074594697090505, + "learning_rate": 3.323580248175223e-06, + "loss": 0.9368, + "step": 3882 + }, + { + "epoch": 0.2918232376371562, + "grad_norm": 2.0045429132996295, + "learning_rate": 3.3232152286541898e-06, + "loss": 0.9823, + "step": 3883 + }, + { + "epoch": 0.29189839170299114, + "grad_norm": 3.134377714024716, + "learning_rate": 3.3228501307275866e-06, + "loss": 0.9491, + "step": 3884 + }, + { + "epoch": 0.2919735457688261, + "grad_norm": 1.613951019839108, + "learning_rate": 3.3224849544170475e-06, + "loss": 0.9298, + "step": 3885 + }, + { + "epoch": 0.29204869983466103, + "grad_norm": 2.742591762902728, + "learning_rate": 3.3221196997442107e-06, + "loss": 0.9214, + "step": 3886 + }, + { + "epoch": 0.292123853900496, + "grad_norm": 1.8794341743472545, + "learning_rate": 3.3217543667307196e-06, + "loss": 0.9755, + "step": 3887 + }, + { + "epoch": 0.29219900796633097, + "grad_norm": 3.4098558850752654, + "learning_rate": 3.3213889553982206e-06, + "loss": 1.0771, + "step": 3888 + }, + { + "epoch": 0.29227416203216594, + "grad_norm": 0.8014414972953647, + "learning_rate": 3.321023465768366e-06, + "loss": 0.882, + "step": 3889 + }, + { + "epoch": 0.2923493160980009, + "grad_norm": 1.842618109904259, + "learning_rate": 3.320657897862812e-06, + "loss": 1.0186, + "step": 3890 + }, + { + "epoch": 0.2924244701638359, + "grad_norm": 4.124140794454097, + "learning_rate": 3.320292251703221e-06, + "loss": 0.8758, + "step": 3891 + }, + { + "epoch": 0.29249962422967085, + "grad_norm": 1.7664313036606796, + "learning_rate": 3.3199265273112583e-06, + "loss": 0.8938, + "step": 3892 + }, + { + "epoch": 0.29257477829550577, + "grad_norm": 2.3504825786439145, + "learning_rate": 3.3195607247085945e-06, + "loss": 0.873, + "step": 3893 + }, + { + "epoch": 0.29264993236134074, + "grad_norm": 1.5273282179567107, + "learning_rate": 3.319194843916905e-06, + "loss": 0.8679, + "step": 3894 + }, + { + "epoch": 0.2927250864271757, + "grad_norm": 3.0458898272060893, + "learning_rate": 3.3188288849578694e-06, + "loss": 0.9825, + "step": 3895 + }, + { + "epoch": 0.2928002404930107, + "grad_norm": 1.923375915979326, + "learning_rate": 3.318462847853172e-06, + "loss": 1.053, + "step": 3896 + }, + { + "epoch": 0.29287539455884565, + "grad_norm": 0.6907143875893501, + "learning_rate": 3.3180967326245018e-06, + "loss": 0.8178, + "step": 3897 + }, + { + "epoch": 0.2929505486246806, + "grad_norm": 1.70971949870043, + "learning_rate": 3.3177305392935536e-06, + "loss": 1.0306, + "step": 3898 + }, + { + "epoch": 0.29302570269051553, + "grad_norm": 1.7632086769755904, + "learning_rate": 3.317364267882025e-06, + "loss": 1.0179, + "step": 3899 + }, + { + "epoch": 0.2931008567563505, + "grad_norm": 1.854132434086056, + "learning_rate": 3.3169979184116182e-06, + "loss": 1.1048, + "step": 3900 + }, + { + "epoch": 0.2931760108221855, + "grad_norm": 1.96834396534933, + "learning_rate": 3.3166314909040427e-06, + "loss": 0.9902, + "step": 3901 + }, + { + "epoch": 0.29325116488802044, + "grad_norm": 2.320390077058315, + "learning_rate": 3.316264985381009e-06, + "loss": 0.9826, + "step": 3902 + }, + { + "epoch": 0.2933263189538554, + "grad_norm": 1.748619079706351, + "learning_rate": 3.315898401864235e-06, + "loss": 0.9834, + "step": 3903 + }, + { + "epoch": 0.2934014730196904, + "grad_norm": 1.637835210603773, + "learning_rate": 3.315531740375441e-06, + "loss": 0.9368, + "step": 3904 + }, + { + "epoch": 0.29347662708552535, + "grad_norm": 2.107231690148375, + "learning_rate": 3.3151650009363544e-06, + "loss": 0.992, + "step": 3905 + }, + { + "epoch": 0.29355178115136027, + "grad_norm": 1.4001079364448956, + "learning_rate": 3.3147981835687054e-06, + "loss": 0.9743, + "step": 3906 + }, + { + "epoch": 0.29362693521719524, + "grad_norm": 1.3089838613647438, + "learning_rate": 3.314431288294229e-06, + "loss": 1.0415, + "step": 3907 + }, + { + "epoch": 0.2937020892830302, + "grad_norm": 2.146876430372181, + "learning_rate": 3.314064315134666e-06, + "loss": 0.9736, + "step": 3908 + }, + { + "epoch": 0.2937772433488652, + "grad_norm": 1.722927772766014, + "learning_rate": 3.31369726411176e-06, + "loss": 0.9302, + "step": 3909 + }, + { + "epoch": 0.29385239741470015, + "grad_norm": 1.7530582136464692, + "learning_rate": 3.313330135247261e-06, + "loss": 1.0435, + "step": 3910 + }, + { + "epoch": 0.2939275514805351, + "grad_norm": 1.452954928382337, + "learning_rate": 3.312962928562922e-06, + "loss": 0.9183, + "step": 3911 + }, + { + "epoch": 0.29400270554637004, + "grad_norm": 1.6950454853222674, + "learning_rate": 3.312595644080502e-06, + "loss": 0.862, + "step": 3912 + }, + { + "epoch": 0.294077859612205, + "grad_norm": 1.8255866270593542, + "learning_rate": 3.312228281821764e-06, + "loss": 0.9577, + "step": 3913 + }, + { + "epoch": 0.29415301367804, + "grad_norm": 2.5527340948079895, + "learning_rate": 3.311860841808475e-06, + "loss": 1.013, + "step": 3914 + }, + { + "epoch": 0.29422816774387495, + "grad_norm": 1.6864530353176252, + "learning_rate": 3.311493324062408e-06, + "loss": 0.93, + "step": 3915 + }, + { + "epoch": 0.2943033218097099, + "grad_norm": 2.123126606641627, + "learning_rate": 3.3111257286053394e-06, + "loss": 0.9448, + "step": 3916 + }, + { + "epoch": 0.2943784758755449, + "grad_norm": 2.788962055182064, + "learning_rate": 3.310758055459051e-06, + "loss": 0.9434, + "step": 3917 + }, + { + "epoch": 0.2944536299413798, + "grad_norm": 2.015500323557467, + "learning_rate": 3.3103903046453282e-06, + "loss": 1.0387, + "step": 3918 + }, + { + "epoch": 0.2945287840072148, + "grad_norm": 1.783301995201109, + "learning_rate": 3.3100224761859626e-06, + "loss": 1.0203, + "step": 3919 + }, + { + "epoch": 0.29460393807304974, + "grad_norm": 1.6945830854733526, + "learning_rate": 3.309654570102748e-06, + "loss": 1.0235, + "step": 3920 + }, + { + "epoch": 0.2946790921388847, + "grad_norm": 1.6566000514513906, + "learning_rate": 3.309286586417486e-06, + "loss": 0.9415, + "step": 3921 + }, + { + "epoch": 0.2947542462047197, + "grad_norm": 7.3814544881973045, + "learning_rate": 3.3089185251519797e-06, + "loss": 0.9309, + "step": 3922 + }, + { + "epoch": 0.29482940027055465, + "grad_norm": 1.6160750686375691, + "learning_rate": 3.3085503863280387e-06, + "loss": 0.9836, + "step": 3923 + }, + { + "epoch": 0.2949045543363896, + "grad_norm": 2.916089373177859, + "learning_rate": 3.3081821699674763e-06, + "loss": 0.9819, + "step": 3924 + }, + { + "epoch": 0.29497970840222454, + "grad_norm": 2.220409256439624, + "learning_rate": 3.307813876092111e-06, + "loss": 0.9832, + "step": 3925 + }, + { + "epoch": 0.2950548624680595, + "grad_norm": 1.904707925888829, + "learning_rate": 3.307445504723766e-06, + "loss": 1.0683, + "step": 3926 + }, + { + "epoch": 0.2951300165338945, + "grad_norm": 0.7347280181394954, + "learning_rate": 3.307077055884268e-06, + "loss": 0.8807, + "step": 3927 + }, + { + "epoch": 0.29520517059972945, + "grad_norm": 1.6672882192660978, + "learning_rate": 3.3067085295954497e-06, + "loss": 1.0584, + "step": 3928 + }, + { + "epoch": 0.2952803246655644, + "grad_norm": 2.6307266983527158, + "learning_rate": 3.306339925879147e-06, + "loss": 0.9448, + "step": 3929 + }, + { + "epoch": 0.2953554787313994, + "grad_norm": 1.9145120520980754, + "learning_rate": 3.305971244757201e-06, + "loss": 1.0078, + "step": 3930 + }, + { + "epoch": 0.2954306327972343, + "grad_norm": 1.9367349406216148, + "learning_rate": 3.305602486251458e-06, + "loss": 1.0138, + "step": 3931 + }, + { + "epoch": 0.2955057868630693, + "grad_norm": 1.707082717084723, + "learning_rate": 3.3052336503837686e-06, + "loss": 0.9268, + "step": 3932 + }, + { + "epoch": 0.29558094092890425, + "grad_norm": 1.9829361873252882, + "learning_rate": 3.304864737175987e-06, + "loss": 1.0059, + "step": 3933 + }, + { + "epoch": 0.2956560949947392, + "grad_norm": 1.251585098424525, + "learning_rate": 3.3044957466499736e-06, + "loss": 0.9068, + "step": 3934 + }, + { + "epoch": 0.2957312490605742, + "grad_norm": 1.6904547219541044, + "learning_rate": 3.3041266788275913e-06, + "loss": 1.0093, + "step": 3935 + }, + { + "epoch": 0.29580640312640916, + "grad_norm": 2.235858451464254, + "learning_rate": 3.303757533730709e-06, + "loss": 1.0508, + "step": 3936 + }, + { + "epoch": 0.2958815571922441, + "grad_norm": 1.4535000224866363, + "learning_rate": 3.3033883113812017e-06, + "loss": 1.0132, + "step": 3937 + }, + { + "epoch": 0.29595671125807904, + "grad_norm": 1.699068610913722, + "learning_rate": 3.303019011800946e-06, + "loss": 1.0095, + "step": 3938 + }, + { + "epoch": 0.296031865323914, + "grad_norm": 1.8018917871302556, + "learning_rate": 3.302649635011823e-06, + "loss": 1.0525, + "step": 3939 + }, + { + "epoch": 0.296107019389749, + "grad_norm": 2.7090052333653096, + "learning_rate": 3.302280181035722e-06, + "loss": 1.0422, + "step": 3940 + }, + { + "epoch": 0.29618217345558395, + "grad_norm": 2.482808416902175, + "learning_rate": 3.301910649894533e-06, + "loss": 0.9917, + "step": 3941 + }, + { + "epoch": 0.2962573275214189, + "grad_norm": 1.7329071982179247, + "learning_rate": 3.3015410416101527e-06, + "loss": 0.9814, + "step": 3942 + }, + { + "epoch": 0.2963324815872539, + "grad_norm": 1.7063709090434402, + "learning_rate": 3.301171356204482e-06, + "loss": 1.0792, + "step": 3943 + }, + { + "epoch": 0.2964076356530888, + "grad_norm": 1.74578083144507, + "learning_rate": 3.300801593699425e-06, + "loss": 0.9182, + "step": 3944 + }, + { + "epoch": 0.2964827897189238, + "grad_norm": 0.6196523290705659, + "learning_rate": 3.300431754116894e-06, + "loss": 0.803, + "step": 3945 + }, + { + "epoch": 0.29655794378475875, + "grad_norm": 2.1932412147115565, + "learning_rate": 3.3000618374788e-06, + "loss": 0.8426, + "step": 3946 + }, + { + "epoch": 0.2966330978505937, + "grad_norm": 1.5169233286624222, + "learning_rate": 3.299691843807065e-06, + "loss": 1.0497, + "step": 3947 + }, + { + "epoch": 0.2967082519164287, + "grad_norm": 4.640256719393335, + "learning_rate": 3.2993217731236118e-06, + "loss": 1.0174, + "step": 3948 + }, + { + "epoch": 0.29678340598226366, + "grad_norm": 2.3146799483394305, + "learning_rate": 3.2989516254503677e-06, + "loss": 0.8307, + "step": 3949 + }, + { + "epoch": 0.29685856004809863, + "grad_norm": 2.2862892406969126, + "learning_rate": 3.298581400809266e-06, + "loss": 0.9089, + "step": 3950 + }, + { + "epoch": 0.29693371411393354, + "grad_norm": 1.845012297343802, + "learning_rate": 3.298211099222243e-06, + "loss": 0.9649, + "step": 3951 + }, + { + "epoch": 0.2970088681797685, + "grad_norm": 1.5954726390064597, + "learning_rate": 3.2978407207112416e-06, + "loss": 0.988, + "step": 3952 + }, + { + "epoch": 0.2970840222456035, + "grad_norm": 2.1422284657284902, + "learning_rate": 3.297470265298208e-06, + "loss": 0.8943, + "step": 3953 + }, + { + "epoch": 0.29715917631143846, + "grad_norm": 2.1123210537753043, + "learning_rate": 3.2970997330050923e-06, + "loss": 0.978, + "step": 3954 + }, + { + "epoch": 0.2972343303772734, + "grad_norm": 2.1669031772151355, + "learning_rate": 3.2967291238538507e-06, + "loss": 0.9416, + "step": 3955 + }, + { + "epoch": 0.2973094844431084, + "grad_norm": 1.9552910962182377, + "learning_rate": 3.296358437866443e-06, + "loss": 0.9745, + "step": 3956 + }, + { + "epoch": 0.2973846385089433, + "grad_norm": 2.312724540737735, + "learning_rate": 3.2959876750648338e-06, + "loss": 1.0085, + "step": 3957 + }, + { + "epoch": 0.2974597925747783, + "grad_norm": 1.7371545461928246, + "learning_rate": 3.2956168354709923e-06, + "loss": 1.0031, + "step": 3958 + }, + { + "epoch": 0.29753494664061325, + "grad_norm": 2.1462784339474297, + "learning_rate": 3.295245919106892e-06, + "loss": 0.932, + "step": 3959 + }, + { + "epoch": 0.2976101007064482, + "grad_norm": 4.220169619726522, + "learning_rate": 3.294874925994511e-06, + "loss": 1.042, + "step": 3960 + }, + { + "epoch": 0.2976852547722832, + "grad_norm": 1.8324214950753044, + "learning_rate": 3.2945038561558324e-06, + "loss": 0.9067, + "step": 3961 + }, + { + "epoch": 0.29776040883811816, + "grad_norm": 1.9016040405650978, + "learning_rate": 3.2941327096128435e-06, + "loss": 1.0212, + "step": 3962 + }, + { + "epoch": 0.2978355629039531, + "grad_norm": 0.7624119660884469, + "learning_rate": 3.2937614863875353e-06, + "loss": 0.9228, + "step": 3963 + }, + { + "epoch": 0.29791071696978805, + "grad_norm": 2.297493883691394, + "learning_rate": 3.293390186501906e-06, + "loss": 0.9358, + "step": 3964 + }, + { + "epoch": 0.297985871035623, + "grad_norm": 2.150305695203523, + "learning_rate": 3.2930188099779546e-06, + "loss": 0.9122, + "step": 3965 + }, + { + "epoch": 0.298061025101458, + "grad_norm": 1.6176224275743485, + "learning_rate": 3.292647356837688e-06, + "loss": 1.0446, + "step": 3966 + }, + { + "epoch": 0.29813617916729296, + "grad_norm": 2.547898724529966, + "learning_rate": 3.2922758271031147e-06, + "loss": 1.0421, + "step": 3967 + }, + { + "epoch": 0.29821133323312793, + "grad_norm": 1.991517496574712, + "learning_rate": 3.2919042207962506e-06, + "loss": 0.9048, + "step": 3968 + }, + { + "epoch": 0.2982864872989629, + "grad_norm": 1.6368561097548397, + "learning_rate": 3.2915325379391147e-06, + "loss": 0.9724, + "step": 3969 + }, + { + "epoch": 0.2983616413647978, + "grad_norm": 1.3201473516524862, + "learning_rate": 3.2911607785537297e-06, + "loss": 1.0234, + "step": 3970 + }, + { + "epoch": 0.2984367954306328, + "grad_norm": 1.4373964189445798, + "learning_rate": 3.290788942662125e-06, + "loss": 1.0608, + "step": 3971 + }, + { + "epoch": 0.29851194949646775, + "grad_norm": 1.6027336490637827, + "learning_rate": 3.290417030286333e-06, + "loss": 1.0267, + "step": 3972 + }, + { + "epoch": 0.2985871035623027, + "grad_norm": 1.860833245343031, + "learning_rate": 3.2900450414483897e-06, + "loss": 0.9534, + "step": 3973 + }, + { + "epoch": 0.2986622576281377, + "grad_norm": 1.6751157065894422, + "learning_rate": 3.2896729761703386e-06, + "loss": 0.9617, + "step": 3974 + }, + { + "epoch": 0.29873741169397267, + "grad_norm": 1.6326785968102167, + "learning_rate": 3.2893008344742244e-06, + "loss": 0.9114, + "step": 3975 + }, + { + "epoch": 0.2988125657598076, + "grad_norm": 1.5905860461428443, + "learning_rate": 3.288928616382099e-06, + "loss": 1.0107, + "step": 3976 + }, + { + "epoch": 0.29888771982564255, + "grad_norm": 3.9130003929069743, + "learning_rate": 3.288556321916018e-06, + "loss": 0.9781, + "step": 3977 + }, + { + "epoch": 0.2989628738914775, + "grad_norm": 2.2808553275888905, + "learning_rate": 3.2881839510980403e-06, + "loss": 1.079, + "step": 3978 + }, + { + "epoch": 0.2990380279573125, + "grad_norm": 1.5049275683085557, + "learning_rate": 3.2878115039502304e-06, + "loss": 0.9384, + "step": 3979 + }, + { + "epoch": 0.29911318202314746, + "grad_norm": 6.401022960448239, + "learning_rate": 3.2874389804946575e-06, + "loss": 1.0056, + "step": 3980 + }, + { + "epoch": 0.29918833608898243, + "grad_norm": 1.4628056296497112, + "learning_rate": 3.287066380753395e-06, + "loss": 1.0612, + "step": 3981 + }, + { + "epoch": 0.2992634901548174, + "grad_norm": 5.366002025477325, + "learning_rate": 3.2866937047485216e-06, + "loss": 1.0597, + "step": 3982 + }, + { + "epoch": 0.2993386442206523, + "grad_norm": 1.7183839581589677, + "learning_rate": 3.2863209525021186e-06, + "loss": 0.976, + "step": 3983 + }, + { + "epoch": 0.2994137982864873, + "grad_norm": 1.5312162123085917, + "learning_rate": 3.285948124036274e-06, + "loss": 1.0061, + "step": 3984 + }, + { + "epoch": 0.29948895235232226, + "grad_norm": 1.7719821419860116, + "learning_rate": 3.2855752193730786e-06, + "loss": 1.0317, + "step": 3985 + }, + { + "epoch": 0.2995641064181572, + "grad_norm": 1.915384700072189, + "learning_rate": 3.2852022385346283e-06, + "loss": 1.0243, + "step": 3986 + }, + { + "epoch": 0.2996392604839922, + "grad_norm": 4.5935517352649535, + "learning_rate": 3.2848291815430245e-06, + "loss": 0.9968, + "step": 3987 + }, + { + "epoch": 0.29971441454982717, + "grad_norm": 1.5172559477179914, + "learning_rate": 3.2844560484203717e-06, + "loss": 1.0473, + "step": 3988 + }, + { + "epoch": 0.2997895686156621, + "grad_norm": 1.5291699664975076, + "learning_rate": 3.2840828391887792e-06, + "loss": 0.8902, + "step": 3989 + }, + { + "epoch": 0.29986472268149705, + "grad_norm": 1.7539004047525788, + "learning_rate": 3.2837095538703613e-06, + "loss": 0.9121, + "step": 3990 + }, + { + "epoch": 0.299939876747332, + "grad_norm": 1.8292833228773109, + "learning_rate": 3.283336192487237e-06, + "loss": 0.8946, + "step": 3991 + }, + { + "epoch": 0.300015030813167, + "grad_norm": 1.4291188928091096, + "learning_rate": 3.282962755061529e-06, + "loss": 0.988, + "step": 3992 + }, + { + "epoch": 0.30009018487900196, + "grad_norm": 1.726797924993248, + "learning_rate": 3.2825892416153656e-06, + "loss": 1.0842, + "step": 3993 + }, + { + "epoch": 0.30016533894483693, + "grad_norm": 1.3647702455516113, + "learning_rate": 3.282215652170877e-06, + "loss": 1.0627, + "step": 3994 + }, + { + "epoch": 0.3002404930106719, + "grad_norm": 1.854759450158009, + "learning_rate": 3.2818419867502024e-06, + "loss": 1.0224, + "step": 3995 + }, + { + "epoch": 0.3003156470765068, + "grad_norm": 1.8478883109628381, + "learning_rate": 3.2814682453754805e-06, + "loss": 1.0119, + "step": 3996 + }, + { + "epoch": 0.3003908011423418, + "grad_norm": 33.13987726054142, + "learning_rate": 3.281094428068858e-06, + "loss": 1.0049, + "step": 3997 + }, + { + "epoch": 0.30046595520817676, + "grad_norm": 1.7988725257057583, + "learning_rate": 3.280720534852486e-06, + "loss": 1.0409, + "step": 3998 + }, + { + "epoch": 0.30054110927401173, + "grad_norm": 0.7873597232071186, + "learning_rate": 3.2803465657485175e-06, + "loss": 0.8571, + "step": 3999 + }, + { + "epoch": 0.3006162633398467, + "grad_norm": 1.5256439662881172, + "learning_rate": 3.279972520779112e-06, + "loss": 0.9943, + "step": 4000 + }, + { + "epoch": 0.30069141740568167, + "grad_norm": 1.783866165751919, + "learning_rate": 3.279598399966433e-06, + "loss": 0.9327, + "step": 4001 + }, + { + "epoch": 0.3007665714715166, + "grad_norm": 2.0312815737104386, + "learning_rate": 3.27922420333265e-06, + "loss": 1.0168, + "step": 4002 + }, + { + "epoch": 0.30084172553735156, + "grad_norm": 1.8039732341134065, + "learning_rate": 3.278849930899934e-06, + "loss": 0.9368, + "step": 4003 + }, + { + "epoch": 0.3009168796031865, + "grad_norm": 1.5695507422414552, + "learning_rate": 3.278475582690462e-06, + "loss": 1.0301, + "step": 4004 + }, + { + "epoch": 0.3009920336690215, + "grad_norm": 4.746801604556742, + "learning_rate": 3.2781011587264173e-06, + "loss": 0.9777, + "step": 4005 + }, + { + "epoch": 0.30106718773485647, + "grad_norm": 0.7761067271429258, + "learning_rate": 3.2777266590299835e-06, + "loss": 0.8784, + "step": 4006 + }, + { + "epoch": 0.30114234180069144, + "grad_norm": 2.7672830557627583, + "learning_rate": 3.277352083623353e-06, + "loss": 0.9491, + "step": 4007 + }, + { + "epoch": 0.30121749586652635, + "grad_norm": 2.2929072981078114, + "learning_rate": 3.2769774325287197e-06, + "loss": 0.9884, + "step": 4008 + }, + { + "epoch": 0.3012926499323613, + "grad_norm": 2.953819717959865, + "learning_rate": 3.2766027057682844e-06, + "loss": 1.068, + "step": 4009 + }, + { + "epoch": 0.3013678039981963, + "grad_norm": 1.6319414215420622, + "learning_rate": 3.27622790336425e-06, + "loss": 0.9703, + "step": 4010 + }, + { + "epoch": 0.30144295806403126, + "grad_norm": 1.7191397485121884, + "learning_rate": 3.2758530253388255e-06, + "loss": 0.96, + "step": 4011 + }, + { + "epoch": 0.30151811212986623, + "grad_norm": 0.7112641474920758, + "learning_rate": 3.2754780717142233e-06, + "loss": 0.8262, + "step": 4012 + }, + { + "epoch": 0.3015932661957012, + "grad_norm": 1.9006076543757215, + "learning_rate": 3.2751030425126616e-06, + "loss": 1.0694, + "step": 4013 + }, + { + "epoch": 0.3016684202615362, + "grad_norm": 1.840815649861469, + "learning_rate": 3.2747279377563616e-06, + "loss": 1.0203, + "step": 4014 + }, + { + "epoch": 0.3017435743273711, + "grad_norm": 3.866208347474821, + "learning_rate": 3.2743527574675507e-06, + "loss": 1.0446, + "step": 4015 + }, + { + "epoch": 0.30181872839320606, + "grad_norm": 1.31615837321656, + "learning_rate": 3.2739775016684584e-06, + "loss": 1.0281, + "step": 4016 + }, + { + "epoch": 0.30189388245904103, + "grad_norm": 2.020446923925472, + "learning_rate": 3.2736021703813214e-06, + "loss": 1.0275, + "step": 4017 + }, + { + "epoch": 0.301969036524876, + "grad_norm": 1.7263275953306623, + "learning_rate": 3.2732267636283782e-06, + "loss": 1.0083, + "step": 4018 + }, + { + "epoch": 0.30204419059071097, + "grad_norm": 0.7339800726422893, + "learning_rate": 3.2728512814318744e-06, + "loss": 0.8945, + "step": 4019 + }, + { + "epoch": 0.30211934465654594, + "grad_norm": 1.5003335899709331, + "learning_rate": 3.2724757238140572e-06, + "loss": 0.9667, + "step": 4020 + }, + { + "epoch": 0.30219449872238086, + "grad_norm": 1.7881066046644913, + "learning_rate": 3.2721000907971813e-06, + "loss": 1.0546, + "step": 4021 + }, + { + "epoch": 0.3022696527882158, + "grad_norm": 1.9211263051169756, + "learning_rate": 3.2717243824035037e-06, + "loss": 0.8966, + "step": 4022 + }, + { + "epoch": 0.3023448068540508, + "grad_norm": 1.9454201051909463, + "learning_rate": 3.2713485986552865e-06, + "loss": 0.9634, + "step": 4023 + }, + { + "epoch": 0.30241996091988577, + "grad_norm": 3.024558478046272, + "learning_rate": 3.2709727395747974e-06, + "loss": 1.01, + "step": 4024 + }, + { + "epoch": 0.30249511498572074, + "grad_norm": 0.6561249205054542, + "learning_rate": 3.2705968051843053e-06, + "loss": 0.7712, + "step": 4025 + }, + { + "epoch": 0.3025702690515557, + "grad_norm": 1.4203650872931128, + "learning_rate": 3.270220795506088e-06, + "loss": 1.0069, + "step": 4026 + }, + { + "epoch": 0.3026454231173907, + "grad_norm": 2.0810383044653693, + "learning_rate": 3.269844710562424e-06, + "loss": 1.047, + "step": 4027 + }, + { + "epoch": 0.3027205771832256, + "grad_norm": 2.5845460403837426, + "learning_rate": 3.269468550375599e-06, + "loss": 0.9497, + "step": 4028 + }, + { + "epoch": 0.30279573124906056, + "grad_norm": 1.6088641939337498, + "learning_rate": 3.2690923149679008e-06, + "loss": 0.961, + "step": 4029 + }, + { + "epoch": 0.30287088531489553, + "grad_norm": 1.5621387827099529, + "learning_rate": 3.268716004361623e-06, + "loss": 0.967, + "step": 4030 + }, + { + "epoch": 0.3029460393807305, + "grad_norm": 2.0657079775212117, + "learning_rate": 3.2683396185790644e-06, + "loss": 1.0353, + "step": 4031 + }, + { + "epoch": 0.3030211934465655, + "grad_norm": 1.4347586822964937, + "learning_rate": 3.2679631576425265e-06, + "loss": 1.0085, + "step": 4032 + }, + { + "epoch": 0.30309634751240044, + "grad_norm": 2.308012242025241, + "learning_rate": 3.267586621574315e-06, + "loss": 0.983, + "step": 4033 + }, + { + "epoch": 0.30317150157823536, + "grad_norm": 1.6816638763273521, + "learning_rate": 3.2672100103967434e-06, + "loss": 0.9335, + "step": 4034 + }, + { + "epoch": 0.30324665564407033, + "grad_norm": 1.733841685185914, + "learning_rate": 3.266833324132126e-06, + "loss": 1.0411, + "step": 4035 + }, + { + "epoch": 0.3033218097099053, + "grad_norm": 2.011681891538859, + "learning_rate": 3.2664565628027833e-06, + "loss": 1.0905, + "step": 4036 + }, + { + "epoch": 0.30339696377574027, + "grad_norm": 4.129252005169021, + "learning_rate": 3.2660797264310393e-06, + "loss": 0.9528, + "step": 4037 + }, + { + "epoch": 0.30347211784157524, + "grad_norm": 1.6501999965027396, + "learning_rate": 3.2657028150392236e-06, + "loss": 1.026, + "step": 4038 + }, + { + "epoch": 0.3035472719074102, + "grad_norm": 4.5691446491336905, + "learning_rate": 3.2653258286496696e-06, + "loss": 1.025, + "step": 4039 + }, + { + "epoch": 0.3036224259732452, + "grad_norm": 2.839706900341335, + "learning_rate": 3.2649487672847143e-06, + "loss": 0.9917, + "step": 4040 + }, + { + "epoch": 0.3036975800390801, + "grad_norm": 1.560602791062533, + "learning_rate": 3.264571630966701e-06, + "loss": 1.0413, + "step": 4041 + }, + { + "epoch": 0.30377273410491507, + "grad_norm": 1.8666107691615317, + "learning_rate": 3.2641944197179767e-06, + "loss": 0.992, + "step": 4042 + }, + { + "epoch": 0.30384788817075004, + "grad_norm": 1.8999682875546577, + "learning_rate": 3.2638171335608914e-06, + "loss": 0.9352, + "step": 4043 + }, + { + "epoch": 0.303923042236585, + "grad_norm": 2.5523619726748956, + "learning_rate": 3.263439772517802e-06, + "loss": 1.0263, + "step": 4044 + }, + { + "epoch": 0.30399819630242, + "grad_norm": 2.755275088524927, + "learning_rate": 3.2630623366110673e-06, + "loss": 0.9314, + "step": 4045 + }, + { + "epoch": 0.30407335036825495, + "grad_norm": 1.777976122692594, + "learning_rate": 3.2626848258630533e-06, + "loss": 0.8741, + "step": 4046 + }, + { + "epoch": 0.30414850443408986, + "grad_norm": 1.6451583717659308, + "learning_rate": 3.2623072402961283e-06, + "loss": 0.8816, + "step": 4047 + }, + { + "epoch": 0.30422365849992483, + "grad_norm": 1.8863159128735012, + "learning_rate": 3.2619295799326657e-06, + "loss": 0.9428, + "step": 4048 + }, + { + "epoch": 0.3042988125657598, + "grad_norm": 2.6897671030753982, + "learning_rate": 3.2615518447950425e-06, + "loss": 1.0373, + "step": 4049 + }, + { + "epoch": 0.3043739666315948, + "grad_norm": 1.6580770959127928, + "learning_rate": 3.2611740349056424e-06, + "loss": 0.9812, + "step": 4050 + }, + { + "epoch": 0.30444912069742974, + "grad_norm": 1.3308162125736465, + "learning_rate": 3.2607961502868507e-06, + "loss": 1.0426, + "step": 4051 + }, + { + "epoch": 0.3045242747632647, + "grad_norm": 1.3976001470232724, + "learning_rate": 3.2604181909610595e-06, + "loss": 1.0026, + "step": 4052 + }, + { + "epoch": 0.3045994288290996, + "grad_norm": 1.461092935544977, + "learning_rate": 3.2600401569506646e-06, + "loss": 0.9565, + "step": 4053 + }, + { + "epoch": 0.3046745828949346, + "grad_norm": 2.1305258093852766, + "learning_rate": 3.2596620482780647e-06, + "loss": 0.9288, + "step": 4054 + }, + { + "epoch": 0.30474973696076957, + "grad_norm": 2.0191305268386244, + "learning_rate": 3.2592838649656648e-06, + "loss": 1.0132, + "step": 4055 + }, + { + "epoch": 0.30482489102660454, + "grad_norm": 1.6529803841979531, + "learning_rate": 3.2589056070358743e-06, + "loss": 1.021, + "step": 4056 + }, + { + "epoch": 0.3049000450924395, + "grad_norm": 1.6583725625898145, + "learning_rate": 3.258527274511105e-06, + "loss": 0.9902, + "step": 4057 + }, + { + "epoch": 0.3049751991582745, + "grad_norm": 1.4771281778534298, + "learning_rate": 3.2581488674137766e-06, + "loss": 1.0726, + "step": 4058 + }, + { + "epoch": 0.30505035322410945, + "grad_norm": 1.6929712257666412, + "learning_rate": 3.2577703857663094e-06, + "loss": 0.9309, + "step": 4059 + }, + { + "epoch": 0.30512550728994436, + "grad_norm": 1.7680374813163198, + "learning_rate": 3.2573918295911306e-06, + "loss": 0.9949, + "step": 4060 + }, + { + "epoch": 0.30520066135577933, + "grad_norm": 1.9035011859285886, + "learning_rate": 3.257013198910671e-06, + "loss": 0.9535, + "step": 4061 + }, + { + "epoch": 0.3052758154216143, + "grad_norm": 2.102387787390646, + "learning_rate": 3.256634493747366e-06, + "loss": 1.0407, + "step": 4062 + }, + { + "epoch": 0.3053509694874493, + "grad_norm": 2.1958772740561034, + "learning_rate": 3.256255714123655e-06, + "loss": 0.9739, + "step": 4063 + }, + { + "epoch": 0.30542612355328425, + "grad_norm": 1.9354506082778578, + "learning_rate": 3.255876860061983e-06, + "loss": 0.978, + "step": 4064 + }, + { + "epoch": 0.3055012776191192, + "grad_norm": 1.5040598605093354, + "learning_rate": 3.2554979315847977e-06, + "loss": 1.0039, + "step": 4065 + }, + { + "epoch": 0.30557643168495413, + "grad_norm": 2.2393312770739047, + "learning_rate": 3.255118928714552e-06, + "loss": 0.8915, + "step": 4066 + }, + { + "epoch": 0.3056515857507891, + "grad_norm": 2.273505350784908, + "learning_rate": 3.2547398514737038e-06, + "loss": 0.9247, + "step": 4067 + }, + { + "epoch": 0.30572673981662407, + "grad_norm": 1.6882760418092677, + "learning_rate": 3.2543606998847145e-06, + "loss": 0.9531, + "step": 4068 + }, + { + "epoch": 0.30580189388245904, + "grad_norm": 1.9210362038325948, + "learning_rate": 3.253981473970051e-06, + "loss": 0.9764, + "step": 4069 + }, + { + "epoch": 0.305877047948294, + "grad_norm": 1.6632343075973974, + "learning_rate": 3.253602173752183e-06, + "loss": 0.9331, + "step": 4070 + }, + { + "epoch": 0.305952202014129, + "grad_norm": 1.9506912937895586, + "learning_rate": 3.253222799253586e-06, + "loss": 0.9562, + "step": 4071 + }, + { + "epoch": 0.30602735607996395, + "grad_norm": 1.8263058210402037, + "learning_rate": 3.2528433504967394e-06, + "loss": 0.8989, + "step": 4072 + }, + { + "epoch": 0.30610251014579887, + "grad_norm": 2.2447525929696126, + "learning_rate": 3.252463827504126e-06, + "loss": 1.0498, + "step": 4073 + }, + { + "epoch": 0.30617766421163384, + "grad_norm": 1.6247882705081031, + "learning_rate": 3.2520842302982356e-06, + "loss": 0.975, + "step": 4074 + }, + { + "epoch": 0.3062528182774688, + "grad_norm": 1.9962991806593449, + "learning_rate": 3.2517045589015602e-06, + "loss": 0.9541, + "step": 4075 + }, + { + "epoch": 0.3063279723433038, + "grad_norm": 1.5293878404010466, + "learning_rate": 3.251324813336596e-06, + "loss": 0.9382, + "step": 4076 + }, + { + "epoch": 0.30640312640913875, + "grad_norm": 1.3144830527448306, + "learning_rate": 3.2509449936258452e-06, + "loss": 0.9631, + "step": 4077 + }, + { + "epoch": 0.3064782804749737, + "grad_norm": 1.7416143458172002, + "learning_rate": 3.2505650997918127e-06, + "loss": 1.0781, + "step": 4078 + }, + { + "epoch": 0.30655343454080863, + "grad_norm": 1.4110391235929416, + "learning_rate": 3.2501851318570103e-06, + "loss": 0.9194, + "step": 4079 + }, + { + "epoch": 0.3066285886066436, + "grad_norm": 1.6632858398144825, + "learning_rate": 3.249805089843951e-06, + "loss": 1.0318, + "step": 4080 + }, + { + "epoch": 0.3067037426724786, + "grad_norm": 1.4852330990112081, + "learning_rate": 3.249424973775155e-06, + "loss": 0.9549, + "step": 4081 + }, + { + "epoch": 0.30677889673831354, + "grad_norm": 1.473868440328913, + "learning_rate": 3.249044783673144e-06, + "loss": 0.9191, + "step": 4082 + }, + { + "epoch": 0.3068540508041485, + "grad_norm": 1.96495835442128, + "learning_rate": 3.2486645195604466e-06, + "loss": 1.0182, + "step": 4083 + }, + { + "epoch": 0.3069292048699835, + "grad_norm": 1.9388800597452853, + "learning_rate": 3.2482841814595954e-06, + "loss": 0.9863, + "step": 4084 + }, + { + "epoch": 0.30700435893581846, + "grad_norm": 1.8876467489832793, + "learning_rate": 3.247903769393127e-06, + "loss": 0.9927, + "step": 4085 + }, + { + "epoch": 0.30707951300165337, + "grad_norm": 5.558336735866864, + "learning_rate": 3.247523283383581e-06, + "loss": 1.0372, + "step": 4086 + }, + { + "epoch": 0.30715466706748834, + "grad_norm": 1.4050604133242608, + "learning_rate": 3.2471427234535034e-06, + "loss": 1.0531, + "step": 4087 + }, + { + "epoch": 0.3072298211333233, + "grad_norm": 2.488122093584193, + "learning_rate": 3.246762089625444e-06, + "loss": 0.8962, + "step": 4088 + }, + { + "epoch": 0.3073049751991583, + "grad_norm": 1.9079281658629446, + "learning_rate": 3.2463813819219565e-06, + "loss": 1.0918, + "step": 4089 + }, + { + "epoch": 0.30738012926499325, + "grad_norm": 1.6976860061333905, + "learning_rate": 3.2460006003655993e-06, + "loss": 1.1196, + "step": 4090 + }, + { + "epoch": 0.3074552833308282, + "grad_norm": 1.7197156101031845, + "learning_rate": 3.245619744978936e-06, + "loss": 0.9133, + "step": 4091 + }, + { + "epoch": 0.30753043739666314, + "grad_norm": 1.6011719018024593, + "learning_rate": 3.2452388157845322e-06, + "loss": 0.886, + "step": 4092 + }, + { + "epoch": 0.3076055914624981, + "grad_norm": 1.6390077885459564, + "learning_rate": 3.2448578128049607e-06, + "loss": 1.1377, + "step": 4093 + }, + { + "epoch": 0.3076807455283331, + "grad_norm": 0.874822632660275, + "learning_rate": 3.2444767360627964e-06, + "loss": 0.9185, + "step": 4094 + }, + { + "epoch": 0.30775589959416805, + "grad_norm": 2.2811197740147007, + "learning_rate": 3.2440955855806203e-06, + "loss": 0.953, + "step": 4095 + }, + { + "epoch": 0.307831053660003, + "grad_norm": 1.5406999736130997, + "learning_rate": 3.2437143613810173e-06, + "loss": 0.9833, + "step": 4096 + }, + { + "epoch": 0.307906207725838, + "grad_norm": 2.08418857821732, + "learning_rate": 3.2433330634865757e-06, + "loss": 1.0714, + "step": 4097 + }, + { + "epoch": 0.3079813617916729, + "grad_norm": 3.139327507168633, + "learning_rate": 3.242951691919888e-06, + "loss": 0.8725, + "step": 4098 + }, + { + "epoch": 0.3080565158575079, + "grad_norm": 1.9177196830428849, + "learning_rate": 3.242570246703554e-06, + "loss": 0.8861, + "step": 4099 + }, + { + "epoch": 0.30813166992334284, + "grad_norm": 1.998657849103849, + "learning_rate": 3.242188727860174e-06, + "loss": 0.9634, + "step": 4100 + }, + { + "epoch": 0.3082068239891778, + "grad_norm": 1.5932075474912033, + "learning_rate": 3.2418071354123566e-06, + "loss": 0.9304, + "step": 4101 + }, + { + "epoch": 0.3082819780550128, + "grad_norm": 1.6614427316664573, + "learning_rate": 3.2414254693827098e-06, + "loss": 1.0025, + "step": 4102 + }, + { + "epoch": 0.30835713212084775, + "grad_norm": 1.508500253633725, + "learning_rate": 3.2410437297938512e-06, + "loss": 0.9554, + "step": 4103 + }, + { + "epoch": 0.3084322861866827, + "grad_norm": 1.6364922196028175, + "learning_rate": 3.240661916668399e-06, + "loss": 0.9558, + "step": 4104 + }, + { + "epoch": 0.30850744025251764, + "grad_norm": 0.8636502474143432, + "learning_rate": 3.2402800300289773e-06, + "loss": 0.9502, + "step": 4105 + }, + { + "epoch": 0.3085825943183526, + "grad_norm": 6.036553931500979, + "learning_rate": 3.2398980698982143e-06, + "loss": 0.8533, + "step": 4106 + }, + { + "epoch": 0.3086577483841876, + "grad_norm": 2.0829518413937644, + "learning_rate": 3.2395160362987432e-06, + "loss": 1.0253, + "step": 4107 + }, + { + "epoch": 0.30873290245002255, + "grad_norm": 1.6328060752340705, + "learning_rate": 3.2391339292532004e-06, + "loss": 1.027, + "step": 4108 + }, + { + "epoch": 0.3088080565158575, + "grad_norm": 1.9389550682634151, + "learning_rate": 3.2387517487842273e-06, + "loss": 1.0934, + "step": 4109 + }, + { + "epoch": 0.3088832105816925, + "grad_norm": 1.1937311720237083, + "learning_rate": 3.2383694949144693e-06, + "loss": 1.0038, + "step": 4110 + }, + { + "epoch": 0.3089583646475274, + "grad_norm": 1.5825541821163995, + "learning_rate": 3.2379871676665767e-06, + "loss": 1.0744, + "step": 4111 + }, + { + "epoch": 0.3090335187133624, + "grad_norm": 1.5647444055021162, + "learning_rate": 3.237604767063204e-06, + "loss": 1.0186, + "step": 4112 + }, + { + "epoch": 0.30910867277919735, + "grad_norm": 1.5960911777380382, + "learning_rate": 3.23722229312701e-06, + "loss": 0.9921, + "step": 4113 + }, + { + "epoch": 0.3091838268450323, + "grad_norm": 1.5660443351627804, + "learning_rate": 3.2368397458806573e-06, + "loss": 0.9582, + "step": 4114 + }, + { + "epoch": 0.3092589809108673, + "grad_norm": 1.6874931476595179, + "learning_rate": 3.236457125346814e-06, + "loss": 0.9442, + "step": 4115 + }, + { + "epoch": 0.30933413497670226, + "grad_norm": 1.757621246105993, + "learning_rate": 3.23607443154815e-06, + "loss": 0.978, + "step": 4116 + }, + { + "epoch": 0.3094092890425372, + "grad_norm": 2.059058352310049, + "learning_rate": 3.2356916645073444e-06, + "loss": 1.0197, + "step": 4117 + }, + { + "epoch": 0.30948444310837214, + "grad_norm": 1.4262965277154482, + "learning_rate": 3.2353088242470744e-06, + "loss": 0.8437, + "step": 4118 + }, + { + "epoch": 0.3095595971742071, + "grad_norm": 1.9476982715050426, + "learning_rate": 3.2349259107900267e-06, + "loss": 0.9753, + "step": 4119 + }, + { + "epoch": 0.3096347512400421, + "grad_norm": 5.046447546841003, + "learning_rate": 3.2345429241588902e-06, + "loss": 0.966, + "step": 4120 + }, + { + "epoch": 0.30970990530587705, + "grad_norm": 2.3003480564971626, + "learning_rate": 3.234159864376358e-06, + "loss": 0.9796, + "step": 4121 + }, + { + "epoch": 0.309785059371712, + "grad_norm": 2.51080884811595, + "learning_rate": 3.233776731465128e-06, + "loss": 0.9072, + "step": 4122 + }, + { + "epoch": 0.309860213437547, + "grad_norm": 2.397143583406416, + "learning_rate": 3.233393525447902e-06, + "loss": 0.9256, + "step": 4123 + }, + { + "epoch": 0.3099353675033819, + "grad_norm": 1.7070110677752508, + "learning_rate": 3.2330102463473867e-06, + "loss": 1.0082, + "step": 4124 + }, + { + "epoch": 0.3100105215692169, + "grad_norm": 2.442020625821673, + "learning_rate": 3.2326268941862927e-06, + "loss": 1.0162, + "step": 4125 + }, + { + "epoch": 0.31008567563505185, + "grad_norm": 1.616889004641156, + "learning_rate": 3.2322434689873353e-06, + "loss": 1.0718, + "step": 4126 + }, + { + "epoch": 0.3101608297008868, + "grad_norm": 1.5070877309827666, + "learning_rate": 3.231859970773234e-06, + "loss": 1.0279, + "step": 4127 + }, + { + "epoch": 0.3102359837667218, + "grad_norm": 1.9287379472990094, + "learning_rate": 3.231476399566712e-06, + "loss": 0.9454, + "step": 4128 + }, + { + "epoch": 0.31031113783255676, + "grad_norm": 1.9302971992547058, + "learning_rate": 3.231092755390498e-06, + "loss": 1.0924, + "step": 4129 + }, + { + "epoch": 0.31038629189839173, + "grad_norm": 0.7208886463427433, + "learning_rate": 3.230709038267324e-06, + "loss": 0.7892, + "step": 4130 + }, + { + "epoch": 0.31046144596422665, + "grad_norm": 1.9479890357733132, + "learning_rate": 3.2303252482199265e-06, + "loss": 0.9639, + "step": 4131 + }, + { + "epoch": 0.3105366000300616, + "grad_norm": 1.7535393210016912, + "learning_rate": 3.2299413852710466e-06, + "loss": 0.9072, + "step": 4132 + }, + { + "epoch": 0.3106117540958966, + "grad_norm": 1.814290116361635, + "learning_rate": 3.2295574494434307e-06, + "loss": 0.9934, + "step": 4133 + }, + { + "epoch": 0.31068690816173156, + "grad_norm": 1.864499507561289, + "learning_rate": 3.229173440759827e-06, + "loss": 1.0445, + "step": 4134 + }, + { + "epoch": 0.3107620622275665, + "grad_norm": 2.246046301299095, + "learning_rate": 3.22878935924299e-06, + "loss": 1.0283, + "step": 4135 + }, + { + "epoch": 0.3108372162934015, + "grad_norm": 2.560513540555806, + "learning_rate": 3.2284052049156783e-06, + "loss": 0.9539, + "step": 4136 + }, + { + "epoch": 0.3109123703592364, + "grad_norm": 2.9607511678157232, + "learning_rate": 3.2280209778006545e-06, + "loss": 1.0205, + "step": 4137 + }, + { + "epoch": 0.3109875244250714, + "grad_norm": 3.118416527175754, + "learning_rate": 3.227636677920685e-06, + "loss": 1.0458, + "step": 4138 + }, + { + "epoch": 0.31106267849090635, + "grad_norm": 1.0316174748288502, + "learning_rate": 3.227252305298542e-06, + "loss": 0.9704, + "step": 4139 + }, + { + "epoch": 0.3111378325567413, + "grad_norm": 1.7126141698049504, + "learning_rate": 3.2268678599570002e-06, + "loss": 1.0619, + "step": 4140 + }, + { + "epoch": 0.3112129866225763, + "grad_norm": 3.853765449604317, + "learning_rate": 3.2264833419188397e-06, + "loss": 1.008, + "step": 4141 + }, + { + "epoch": 0.31128814068841126, + "grad_norm": 3.0176271107472177, + "learning_rate": 3.2260987512068443e-06, + "loss": 0.8468, + "step": 4142 + }, + { + "epoch": 0.3113632947542462, + "grad_norm": 1.581797490687158, + "learning_rate": 3.225714087843803e-06, + "loss": 1.0441, + "step": 4143 + }, + { + "epoch": 0.31143844882008115, + "grad_norm": 1.6814307814050062, + "learning_rate": 3.225329351852509e-06, + "loss": 1.0211, + "step": 4144 + }, + { + "epoch": 0.3115136028859161, + "grad_norm": 2.4112073686706097, + "learning_rate": 3.2249445432557584e-06, + "loss": 0.9946, + "step": 4145 + }, + { + "epoch": 0.3115887569517511, + "grad_norm": 1.8368113933711636, + "learning_rate": 3.224559662076353e-06, + "loss": 0.9703, + "step": 4146 + }, + { + "epoch": 0.31166391101758606, + "grad_norm": 1.6209364481522293, + "learning_rate": 3.224174708337098e-06, + "loss": 0.987, + "step": 4147 + }, + { + "epoch": 0.31173906508342103, + "grad_norm": 1.654619529992313, + "learning_rate": 3.2237896820608047e-06, + "loss": 1.0137, + "step": 4148 + }, + { + "epoch": 0.311814219149256, + "grad_norm": 1.4939156157167393, + "learning_rate": 3.223404583270286e-06, + "loss": 1.1012, + "step": 4149 + }, + { + "epoch": 0.3118893732150909, + "grad_norm": 3.201786412851505, + "learning_rate": 3.223019411988361e-06, + "loss": 1.0697, + "step": 4150 + }, + { + "epoch": 0.3119645272809259, + "grad_norm": 2.8849011636659396, + "learning_rate": 3.2226341682378525e-06, + "loss": 1.0108, + "step": 4151 + }, + { + "epoch": 0.31203968134676086, + "grad_norm": 1.5832393601621875, + "learning_rate": 3.2222488520415876e-06, + "loss": 1.021, + "step": 4152 + }, + { + "epoch": 0.3121148354125958, + "grad_norm": 1.738072895035836, + "learning_rate": 3.221863463422399e-06, + "loss": 1.0678, + "step": 4153 + }, + { + "epoch": 0.3121899894784308, + "grad_norm": 1.822416800129143, + "learning_rate": 3.2214780024031204e-06, + "loss": 1.0321, + "step": 4154 + }, + { + "epoch": 0.31226514354426577, + "grad_norm": 1.763528980705958, + "learning_rate": 3.221092469006593e-06, + "loss": 0.9299, + "step": 4155 + }, + { + "epoch": 0.3123402976101007, + "grad_norm": 2.061437043910058, + "learning_rate": 3.220706863255661e-06, + "loss": 1.032, + "step": 4156 + }, + { + "epoch": 0.31241545167593565, + "grad_norm": 1.6267305842463693, + "learning_rate": 3.220321185173173e-06, + "loss": 1.0228, + "step": 4157 + }, + { + "epoch": 0.3124906057417706, + "grad_norm": 1.5558240560382037, + "learning_rate": 3.219935434781982e-06, + "loss": 0.9622, + "step": 4158 + }, + { + "epoch": 0.3125657598076056, + "grad_norm": 2.267936947580215, + "learning_rate": 3.2195496121049447e-06, + "loss": 1.0927, + "step": 4159 + }, + { + "epoch": 0.31264091387344056, + "grad_norm": 1.7625975859119605, + "learning_rate": 3.219163717164923e-06, + "loss": 0.9837, + "step": 4160 + }, + { + "epoch": 0.31271606793927553, + "grad_norm": 1.961303615967752, + "learning_rate": 3.218777749984782e-06, + "loss": 1.0603, + "step": 4161 + }, + { + "epoch": 0.3127912220051105, + "grad_norm": 1.967288489018181, + "learning_rate": 3.2183917105873934e-06, + "loss": 0.8512, + "step": 4162 + }, + { + "epoch": 0.3128663760709454, + "grad_norm": 2.0105419086982694, + "learning_rate": 3.21800559899563e-06, + "loss": 0.9429, + "step": 4163 + }, + { + "epoch": 0.3129415301367804, + "grad_norm": 2.464447908612129, + "learning_rate": 3.217619415232371e-06, + "loss": 1.0787, + "step": 4164 + }, + { + "epoch": 0.31301668420261536, + "grad_norm": 2.3826907267487982, + "learning_rate": 3.217233159320498e-06, + "loss": 1.0486, + "step": 4165 + }, + { + "epoch": 0.31309183826845033, + "grad_norm": 2.1664963679755322, + "learning_rate": 3.2168468312829005e-06, + "loss": 0.9362, + "step": 4166 + }, + { + "epoch": 0.3131669923342853, + "grad_norm": 2.457802655975747, + "learning_rate": 3.2164604311424677e-06, + "loss": 0.9825, + "step": 4167 + }, + { + "epoch": 0.31324214640012027, + "grad_norm": 0.7692836227507469, + "learning_rate": 3.2160739589220968e-06, + "loss": 0.8611, + "step": 4168 + }, + { + "epoch": 0.3133173004659552, + "grad_norm": 2.631174908289119, + "learning_rate": 3.2156874146446872e-06, + "loss": 0.9847, + "step": 4169 + }, + { + "epoch": 0.31339245453179015, + "grad_norm": 2.233735639956126, + "learning_rate": 3.2153007983331437e-06, + "loss": 1.081, + "step": 4170 + }, + { + "epoch": 0.3134676085976251, + "grad_norm": 2.582543026620173, + "learning_rate": 3.214914110010373e-06, + "loss": 1.0056, + "step": 4171 + }, + { + "epoch": 0.3135427626634601, + "grad_norm": 1.8748833302121357, + "learning_rate": 3.214527349699289e-06, + "loss": 1.0403, + "step": 4172 + }, + { + "epoch": 0.31361791672929507, + "grad_norm": 2.4288037032758805, + "learning_rate": 3.21414051742281e-06, + "loss": 1.0011, + "step": 4173 + }, + { + "epoch": 0.31369307079513004, + "grad_norm": 2.9340377857606885, + "learning_rate": 3.2137536132038552e-06, + "loss": 1.0308, + "step": 4174 + }, + { + "epoch": 0.31376822486096495, + "grad_norm": 2.144624301580111, + "learning_rate": 3.213366637065351e-06, + "loss": 1.0267, + "step": 4175 + }, + { + "epoch": 0.3138433789267999, + "grad_norm": 2.4320752349120913, + "learning_rate": 3.212979589030228e-06, + "loss": 1.0668, + "step": 4176 + }, + { + "epoch": 0.3139185329926349, + "grad_norm": 3.072339192241413, + "learning_rate": 3.212592469121419e-06, + "loss": 1.0859, + "step": 4177 + }, + { + "epoch": 0.31399368705846986, + "grad_norm": 1.7676570305678827, + "learning_rate": 3.2122052773618625e-06, + "loss": 1.1038, + "step": 4178 + }, + { + "epoch": 0.31406884112430483, + "grad_norm": 3.080363432757351, + "learning_rate": 3.211818013774502e-06, + "loss": 1.0469, + "step": 4179 + }, + { + "epoch": 0.3141439951901398, + "grad_norm": 1.5488221984980797, + "learning_rate": 3.211430678382284e-06, + "loss": 1.0439, + "step": 4180 + }, + { + "epoch": 0.3142191492559748, + "grad_norm": 1.3453232626312783, + "learning_rate": 3.211043271208159e-06, + "loss": 0.9307, + "step": 4181 + }, + { + "epoch": 0.3142943033218097, + "grad_norm": 0.7319995527213976, + "learning_rate": 3.2106557922750826e-06, + "loss": 0.8346, + "step": 4182 + }, + { + "epoch": 0.31436945738764466, + "grad_norm": 4.658450022937988, + "learning_rate": 3.210268241606015e-06, + "loss": 0.9355, + "step": 4183 + }, + { + "epoch": 0.3144446114534796, + "grad_norm": 0.7212902458505558, + "learning_rate": 3.2098806192239196e-06, + "loss": 0.8921, + "step": 4184 + }, + { + "epoch": 0.3145197655193146, + "grad_norm": 3.2879449753531014, + "learning_rate": 3.209492925151765e-06, + "loss": 0.8698, + "step": 4185 + }, + { + "epoch": 0.31459491958514957, + "grad_norm": 2.0180162543104387, + "learning_rate": 3.209105159412522e-06, + "loss": 0.9546, + "step": 4186 + }, + { + "epoch": 0.31467007365098454, + "grad_norm": 1.9666590410289448, + "learning_rate": 3.2087173220291695e-06, + "loss": 1.077, + "step": 4187 + }, + { + "epoch": 0.31474522771681945, + "grad_norm": 2.248232040932494, + "learning_rate": 3.2083294130246865e-06, + "loss": 0.9699, + "step": 4188 + }, + { + "epoch": 0.3148203817826544, + "grad_norm": 1.602116930757726, + "learning_rate": 3.207941432422059e-06, + "loss": 1.025, + "step": 4189 + }, + { + "epoch": 0.3148955358484894, + "grad_norm": 1.7430672242009189, + "learning_rate": 3.2075533802442757e-06, + "loss": 0.9413, + "step": 4190 + }, + { + "epoch": 0.31497068991432436, + "grad_norm": 0.7548477539764887, + "learning_rate": 3.2071652565143313e-06, + "loss": 0.9146, + "step": 4191 + }, + { + "epoch": 0.31504584398015933, + "grad_norm": 2.730135874179066, + "learning_rate": 3.206777061255223e-06, + "loss": 1.0083, + "step": 4192 + }, + { + "epoch": 0.3151209980459943, + "grad_norm": 1.9729991652558965, + "learning_rate": 3.206388794489952e-06, + "loss": 0.9994, + "step": 4193 + }, + { + "epoch": 0.3151961521118293, + "grad_norm": 2.040277345728138, + "learning_rate": 3.2060004562415258e-06, + "loss": 1.0331, + "step": 4194 + }, + { + "epoch": 0.3152713061776642, + "grad_norm": 1.4944469661447142, + "learning_rate": 3.205612046532954e-06, + "loss": 0.8911, + "step": 4195 + }, + { + "epoch": 0.31534646024349916, + "grad_norm": 2.4256478732140576, + "learning_rate": 3.2052235653872525e-06, + "loss": 1.1022, + "step": 4196 + }, + { + "epoch": 0.31542161430933413, + "grad_norm": 3.3623764469179496, + "learning_rate": 3.2048350128274395e-06, + "loss": 1.0185, + "step": 4197 + }, + { + "epoch": 0.3154967683751691, + "grad_norm": 2.5207905305953933, + "learning_rate": 3.2044463888765384e-06, + "loss": 1.048, + "step": 4198 + }, + { + "epoch": 0.31557192244100407, + "grad_norm": 1.5507119881563411, + "learning_rate": 3.204057693557576e-06, + "loss": 1.0606, + "step": 4199 + }, + { + "epoch": 0.31564707650683904, + "grad_norm": 1.521928480019369, + "learning_rate": 3.203668926893585e-06, + "loss": 1.0139, + "step": 4200 + }, + { + "epoch": 0.31572223057267396, + "grad_norm": 0.7138982471227066, + "learning_rate": 3.2032800889076014e-06, + "loss": 0.8671, + "step": 4201 + }, + { + "epoch": 0.3157973846385089, + "grad_norm": 1.5645015866206637, + "learning_rate": 3.2028911796226642e-06, + "loss": 0.932, + "step": 4202 + }, + { + "epoch": 0.3158725387043439, + "grad_norm": 0.7990726758825902, + "learning_rate": 3.2025021990618193e-06, + "loss": 0.9041, + "step": 4203 + }, + { + "epoch": 0.31594769277017887, + "grad_norm": 1.6992406383288514, + "learning_rate": 3.2021131472481135e-06, + "loss": 0.9604, + "step": 4204 + }, + { + "epoch": 0.31602284683601384, + "grad_norm": 1.3980240024039672, + "learning_rate": 3.2017240242046005e-06, + "loss": 0.9777, + "step": 4205 + }, + { + "epoch": 0.3160980009018488, + "grad_norm": 1.9388082455925058, + "learning_rate": 3.2013348299543382e-06, + "loss": 1.0541, + "step": 4206 + }, + { + "epoch": 0.3161731549676838, + "grad_norm": 1.8106004034234267, + "learning_rate": 3.2009455645203864e-06, + "loss": 0.9426, + "step": 4207 + }, + { + "epoch": 0.3162483090335187, + "grad_norm": 3.3619006221863597, + "learning_rate": 3.2005562279258113e-06, + "loss": 1.0231, + "step": 4208 + }, + { + "epoch": 0.31632346309935366, + "grad_norm": 2.1652320978798576, + "learning_rate": 3.200166820193682e-06, + "loss": 0.9175, + "step": 4209 + }, + { + "epoch": 0.31639861716518863, + "grad_norm": 1.500017324983363, + "learning_rate": 3.1997773413470736e-06, + "loss": 1.0348, + "step": 4210 + }, + { + "epoch": 0.3164737712310236, + "grad_norm": 1.725537509378977, + "learning_rate": 3.1993877914090632e-06, + "loss": 1.0118, + "step": 4211 + }, + { + "epoch": 0.3165489252968586, + "grad_norm": 2.08255622357403, + "learning_rate": 3.198998170402733e-06, + "loss": 1.038, + "step": 4212 + }, + { + "epoch": 0.31662407936269354, + "grad_norm": 3.9017719693797583, + "learning_rate": 3.1986084783511708e-06, + "loss": 0.9538, + "step": 4213 + }, + { + "epoch": 0.31669923342852846, + "grad_norm": 1.6517372551367586, + "learning_rate": 3.198218715277466e-06, + "loss": 1.0223, + "step": 4214 + }, + { + "epoch": 0.31677438749436343, + "grad_norm": 2.14120753917026, + "learning_rate": 3.1978288812047136e-06, + "loss": 0.9692, + "step": 4215 + }, + { + "epoch": 0.3168495415601984, + "grad_norm": 1.4866352741422406, + "learning_rate": 3.1974389761560137e-06, + "loss": 1.0242, + "step": 4216 + }, + { + "epoch": 0.31692469562603337, + "grad_norm": 2.083416962534531, + "learning_rate": 3.197049000154469e-06, + "loss": 0.9827, + "step": 4217 + }, + { + "epoch": 0.31699984969186834, + "grad_norm": 1.8805294519951468, + "learning_rate": 3.1966589532231876e-06, + "loss": 0.8765, + "step": 4218 + }, + { + "epoch": 0.3170750037577033, + "grad_norm": 1.842844207164866, + "learning_rate": 3.1962688353852805e-06, + "loss": 1.1216, + "step": 4219 + }, + { + "epoch": 0.3171501578235382, + "grad_norm": 1.655872661863934, + "learning_rate": 3.195878646663864e-06, + "loss": 0.9555, + "step": 4220 + }, + { + "epoch": 0.3172253118893732, + "grad_norm": 6.550789402573937, + "learning_rate": 3.195488387082059e-06, + "loss": 0.9841, + "step": 4221 + }, + { + "epoch": 0.31730046595520817, + "grad_norm": 1.7283096794243626, + "learning_rate": 3.1950980566629886e-06, + "loss": 0.8887, + "step": 4222 + }, + { + "epoch": 0.31737562002104314, + "grad_norm": 0.7920750434705228, + "learning_rate": 3.194707655429782e-06, + "loss": 0.8928, + "step": 4223 + }, + { + "epoch": 0.3174507740868781, + "grad_norm": 2.668124237668073, + "learning_rate": 3.1943171834055723e-06, + "loss": 1.077, + "step": 4224 + }, + { + "epoch": 0.3175259281527131, + "grad_norm": 1.6874516091649312, + "learning_rate": 3.1939266406134963e-06, + "loss": 1.0199, + "step": 4225 + }, + { + "epoch": 0.31760108221854805, + "grad_norm": 2.0230393183644955, + "learning_rate": 3.1935360270766945e-06, + "loss": 1.0059, + "step": 4226 + }, + { + "epoch": 0.31767623628438296, + "grad_norm": 1.7086853812109797, + "learning_rate": 3.1931453428183125e-06, + "loss": 1.0636, + "step": 4227 + }, + { + "epoch": 0.31775139035021793, + "grad_norm": 1.5274006474426505, + "learning_rate": 3.1927545878615005e-06, + "loss": 1.0542, + "step": 4228 + }, + { + "epoch": 0.3178265444160529, + "grad_norm": 2.608502636209669, + "learning_rate": 3.1923637622294123e-06, + "loss": 1.0475, + "step": 4229 + }, + { + "epoch": 0.3179016984818879, + "grad_norm": 1.5202745517177836, + "learning_rate": 3.191972865945205e-06, + "loss": 0.9618, + "step": 4230 + }, + { + "epoch": 0.31797685254772284, + "grad_norm": 1.5225511368806317, + "learning_rate": 3.191581899032041e-06, + "loss": 1.0088, + "step": 4231 + }, + { + "epoch": 0.3180520066135578, + "grad_norm": 1.6281536019241982, + "learning_rate": 3.1911908615130862e-06, + "loss": 0.9586, + "step": 4232 + }, + { + "epoch": 0.31812716067939273, + "grad_norm": 1.8405830210123013, + "learning_rate": 3.1907997534115118e-06, + "loss": 1.0147, + "step": 4233 + }, + { + "epoch": 0.3182023147452277, + "grad_norm": 1.5859937986707253, + "learning_rate": 3.190408574750492e-06, + "loss": 1.0469, + "step": 4234 + }, + { + "epoch": 0.31827746881106267, + "grad_norm": 1.9070676082034719, + "learning_rate": 3.1900173255532057e-06, + "loss": 1.0528, + "step": 4235 + }, + { + "epoch": 0.31835262287689764, + "grad_norm": 1.3745471902339081, + "learning_rate": 3.1896260058428364e-06, + "loss": 0.9535, + "step": 4236 + }, + { + "epoch": 0.3184277769427326, + "grad_norm": 1.496616202581948, + "learning_rate": 3.1892346156425704e-06, + "loss": 1.0323, + "step": 4237 + }, + { + "epoch": 0.3185029310085676, + "grad_norm": 2.5665993813782015, + "learning_rate": 3.1888431549755998e-06, + "loss": 1.0399, + "step": 4238 + }, + { + "epoch": 0.31857808507440255, + "grad_norm": 1.6841943647525246, + "learning_rate": 3.18845162386512e-06, + "loss": 0.9717, + "step": 4239 + }, + { + "epoch": 0.31865323914023747, + "grad_norm": 1.476506469309347, + "learning_rate": 3.1880600223343303e-06, + "loss": 1.0152, + "step": 4240 + }, + { + "epoch": 0.31872839320607244, + "grad_norm": 1.9854808939752042, + "learning_rate": 3.187668350406435e-06, + "loss": 1.0512, + "step": 4241 + }, + { + "epoch": 0.3188035472719074, + "grad_norm": 2.1061563306859084, + "learning_rate": 3.187276608104642e-06, + "loss": 1.0507, + "step": 4242 + }, + { + "epoch": 0.3188787013377424, + "grad_norm": 2.020827683378626, + "learning_rate": 3.1868847954521635e-06, + "loss": 1.0104, + "step": 4243 + }, + { + "epoch": 0.31895385540357735, + "grad_norm": 2.70749403835778, + "learning_rate": 3.1864929124722162e-06, + "loss": 0.9888, + "step": 4244 + }, + { + "epoch": 0.3190290094694123, + "grad_norm": 2.028646120619029, + "learning_rate": 3.1861009591880206e-06, + "loss": 1.0389, + "step": 4245 + }, + { + "epoch": 0.31910416353524723, + "grad_norm": 1.8310194007016556, + "learning_rate": 3.1857089356228015e-06, + "loss": 0.9496, + "step": 4246 + }, + { + "epoch": 0.3191793176010822, + "grad_norm": 1.6699467171946065, + "learning_rate": 3.185316841799787e-06, + "loss": 0.9938, + "step": 4247 + }, + { + "epoch": 0.3192544716669172, + "grad_norm": 2.0117143649451603, + "learning_rate": 3.1849246777422108e-06, + "loss": 1.0229, + "step": 4248 + }, + { + "epoch": 0.31932962573275214, + "grad_norm": 1.7543129905991648, + "learning_rate": 3.1845324434733104e-06, + "loss": 0.9986, + "step": 4249 + }, + { + "epoch": 0.3194047797985871, + "grad_norm": 1.5700943069318836, + "learning_rate": 3.1841401390163263e-06, + "loss": 1.0045, + "step": 4250 + }, + { + "epoch": 0.3194799338644221, + "grad_norm": 1.8343393715512932, + "learning_rate": 3.183747764394505e-06, + "loss": 1.0081, + "step": 4251 + }, + { + "epoch": 0.31955508793025705, + "grad_norm": 4.0039019626501435, + "learning_rate": 3.1833553196310956e-06, + "loss": 1.0312, + "step": 4252 + }, + { + "epoch": 0.31963024199609197, + "grad_norm": 1.4903370833094125, + "learning_rate": 3.1829628047493523e-06, + "loss": 0.9191, + "step": 4253 + }, + { + "epoch": 0.31970539606192694, + "grad_norm": 1.8830002180677745, + "learning_rate": 3.1825702197725325e-06, + "loss": 0.9676, + "step": 4254 + }, + { + "epoch": 0.3197805501277619, + "grad_norm": 1.6004529133156271, + "learning_rate": 3.1821775647239e-06, + "loss": 0.9461, + "step": 4255 + }, + { + "epoch": 0.3198557041935969, + "grad_norm": 2.1439386045718147, + "learning_rate": 3.1817848396267188e-06, + "loss": 0.9746, + "step": 4256 + }, + { + "epoch": 0.31993085825943185, + "grad_norm": 1.7333770492738563, + "learning_rate": 3.181392044504261e-06, + "loss": 1.0954, + "step": 4257 + }, + { + "epoch": 0.3200060123252668, + "grad_norm": 1.9148740955128773, + "learning_rate": 3.1809991793798e-06, + "loss": 0.9169, + "step": 4258 + }, + { + "epoch": 0.32008116639110173, + "grad_norm": 1.4958154006067261, + "learning_rate": 3.180606244276616e-06, + "loss": 0.9824, + "step": 4259 + }, + { + "epoch": 0.3201563204569367, + "grad_norm": 1.6622404053586561, + "learning_rate": 3.180213239217991e-06, + "loss": 0.938, + "step": 4260 + }, + { + "epoch": 0.3202314745227717, + "grad_norm": 2.0786724086647097, + "learning_rate": 3.1798201642272123e-06, + "loss": 0.9515, + "step": 4261 + }, + { + "epoch": 0.32030662858860665, + "grad_norm": 0.7661841549982761, + "learning_rate": 3.179427019327571e-06, + "loss": 0.8304, + "step": 4262 + }, + { + "epoch": 0.3203817826544416, + "grad_norm": 1.6084421657971142, + "learning_rate": 3.179033804542363e-06, + "loss": 1.0276, + "step": 4263 + }, + { + "epoch": 0.3204569367202766, + "grad_norm": 1.5086836595970368, + "learning_rate": 3.178640519894886e-06, + "loss": 1.0241, + "step": 4264 + }, + { + "epoch": 0.3205320907861115, + "grad_norm": 1.828435366115391, + "learning_rate": 3.178247165408446e-06, + "loss": 1.0708, + "step": 4265 + }, + { + "epoch": 0.32060724485194647, + "grad_norm": 1.6830043247731719, + "learning_rate": 3.1778537411063487e-06, + "loss": 0.9922, + "step": 4266 + }, + { + "epoch": 0.32068239891778144, + "grad_norm": 1.874044810814804, + "learning_rate": 3.1774602470119076e-06, + "loss": 1.0776, + "step": 4267 + }, + { + "epoch": 0.3207575529836164, + "grad_norm": 2.1106122944556454, + "learning_rate": 3.177066683148438e-06, + "loss": 1.0443, + "step": 4268 + }, + { + "epoch": 0.3208327070494514, + "grad_norm": 2.056580806199153, + "learning_rate": 3.17667304953926e-06, + "loss": 1.0539, + "step": 4269 + }, + { + "epoch": 0.32090786111528635, + "grad_norm": 1.4518261098327645, + "learning_rate": 3.176279346207698e-06, + "loss": 1.0176, + "step": 4270 + }, + { + "epoch": 0.3209830151811213, + "grad_norm": 1.7711695482804868, + "learning_rate": 3.1758855731770807e-06, + "loss": 1.0012, + "step": 4271 + }, + { + "epoch": 0.32105816924695624, + "grad_norm": 0.8185562588879833, + "learning_rate": 3.1754917304707405e-06, + "loss": 0.8818, + "step": 4272 + }, + { + "epoch": 0.3211333233127912, + "grad_norm": 0.59958303188646, + "learning_rate": 3.1750978181120136e-06, + "loss": 0.8188, + "step": 4273 + }, + { + "epoch": 0.3212084773786262, + "grad_norm": 2.7234403994310084, + "learning_rate": 3.1747038361242417e-06, + "loss": 0.9804, + "step": 4274 + }, + { + "epoch": 0.32128363144446115, + "grad_norm": 6.225565027538414, + "learning_rate": 3.1743097845307692e-06, + "loss": 0.9834, + "step": 4275 + }, + { + "epoch": 0.3213587855102961, + "grad_norm": 1.5851231631251879, + "learning_rate": 3.1739156633549445e-06, + "loss": 0.8858, + "step": 4276 + }, + { + "epoch": 0.3214339395761311, + "grad_norm": 1.730580896743443, + "learning_rate": 3.1735214726201223e-06, + "loss": 0.9721, + "step": 4277 + }, + { + "epoch": 0.321509093641966, + "grad_norm": 2.1532135744520535, + "learning_rate": 3.1731272123496588e-06, + "loss": 0.834, + "step": 4278 + }, + { + "epoch": 0.321584247707801, + "grad_norm": 1.6436646413940834, + "learning_rate": 3.1727328825669164e-06, + "loss": 0.9621, + "step": 4279 + }, + { + "epoch": 0.32165940177363594, + "grad_norm": 2.0814557452073275, + "learning_rate": 3.172338483295259e-06, + "loss": 1.0449, + "step": 4280 + }, + { + "epoch": 0.3217345558394709, + "grad_norm": 1.6636450158750289, + "learning_rate": 3.1719440145580578e-06, + "loss": 0.9588, + "step": 4281 + }, + { + "epoch": 0.3218097099053059, + "grad_norm": 1.4566594280933087, + "learning_rate": 3.1715494763786855e-06, + "loss": 1.1039, + "step": 4282 + }, + { + "epoch": 0.32188486397114086, + "grad_norm": 2.428741761653512, + "learning_rate": 3.1711548687805214e-06, + "loss": 0.8882, + "step": 4283 + }, + { + "epoch": 0.3219600180369758, + "grad_norm": 1.8025542362428844, + "learning_rate": 3.170760191786946e-06, + "loss": 1.0816, + "step": 4284 + }, + { + "epoch": 0.32203517210281074, + "grad_norm": 1.4827213914575013, + "learning_rate": 3.1703654454213473e-06, + "loss": 1.0266, + "step": 4285 + }, + { + "epoch": 0.3221103261686457, + "grad_norm": 1.6662443659144142, + "learning_rate": 3.169970629707113e-06, + "loss": 0.9698, + "step": 4286 + }, + { + "epoch": 0.3221854802344807, + "grad_norm": 1.8375567498818246, + "learning_rate": 3.1695757446676396e-06, + "loss": 1.1174, + "step": 4287 + }, + { + "epoch": 0.32226063430031565, + "grad_norm": 2.2467462537273084, + "learning_rate": 3.169180790326324e-06, + "loss": 0.9112, + "step": 4288 + }, + { + "epoch": 0.3223357883661506, + "grad_norm": 1.8206539897922023, + "learning_rate": 3.1687857667065698e-06, + "loss": 1.0091, + "step": 4289 + }, + { + "epoch": 0.3224109424319856, + "grad_norm": 2.2281910077305205, + "learning_rate": 3.1683906738317838e-06, + "loss": 0.9813, + "step": 4290 + }, + { + "epoch": 0.3224860964978205, + "grad_norm": 2.169057981406417, + "learning_rate": 3.167995511725375e-06, + "loss": 1.0711, + "step": 4291 + }, + { + "epoch": 0.3225612505636555, + "grad_norm": 1.9629156462180133, + "learning_rate": 3.1676002804107607e-06, + "loss": 0.9598, + "step": 4292 + }, + { + "epoch": 0.32263640462949045, + "grad_norm": 1.6950531510653355, + "learning_rate": 3.167204979911359e-06, + "loss": 1.0377, + "step": 4293 + }, + { + "epoch": 0.3227115586953254, + "grad_norm": 1.9140595728014926, + "learning_rate": 3.166809610250592e-06, + "loss": 0.9482, + "step": 4294 + }, + { + "epoch": 0.3227867127611604, + "grad_norm": 2.1828056100505857, + "learning_rate": 3.1664141714518876e-06, + "loss": 0.9996, + "step": 4295 + }, + { + "epoch": 0.32286186682699536, + "grad_norm": 1.445508046711363, + "learning_rate": 3.1660186635386773e-06, + "loss": 0.8227, + "step": 4296 + }, + { + "epoch": 0.32293702089283033, + "grad_norm": 2.034247548788543, + "learning_rate": 3.1656230865343964e-06, + "loss": 0.8519, + "step": 4297 + }, + { + "epoch": 0.32301217495866524, + "grad_norm": 1.3582488909414403, + "learning_rate": 3.165227440462484e-06, + "loss": 1.0269, + "step": 4298 + }, + { + "epoch": 0.3230873290245002, + "grad_norm": 0.8678251791533654, + "learning_rate": 3.164831725346383e-06, + "loss": 0.9347, + "step": 4299 + }, + { + "epoch": 0.3231624830903352, + "grad_norm": 2.989437263633114, + "learning_rate": 3.1644359412095432e-06, + "loss": 1.0709, + "step": 4300 + }, + { + "epoch": 0.32323763715617015, + "grad_norm": 2.0254842061538425, + "learning_rate": 3.164040088075414e-06, + "loss": 0.9882, + "step": 4301 + }, + { + "epoch": 0.3233127912220051, + "grad_norm": 1.6369118952079675, + "learning_rate": 3.1636441659674528e-06, + "loss": 1.0687, + "step": 4302 + }, + { + "epoch": 0.3233879452878401, + "grad_norm": 1.5798650534258085, + "learning_rate": 3.1632481749091185e-06, + "loss": 1.0458, + "step": 4303 + }, + { + "epoch": 0.323463099353675, + "grad_norm": 1.5241235215448883, + "learning_rate": 3.1628521149238757e-06, + "loss": 0.9926, + "step": 4304 + }, + { + "epoch": 0.32353825341951, + "grad_norm": 1.7028905724689385, + "learning_rate": 3.1624559860351917e-06, + "loss": 1.0657, + "step": 4305 + }, + { + "epoch": 0.32361340748534495, + "grad_norm": 1.7622156901370114, + "learning_rate": 3.1620597882665393e-06, + "loss": 0.9395, + "step": 4306 + }, + { + "epoch": 0.3236885615511799, + "grad_norm": 1.441426843015371, + "learning_rate": 3.1616635216413952e-06, + "loss": 1.0638, + "step": 4307 + }, + { + "epoch": 0.3237637156170149, + "grad_norm": 1.333701822899823, + "learning_rate": 3.161267186183239e-06, + "loss": 1.1002, + "step": 4308 + }, + { + "epoch": 0.32383886968284986, + "grad_norm": 1.544682885745174, + "learning_rate": 3.160870781915555e-06, + "loss": 0.9761, + "step": 4309 + }, + { + "epoch": 0.3239140237486848, + "grad_norm": 4.3439225635967995, + "learning_rate": 3.160474308861832e-06, + "loss": 1.1043, + "step": 4310 + }, + { + "epoch": 0.32398917781451975, + "grad_norm": 2.0529947596658236, + "learning_rate": 3.160077767045562e-06, + "loss": 1.0925, + "step": 4311 + }, + { + "epoch": 0.3240643318803547, + "grad_norm": 2.0423427811699777, + "learning_rate": 3.1596811564902426e-06, + "loss": 1.047, + "step": 4312 + }, + { + "epoch": 0.3241394859461897, + "grad_norm": 1.617028417456709, + "learning_rate": 3.159284477219374e-06, + "loss": 1.0012, + "step": 4313 + }, + { + "epoch": 0.32421464001202466, + "grad_norm": 3.574192468098386, + "learning_rate": 3.1588877292564606e-06, + "loss": 0.8688, + "step": 4314 + }, + { + "epoch": 0.3242897940778596, + "grad_norm": 1.4221503860904967, + "learning_rate": 3.1584909126250116e-06, + "loss": 1.0119, + "step": 4315 + }, + { + "epoch": 0.3243649481436946, + "grad_norm": 1.5943766839090665, + "learning_rate": 3.15809402734854e-06, + "loss": 0.9316, + "step": 4316 + }, + { + "epoch": 0.3244401022095295, + "grad_norm": 2.100056615702075, + "learning_rate": 3.1576970734505624e-06, + "loss": 1.0177, + "step": 4317 + }, + { + "epoch": 0.3245152562753645, + "grad_norm": 3.1122185821160846, + "learning_rate": 3.1573000509546004e-06, + "loss": 1.0475, + "step": 4318 + }, + { + "epoch": 0.32459041034119945, + "grad_norm": 2.1512097038294185, + "learning_rate": 3.1569029598841788e-06, + "loss": 0.9125, + "step": 4319 + }, + { + "epoch": 0.3246655644070344, + "grad_norm": 1.5192235326176744, + "learning_rate": 3.1565058002628268e-06, + "loss": 0.9374, + "step": 4320 + }, + { + "epoch": 0.3247407184728694, + "grad_norm": 4.603496151714222, + "learning_rate": 3.156108572114077e-06, + "loss": 0.9989, + "step": 4321 + }, + { + "epoch": 0.32481587253870436, + "grad_norm": 1.5637836524031825, + "learning_rate": 3.155711275461468e-06, + "loss": 0.9267, + "step": 4322 + }, + { + "epoch": 0.3248910266045393, + "grad_norm": 3.604605425696879, + "learning_rate": 3.15531391032854e-06, + "loss": 0.8976, + "step": 4323 + }, + { + "epoch": 0.32496618067037425, + "grad_norm": 1.850270718020801, + "learning_rate": 3.1549164767388386e-06, + "loss": 1.1168, + "step": 4324 + }, + { + "epoch": 0.3250413347362092, + "grad_norm": 2.3582135303742295, + "learning_rate": 3.1545189747159136e-06, + "loss": 1.0585, + "step": 4325 + }, + { + "epoch": 0.3251164888020442, + "grad_norm": 1.9360730238454615, + "learning_rate": 3.1541214042833187e-06, + "loss": 0.9616, + "step": 4326 + }, + { + "epoch": 0.32519164286787916, + "grad_norm": 1.8220758369943517, + "learning_rate": 3.153723765464611e-06, + "loss": 1.0134, + "step": 4327 + }, + { + "epoch": 0.32526679693371413, + "grad_norm": 2.0838047511477082, + "learning_rate": 3.1533260582833527e-06, + "loss": 0.9993, + "step": 4328 + }, + { + "epoch": 0.3253419509995491, + "grad_norm": 2.1107516847228966, + "learning_rate": 3.152928282763109e-06, + "loss": 0.8919, + "step": 4329 + }, + { + "epoch": 0.325417105065384, + "grad_norm": 1.8337521797068408, + "learning_rate": 3.152530438927449e-06, + "loss": 1.0751, + "step": 4330 + }, + { + "epoch": 0.325492259131219, + "grad_norm": 1.809458449684334, + "learning_rate": 3.1521325267999476e-06, + "loss": 0.9615, + "step": 4331 + }, + { + "epoch": 0.32556741319705396, + "grad_norm": 1.8356719129293568, + "learning_rate": 3.1517345464041817e-06, + "loss": 1.0918, + "step": 4332 + }, + { + "epoch": 0.3256425672628889, + "grad_norm": 3.2081033863712194, + "learning_rate": 3.1513364977637344e-06, + "loss": 1.0805, + "step": 4333 + }, + { + "epoch": 0.3257177213287239, + "grad_norm": 1.9026968939575948, + "learning_rate": 3.1509383809021905e-06, + "loss": 1.0082, + "step": 4334 + }, + { + "epoch": 0.32579287539455887, + "grad_norm": 2.075241497647821, + "learning_rate": 3.15054019584314e-06, + "loss": 1.0536, + "step": 4335 + }, + { + "epoch": 0.3258680294603938, + "grad_norm": 1.9873116457955577, + "learning_rate": 3.150141942610178e-06, + "loss": 1.004, + "step": 4336 + }, + { + "epoch": 0.32594318352622875, + "grad_norm": 2.126406148282212, + "learning_rate": 3.149743621226901e-06, + "loss": 1.0261, + "step": 4337 + }, + { + "epoch": 0.3260183375920637, + "grad_norm": 1.5489352599295831, + "learning_rate": 3.149345231716912e-06, + "loss": 1.0045, + "step": 4338 + }, + { + "epoch": 0.3260934916578987, + "grad_norm": 1.9855351098173775, + "learning_rate": 3.148946774103817e-06, + "loss": 1.0291, + "step": 4339 + }, + { + "epoch": 0.32616864572373366, + "grad_norm": 1.6925022723634868, + "learning_rate": 3.1485482484112257e-06, + "loss": 0.9539, + "step": 4340 + }, + { + "epoch": 0.32624379978956863, + "grad_norm": 2.024783596202368, + "learning_rate": 3.148149654662753e-06, + "loss": 0.8389, + "step": 4341 + }, + { + "epoch": 0.3263189538554036, + "grad_norm": 1.9924543613390604, + "learning_rate": 3.1477509928820165e-06, + "loss": 1.0321, + "step": 4342 + }, + { + "epoch": 0.3263941079212385, + "grad_norm": 2.540556482796447, + "learning_rate": 3.147352263092638e-06, + "loss": 0.9773, + "step": 4343 + }, + { + "epoch": 0.3264692619870735, + "grad_norm": 1.6720474011669848, + "learning_rate": 3.1469534653182453e-06, + "loss": 1.0465, + "step": 4344 + }, + { + "epoch": 0.32654441605290846, + "grad_norm": 1.540463811709963, + "learning_rate": 3.146554599582468e-06, + "loss": 1.0173, + "step": 4345 + }, + { + "epoch": 0.32661957011874343, + "grad_norm": 1.7241811080163116, + "learning_rate": 3.1461556659089397e-06, + "loss": 1.0041, + "step": 4346 + }, + { + "epoch": 0.3266947241845784, + "grad_norm": 2.353377139500568, + "learning_rate": 3.145756664321299e-06, + "loss": 1.0754, + "step": 4347 + }, + { + "epoch": 0.32676987825041337, + "grad_norm": 1.814704343817127, + "learning_rate": 3.145357594843189e-06, + "loss": 1.0263, + "step": 4348 + }, + { + "epoch": 0.3268450323162483, + "grad_norm": 1.6701335928928354, + "learning_rate": 3.1449584574982556e-06, + "loss": 0.9823, + "step": 4349 + }, + { + "epoch": 0.32692018638208326, + "grad_norm": 1.4862564209975002, + "learning_rate": 3.144559252310149e-06, + "loss": 0.9649, + "step": 4350 + }, + { + "epoch": 0.3269953404479182, + "grad_norm": 1.9275953557712315, + "learning_rate": 3.1441599793025243e-06, + "loss": 0.9987, + "step": 4351 + }, + { + "epoch": 0.3270704945137532, + "grad_norm": 2.229578057472214, + "learning_rate": 3.1437606384990396e-06, + "loss": 0.9285, + "step": 4352 + }, + { + "epoch": 0.32714564857958817, + "grad_norm": 1.4863003741968333, + "learning_rate": 3.1433612299233567e-06, + "loss": 0.9704, + "step": 4353 + }, + { + "epoch": 0.32722080264542314, + "grad_norm": 1.811485006330889, + "learning_rate": 3.1429617535991427e-06, + "loss": 0.8713, + "step": 4354 + }, + { + "epoch": 0.32729595671125805, + "grad_norm": 2.0204473959369826, + "learning_rate": 3.1425622095500685e-06, + "loss": 0.8815, + "step": 4355 + }, + { + "epoch": 0.327371110777093, + "grad_norm": 3.76915958133706, + "learning_rate": 3.1421625977998087e-06, + "loss": 1.0298, + "step": 4356 + }, + { + "epoch": 0.327446264842928, + "grad_norm": 1.4124890419864111, + "learning_rate": 3.1417629183720403e-06, + "loss": 0.9666, + "step": 4357 + }, + { + "epoch": 0.32752141890876296, + "grad_norm": 3.2920488224233173, + "learning_rate": 3.1413631712904476e-06, + "loss": 0.9582, + "step": 4358 + }, + { + "epoch": 0.32759657297459793, + "grad_norm": 2.1080932113553357, + "learning_rate": 3.140963356578716e-06, + "loss": 1.1105, + "step": 4359 + }, + { + "epoch": 0.3276717270404329, + "grad_norm": 2.1914311643873945, + "learning_rate": 3.1405634742605366e-06, + "loss": 1.1239, + "step": 4360 + }, + { + "epoch": 0.3277468811062679, + "grad_norm": 1.6821936075863981, + "learning_rate": 3.1401635243596043e-06, + "loss": 1.0799, + "step": 4361 + }, + { + "epoch": 0.3278220351721028, + "grad_norm": 2.920378276151545, + "learning_rate": 3.1397635068996167e-06, + "loss": 0.9556, + "step": 4362 + }, + { + "epoch": 0.32789718923793776, + "grad_norm": 1.4768188092636474, + "learning_rate": 3.139363421904277e-06, + "loss": 0.9239, + "step": 4363 + }, + { + "epoch": 0.32797234330377273, + "grad_norm": 0.7234196934661042, + "learning_rate": 3.138963269397292e-06, + "loss": 0.8058, + "step": 4364 + }, + { + "epoch": 0.3280474973696077, + "grad_norm": 1.8572010827630934, + "learning_rate": 3.1385630494023716e-06, + "loss": 1.0475, + "step": 4365 + }, + { + "epoch": 0.32812265143544267, + "grad_norm": 1.4773394442878325, + "learning_rate": 3.1381627619432307e-06, + "loss": 1.0191, + "step": 4366 + }, + { + "epoch": 0.32819780550127764, + "grad_norm": 1.9180113871199582, + "learning_rate": 3.1377624070435874e-06, + "loss": 1.0158, + "step": 4367 + }, + { + "epoch": 0.32827295956711255, + "grad_norm": 1.8646634970824318, + "learning_rate": 3.137361984727165e-06, + "loss": 1.0299, + "step": 4368 + }, + { + "epoch": 0.3283481136329475, + "grad_norm": 1.3613619697555968, + "learning_rate": 3.1369614950176903e-06, + "loss": 0.9452, + "step": 4369 + }, + { + "epoch": 0.3284232676987825, + "grad_norm": 1.6984357761911244, + "learning_rate": 3.1365609379388922e-06, + "loss": 1.0244, + "step": 4370 + }, + { + "epoch": 0.32849842176461747, + "grad_norm": 8.034220936387516, + "learning_rate": 3.1361603135145074e-06, + "loss": 0.9706, + "step": 4371 + }, + { + "epoch": 0.32857357583045244, + "grad_norm": 2.584910208878418, + "learning_rate": 3.135759621768273e-06, + "loss": 1.0403, + "step": 4372 + }, + { + "epoch": 0.3286487298962874, + "grad_norm": 1.808054569691023, + "learning_rate": 3.1353588627239317e-06, + "loss": 1.0597, + "step": 4373 + }, + { + "epoch": 0.3287238839621224, + "grad_norm": 1.804270047873906, + "learning_rate": 3.13495803640523e-06, + "loss": 0.9979, + "step": 4374 + }, + { + "epoch": 0.3287990380279573, + "grad_norm": 1.9494079620323967, + "learning_rate": 3.134557142835919e-06, + "loss": 0.8929, + "step": 4375 + }, + { + "epoch": 0.32887419209379226, + "grad_norm": 4.28513354953536, + "learning_rate": 3.134156182039753e-06, + "loss": 0.9364, + "step": 4376 + }, + { + "epoch": 0.32894934615962723, + "grad_norm": 1.6160190802618593, + "learning_rate": 3.13375515404049e-06, + "loss": 0.98, + "step": 4377 + }, + { + "epoch": 0.3290245002254622, + "grad_norm": 1.7301931743045793, + "learning_rate": 3.133354058861893e-06, + "loss": 1.0523, + "step": 4378 + }, + { + "epoch": 0.3290996542912972, + "grad_norm": 1.257628670284708, + "learning_rate": 3.1329528965277275e-06, + "loss": 1.0104, + "step": 4379 + }, + { + "epoch": 0.32917480835713214, + "grad_norm": 2.316830472392224, + "learning_rate": 3.1325516670617648e-06, + "loss": 1.0149, + "step": 4380 + }, + { + "epoch": 0.32924996242296706, + "grad_norm": 4.573242500300277, + "learning_rate": 3.132150370487779e-06, + "loss": 0.9861, + "step": 4381 + }, + { + "epoch": 0.329325116488802, + "grad_norm": 1.6051660702550825, + "learning_rate": 3.1317490068295486e-06, + "loss": 0.9816, + "step": 4382 + }, + { + "epoch": 0.329400270554637, + "grad_norm": 2.171973576126806, + "learning_rate": 3.131347576110855e-06, + "loss": 1.0609, + "step": 4383 + }, + { + "epoch": 0.32947542462047197, + "grad_norm": 2.677789844742526, + "learning_rate": 3.130946078355486e-06, + "loss": 1.0312, + "step": 4384 + }, + { + "epoch": 0.32955057868630694, + "grad_norm": 1.8392064636976397, + "learning_rate": 3.1305445135872318e-06, + "loss": 1.0298, + "step": 4385 + }, + { + "epoch": 0.3296257327521419, + "grad_norm": 1.4458769649697722, + "learning_rate": 3.1301428818298847e-06, + "loss": 0.9082, + "step": 4386 + }, + { + "epoch": 0.3297008868179769, + "grad_norm": 1.8908465547240905, + "learning_rate": 3.129741183107245e-06, + "loss": 1.0371, + "step": 4387 + }, + { + "epoch": 0.3297760408838118, + "grad_norm": 1.6165628009871817, + "learning_rate": 3.129339417443114e-06, + "loss": 0.9132, + "step": 4388 + }, + { + "epoch": 0.32985119494964676, + "grad_norm": 1.5559082605878793, + "learning_rate": 3.128937584861298e-06, + "loss": 1.0087, + "step": 4389 + }, + { + "epoch": 0.32992634901548173, + "grad_norm": 1.6488891272160535, + "learning_rate": 3.128535685385607e-06, + "loss": 0.914, + "step": 4390 + }, + { + "epoch": 0.3300015030813167, + "grad_norm": 1.7321168788034882, + "learning_rate": 3.1281337190398552e-06, + "loss": 1.039, + "step": 4391 + }, + { + "epoch": 0.3300766571471517, + "grad_norm": 2.9981194005885667, + "learning_rate": 3.1277316858478607e-06, + "loss": 0.9725, + "step": 4392 + }, + { + "epoch": 0.33015181121298665, + "grad_norm": 0.8488112636248465, + "learning_rate": 3.1273295858334454e-06, + "loss": 0.8081, + "step": 4393 + }, + { + "epoch": 0.33022696527882156, + "grad_norm": 2.3003524095648733, + "learning_rate": 3.1269274190204352e-06, + "loss": 1.0349, + "step": 4394 + }, + { + "epoch": 0.33030211934465653, + "grad_norm": 1.8407920126927784, + "learning_rate": 3.1265251854326613e-06, + "loss": 1.0169, + "step": 4395 + }, + { + "epoch": 0.3303772734104915, + "grad_norm": 1.8365297701380703, + "learning_rate": 3.126122885093955e-06, + "loss": 0.9361, + "step": 4396 + }, + { + "epoch": 0.33045242747632647, + "grad_norm": 1.968905972554242, + "learning_rate": 3.1257205180281555e-06, + "loss": 1.0198, + "step": 4397 + }, + { + "epoch": 0.33052758154216144, + "grad_norm": 10.199789860374613, + "learning_rate": 3.125318084259105e-06, + "loss": 1.0476, + "step": 4398 + }, + { + "epoch": 0.3306027356079964, + "grad_norm": 2.0158965644943776, + "learning_rate": 3.1249155838106493e-06, + "loss": 0.9414, + "step": 4399 + }, + { + "epoch": 0.3306778896738313, + "grad_norm": 4.328307440593028, + "learning_rate": 3.1245130167066373e-06, + "loss": 0.9747, + "step": 4400 + }, + { + "epoch": 0.3307530437396663, + "grad_norm": 1.644872199662532, + "learning_rate": 3.1241103829709234e-06, + "loss": 0.924, + "step": 4401 + }, + { + "epoch": 0.33082819780550127, + "grad_norm": 1.650929524396831, + "learning_rate": 3.123707682627364e-06, + "loss": 0.9149, + "step": 4402 + }, + { + "epoch": 0.33090335187133624, + "grad_norm": 1.9237389558372229, + "learning_rate": 3.1233049156998215e-06, + "loss": 0.8843, + "step": 4403 + }, + { + "epoch": 0.3309785059371712, + "grad_norm": 3.3528048829130728, + "learning_rate": 3.122902082212162e-06, + "loss": 1.0332, + "step": 4404 + }, + { + "epoch": 0.3310536600030062, + "grad_norm": 2.1947047794079513, + "learning_rate": 3.122499182188254e-06, + "loss": 0.9666, + "step": 4405 + }, + { + "epoch": 0.33112881406884115, + "grad_norm": 1.665583314069815, + "learning_rate": 3.1220962156519715e-06, + "loss": 0.9997, + "step": 4406 + }, + { + "epoch": 0.33120396813467606, + "grad_norm": 1.6713919253634577, + "learning_rate": 3.121693182627191e-06, + "loss": 1.039, + "step": 4407 + }, + { + "epoch": 0.33127912220051103, + "grad_norm": 1.7846494660677317, + "learning_rate": 3.1212900831377934e-06, + "loss": 1.0433, + "step": 4408 + }, + { + "epoch": 0.331354276266346, + "grad_norm": 1.4322201335267972, + "learning_rate": 3.1208869172076657e-06, + "loss": 0.9528, + "step": 4409 + }, + { + "epoch": 0.331429430332181, + "grad_norm": 2.7971102652883815, + "learning_rate": 3.120483684860696e-06, + "loss": 0.9584, + "step": 4410 + }, + { + "epoch": 0.33150458439801594, + "grad_norm": 0.7769809432153677, + "learning_rate": 3.1200803861207774e-06, + "loss": 0.8891, + "step": 4411 + }, + { + "epoch": 0.3315797384638509, + "grad_norm": 1.871457695656375, + "learning_rate": 3.1196770210118063e-06, + "loss": 1.0162, + "step": 4412 + }, + { + "epoch": 0.33165489252968583, + "grad_norm": 1.610739120567439, + "learning_rate": 3.1192735895576845e-06, + "loss": 1.0382, + "step": 4413 + }, + { + "epoch": 0.3317300465955208, + "grad_norm": 1.744058468241048, + "learning_rate": 3.118870091782316e-06, + "loss": 0.8912, + "step": 4414 + }, + { + "epoch": 0.33180520066135577, + "grad_norm": 1.8414138606047314, + "learning_rate": 3.118466527709611e-06, + "loss": 0.958, + "step": 4415 + }, + { + "epoch": 0.33188035472719074, + "grad_norm": 1.636378068450902, + "learning_rate": 3.1180628973634807e-06, + "loss": 0.9639, + "step": 4416 + }, + { + "epoch": 0.3319555087930257, + "grad_norm": 1.5791402138513355, + "learning_rate": 3.117659200767843e-06, + "loss": 0.9404, + "step": 4417 + }, + { + "epoch": 0.3320306628588607, + "grad_norm": 1.7693758932129584, + "learning_rate": 3.1172554379466176e-06, + "loss": 0.9184, + "step": 4418 + }, + { + "epoch": 0.33210581692469565, + "grad_norm": 2.0278538409501676, + "learning_rate": 3.1168516089237288e-06, + "loss": 0.9589, + "step": 4419 + }, + { + "epoch": 0.33218097099053057, + "grad_norm": 0.6819851067698366, + "learning_rate": 3.1164477137231054e-06, + "loss": 0.8125, + "step": 4420 + }, + { + "epoch": 0.33225612505636554, + "grad_norm": 1.77907348972493, + "learning_rate": 3.1160437523686806e-06, + "loss": 1.0414, + "step": 4421 + }, + { + "epoch": 0.3323312791222005, + "grad_norm": 4.787241454152531, + "learning_rate": 3.1156397248843896e-06, + "loss": 1.0433, + "step": 4422 + }, + { + "epoch": 0.3324064331880355, + "grad_norm": 1.4908703488139592, + "learning_rate": 3.1152356312941724e-06, + "loss": 1.0111, + "step": 4423 + }, + { + "epoch": 0.33248158725387045, + "grad_norm": 2.158285424229758, + "learning_rate": 3.114831471621974e-06, + "loss": 1.0475, + "step": 4424 + }, + { + "epoch": 0.3325567413197054, + "grad_norm": 1.7549895953008547, + "learning_rate": 3.1144272458917417e-06, + "loss": 0.842, + "step": 4425 + }, + { + "epoch": 0.33263189538554033, + "grad_norm": 1.9220683923197526, + "learning_rate": 3.114022954127427e-06, + "loss": 0.9383, + "step": 4426 + }, + { + "epoch": 0.3327070494513753, + "grad_norm": 1.4557755610170702, + "learning_rate": 3.1136185963529873e-06, + "loss": 0.9852, + "step": 4427 + }, + { + "epoch": 0.3327822035172103, + "grad_norm": 1.8907051384705755, + "learning_rate": 3.1132141725923812e-06, + "loss": 1.0594, + "step": 4428 + }, + { + "epoch": 0.33285735758304524, + "grad_norm": 2.0603553719141363, + "learning_rate": 3.1128096828695728e-06, + "loss": 1.0249, + "step": 4429 + }, + { + "epoch": 0.3329325116488802, + "grad_norm": 1.6794226859065653, + "learning_rate": 3.1124051272085286e-06, + "loss": 0.982, + "step": 4430 + }, + { + "epoch": 0.3330076657147152, + "grad_norm": 2.0484124153748713, + "learning_rate": 3.1120005056332216e-06, + "loss": 1.0081, + "step": 4431 + }, + { + "epoch": 0.33308281978055015, + "grad_norm": 1.8504995522177732, + "learning_rate": 3.111595818167627e-06, + "loss": 0.989, + "step": 4432 + }, + { + "epoch": 0.33315797384638507, + "grad_norm": 1.5006887920759024, + "learning_rate": 3.111191064835723e-06, + "loss": 0.9545, + "step": 4433 + }, + { + "epoch": 0.33323312791222004, + "grad_norm": 0.6818880872580662, + "learning_rate": 3.1107862456614932e-06, + "loss": 0.8157, + "step": 4434 + }, + { + "epoch": 0.333308281978055, + "grad_norm": 3.0113609091483013, + "learning_rate": 3.1103813606689253e-06, + "loss": 1.0097, + "step": 4435 + }, + { + "epoch": 0.33338343604389, + "grad_norm": 2.3085309583254183, + "learning_rate": 3.1099764098820096e-06, + "loss": 1.0159, + "step": 4436 + }, + { + "epoch": 0.33345859010972495, + "grad_norm": 1.8813894758766923, + "learning_rate": 3.1095713933247416e-06, + "loss": 1.0099, + "step": 4437 + }, + { + "epoch": 0.3335337441755599, + "grad_norm": 2.6568964452064248, + "learning_rate": 3.1091663110211188e-06, + "loss": 0.9622, + "step": 4438 + }, + { + "epoch": 0.33360889824139484, + "grad_norm": 2.0559842619163704, + "learning_rate": 3.1087611629951457e-06, + "loss": 0.83, + "step": 4439 + }, + { + "epoch": 0.3336840523072298, + "grad_norm": 1.6124454577780551, + "learning_rate": 3.1083559492708277e-06, + "loss": 1.0288, + "step": 4440 + }, + { + "epoch": 0.3337592063730648, + "grad_norm": 2.265642152918635, + "learning_rate": 3.1079506698721752e-06, + "loss": 0.9674, + "step": 4441 + }, + { + "epoch": 0.33383436043889975, + "grad_norm": 0.8035647142220591, + "learning_rate": 3.107545324823203e-06, + "loss": 0.9804, + "step": 4442 + }, + { + "epoch": 0.3339095145047347, + "grad_norm": 1.562139545825636, + "learning_rate": 3.1071399141479292e-06, + "loss": 0.9442, + "step": 4443 + }, + { + "epoch": 0.3339846685705697, + "grad_norm": 1.610473554164908, + "learning_rate": 3.1067344378703765e-06, + "loss": 0.9267, + "step": 4444 + }, + { + "epoch": 0.3340598226364046, + "grad_norm": 1.370413848017323, + "learning_rate": 3.10632889601457e-06, + "loss": 0.9321, + "step": 4445 + }, + { + "epoch": 0.3341349767022396, + "grad_norm": 1.6116060514336934, + "learning_rate": 3.10592328860454e-06, + "loss": 1.0458, + "step": 4446 + }, + { + "epoch": 0.33421013076807454, + "grad_norm": 1.9513448308252572, + "learning_rate": 3.10551761566432e-06, + "loss": 0.9143, + "step": 4447 + }, + { + "epoch": 0.3342852848339095, + "grad_norm": 1.9848347529276695, + "learning_rate": 3.1051118772179483e-06, + "loss": 1.0827, + "step": 4448 + }, + { + "epoch": 0.3343604388997445, + "grad_norm": 1.5701033419717396, + "learning_rate": 3.104706073289466e-06, + "loss": 1.0648, + "step": 4449 + }, + { + "epoch": 0.33443559296557945, + "grad_norm": 1.8187964023818777, + "learning_rate": 3.104300203902919e-06, + "loss": 1.0362, + "step": 4450 + }, + { + "epoch": 0.3345107470314144, + "grad_norm": 1.49292141905715, + "learning_rate": 3.1038942690823556e-06, + "loss": 1.0213, + "step": 4451 + }, + { + "epoch": 0.33458590109724934, + "grad_norm": 1.5828992181579251, + "learning_rate": 3.10348826885183e-06, + "loss": 0.9387, + "step": 4452 + }, + { + "epoch": 0.3346610551630843, + "grad_norm": 1.8796245129972713, + "learning_rate": 3.1030822032353997e-06, + "loss": 1.1273, + "step": 4453 + }, + { + "epoch": 0.3347362092289193, + "grad_norm": 2.554586239362375, + "learning_rate": 3.1026760722571236e-06, + "loss": 0.9303, + "step": 4454 + }, + { + "epoch": 0.33481136329475425, + "grad_norm": 1.6086660508761839, + "learning_rate": 3.1022698759410684e-06, + "loss": 1.0015, + "step": 4455 + }, + { + "epoch": 0.3348865173605892, + "grad_norm": 2.1242611666531004, + "learning_rate": 3.1018636143113022e-06, + "loss": 1.0727, + "step": 4456 + }, + { + "epoch": 0.3349616714264242, + "grad_norm": 1.9380814079886037, + "learning_rate": 3.1014572873918976e-06, + "loss": 0.9862, + "step": 4457 + }, + { + "epoch": 0.3350368254922591, + "grad_norm": 1.8071056860251637, + "learning_rate": 3.101050895206931e-06, + "loss": 1.0581, + "step": 4458 + }, + { + "epoch": 0.3351119795580941, + "grad_norm": 3.525827345915504, + "learning_rate": 3.100644437780482e-06, + "loss": 0.9828, + "step": 4459 + }, + { + "epoch": 0.33518713362392905, + "grad_norm": 1.3960878889732526, + "learning_rate": 3.100237915136636e-06, + "loss": 1.0731, + "step": 4460 + }, + { + "epoch": 0.335262287689764, + "grad_norm": 1.394018554473713, + "learning_rate": 3.0998313272994805e-06, + "loss": 0.8704, + "step": 4461 + }, + { + "epoch": 0.335337441755599, + "grad_norm": 1.5489922108266587, + "learning_rate": 3.0994246742931076e-06, + "loss": 1.0364, + "step": 4462 + }, + { + "epoch": 0.33541259582143396, + "grad_norm": 8.269806638807326, + "learning_rate": 3.099017956141612e-06, + "loss": 1.0076, + "step": 4463 + }, + { + "epoch": 0.3354877498872689, + "grad_norm": 1.7323157655374326, + "learning_rate": 3.098611172869094e-06, + "loss": 1.0004, + "step": 4464 + }, + { + "epoch": 0.33556290395310384, + "grad_norm": 1.8608619450398018, + "learning_rate": 3.0982043244996582e-06, + "loss": 0.9405, + "step": 4465 + }, + { + "epoch": 0.3356380580189388, + "grad_norm": 2.1517011787528686, + "learning_rate": 3.09779741105741e-06, + "loss": 0.9583, + "step": 4466 + }, + { + "epoch": 0.3357132120847738, + "grad_norm": 1.7061485260353086, + "learning_rate": 3.0973904325664615e-06, + "loss": 0.9756, + "step": 4467 + }, + { + "epoch": 0.33578836615060875, + "grad_norm": 2.0406547555900905, + "learning_rate": 3.0969833890509282e-06, + "loss": 0.9338, + "step": 4468 + }, + { + "epoch": 0.3358635202164437, + "grad_norm": 1.6397624336861187, + "learning_rate": 3.096576280534928e-06, + "loss": 0.9881, + "step": 4469 + }, + { + "epoch": 0.3359386742822787, + "grad_norm": 1.7372806712522841, + "learning_rate": 3.096169107042584e-06, + "loss": 0.9689, + "step": 4470 + }, + { + "epoch": 0.3360138283481136, + "grad_norm": 2.4393658465859502, + "learning_rate": 3.0957618685980233e-06, + "loss": 0.9894, + "step": 4471 + }, + { + "epoch": 0.3360889824139486, + "grad_norm": 0.7223707305206428, + "learning_rate": 3.0953545652253763e-06, + "loss": 0.8622, + "step": 4472 + }, + { + "epoch": 0.33616413647978355, + "grad_norm": 2.006202616384567, + "learning_rate": 3.094947196948776e-06, + "loss": 1.0353, + "step": 4473 + }, + { + "epoch": 0.3362392905456185, + "grad_norm": 2.8611429720918333, + "learning_rate": 3.0945397637923617e-06, + "loss": 0.9795, + "step": 4474 + }, + { + "epoch": 0.3363144446114535, + "grad_norm": 1.5774802222122937, + "learning_rate": 3.094132265780275e-06, + "loss": 1.0335, + "step": 4475 + }, + { + "epoch": 0.33638959867728846, + "grad_norm": 1.9548384817773683, + "learning_rate": 3.0937247029366623e-06, + "loss": 1.0438, + "step": 4476 + }, + { + "epoch": 0.33646475274312343, + "grad_norm": 1.625550176943745, + "learning_rate": 3.0933170752856723e-06, + "loss": 0.9771, + "step": 4477 + }, + { + "epoch": 0.33653990680895834, + "grad_norm": 1.9285344682916077, + "learning_rate": 3.0929093828514595e-06, + "loss": 1.0233, + "step": 4478 + }, + { + "epoch": 0.3366150608747933, + "grad_norm": 2.4853380364010618, + "learning_rate": 3.0925016256581805e-06, + "loss": 0.9533, + "step": 4479 + }, + { + "epoch": 0.3366902149406283, + "grad_norm": 2.1166036441261564, + "learning_rate": 3.092093803729997e-06, + "loss": 0.965, + "step": 4480 + }, + { + "epoch": 0.33676536900646326, + "grad_norm": 0.6879712786986494, + "learning_rate": 3.091685917091073e-06, + "loss": 0.7952, + "step": 4481 + }, + { + "epoch": 0.3368405230722982, + "grad_norm": 2.458059220198912, + "learning_rate": 3.0912779657655784e-06, + "loss": 0.9851, + "step": 4482 + }, + { + "epoch": 0.3369156771381332, + "grad_norm": 0.7315984140450884, + "learning_rate": 3.0908699497776864e-06, + "loss": 0.9063, + "step": 4483 + }, + { + "epoch": 0.3369908312039681, + "grad_norm": 1.5097128795969121, + "learning_rate": 3.0904618691515714e-06, + "loss": 0.9554, + "step": 4484 + }, + { + "epoch": 0.3370659852698031, + "grad_norm": 2.0164515257064033, + "learning_rate": 3.0900537239114157e-06, + "loss": 1.0382, + "step": 4485 + }, + { + "epoch": 0.33714113933563805, + "grad_norm": 1.5509591180761675, + "learning_rate": 3.089645514081402e-06, + "loss": 1.1063, + "step": 4486 + }, + { + "epoch": 0.337216293401473, + "grad_norm": 1.783657120636986, + "learning_rate": 3.08923723968572e-06, + "loss": 0.9703, + "step": 4487 + }, + { + "epoch": 0.337291447467308, + "grad_norm": 2.1059603709869474, + "learning_rate": 3.0888289007485605e-06, + "loss": 1.0613, + "step": 4488 + }, + { + "epoch": 0.33736660153314296, + "grad_norm": 1.9883741195963578, + "learning_rate": 3.0884204972941187e-06, + "loss": 1.0295, + "step": 4489 + }, + { + "epoch": 0.3374417555989779, + "grad_norm": 1.9421749933995578, + "learning_rate": 3.088012029346595e-06, + "loss": 1.0081, + "step": 4490 + }, + { + "epoch": 0.33751690966481285, + "grad_norm": 2.343708495726261, + "learning_rate": 3.087603496930192e-06, + "loss": 1.0344, + "step": 4491 + }, + { + "epoch": 0.3375920637306478, + "grad_norm": 5.2313050334410125, + "learning_rate": 3.087194900069117e-06, + "loss": 0.9905, + "step": 4492 + }, + { + "epoch": 0.3376672177964828, + "grad_norm": 1.5614643479380474, + "learning_rate": 3.0867862387875815e-06, + "loss": 0.9551, + "step": 4493 + }, + { + "epoch": 0.33774237186231776, + "grad_norm": 1.4452338944788024, + "learning_rate": 3.0863775131097995e-06, + "loss": 0.9417, + "step": 4494 + }, + { + "epoch": 0.33781752592815273, + "grad_norm": 1.8381722991471525, + "learning_rate": 3.0859687230599897e-06, + "loss": 1.0358, + "step": 4495 + }, + { + "epoch": 0.3378926799939877, + "grad_norm": 1.6284009117594307, + "learning_rate": 3.0855598686623745e-06, + "loss": 1.1321, + "step": 4496 + }, + { + "epoch": 0.3379678340598226, + "grad_norm": 1.6492837478355973, + "learning_rate": 3.085150949941181e-06, + "loss": 1.059, + "step": 4497 + }, + { + "epoch": 0.3380429881256576, + "grad_norm": 1.4867499375793125, + "learning_rate": 3.084741966920638e-06, + "loss": 0.9649, + "step": 4498 + }, + { + "epoch": 0.33811814219149255, + "grad_norm": 1.6618152900634215, + "learning_rate": 3.0843329196249794e-06, + "loss": 1.0339, + "step": 4499 + }, + { + "epoch": 0.3381932962573275, + "grad_norm": 1.64063241139509, + "learning_rate": 3.0839238080784435e-06, + "loss": 0.9925, + "step": 4500 + }, + { + "epoch": 0.3382684503231625, + "grad_norm": 1.9619316833221265, + "learning_rate": 3.083514632305271e-06, + "loss": 0.991, + "step": 4501 + }, + { + "epoch": 0.33834360438899747, + "grad_norm": 1.723541655034599, + "learning_rate": 3.0831053923297074e-06, + "loss": 0.9367, + "step": 4502 + }, + { + "epoch": 0.3384187584548324, + "grad_norm": 1.6919690048318756, + "learning_rate": 3.082696088176002e-06, + "loss": 0.9411, + "step": 4503 + }, + { + "epoch": 0.33849391252066735, + "grad_norm": 1.6948487662755205, + "learning_rate": 3.0822867198684073e-06, + "loss": 0.9865, + "step": 4504 + }, + { + "epoch": 0.3385690665865023, + "grad_norm": 1.538687878244875, + "learning_rate": 3.0818772874311804e-06, + "loss": 0.9641, + "step": 4505 + }, + { + "epoch": 0.3386442206523373, + "grad_norm": 1.7518392162227245, + "learning_rate": 3.081467790888581e-06, + "loss": 1.0024, + "step": 4506 + }, + { + "epoch": 0.33871937471817226, + "grad_norm": 2.242004706079204, + "learning_rate": 3.0810582302648743e-06, + "loss": 0.9816, + "step": 4507 + }, + { + "epoch": 0.33879452878400723, + "grad_norm": 1.7309201859398353, + "learning_rate": 3.0806486055843276e-06, + "loss": 0.9881, + "step": 4508 + }, + { + "epoch": 0.3388696828498422, + "grad_norm": 2.2831058008288445, + "learning_rate": 3.080238916871213e-06, + "loss": 0.9489, + "step": 4509 + }, + { + "epoch": 0.3389448369156771, + "grad_norm": 2.8155537451260124, + "learning_rate": 3.079829164149806e-06, + "loss": 1.0265, + "step": 4510 + }, + { + "epoch": 0.3390199909815121, + "grad_norm": 1.6444689813636237, + "learning_rate": 3.0794193474443866e-06, + "loss": 0.9777, + "step": 4511 + }, + { + "epoch": 0.33909514504734706, + "grad_norm": 1.6892104487241177, + "learning_rate": 3.0790094667792368e-06, + "loss": 1.0145, + "step": 4512 + }, + { + "epoch": 0.339170299113182, + "grad_norm": 2.518130742660871, + "learning_rate": 3.078599522178644e-06, + "loss": 0.9332, + "step": 4513 + }, + { + "epoch": 0.339245453179017, + "grad_norm": 2.3799330025547296, + "learning_rate": 3.0781895136669e-06, + "loss": 0.9105, + "step": 4514 + }, + { + "epoch": 0.33932060724485197, + "grad_norm": 2.0726362898203754, + "learning_rate": 3.077779441268299e-06, + "loss": 0.971, + "step": 4515 + }, + { + "epoch": 0.3393957613106869, + "grad_norm": 1.7279584945842124, + "learning_rate": 3.077369305007138e-06, + "loss": 0.9232, + "step": 4516 + }, + { + "epoch": 0.33947091537652185, + "grad_norm": 3.472299357934987, + "learning_rate": 3.07695910490772e-06, + "loss": 0.9168, + "step": 4517 + }, + { + "epoch": 0.3395460694423568, + "grad_norm": 2.6379873431164578, + "learning_rate": 3.076548840994352e-06, + "loss": 0.8271, + "step": 4518 + }, + { + "epoch": 0.3396212235081918, + "grad_norm": 2.595657750085441, + "learning_rate": 3.076138513291342e-06, + "loss": 0.8613, + "step": 4519 + }, + { + "epoch": 0.33969637757402676, + "grad_norm": 11.935227102828573, + "learning_rate": 3.0757281218230046e-06, + "loss": 0.9135, + "step": 4520 + }, + { + "epoch": 0.33977153163986173, + "grad_norm": 1.4904816948484214, + "learning_rate": 3.0753176666136575e-06, + "loss": 0.979, + "step": 4521 + }, + { + "epoch": 0.3398466857056967, + "grad_norm": 1.6363185493442218, + "learning_rate": 3.0749071476876203e-06, + "loss": 0.9967, + "step": 4522 + }, + { + "epoch": 0.3399218397715316, + "grad_norm": 1.6209830740410323, + "learning_rate": 3.0744965650692184e-06, + "loss": 1.0119, + "step": 4523 + }, + { + "epoch": 0.3399969938373666, + "grad_norm": 1.8310559244787488, + "learning_rate": 3.0740859187827807e-06, + "loss": 0.9517, + "step": 4524 + }, + { + "epoch": 0.34007214790320156, + "grad_norm": 1.9377882343133088, + "learning_rate": 3.0736752088526388e-06, + "loss": 1.0568, + "step": 4525 + }, + { + "epoch": 0.34014730196903653, + "grad_norm": 1.8570201290890518, + "learning_rate": 3.0732644353031304e-06, + "loss": 0.9948, + "step": 4526 + }, + { + "epoch": 0.3402224560348715, + "grad_norm": 1.3897889442209417, + "learning_rate": 3.072853598158594e-06, + "loss": 0.9767, + "step": 4527 + }, + { + "epoch": 0.34029761010070647, + "grad_norm": 1.646898464202211, + "learning_rate": 3.0724426974433737e-06, + "loss": 1.0206, + "step": 4528 + }, + { + "epoch": 0.3403727641665414, + "grad_norm": 1.5868999398410042, + "learning_rate": 3.0720317331818163e-06, + "loss": 0.9411, + "step": 4529 + }, + { + "epoch": 0.34044791823237636, + "grad_norm": 0.7929794611465819, + "learning_rate": 3.071620705398274e-06, + "loss": 0.8834, + "step": 4530 + }, + { + "epoch": 0.3405230722982113, + "grad_norm": 1.457556008756267, + "learning_rate": 3.0712096141171017e-06, + "loss": 1.0719, + "step": 4531 + }, + { + "epoch": 0.3405982263640463, + "grad_norm": 2.195497226748955, + "learning_rate": 3.070798459362658e-06, + "loss": 1.0985, + "step": 4532 + }, + { + "epoch": 0.34067338042988127, + "grad_norm": 2.429656000945086, + "learning_rate": 3.070387241159305e-06, + "loss": 0.8483, + "step": 4533 + }, + { + "epoch": 0.34074853449571624, + "grad_norm": 2.0596742635425183, + "learning_rate": 3.069975959531408e-06, + "loss": 1.0318, + "step": 4534 + }, + { + "epoch": 0.34082368856155115, + "grad_norm": 1.387638356639268, + "learning_rate": 3.0695646145033404e-06, + "loss": 0.9717, + "step": 4535 + }, + { + "epoch": 0.3408988426273861, + "grad_norm": 1.442426449703823, + "learning_rate": 3.0691532060994722e-06, + "loss": 0.9216, + "step": 4536 + }, + { + "epoch": 0.3409739966932211, + "grad_norm": 1.7378269244759657, + "learning_rate": 3.068741734344183e-06, + "loss": 0.9643, + "step": 4537 + }, + { + "epoch": 0.34104915075905606, + "grad_norm": 1.4534673800172366, + "learning_rate": 3.0683301992618538e-06, + "loss": 1.0268, + "step": 4538 + }, + { + "epoch": 0.34112430482489103, + "grad_norm": 2.297934774245033, + "learning_rate": 3.067918600876869e-06, + "loss": 1.0336, + "step": 4539 + }, + { + "epoch": 0.341199458890726, + "grad_norm": 0.8077923032497364, + "learning_rate": 3.067506939213617e-06, + "loss": 0.8824, + "step": 4540 + }, + { + "epoch": 0.341274612956561, + "grad_norm": 2.0371682441083423, + "learning_rate": 3.067095214296492e-06, + "loss": 1.0034, + "step": 4541 + }, + { + "epoch": 0.3413497670223959, + "grad_norm": 4.700141441976503, + "learning_rate": 3.066683426149889e-06, + "loss": 0.9872, + "step": 4542 + }, + { + "epoch": 0.34142492108823086, + "grad_norm": 1.6695769729029795, + "learning_rate": 3.066271574798209e-06, + "loss": 0.9938, + "step": 4543 + }, + { + "epoch": 0.34150007515406583, + "grad_norm": 2.397244930442186, + "learning_rate": 3.0658596602658548e-06, + "loss": 0.9142, + "step": 4544 + }, + { + "epoch": 0.3415752292199008, + "grad_norm": 1.8421185274291936, + "learning_rate": 3.0654476825772338e-06, + "loss": 1.0225, + "step": 4545 + }, + { + "epoch": 0.34165038328573577, + "grad_norm": 2.4325039843755007, + "learning_rate": 3.0650356417567586e-06, + "loss": 1.0011, + "step": 4546 + }, + { + "epoch": 0.34172553735157074, + "grad_norm": 1.8638702677165946, + "learning_rate": 3.064623537828843e-06, + "loss": 1.0381, + "step": 4547 + }, + { + "epoch": 0.34180069141740566, + "grad_norm": 2.1628249172366285, + "learning_rate": 3.0642113708179062e-06, + "loss": 1.007, + "step": 4548 + }, + { + "epoch": 0.3418758454832406, + "grad_norm": 1.8976630997956216, + "learning_rate": 3.0637991407483706e-06, + "loss": 1.0321, + "step": 4549 + }, + { + "epoch": 0.3419509995490756, + "grad_norm": 2.359177739063377, + "learning_rate": 3.0633868476446615e-06, + "loss": 0.9466, + "step": 4550 + }, + { + "epoch": 0.34202615361491057, + "grad_norm": 1.7105805882635556, + "learning_rate": 3.062974491531211e-06, + "loss": 1.0033, + "step": 4551 + }, + { + "epoch": 0.34210130768074554, + "grad_norm": 1.741664585617258, + "learning_rate": 3.06256207243245e-06, + "loss": 1.1033, + "step": 4552 + }, + { + "epoch": 0.3421764617465805, + "grad_norm": 1.797238255147091, + "learning_rate": 3.0621495903728177e-06, + "loss": 0.9697, + "step": 4553 + }, + { + "epoch": 0.3422516158124155, + "grad_norm": 1.4692758471071994, + "learning_rate": 3.061737045376756e-06, + "loss": 0.9064, + "step": 4554 + }, + { + "epoch": 0.3423267698782504, + "grad_norm": 1.6970621388691056, + "learning_rate": 3.061324437468708e-06, + "loss": 0.8927, + "step": 4555 + }, + { + "epoch": 0.34240192394408536, + "grad_norm": 1.382514274895973, + "learning_rate": 3.060911766673123e-06, + "loss": 1.1044, + "step": 4556 + }, + { + "epoch": 0.34247707800992033, + "grad_norm": 1.8377411119736267, + "learning_rate": 3.0604990330144537e-06, + "loss": 1.0104, + "step": 4557 + }, + { + "epoch": 0.3425522320757553, + "grad_norm": 1.6887179324072343, + "learning_rate": 3.0600862365171553e-06, + "loss": 1.0413, + "step": 4558 + }, + { + "epoch": 0.3426273861415903, + "grad_norm": 1.7752903741076533, + "learning_rate": 3.0596733772056884e-06, + "loss": 1.0134, + "step": 4559 + }, + { + "epoch": 0.34270254020742524, + "grad_norm": 2.4953490864287646, + "learning_rate": 3.0592604551045157e-06, + "loss": 0.9734, + "step": 4560 + }, + { + "epoch": 0.34277769427326016, + "grad_norm": 2.0906378547489934, + "learning_rate": 3.0588474702381055e-06, + "loss": 0.9833, + "step": 4561 + }, + { + "epoch": 0.34285284833909513, + "grad_norm": 2.007180556063019, + "learning_rate": 3.0584344226309277e-06, + "loss": 0.9868, + "step": 4562 + }, + { + "epoch": 0.3429280024049301, + "grad_norm": 1.9010780087136991, + "learning_rate": 3.0580213123074573e-06, + "loss": 0.9028, + "step": 4563 + }, + { + "epoch": 0.34300315647076507, + "grad_norm": 0.7143901365450398, + "learning_rate": 3.0576081392921723e-06, + "loss": 0.8665, + "step": 4564 + }, + { + "epoch": 0.34307831053660004, + "grad_norm": 1.3597244602756873, + "learning_rate": 3.057194903609556e-06, + "loss": 0.9953, + "step": 4565 + }, + { + "epoch": 0.343153464602435, + "grad_norm": 1.337899312457538, + "learning_rate": 3.056781605284093e-06, + "loss": 0.8086, + "step": 4566 + }, + { + "epoch": 0.34322861866827, + "grad_norm": 1.7922454239282284, + "learning_rate": 3.056368244340273e-06, + "loss": 1.0921, + "step": 4567 + }, + { + "epoch": 0.3433037727341049, + "grad_norm": 4.24085620764267, + "learning_rate": 3.05595482080259e-06, + "loss": 0.9898, + "step": 4568 + }, + { + "epoch": 0.34337892679993987, + "grad_norm": 1.7623459741913372, + "learning_rate": 3.05554133469554e-06, + "loss": 1.0429, + "step": 4569 + }, + { + "epoch": 0.34345408086577484, + "grad_norm": 1.9463346316165626, + "learning_rate": 3.055127786043624e-06, + "loss": 0.975, + "step": 4570 + }, + { + "epoch": 0.3435292349316098, + "grad_norm": 1.8493655405579368, + "learning_rate": 3.0547141748713463e-06, + "loss": 0.9431, + "step": 4571 + }, + { + "epoch": 0.3436043889974448, + "grad_norm": 2.500314502007248, + "learning_rate": 3.0543005012032152e-06, + "loss": 0.8894, + "step": 4572 + }, + { + "epoch": 0.34367954306327975, + "grad_norm": 1.6424823238635178, + "learning_rate": 3.0538867650637416e-06, + "loss": 1.0547, + "step": 4573 + }, + { + "epoch": 0.34375469712911466, + "grad_norm": 0.7353412683179277, + "learning_rate": 3.053472966477442e-06, + "loss": 0.9113, + "step": 4574 + }, + { + "epoch": 0.34382985119494963, + "grad_norm": 1.3605695277389196, + "learning_rate": 3.053059105468835e-06, + "loss": 1.0746, + "step": 4575 + }, + { + "epoch": 0.3439050052607846, + "grad_norm": 2.2152523356869755, + "learning_rate": 3.052645182062444e-06, + "loss": 0.9731, + "step": 4576 + }, + { + "epoch": 0.3439801593266196, + "grad_norm": 2.053609703133771, + "learning_rate": 3.052231196282795e-06, + "loss": 0.9796, + "step": 4577 + }, + { + "epoch": 0.34405531339245454, + "grad_norm": 0.7095558238525606, + "learning_rate": 3.051817148154418e-06, + "loss": 0.8392, + "step": 4578 + }, + { + "epoch": 0.3441304674582895, + "grad_norm": 1.4893874015083686, + "learning_rate": 3.0514030377018473e-06, + "loss": 0.9988, + "step": 4579 + }, + { + "epoch": 0.3442056215241244, + "grad_norm": 1.4038379752360033, + "learning_rate": 3.0509888649496204e-06, + "loss": 0.9297, + "step": 4580 + }, + { + "epoch": 0.3442807755899594, + "grad_norm": 2.140507381577073, + "learning_rate": 3.05057462992228e-06, + "loss": 1.1101, + "step": 4581 + }, + { + "epoch": 0.34435592965579437, + "grad_norm": 2.1363156791041282, + "learning_rate": 3.050160332644368e-06, + "loss": 0.9337, + "step": 4582 + }, + { + "epoch": 0.34443108372162934, + "grad_norm": 1.1970535762623342, + "learning_rate": 3.0497459731404364e-06, + "loss": 0.9854, + "step": 4583 + }, + { + "epoch": 0.3445062377874643, + "grad_norm": 1.5860819633345822, + "learning_rate": 3.049331551435035e-06, + "loss": 1.0104, + "step": 4584 + }, + { + "epoch": 0.3445813918532993, + "grad_norm": 1.647721410758853, + "learning_rate": 3.048917067552722e-06, + "loss": 0.9836, + "step": 4585 + }, + { + "epoch": 0.34465654591913425, + "grad_norm": 2.056849281616219, + "learning_rate": 3.0485025215180554e-06, + "loss": 0.983, + "step": 4586 + }, + { + "epoch": 0.34473169998496916, + "grad_norm": 2.007766544988741, + "learning_rate": 3.0480879133556e-06, + "loss": 1.0177, + "step": 4587 + }, + { + "epoch": 0.34480685405080413, + "grad_norm": 1.8015843253620911, + "learning_rate": 3.047673243089922e-06, + "loss": 0.9727, + "step": 4588 + }, + { + "epoch": 0.3448820081166391, + "grad_norm": 1.5800327817920348, + "learning_rate": 3.047258510745593e-06, + "loss": 1.0255, + "step": 4589 + }, + { + "epoch": 0.3449571621824741, + "grad_norm": 3.5576813886003014, + "learning_rate": 3.046843716347187e-06, + "loss": 0.9814, + "step": 4590 + }, + { + "epoch": 0.34503231624830905, + "grad_norm": 1.52561466462019, + "learning_rate": 3.046428859919281e-06, + "loss": 1.0559, + "step": 4591 + }, + { + "epoch": 0.345107470314144, + "grad_norm": 1.6319575649662916, + "learning_rate": 3.0460139414864593e-06, + "loss": 0.8723, + "step": 4592 + }, + { + "epoch": 0.34518262437997893, + "grad_norm": 1.992976312985607, + "learning_rate": 3.0455989610733057e-06, + "loss": 0.9874, + "step": 4593 + }, + { + "epoch": 0.3452577784458139, + "grad_norm": 1.70411723219873, + "learning_rate": 3.0451839187044095e-06, + "loss": 1.0065, + "step": 4594 + }, + { + "epoch": 0.34533293251164887, + "grad_norm": 1.523296956157329, + "learning_rate": 3.0447688144043636e-06, + "loss": 1.0062, + "step": 4595 + }, + { + "epoch": 0.34540808657748384, + "grad_norm": 3.4797894192703414, + "learning_rate": 3.0443536481977657e-06, + "loss": 0.9742, + "step": 4596 + }, + { + "epoch": 0.3454832406433188, + "grad_norm": 1.8605439174259377, + "learning_rate": 3.0439384201092145e-06, + "loss": 1.0375, + "step": 4597 + }, + { + "epoch": 0.3455583947091538, + "grad_norm": 1.4456918579421125, + "learning_rate": 3.0435231301633147e-06, + "loss": 0.9808, + "step": 4598 + }, + { + "epoch": 0.34563354877498875, + "grad_norm": 1.7485400649627412, + "learning_rate": 3.043107778384673e-06, + "loss": 0.9967, + "step": 4599 + }, + { + "epoch": 0.34570870284082367, + "grad_norm": 1.6399516585831497, + "learning_rate": 3.0426923647979016e-06, + "loss": 1.0505, + "step": 4600 + }, + { + "epoch": 0.34578385690665864, + "grad_norm": 0.8314670207742965, + "learning_rate": 3.042276889427615e-06, + "loss": 0.8962, + "step": 4601 + }, + { + "epoch": 0.3458590109724936, + "grad_norm": 1.7376323045813469, + "learning_rate": 3.041861352298431e-06, + "loss": 1.0082, + "step": 4602 + }, + { + "epoch": 0.3459341650383286, + "grad_norm": 1.8259718515121333, + "learning_rate": 3.0414457534349727e-06, + "loss": 1.0304, + "step": 4603 + }, + { + "epoch": 0.34600931910416355, + "grad_norm": 2.9470730889935726, + "learning_rate": 3.041030092861866e-06, + "loss": 0.8972, + "step": 4604 + }, + { + "epoch": 0.3460844731699985, + "grad_norm": 2.364539449903542, + "learning_rate": 3.0406143706037384e-06, + "loss": 0.9043, + "step": 4605 + }, + { + "epoch": 0.34615962723583343, + "grad_norm": 1.756712505760723, + "learning_rate": 3.040198586685226e-06, + "loss": 0.9935, + "step": 4606 + }, + { + "epoch": 0.3462347813016684, + "grad_norm": 1.5962206072629126, + "learning_rate": 3.0397827411309632e-06, + "loss": 0.8538, + "step": 4607 + }, + { + "epoch": 0.3463099353675034, + "grad_norm": 1.664151113237196, + "learning_rate": 3.0393668339655917e-06, + "loss": 0.9676, + "step": 4608 + }, + { + "epoch": 0.34638508943333834, + "grad_norm": 2.1940068064734453, + "learning_rate": 3.0389508652137555e-06, + "loss": 0.9178, + "step": 4609 + }, + { + "epoch": 0.3464602434991733, + "grad_norm": 1.8189017268598864, + "learning_rate": 3.0385348349001023e-06, + "loss": 1.0475, + "step": 4610 + }, + { + "epoch": 0.3465353975650083, + "grad_norm": 2.064236690035276, + "learning_rate": 3.038118743049283e-06, + "loss": 1.1048, + "step": 4611 + }, + { + "epoch": 0.34661055163084326, + "grad_norm": 1.287108170967523, + "learning_rate": 3.0377025896859532e-06, + "loss": 1.0357, + "step": 4612 + }, + { + "epoch": 0.34668570569667817, + "grad_norm": 0.650964890085544, + "learning_rate": 3.037286374834771e-06, + "loss": 0.7657, + "step": 4613 + }, + { + "epoch": 0.34676085976251314, + "grad_norm": 1.7577067025453512, + "learning_rate": 3.036870098520399e-06, + "loss": 1.1141, + "step": 4614 + }, + { + "epoch": 0.3468360138283481, + "grad_norm": 1.3022051741497096, + "learning_rate": 3.036453760767504e-06, + "loss": 0.9207, + "step": 4615 + }, + { + "epoch": 0.3469111678941831, + "grad_norm": 1.4480275813737344, + "learning_rate": 3.036037361600754e-06, + "loss": 0.9415, + "step": 4616 + }, + { + "epoch": 0.34698632196001805, + "grad_norm": 1.6892361363768915, + "learning_rate": 3.0356209010448234e-06, + "loss": 0.9311, + "step": 4617 + }, + { + "epoch": 0.347061476025853, + "grad_norm": 2.2483214368946123, + "learning_rate": 3.0352043791243886e-06, + "loss": 1.0532, + "step": 4618 + }, + { + "epoch": 0.34713663009168794, + "grad_norm": 1.420321275874892, + "learning_rate": 3.0347877958641303e-06, + "loss": 0.9875, + "step": 4619 + }, + { + "epoch": 0.3472117841575229, + "grad_norm": 1.7204813473307583, + "learning_rate": 3.0343711512887325e-06, + "loss": 0.9954, + "step": 4620 + }, + { + "epoch": 0.3472869382233579, + "grad_norm": 1.4841471446982906, + "learning_rate": 3.0339544454228836e-06, + "loss": 1.0825, + "step": 4621 + }, + { + "epoch": 0.34736209228919285, + "grad_norm": 3.2365943555156496, + "learning_rate": 3.0335376782912742e-06, + "loss": 1.0171, + "step": 4622 + }, + { + "epoch": 0.3474372463550278, + "grad_norm": 1.4381742140245923, + "learning_rate": 3.0331208499185996e-06, + "loss": 0.969, + "step": 4623 + }, + { + "epoch": 0.3475124004208628, + "grad_norm": 1.4733558834444567, + "learning_rate": 3.0327039603295587e-06, + "loss": 1.0297, + "step": 4624 + }, + { + "epoch": 0.3475875544866977, + "grad_norm": 2.5118871368395945, + "learning_rate": 3.032287009548853e-06, + "loss": 1.0071, + "step": 4625 + }, + { + "epoch": 0.3476627085525327, + "grad_norm": 2.0949640881867726, + "learning_rate": 3.03186999760119e-06, + "loss": 0.9448, + "step": 4626 + }, + { + "epoch": 0.34773786261836764, + "grad_norm": 1.5449368443589728, + "learning_rate": 3.031452924511279e-06, + "loss": 1.0444, + "step": 4627 + }, + { + "epoch": 0.3478130166842026, + "grad_norm": 4.356414774291923, + "learning_rate": 3.031035790303831e-06, + "loss": 0.9917, + "step": 4628 + }, + { + "epoch": 0.3478881707500376, + "grad_norm": 1.9323404585306805, + "learning_rate": 3.030618595003565e-06, + "loss": 0.9641, + "step": 4629 + }, + { + "epoch": 0.34796332481587255, + "grad_norm": 2.2489931184859304, + "learning_rate": 3.0302013386352004e-06, + "loss": 0.9227, + "step": 4630 + }, + { + "epoch": 0.3480384788817075, + "grad_norm": 1.7935150577454373, + "learning_rate": 3.0297840212234623e-06, + "loss": 0.9147, + "step": 4631 + }, + { + "epoch": 0.34811363294754244, + "grad_norm": 2.359591139436321, + "learning_rate": 3.029366642793077e-06, + "loss": 1.0098, + "step": 4632 + }, + { + "epoch": 0.3481887870133774, + "grad_norm": 1.5241067052315873, + "learning_rate": 3.0289492033687768e-06, + "loss": 1.0359, + "step": 4633 + }, + { + "epoch": 0.3482639410792124, + "grad_norm": 1.7435370722013637, + "learning_rate": 3.0285317029752957e-06, + "loss": 0.9283, + "step": 4634 + }, + { + "epoch": 0.34833909514504735, + "grad_norm": 1.740837497855331, + "learning_rate": 3.028114141637373e-06, + "loss": 0.9873, + "step": 4635 + }, + { + "epoch": 0.3484142492108823, + "grad_norm": 1.6707321967181439, + "learning_rate": 3.0276965193797503e-06, + "loss": 1.135, + "step": 4636 + }, + { + "epoch": 0.3484894032767173, + "grad_norm": 2.2976033619407135, + "learning_rate": 3.0272788362271743e-06, + "loss": 1.0027, + "step": 4637 + }, + { + "epoch": 0.3485645573425522, + "grad_norm": 0.7678007235029528, + "learning_rate": 3.0268610922043925e-06, + "loss": 0.8424, + "step": 4638 + }, + { + "epoch": 0.3486397114083872, + "grad_norm": 1.6098391409890087, + "learning_rate": 3.0264432873361594e-06, + "loss": 0.968, + "step": 4639 + }, + { + "epoch": 0.34871486547422215, + "grad_norm": 1.624506068356006, + "learning_rate": 3.026025421647231e-06, + "loss": 1.0595, + "step": 4640 + }, + { + "epoch": 0.3487900195400571, + "grad_norm": 2.3616522100414885, + "learning_rate": 3.025607495162367e-06, + "loss": 0.996, + "step": 4641 + }, + { + "epoch": 0.3488651736058921, + "grad_norm": 2.212349807900159, + "learning_rate": 3.025189507906332e-06, + "loss": 1.0131, + "step": 4642 + }, + { + "epoch": 0.34894032767172706, + "grad_norm": 1.8680116757797873, + "learning_rate": 3.0247714599038936e-06, + "loss": 0.9837, + "step": 4643 + }, + { + "epoch": 0.349015481737562, + "grad_norm": 1.3245171404860279, + "learning_rate": 3.0243533511798205e-06, + "loss": 0.9925, + "step": 4644 + }, + { + "epoch": 0.34909063580339694, + "grad_norm": 1.6641756834124306, + "learning_rate": 3.0239351817588903e-06, + "loss": 0.9141, + "step": 4645 + }, + { + "epoch": 0.3491657898692319, + "grad_norm": 1.444360698000238, + "learning_rate": 3.023516951665879e-06, + "loss": 1.0098, + "step": 4646 + }, + { + "epoch": 0.3492409439350669, + "grad_norm": 1.8072641977060406, + "learning_rate": 3.0230986609255687e-06, + "loss": 1.0147, + "step": 4647 + }, + { + "epoch": 0.34931609800090185, + "grad_norm": 1.9782962357348386, + "learning_rate": 3.022680309562746e-06, + "loss": 1.0937, + "step": 4648 + }, + { + "epoch": 0.3493912520667368, + "grad_norm": 1.681598233033113, + "learning_rate": 3.022261897602198e-06, + "loss": 0.8924, + "step": 4649 + }, + { + "epoch": 0.3494664061325718, + "grad_norm": 1.633381648549262, + "learning_rate": 3.0218434250687184e-06, + "loss": 0.9667, + "step": 4650 + }, + { + "epoch": 0.3495415601984067, + "grad_norm": 1.857411221093591, + "learning_rate": 3.021424891987103e-06, + "loss": 0.9562, + "step": 4651 + }, + { + "epoch": 0.3496167142642417, + "grad_norm": 2.079308079845761, + "learning_rate": 3.0210062983821513e-06, + "loss": 0.9615, + "step": 4652 + }, + { + "epoch": 0.34969186833007665, + "grad_norm": 1.8419856665443965, + "learning_rate": 3.0205876442786666e-06, + "loss": 0.9588, + "step": 4653 + }, + { + "epoch": 0.3497670223959116, + "grad_norm": 2.60087131792333, + "learning_rate": 3.0201689297014565e-06, + "loss": 0.9698, + "step": 4654 + }, + { + "epoch": 0.3498421764617466, + "grad_norm": 2.435421277366257, + "learning_rate": 3.01975015467533e-06, + "loss": 1.0169, + "step": 4655 + }, + { + "epoch": 0.34991733052758156, + "grad_norm": 1.4943581017687901, + "learning_rate": 3.019331319225103e-06, + "loss": 1.0263, + "step": 4656 + }, + { + "epoch": 0.34999248459341653, + "grad_norm": 3.247451222958646, + "learning_rate": 3.018912423375591e-06, + "loss": 0.9667, + "step": 4657 + }, + { + "epoch": 0.35006763865925145, + "grad_norm": 1.5920955073591605, + "learning_rate": 3.018493467151616e-06, + "loss": 0.9672, + "step": 4658 + }, + { + "epoch": 0.3501427927250864, + "grad_norm": 3.732893390505597, + "learning_rate": 3.0180744505780045e-06, + "loss": 0.8768, + "step": 4659 + }, + { + "epoch": 0.3502179467909214, + "grad_norm": 1.8747665259913138, + "learning_rate": 3.0176553736795827e-06, + "loss": 0.9822, + "step": 4660 + }, + { + "epoch": 0.35029310085675636, + "grad_norm": 1.4613358877506577, + "learning_rate": 3.0172362364811827e-06, + "loss": 0.9924, + "step": 4661 + }, + { + "epoch": 0.3503682549225913, + "grad_norm": 1.8848753585890983, + "learning_rate": 3.016817039007641e-06, + "loss": 1.0049, + "step": 4662 + }, + { + "epoch": 0.3504434089884263, + "grad_norm": 1.7678529304465218, + "learning_rate": 3.0163977812837954e-06, + "loss": 1.0181, + "step": 4663 + }, + { + "epoch": 0.3505185630542612, + "grad_norm": 1.7071487072093665, + "learning_rate": 3.0159784633344894e-06, + "loss": 0.9111, + "step": 4664 + }, + { + "epoch": 0.3505937171200962, + "grad_norm": 0.6901779255346696, + "learning_rate": 3.0155590851845694e-06, + "loss": 0.8317, + "step": 4665 + }, + { + "epoch": 0.35066887118593115, + "grad_norm": 1.5879214801113124, + "learning_rate": 3.0151396468588844e-06, + "loss": 0.9903, + "step": 4666 + }, + { + "epoch": 0.3507440252517661, + "grad_norm": 2.3667778441234906, + "learning_rate": 3.0147201483822884e-06, + "loss": 1.0433, + "step": 4667 + }, + { + "epoch": 0.3508191793176011, + "grad_norm": 1.4138040042886866, + "learning_rate": 3.014300589779638e-06, + "loss": 1.0001, + "step": 4668 + }, + { + "epoch": 0.35089433338343606, + "grad_norm": 4.011141281390181, + "learning_rate": 3.0138809710757927e-06, + "loss": 1.0109, + "step": 4669 + }, + { + "epoch": 0.350969487449271, + "grad_norm": 1.9989837806543638, + "learning_rate": 3.013461292295619e-06, + "loss": 0.9884, + "step": 4670 + }, + { + "epoch": 0.35104464151510595, + "grad_norm": 1.9551536705520163, + "learning_rate": 3.013041553463982e-06, + "loss": 1.0514, + "step": 4671 + }, + { + "epoch": 0.3511197955809409, + "grad_norm": 1.5200729385242795, + "learning_rate": 3.012621754605754e-06, + "loss": 0.8502, + "step": 4672 + }, + { + "epoch": 0.3511949496467759, + "grad_norm": 1.3346968126519168, + "learning_rate": 3.012201895745809e-06, + "loss": 0.9047, + "step": 4673 + }, + { + "epoch": 0.35127010371261086, + "grad_norm": 1.8521829201435884, + "learning_rate": 3.011781976909026e-06, + "loss": 0.9731, + "step": 4674 + }, + { + "epoch": 0.35134525777844583, + "grad_norm": 2.505569453129237, + "learning_rate": 3.011361998120287e-06, + "loss": 0.9461, + "step": 4675 + }, + { + "epoch": 0.3514204118442808, + "grad_norm": 1.6628179776697065, + "learning_rate": 3.0109419594044765e-06, + "loss": 1.0498, + "step": 4676 + }, + { + "epoch": 0.3514955659101157, + "grad_norm": 1.8908853272840505, + "learning_rate": 3.0105218607864835e-06, + "loss": 1.0763, + "step": 4677 + }, + { + "epoch": 0.3515707199759507, + "grad_norm": 1.621730449607481, + "learning_rate": 3.010101702291201e-06, + "loss": 1.0372, + "step": 4678 + }, + { + "epoch": 0.35164587404178566, + "grad_norm": 5.324892019690381, + "learning_rate": 3.0096814839435244e-06, + "loss": 0.949, + "step": 4679 + }, + { + "epoch": 0.3517210281076206, + "grad_norm": 1.571551237644355, + "learning_rate": 3.0092612057683532e-06, + "loss": 0.9289, + "step": 4680 + }, + { + "epoch": 0.3517961821734556, + "grad_norm": 1.2487892963385698, + "learning_rate": 3.0088408677905913e-06, + "loss": 1.0932, + "step": 4681 + }, + { + "epoch": 0.35187133623929057, + "grad_norm": 2.50556612268641, + "learning_rate": 3.0084204700351453e-06, + "loss": 0.8776, + "step": 4682 + }, + { + "epoch": 0.3519464903051255, + "grad_norm": 1.6153786535991443, + "learning_rate": 3.0080000125269242e-06, + "loss": 0.9769, + "step": 4683 + }, + { + "epoch": 0.35202164437096045, + "grad_norm": 1.5287529963591087, + "learning_rate": 3.0075794952908436e-06, + "loss": 0.8977, + "step": 4684 + }, + { + "epoch": 0.3520967984367954, + "grad_norm": 3.2056551732878034, + "learning_rate": 3.007158918351818e-06, + "loss": 0.9761, + "step": 4685 + }, + { + "epoch": 0.3521719525026304, + "grad_norm": 1.7842410222677914, + "learning_rate": 3.0067382817347712e-06, + "loss": 0.9073, + "step": 4686 + }, + { + "epoch": 0.35224710656846536, + "grad_norm": 1.4509102956541304, + "learning_rate": 3.006317585464626e-06, + "loss": 1.0592, + "step": 4687 + }, + { + "epoch": 0.35232226063430033, + "grad_norm": 1.6244053119269162, + "learning_rate": 3.0058968295663094e-06, + "loss": 1.0567, + "step": 4688 + }, + { + "epoch": 0.3523974147001353, + "grad_norm": 1.6679484524709083, + "learning_rate": 3.0054760140647547e-06, + "loss": 1.0158, + "step": 4689 + }, + { + "epoch": 0.3524725687659702, + "grad_norm": 1.8494108550736523, + "learning_rate": 3.005055138984896e-06, + "loss": 1.0356, + "step": 4690 + }, + { + "epoch": 0.3525477228318052, + "grad_norm": 1.477745510687581, + "learning_rate": 3.0046342043516707e-06, + "loss": 1.024, + "step": 4691 + }, + { + "epoch": 0.35262287689764016, + "grad_norm": 2.038120329803853, + "learning_rate": 3.0042132101900228e-06, + "loss": 1.0218, + "step": 4692 + }, + { + "epoch": 0.35269803096347513, + "grad_norm": 1.715217497466106, + "learning_rate": 3.003792156524897e-06, + "loss": 0.9559, + "step": 4693 + }, + { + "epoch": 0.3527731850293101, + "grad_norm": 0.7360318749045787, + "learning_rate": 3.003371043381241e-06, + "loss": 0.852, + "step": 4694 + }, + { + "epoch": 0.35284833909514507, + "grad_norm": 1.801615755430512, + "learning_rate": 3.0029498707840094e-06, + "loss": 1.0393, + "step": 4695 + }, + { + "epoch": 0.35292349316098, + "grad_norm": 2.00876057724699, + "learning_rate": 3.002528638758157e-06, + "loss": 1.0129, + "step": 4696 + }, + { + "epoch": 0.35299864722681495, + "grad_norm": 2.1662505068065756, + "learning_rate": 3.0021073473286446e-06, + "loss": 0.8403, + "step": 4697 + }, + { + "epoch": 0.3530738012926499, + "grad_norm": 1.6191018977670255, + "learning_rate": 3.0016859965204336e-06, + "loss": 1.0244, + "step": 4698 + }, + { + "epoch": 0.3531489553584849, + "grad_norm": 0.6471492655837383, + "learning_rate": 3.001264586358492e-06, + "loss": 0.8138, + "step": 4699 + }, + { + "epoch": 0.35322410942431987, + "grad_norm": 2.1036495920519025, + "learning_rate": 3.0008431168677898e-06, + "loss": 1.0348, + "step": 4700 + }, + { + "epoch": 0.35329926349015484, + "grad_norm": 2.0890442076333695, + "learning_rate": 3.0004215880732993e-06, + "loss": 1.037, + "step": 4701 + }, + { + "epoch": 0.3533744175559898, + "grad_norm": 13.833748746096164, + "learning_rate": 3e-06, + "loss": 1.0237, + "step": 4702 + }, + { + "epoch": 0.3534495716218247, + "grad_norm": 1.368777500381467, + "learning_rate": 2.999578352672871e-06, + "loss": 1.0541, + "step": 4703 + }, + { + "epoch": 0.3535247256876597, + "grad_norm": 1.343199018140301, + "learning_rate": 2.9991566461168974e-06, + "loss": 0.9733, + "step": 4704 + }, + { + "epoch": 0.35359987975349466, + "grad_norm": 13.730581547367404, + "learning_rate": 2.998734880357066e-06, + "loss": 0.8867, + "step": 4705 + }, + { + "epoch": 0.35367503381932963, + "grad_norm": 1.7658973416292056, + "learning_rate": 2.998313055418369e-06, + "loss": 1.1093, + "step": 4706 + }, + { + "epoch": 0.3537501878851646, + "grad_norm": 2.351700281308026, + "learning_rate": 2.9978911713257998e-06, + "loss": 1.0685, + "step": 4707 + }, + { + "epoch": 0.3538253419509996, + "grad_norm": 1.6188824750385868, + "learning_rate": 2.997469228104358e-06, + "loss": 0.973, + "step": 4708 + }, + { + "epoch": 0.3539004960168345, + "grad_norm": 1.4382431762894154, + "learning_rate": 2.9970472257790454e-06, + "loss": 0.993, + "step": 4709 + }, + { + "epoch": 0.35397565008266946, + "grad_norm": 1.9936867252461836, + "learning_rate": 2.996625164374866e-06, + "loss": 1.0249, + "step": 4710 + }, + { + "epoch": 0.3540508041485044, + "grad_norm": 0.6223187392196825, + "learning_rate": 2.9962030439168297e-06, + "loss": 0.8492, + "step": 4711 + }, + { + "epoch": 0.3541259582143394, + "grad_norm": 1.6817471673158186, + "learning_rate": 2.995780864429948e-06, + "loss": 0.7715, + "step": 4712 + }, + { + "epoch": 0.35420111228017437, + "grad_norm": 1.2719859175622281, + "learning_rate": 2.9953586259392366e-06, + "loss": 0.9238, + "step": 4713 + }, + { + "epoch": 0.35427626634600934, + "grad_norm": 1.79614096866114, + "learning_rate": 2.994936328469716e-06, + "loss": 1.032, + "step": 4714 + }, + { + "epoch": 0.35435142041184425, + "grad_norm": 2.41599648050974, + "learning_rate": 2.9945139720464082e-06, + "loss": 0.9702, + "step": 4715 + }, + { + "epoch": 0.3544265744776792, + "grad_norm": 1.899945574533056, + "learning_rate": 2.9940915566943384e-06, + "loss": 0.9439, + "step": 4716 + }, + { + "epoch": 0.3545017285435142, + "grad_norm": 1.5923987251309215, + "learning_rate": 2.9936690824385383e-06, + "loss": 1.0874, + "step": 4717 + }, + { + "epoch": 0.35457688260934916, + "grad_norm": 1.7836141457312624, + "learning_rate": 2.9932465493040393e-06, + "loss": 0.9656, + "step": 4718 + }, + { + "epoch": 0.35465203667518413, + "grad_norm": 1.6172385829652864, + "learning_rate": 2.992823957315879e-06, + "loss": 1.0115, + "step": 4719 + }, + { + "epoch": 0.3547271907410191, + "grad_norm": 2.489649326189337, + "learning_rate": 2.9924013064990974e-06, + "loss": 1.0655, + "step": 4720 + }, + { + "epoch": 0.3548023448068541, + "grad_norm": 2.068037978895291, + "learning_rate": 2.9919785968787384e-06, + "loss": 1.0606, + "step": 4721 + }, + { + "epoch": 0.354877498872689, + "grad_norm": 1.9600168323767204, + "learning_rate": 2.991555828479849e-06, + "loss": 1.0032, + "step": 4722 + }, + { + "epoch": 0.35495265293852396, + "grad_norm": 1.482562032112616, + "learning_rate": 2.9911330013274792e-06, + "loss": 0.9275, + "step": 4723 + }, + { + "epoch": 0.35502780700435893, + "grad_norm": 0.6169553634025212, + "learning_rate": 2.990710115446684e-06, + "loss": 0.7719, + "step": 4724 + }, + { + "epoch": 0.3551029610701939, + "grad_norm": 3.767173860098829, + "learning_rate": 2.9902871708625216e-06, + "loss": 1.0901, + "step": 4725 + }, + { + "epoch": 0.35517811513602887, + "grad_norm": 1.798508672689611, + "learning_rate": 2.9898641676000518e-06, + "loss": 0.9428, + "step": 4726 + }, + { + "epoch": 0.35525326920186384, + "grad_norm": 1.6226610076202748, + "learning_rate": 2.9894411056843396e-06, + "loss": 0.9967, + "step": 4727 + }, + { + "epoch": 0.35532842326769876, + "grad_norm": 1.7402204005646824, + "learning_rate": 2.9890179851404533e-06, + "loss": 0.9535, + "step": 4728 + }, + { + "epoch": 0.3554035773335337, + "grad_norm": 2.0392085348283646, + "learning_rate": 2.9885948059934635e-06, + "loss": 1.0415, + "step": 4729 + }, + { + "epoch": 0.3554787313993687, + "grad_norm": 7.7429260455953335, + "learning_rate": 2.988171568268446e-06, + "loss": 1.0426, + "step": 4730 + }, + { + "epoch": 0.35555388546520367, + "grad_norm": 2.937978664925782, + "learning_rate": 2.98774827199048e-06, + "loss": 1.0221, + "step": 4731 + }, + { + "epoch": 0.35562903953103864, + "grad_norm": 1.423211528038995, + "learning_rate": 2.9873249171846454e-06, + "loss": 0.9506, + "step": 4732 + }, + { + "epoch": 0.3557041935968736, + "grad_norm": 2.2794529355581803, + "learning_rate": 2.9869015038760296e-06, + "loss": 0.9534, + "step": 4733 + }, + { + "epoch": 0.3557793476627086, + "grad_norm": 1.6295449447824506, + "learning_rate": 2.98647803208972e-06, + "loss": 1.1166, + "step": 4734 + }, + { + "epoch": 0.3558545017285435, + "grad_norm": 4.258890613936288, + "learning_rate": 2.98605450185081e-06, + "loss": 1.0102, + "step": 4735 + }, + { + "epoch": 0.35592965579437846, + "grad_norm": 1.4728304408336654, + "learning_rate": 2.9856309131843945e-06, + "loss": 1.0138, + "step": 4736 + }, + { + "epoch": 0.35600480986021343, + "grad_norm": 2.0319483803552854, + "learning_rate": 2.985207266115574e-06, + "loss": 0.978, + "step": 4737 + }, + { + "epoch": 0.3560799639260484, + "grad_norm": 1.7651555610379177, + "learning_rate": 2.9847835606694494e-06, + "loss": 1.0681, + "step": 4738 + }, + { + "epoch": 0.3561551179918834, + "grad_norm": 2.297305838731252, + "learning_rate": 2.9843597968711285e-06, + "loss": 0.9086, + "step": 4739 + }, + { + "epoch": 0.35623027205771834, + "grad_norm": 0.7098563220829259, + "learning_rate": 2.9839359747457195e-06, + "loss": 0.7821, + "step": 4740 + }, + { + "epoch": 0.35630542612355326, + "grad_norm": 2.1588245446566092, + "learning_rate": 2.9835120943183374e-06, + "loss": 1.0179, + "step": 4741 + }, + { + "epoch": 0.35638058018938823, + "grad_norm": 1.4571102826567297, + "learning_rate": 2.9830881556140965e-06, + "loss": 0.9463, + "step": 4742 + }, + { + "epoch": 0.3564557342552232, + "grad_norm": 1.6497655644186249, + "learning_rate": 2.9826641586581184e-06, + "loss": 0.9899, + "step": 4743 + }, + { + "epoch": 0.35653088832105817, + "grad_norm": 1.9790831163405638, + "learning_rate": 2.9822401034755255e-06, + "loss": 1.0827, + "step": 4744 + }, + { + "epoch": 0.35660604238689314, + "grad_norm": 1.8630254477924648, + "learning_rate": 2.981815990091446e-06, + "loss": 0.971, + "step": 4745 + }, + { + "epoch": 0.3566811964527281, + "grad_norm": 1.6461970575229643, + "learning_rate": 2.9813918185310085e-06, + "loss": 1.102, + "step": 4746 + }, + { + "epoch": 0.3567563505185631, + "grad_norm": 1.7156548808401872, + "learning_rate": 2.9809675888193486e-06, + "loss": 1.0277, + "step": 4747 + }, + { + "epoch": 0.356831504584398, + "grad_norm": 0.790352475085618, + "learning_rate": 2.9805433009816024e-06, + "loss": 0.872, + "step": 4748 + }, + { + "epoch": 0.35690665865023297, + "grad_norm": 1.4273524993477993, + "learning_rate": 2.980118955042911e-06, + "loss": 0.9859, + "step": 4749 + }, + { + "epoch": 0.35698181271606794, + "grad_norm": 1.6165238645085809, + "learning_rate": 2.9796945510284187e-06, + "loss": 0.9011, + "step": 4750 + }, + { + "epoch": 0.3570569667819029, + "grad_norm": 2.4750616778770693, + "learning_rate": 2.9792700889632716e-06, + "loss": 1.0302, + "step": 4751 + }, + { + "epoch": 0.3571321208477379, + "grad_norm": 1.5525873462474378, + "learning_rate": 2.9788455688726234e-06, + "loss": 0.9841, + "step": 4752 + }, + { + "epoch": 0.35720727491357285, + "grad_norm": 1.6046999833346676, + "learning_rate": 2.978420990781626e-06, + "loss": 1.0578, + "step": 4753 + }, + { + "epoch": 0.35728242897940776, + "grad_norm": 1.6505526310786989, + "learning_rate": 2.977996354715438e-06, + "loss": 0.887, + "step": 4754 + }, + { + "epoch": 0.35735758304524273, + "grad_norm": 1.5645242167689726, + "learning_rate": 2.9775716606992217e-06, + "loss": 1.0217, + "step": 4755 + }, + { + "epoch": 0.3574327371110777, + "grad_norm": 1.7129138001077964, + "learning_rate": 2.977146908758141e-06, + "loss": 1.0834, + "step": 4756 + }, + { + "epoch": 0.3575078911769127, + "grad_norm": 2.2957391330711463, + "learning_rate": 2.9767220989173635e-06, + "loss": 0.9903, + "step": 4757 + }, + { + "epoch": 0.35758304524274764, + "grad_norm": 1.5107771432355028, + "learning_rate": 2.9762972312020623e-06, + "loss": 1.0093, + "step": 4758 + }, + { + "epoch": 0.3576581993085826, + "grad_norm": 1.5882419324249961, + "learning_rate": 2.975872305637412e-06, + "loss": 0.9198, + "step": 4759 + }, + { + "epoch": 0.35773335337441753, + "grad_norm": 1.5616834414194234, + "learning_rate": 2.97544732224859e-06, + "loss": 0.9341, + "step": 4760 + }, + { + "epoch": 0.3578085074402525, + "grad_norm": 1.812999459213454, + "learning_rate": 2.975022281060779e-06, + "loss": 0.9262, + "step": 4761 + }, + { + "epoch": 0.35788366150608747, + "grad_norm": 1.7053506290422664, + "learning_rate": 2.9745971820991643e-06, + "loss": 0.9785, + "step": 4762 + }, + { + "epoch": 0.35795881557192244, + "grad_norm": 1.3904228492136765, + "learning_rate": 2.9741720253889346e-06, + "loss": 0.9548, + "step": 4763 + }, + { + "epoch": 0.3580339696377574, + "grad_norm": 3.4872870486199807, + "learning_rate": 2.9737468109552827e-06, + "loss": 1.0027, + "step": 4764 + }, + { + "epoch": 0.3581091237035924, + "grad_norm": 6.546046675765369, + "learning_rate": 2.973321538823402e-06, + "loss": 1.0233, + "step": 4765 + }, + { + "epoch": 0.35818427776942735, + "grad_norm": 1.5262220707867808, + "learning_rate": 2.9728962090184938e-06, + "loss": 1.1792, + "step": 4766 + }, + { + "epoch": 0.35825943183526227, + "grad_norm": 0.7714700045635502, + "learning_rate": 2.9724708215657603e-06, + "loss": 0.9074, + "step": 4767 + }, + { + "epoch": 0.35833458590109724, + "grad_norm": 1.7273760643761222, + "learning_rate": 2.972045376490406e-06, + "loss": 0.9611, + "step": 4768 + }, + { + "epoch": 0.3584097399669322, + "grad_norm": 1.8222689613238148, + "learning_rate": 2.971619873817642e-06, + "loss": 0.891, + "step": 4769 + }, + { + "epoch": 0.3584848940327672, + "grad_norm": 1.9884269975705315, + "learning_rate": 2.971194313572679e-06, + "loss": 0.9892, + "step": 4770 + }, + { + "epoch": 0.35856004809860215, + "grad_norm": 1.5731584561002996, + "learning_rate": 2.970768695780734e-06, + "loss": 0.9711, + "step": 4771 + }, + { + "epoch": 0.3586352021644371, + "grad_norm": 2.0974218709987382, + "learning_rate": 2.970343020467027e-06, + "loss": 0.8785, + "step": 4772 + }, + { + "epoch": 0.35871035623027203, + "grad_norm": 0.793801044715142, + "learning_rate": 2.9699172876567795e-06, + "loss": 0.8465, + "step": 4773 + }, + { + "epoch": 0.358785510296107, + "grad_norm": 1.525846952790431, + "learning_rate": 2.969491497375219e-06, + "loss": 0.9622, + "step": 4774 + }, + { + "epoch": 0.358860664361942, + "grad_norm": 1.7173554051152713, + "learning_rate": 2.969065649647575e-06, + "loss": 1.029, + "step": 4775 + }, + { + "epoch": 0.35893581842777694, + "grad_norm": 8.160205802566157, + "learning_rate": 2.9686397444990803e-06, + "loss": 1.024, + "step": 4776 + }, + { + "epoch": 0.3590109724936119, + "grad_norm": 1.8665672133836642, + "learning_rate": 2.9682137819549718e-06, + "loss": 0.9821, + "step": 4777 + }, + { + "epoch": 0.3590861265594469, + "grad_norm": 1.7310423578449259, + "learning_rate": 2.9677877620404887e-06, + "loss": 1.0533, + "step": 4778 + }, + { + "epoch": 0.35916128062528185, + "grad_norm": 1.813272114426241, + "learning_rate": 2.9673616847808755e-06, + "loss": 0.9704, + "step": 4779 + }, + { + "epoch": 0.35923643469111677, + "grad_norm": 2.6402584927616184, + "learning_rate": 2.966935550201378e-06, + "loss": 0.9608, + "step": 4780 + }, + { + "epoch": 0.35931158875695174, + "grad_norm": 1.6082476769336709, + "learning_rate": 2.9665093583272463e-06, + "loss": 0.93, + "step": 4781 + }, + { + "epoch": 0.3593867428227867, + "grad_norm": 0.7980209413591525, + "learning_rate": 2.966083109183734e-06, + "loss": 0.9384, + "step": 4782 + }, + { + "epoch": 0.3594618968886217, + "grad_norm": 1.6184635011919644, + "learning_rate": 2.9656568027960984e-06, + "loss": 0.9692, + "step": 4783 + }, + { + "epoch": 0.35953705095445665, + "grad_norm": 6.382120083216219, + "learning_rate": 2.9652304391895994e-06, + "loss": 0.9929, + "step": 4784 + }, + { + "epoch": 0.3596122050202916, + "grad_norm": 1.6713131111913946, + "learning_rate": 2.9648040183895004e-06, + "loss": 1.0593, + "step": 4785 + }, + { + "epoch": 0.35968735908612653, + "grad_norm": 2.4145788733040994, + "learning_rate": 2.964377540421069e-06, + "loss": 1.1243, + "step": 4786 + }, + { + "epoch": 0.3597625131519615, + "grad_norm": 1.9656244323438543, + "learning_rate": 2.963951005309576e-06, + "loss": 0.9554, + "step": 4787 + }, + { + "epoch": 0.3598376672177965, + "grad_norm": 2.3182012026787224, + "learning_rate": 2.963524413080294e-06, + "loss": 0.9738, + "step": 4788 + }, + { + "epoch": 0.35991282128363145, + "grad_norm": 2.796506292665148, + "learning_rate": 2.9630977637585016e-06, + "loss": 0.9448, + "step": 4789 + }, + { + "epoch": 0.3599879753494664, + "grad_norm": 1.6516289213135869, + "learning_rate": 2.9626710573694783e-06, + "loss": 0.9726, + "step": 4790 + }, + { + "epoch": 0.3600631294153014, + "grad_norm": 1.5750126126707753, + "learning_rate": 2.9622442939385085e-06, + "loss": 1.07, + "step": 4791 + }, + { + "epoch": 0.36013828348113636, + "grad_norm": 1.395902285604589, + "learning_rate": 2.96181747349088e-06, + "loss": 1.0901, + "step": 4792 + }, + { + "epoch": 0.36021343754697127, + "grad_norm": 2.1095415967921065, + "learning_rate": 2.9613905960518832e-06, + "loss": 0.9159, + "step": 4793 + }, + { + "epoch": 0.36028859161280624, + "grad_norm": 1.6609244689717315, + "learning_rate": 2.960963661646812e-06, + "loss": 1.0274, + "step": 4794 + }, + { + "epoch": 0.3603637456786412, + "grad_norm": 1.4880219481449226, + "learning_rate": 2.960536670300963e-06, + "loss": 1.0086, + "step": 4795 + }, + { + "epoch": 0.3604388997444762, + "grad_norm": 2.569484681739435, + "learning_rate": 2.9601096220396392e-06, + "loss": 0.9471, + "step": 4796 + }, + { + "epoch": 0.36051405381031115, + "grad_norm": 1.8893837755752165, + "learning_rate": 2.9596825168881444e-06, + "loss": 1.1086, + "step": 4797 + }, + { + "epoch": 0.3605892078761461, + "grad_norm": 2.695401397565907, + "learning_rate": 2.9592553548717848e-06, + "loss": 1.1295, + "step": 4798 + }, + { + "epoch": 0.36066436194198104, + "grad_norm": 1.5048915261245945, + "learning_rate": 2.958828136015872e-06, + "loss": 1.0387, + "step": 4799 + }, + { + "epoch": 0.360739516007816, + "grad_norm": 1.9344102156827652, + "learning_rate": 2.958400860345721e-06, + "loss": 1.0027, + "step": 4800 + }, + { + "epoch": 0.360814670073651, + "grad_norm": 1.7133877423227915, + "learning_rate": 2.9579735278866488e-06, + "loss": 0.9409, + "step": 4801 + }, + { + "epoch": 0.36088982413948595, + "grad_norm": 1.9308316935122192, + "learning_rate": 2.9575461386639768e-06, + "loss": 0.9741, + "step": 4802 + }, + { + "epoch": 0.3609649782053209, + "grad_norm": 1.8284276076045807, + "learning_rate": 2.95711869270303e-06, + "loss": 0.9297, + "step": 4803 + }, + { + "epoch": 0.3610401322711559, + "grad_norm": 1.5675373512210624, + "learning_rate": 2.9566911900291346e-06, + "loss": 0.98, + "step": 4804 + }, + { + "epoch": 0.3611152863369908, + "grad_norm": 1.6965905219767894, + "learning_rate": 2.9562636306676237e-06, + "loss": 1.0997, + "step": 4805 + }, + { + "epoch": 0.3611904404028258, + "grad_norm": 1.4376087976830767, + "learning_rate": 2.9558360146438303e-06, + "loss": 0.9625, + "step": 4806 + }, + { + "epoch": 0.36126559446866074, + "grad_norm": 1.3609397914442356, + "learning_rate": 2.9554083419830925e-06, + "loss": 0.8654, + "step": 4807 + }, + { + "epoch": 0.3613407485344957, + "grad_norm": 2.1139729437851527, + "learning_rate": 2.954980612710753e-06, + "loss": 0.9884, + "step": 4808 + }, + { + "epoch": 0.3614159026003307, + "grad_norm": 2.2467849861970413, + "learning_rate": 2.9545528268521548e-06, + "loss": 1.0392, + "step": 4809 + }, + { + "epoch": 0.36149105666616566, + "grad_norm": 1.757285823333679, + "learning_rate": 2.954124984432646e-06, + "loss": 0.9687, + "step": 4810 + }, + { + "epoch": 0.3615662107320006, + "grad_norm": 0.7936481891127644, + "learning_rate": 2.953697085477579e-06, + "loss": 0.9441, + "step": 4811 + }, + { + "epoch": 0.36164136479783554, + "grad_norm": 2.114166582002539, + "learning_rate": 2.953269130012307e-06, + "loss": 0.9503, + "step": 4812 + }, + { + "epoch": 0.3617165188636705, + "grad_norm": 2.1272977859737314, + "learning_rate": 2.9528411180621894e-06, + "loss": 0.8862, + "step": 4813 + }, + { + "epoch": 0.3617916729295055, + "grad_norm": 1.5082139187465842, + "learning_rate": 2.952413049652587e-06, + "loss": 0.9598, + "step": 4814 + }, + { + "epoch": 0.36186682699534045, + "grad_norm": 1.7482727245859697, + "learning_rate": 2.9519849248088633e-06, + "loss": 0.9535, + "step": 4815 + }, + { + "epoch": 0.3619419810611754, + "grad_norm": 4.582038950933959, + "learning_rate": 2.9515567435563886e-06, + "loss": 1.0072, + "step": 4816 + }, + { + "epoch": 0.3620171351270104, + "grad_norm": 1.916576804943182, + "learning_rate": 2.951128505920532e-06, + "loss": 0.9762, + "step": 4817 + }, + { + "epoch": 0.3620922891928453, + "grad_norm": 1.9693456081859275, + "learning_rate": 2.95070021192667e-06, + "loss": 1.0526, + "step": 4818 + }, + { + "epoch": 0.3621674432586803, + "grad_norm": 3.437090623927784, + "learning_rate": 2.9502718616001803e-06, + "loss": 0.9997, + "step": 4819 + }, + { + "epoch": 0.36224259732451525, + "grad_norm": 0.7626661619899432, + "learning_rate": 2.9498434549664434e-06, + "loss": 0.8087, + "step": 4820 + }, + { + "epoch": 0.3623177513903502, + "grad_norm": 2.0600652484374917, + "learning_rate": 2.9494149920508443e-06, + "loss": 1.0427, + "step": 4821 + }, + { + "epoch": 0.3623929054561852, + "grad_norm": 1.4801218235293314, + "learning_rate": 2.9489864728787722e-06, + "loss": 0.9796, + "step": 4822 + }, + { + "epoch": 0.36246805952202016, + "grad_norm": 1.5688941585370022, + "learning_rate": 2.9485578974756167e-06, + "loss": 0.9176, + "step": 4823 + }, + { + "epoch": 0.36254321358785513, + "grad_norm": 1.7480101171283677, + "learning_rate": 2.9481292658667743e-06, + "loss": 1.033, + "step": 4824 + }, + { + "epoch": 0.36261836765369004, + "grad_norm": 1.5405576772469243, + "learning_rate": 2.947700578077643e-06, + "loss": 1.0562, + "step": 4825 + }, + { + "epoch": 0.362693521719525, + "grad_norm": 1.901699074422508, + "learning_rate": 2.947271834133622e-06, + "loss": 0.913, + "step": 4826 + }, + { + "epoch": 0.36276867578536, + "grad_norm": 1.6188721658568308, + "learning_rate": 2.946843034060118e-06, + "loss": 1.0347, + "step": 4827 + }, + { + "epoch": 0.36284382985119495, + "grad_norm": 1.9480197559623853, + "learning_rate": 2.9464141778825384e-06, + "loss": 1.0676, + "step": 4828 + }, + { + "epoch": 0.3629189839170299, + "grad_norm": 1.9279862290559342, + "learning_rate": 2.9459852656262945e-06, + "loss": 0.9593, + "step": 4829 + }, + { + "epoch": 0.3629941379828649, + "grad_norm": 1.9307511213063007, + "learning_rate": 2.945556297316802e-06, + "loss": 0.9233, + "step": 4830 + }, + { + "epoch": 0.3630692920486998, + "grad_norm": 1.6818072051550021, + "learning_rate": 2.9451272729794774e-06, + "loss": 1.0408, + "step": 4831 + }, + { + "epoch": 0.3631444461145348, + "grad_norm": 3.554619891183976, + "learning_rate": 2.944698192639743e-06, + "loss": 0.9406, + "step": 4832 + }, + { + "epoch": 0.36321960018036975, + "grad_norm": 1.7664462179920308, + "learning_rate": 2.944269056323023e-06, + "loss": 1.0086, + "step": 4833 + }, + { + "epoch": 0.3632947542462047, + "grad_norm": 1.7592664342960733, + "learning_rate": 2.9438398640547453e-06, + "loss": 0.8769, + "step": 4834 + }, + { + "epoch": 0.3633699083120397, + "grad_norm": 3.246635309207619, + "learning_rate": 2.943410615860342e-06, + "loss": 0.9736, + "step": 4835 + }, + { + "epoch": 0.36344506237787466, + "grad_norm": 1.696354840045905, + "learning_rate": 2.9429813117652478e-06, + "loss": 1.0318, + "step": 4836 + }, + { + "epoch": 0.36352021644370963, + "grad_norm": 2.961780599171207, + "learning_rate": 2.942551951794899e-06, + "loss": 1.0764, + "step": 4837 + }, + { + "epoch": 0.36359537050954455, + "grad_norm": 2.3485272358027123, + "learning_rate": 2.942122535974738e-06, + "loss": 0.9846, + "step": 4838 + }, + { + "epoch": 0.3636705245753795, + "grad_norm": 1.3251436767413443, + "learning_rate": 2.9416930643302086e-06, + "loss": 1.0296, + "step": 4839 + }, + { + "epoch": 0.3637456786412145, + "grad_norm": 2.7641941830593115, + "learning_rate": 2.9412635368867596e-06, + "loss": 1.0357, + "step": 4840 + }, + { + "epoch": 0.36382083270704946, + "grad_norm": 1.3191261604742814, + "learning_rate": 2.9408339536698422e-06, + "loss": 0.9193, + "step": 4841 + }, + { + "epoch": 0.3638959867728844, + "grad_norm": 1.9297901079952031, + "learning_rate": 2.9404043147049097e-06, + "loss": 0.9939, + "step": 4842 + }, + { + "epoch": 0.3639711408387194, + "grad_norm": 1.964491583204407, + "learning_rate": 2.9399746200174206e-06, + "loss": 1.0028, + "step": 4843 + }, + { + "epoch": 0.3640462949045543, + "grad_norm": 1.2942273793387047, + "learning_rate": 2.939544869632836e-06, + "loss": 0.9678, + "step": 4844 + }, + { + "epoch": 0.3641214489703893, + "grad_norm": 1.6945609259392154, + "learning_rate": 2.9391150635766194e-06, + "loss": 0.981, + "step": 4845 + }, + { + "epoch": 0.36419660303622425, + "grad_norm": 1.4891496662833856, + "learning_rate": 2.9386852018742404e-06, + "loss": 0.8888, + "step": 4846 + }, + { + "epoch": 0.3642717571020592, + "grad_norm": 1.6313309221913725, + "learning_rate": 2.938255284551168e-06, + "loss": 1.0028, + "step": 4847 + }, + { + "epoch": 0.3643469111678942, + "grad_norm": 2.356572893517134, + "learning_rate": 2.9378253116328777e-06, + "loss": 1.0451, + "step": 4848 + }, + { + "epoch": 0.36442206523372916, + "grad_norm": 1.7102116826302256, + "learning_rate": 2.937395283144846e-06, + "loss": 1.005, + "step": 4849 + }, + { + "epoch": 0.3644972192995641, + "grad_norm": 1.9314263420704787, + "learning_rate": 2.9369651991125542e-06, + "loss": 0.9737, + "step": 4850 + }, + { + "epoch": 0.36457237336539905, + "grad_norm": 1.6640066934460118, + "learning_rate": 2.9365350595614863e-06, + "loss": 1.0785, + "step": 4851 + }, + { + "epoch": 0.364647527431234, + "grad_norm": 2.3121669632574124, + "learning_rate": 2.936104864517131e-06, + "loss": 1.0039, + "step": 4852 + }, + { + "epoch": 0.364722681497069, + "grad_norm": 2.1108157288662714, + "learning_rate": 2.935674614004977e-06, + "loss": 0.9331, + "step": 4853 + }, + { + "epoch": 0.36479783556290396, + "grad_norm": 1.924875413900678, + "learning_rate": 2.9352443080505192e-06, + "loss": 0.9371, + "step": 4854 + }, + { + "epoch": 0.36487298962873893, + "grad_norm": 1.3661136006275763, + "learning_rate": 2.934813946679255e-06, + "loss": 1.0523, + "step": 4855 + }, + { + "epoch": 0.3649481436945739, + "grad_norm": 1.8574131465028878, + "learning_rate": 2.9343835299166846e-06, + "loss": 0.9595, + "step": 4856 + }, + { + "epoch": 0.3650232977604088, + "grad_norm": 2.7828362902438837, + "learning_rate": 2.9339530577883125e-06, + "loss": 1.0945, + "step": 4857 + }, + { + "epoch": 0.3650984518262438, + "grad_norm": 1.7675622760209686, + "learning_rate": 2.9335225303196454e-06, + "loss": 0.8635, + "step": 4858 + }, + { + "epoch": 0.36517360589207876, + "grad_norm": 1.8809557104742325, + "learning_rate": 2.933091947536193e-06, + "loss": 1.1727, + "step": 4859 + }, + { + "epoch": 0.3652487599579137, + "grad_norm": 1.8327557202948344, + "learning_rate": 2.93266130946347e-06, + "loss": 0.9569, + "step": 4860 + }, + { + "epoch": 0.3653239140237487, + "grad_norm": 3.3678477761760037, + "learning_rate": 2.9322306161269933e-06, + "loss": 1.0089, + "step": 4861 + }, + { + "epoch": 0.36539906808958367, + "grad_norm": 1.5243001989426432, + "learning_rate": 2.931799867552282e-06, + "loss": 1.0243, + "step": 4862 + }, + { + "epoch": 0.3654742221554186, + "grad_norm": 1.3324162438641864, + "learning_rate": 2.931369063764862e-06, + "loss": 1.0087, + "step": 4863 + }, + { + "epoch": 0.36554937622125355, + "grad_norm": 1.5702486547140315, + "learning_rate": 2.9309382047902574e-06, + "loss": 0.8879, + "step": 4864 + }, + { + "epoch": 0.3656245302870885, + "grad_norm": 1.3790781924325015, + "learning_rate": 2.9305072906539993e-06, + "loss": 0.9774, + "step": 4865 + }, + { + "epoch": 0.3656996843529235, + "grad_norm": 0.6248129564783296, + "learning_rate": 2.930076321381622e-06, + "loss": 0.8445, + "step": 4866 + }, + { + "epoch": 0.36577483841875846, + "grad_norm": 1.5640203327359743, + "learning_rate": 2.92964529699866e-06, + "loss": 1.012, + "step": 4867 + }, + { + "epoch": 0.36584999248459343, + "grad_norm": 1.8997793596551427, + "learning_rate": 2.9292142175306548e-06, + "loss": 1.0056, + "step": 4868 + }, + { + "epoch": 0.3659251465504284, + "grad_norm": 1.471019533526859, + "learning_rate": 2.9287830830031492e-06, + "loss": 0.9366, + "step": 4869 + }, + { + "epoch": 0.3660003006162633, + "grad_norm": 1.8580411085379263, + "learning_rate": 2.9283518934416892e-06, + "loss": 1.0464, + "step": 4870 + }, + { + "epoch": 0.3660754546820983, + "grad_norm": 1.6025045462081147, + "learning_rate": 2.927920648871825e-06, + "loss": 0.8823, + "step": 4871 + }, + { + "epoch": 0.36615060874793326, + "grad_norm": 1.2831502639357002, + "learning_rate": 2.9274893493191084e-06, + "loss": 1.0451, + "step": 4872 + }, + { + "epoch": 0.36622576281376823, + "grad_norm": 0.670634832887831, + "learning_rate": 2.9270579948090962e-06, + "loss": 0.8317, + "step": 4873 + }, + { + "epoch": 0.3663009168796032, + "grad_norm": 1.5739937519739562, + "learning_rate": 2.9266265853673483e-06, + "loss": 0.9516, + "step": 4874 + }, + { + "epoch": 0.36637607094543817, + "grad_norm": 1.4792758292450559, + "learning_rate": 2.926195121019427e-06, + "loss": 0.8408, + "step": 4875 + }, + { + "epoch": 0.3664512250112731, + "grad_norm": 1.675267209616888, + "learning_rate": 2.9257636017908984e-06, + "loss": 0.9933, + "step": 4876 + }, + { + "epoch": 0.36652637907710806, + "grad_norm": 2.1587591637360015, + "learning_rate": 2.925332027707331e-06, + "loss": 1.0198, + "step": 4877 + }, + { + "epoch": 0.366601533142943, + "grad_norm": 1.4738184544629973, + "learning_rate": 2.9249003987942976e-06, + "loss": 1.0176, + "step": 4878 + }, + { + "epoch": 0.366676687208778, + "grad_norm": 3.698805337783924, + "learning_rate": 2.924468715077374e-06, + "loss": 1.0559, + "step": 4879 + }, + { + "epoch": 0.36675184127461297, + "grad_norm": 2.312961841550515, + "learning_rate": 2.9240369765821392e-06, + "loss": 1.039, + "step": 4880 + }, + { + "epoch": 0.36682699534044794, + "grad_norm": 1.5485825794976777, + "learning_rate": 2.9236051833341745e-06, + "loss": 1.0267, + "step": 4881 + }, + { + "epoch": 0.3669021494062829, + "grad_norm": 1.5954080820172292, + "learning_rate": 2.9231733353590668e-06, + "loss": 1.0147, + "step": 4882 + }, + { + "epoch": 0.3669773034721178, + "grad_norm": 1.4552484399852377, + "learning_rate": 2.9227414326824027e-06, + "loss": 1.089, + "step": 4883 + }, + { + "epoch": 0.3670524575379528, + "grad_norm": 3.206106296464376, + "learning_rate": 2.9223094753297767e-06, + "loss": 0.9511, + "step": 4884 + }, + { + "epoch": 0.36712761160378776, + "grad_norm": 1.6034226942373193, + "learning_rate": 2.9218774633267815e-06, + "loss": 0.9404, + "step": 4885 + }, + { + "epoch": 0.36720276566962273, + "grad_norm": 2.0962529949787836, + "learning_rate": 2.9214453966990174e-06, + "loss": 1.0858, + "step": 4886 + }, + { + "epoch": 0.3672779197354577, + "grad_norm": 2.950486068767685, + "learning_rate": 2.9210132754720845e-06, + "loss": 0.9908, + "step": 4887 + }, + { + "epoch": 0.3673530738012927, + "grad_norm": 1.5793214549327392, + "learning_rate": 2.9205810996715885e-06, + "loss": 1.0023, + "step": 4888 + }, + { + "epoch": 0.3674282278671276, + "grad_norm": 1.6734997706576948, + "learning_rate": 2.9201488693231366e-06, + "loss": 1.0339, + "step": 4889 + }, + { + "epoch": 0.36750338193296256, + "grad_norm": 1.8544556753251358, + "learning_rate": 2.9197165844523416e-06, + "loss": 0.9092, + "step": 4890 + }, + { + "epoch": 0.36757853599879753, + "grad_norm": 1.918339400205898, + "learning_rate": 2.9192842450848164e-06, + "loss": 0.9425, + "step": 4891 + }, + { + "epoch": 0.3676536900646325, + "grad_norm": 1.9368395141029375, + "learning_rate": 2.91885185124618e-06, + "loss": 1.004, + "step": 4892 + }, + { + "epoch": 0.36772884413046747, + "grad_norm": 2.1951756553490007, + "learning_rate": 2.918419402962053e-06, + "loss": 0.9463, + "step": 4893 + }, + { + "epoch": 0.36780399819630244, + "grad_norm": 9.635042574382132, + "learning_rate": 2.917986900258059e-06, + "loss": 0.9427, + "step": 4894 + }, + { + "epoch": 0.36787915226213735, + "grad_norm": 1.90093265029261, + "learning_rate": 2.9175543431598257e-06, + "loss": 0.9164, + "step": 4895 + }, + { + "epoch": 0.3679543063279723, + "grad_norm": 1.5830634623305508, + "learning_rate": 2.917121731692985e-06, + "loss": 0.9134, + "step": 4896 + }, + { + "epoch": 0.3680294603938073, + "grad_norm": 1.857512174029907, + "learning_rate": 2.9166890658831695e-06, + "loss": 0.9252, + "step": 4897 + }, + { + "epoch": 0.36810461445964227, + "grad_norm": 1.483658065325875, + "learning_rate": 2.9162563457560157e-06, + "loss": 1.0052, + "step": 4898 + }, + { + "epoch": 0.36817976852547724, + "grad_norm": 1.4497829274771445, + "learning_rate": 2.915823571337166e-06, + "loss": 1.0371, + "step": 4899 + }, + { + "epoch": 0.3682549225913122, + "grad_norm": 4.893736608104481, + "learning_rate": 2.915390742652262e-06, + "loss": 0.9573, + "step": 4900 + }, + { + "epoch": 0.3683300766571472, + "grad_norm": 1.9348157315896801, + "learning_rate": 2.914957859726952e-06, + "loss": 1.0428, + "step": 4901 + }, + { + "epoch": 0.3684052307229821, + "grad_norm": 1.628207049800941, + "learning_rate": 2.9145249225868848e-06, + "loss": 1.0288, + "step": 4902 + }, + { + "epoch": 0.36848038478881706, + "grad_norm": 1.8245800893887871, + "learning_rate": 2.9140919312577134e-06, + "loss": 1.1233, + "step": 4903 + }, + { + "epoch": 0.36855553885465203, + "grad_norm": 1.6760519027002247, + "learning_rate": 2.9136588857650956e-06, + "loss": 0.9632, + "step": 4904 + }, + { + "epoch": 0.368630692920487, + "grad_norm": 2.426766163372201, + "learning_rate": 2.9132257861346897e-06, + "loss": 1.1233, + "step": 4905 + }, + { + "epoch": 0.368705846986322, + "grad_norm": 1.83300748732599, + "learning_rate": 2.912792632392159e-06, + "loss": 1.0431, + "step": 4906 + }, + { + "epoch": 0.36878100105215694, + "grad_norm": 1.767727300766577, + "learning_rate": 2.9123594245631702e-06, + "loss": 0.9449, + "step": 4907 + }, + { + "epoch": 0.36885615511799186, + "grad_norm": 1.5068985140471713, + "learning_rate": 2.9119261626733915e-06, + "loss": 0.9934, + "step": 4908 + }, + { + "epoch": 0.3689313091838268, + "grad_norm": 1.7104782115102757, + "learning_rate": 2.911492846748495e-06, + "loss": 0.9558, + "step": 4909 + }, + { + "epoch": 0.3690064632496618, + "grad_norm": 1.8780636871028444, + "learning_rate": 2.911059476814158e-06, + "loss": 1.009, + "step": 4910 + }, + { + "epoch": 0.36908161731549677, + "grad_norm": 1.446192870202767, + "learning_rate": 2.9106260528960573e-06, + "loss": 1.0079, + "step": 4911 + }, + { + "epoch": 0.36915677138133174, + "grad_norm": 2.5151527861946743, + "learning_rate": 2.910192575019877e-06, + "loss": 1.008, + "step": 4912 + }, + { + "epoch": 0.3692319254471667, + "grad_norm": 1.5927109416626786, + "learning_rate": 2.9097590432113007e-06, + "loss": 0.9255, + "step": 4913 + }, + { + "epoch": 0.3693070795130017, + "grad_norm": 1.8229621300250571, + "learning_rate": 2.909325457496017e-06, + "loss": 1.0383, + "step": 4914 + }, + { + "epoch": 0.3693822335788366, + "grad_norm": 8.015705903892389, + "learning_rate": 2.908891817899718e-06, + "loss": 0.9489, + "step": 4915 + }, + { + "epoch": 0.36945738764467156, + "grad_norm": 1.9271759492080458, + "learning_rate": 2.9084581244480994e-06, + "loss": 1.0667, + "step": 4916 + }, + { + "epoch": 0.36953254171050653, + "grad_norm": 1.3904257213666174, + "learning_rate": 2.908024377166857e-06, + "loss": 0.9929, + "step": 4917 + }, + { + "epoch": 0.3696076957763415, + "grad_norm": 1.5799045160872864, + "learning_rate": 2.9075905760816942e-06, + "loss": 1.0335, + "step": 4918 + }, + { + "epoch": 0.3696828498421765, + "grad_norm": 1.6566979144479925, + "learning_rate": 2.9071567212183138e-06, + "loss": 0.8692, + "step": 4919 + }, + { + "epoch": 0.36975800390801145, + "grad_norm": 2.2009602185146115, + "learning_rate": 2.906722812602424e-06, + "loss": 0.9581, + "step": 4920 + }, + { + "epoch": 0.36983315797384636, + "grad_norm": 1.6341160585515835, + "learning_rate": 2.906288850259736e-06, + "loss": 0.9211, + "step": 4921 + }, + { + "epoch": 0.36990831203968133, + "grad_norm": 1.5375919965807636, + "learning_rate": 2.9058548342159628e-06, + "loss": 0.8493, + "step": 4922 + }, + { + "epoch": 0.3699834661055163, + "grad_norm": 4.3041568712909, + "learning_rate": 2.9054207644968218e-06, + "loss": 1.0019, + "step": 4923 + }, + { + "epoch": 0.37005862017135127, + "grad_norm": 1.7933248860161422, + "learning_rate": 2.904986641128033e-06, + "loss": 1.064, + "step": 4924 + }, + { + "epoch": 0.37013377423718624, + "grad_norm": 1.6860714976716202, + "learning_rate": 2.9045524641353208e-06, + "loss": 1.0081, + "step": 4925 + }, + { + "epoch": 0.3702089283030212, + "grad_norm": 1.4030665073224546, + "learning_rate": 2.904118233544411e-06, + "loss": 0.9975, + "step": 4926 + }, + { + "epoch": 0.3702840823688562, + "grad_norm": 2.0545545179607547, + "learning_rate": 2.9036839493810348e-06, + "loss": 1.1017, + "step": 4927 + }, + { + "epoch": 0.3703592364346911, + "grad_norm": 1.7415098232240858, + "learning_rate": 2.903249611670923e-06, + "loss": 0.9888, + "step": 4928 + }, + { + "epoch": 0.37043439050052607, + "grad_norm": 1.8697772083059954, + "learning_rate": 2.9028152204398135e-06, + "loss": 0.9366, + "step": 4929 + }, + { + "epoch": 0.37050954456636104, + "grad_norm": 1.5877377669929753, + "learning_rate": 2.9023807757134455e-06, + "loss": 1.0776, + "step": 4930 + }, + { + "epoch": 0.370584698632196, + "grad_norm": 1.6872271564456014, + "learning_rate": 2.90194627751756e-06, + "loss": 1.0374, + "step": 4931 + }, + { + "epoch": 0.370659852698031, + "grad_norm": 1.5579792235755712, + "learning_rate": 2.9015117258779045e-06, + "loss": 0.9665, + "step": 4932 + }, + { + "epoch": 0.37073500676386595, + "grad_norm": 1.5391352292609075, + "learning_rate": 2.9010771208202265e-06, + "loss": 1.0315, + "step": 4933 + }, + { + "epoch": 0.37081016082970086, + "grad_norm": 0.8130968176072324, + "learning_rate": 2.900642462370279e-06, + "loss": 0.8895, + "step": 4934 + }, + { + "epoch": 0.37088531489553583, + "grad_norm": 1.6334730938393565, + "learning_rate": 2.900207750553817e-06, + "loss": 0.9246, + "step": 4935 + }, + { + "epoch": 0.3709604689613708, + "grad_norm": 1.4461614641341678, + "learning_rate": 2.899772985396599e-06, + "loss": 1.0031, + "step": 4936 + }, + { + "epoch": 0.3710356230272058, + "grad_norm": 1.170413364601751, + "learning_rate": 2.8993381669243854e-06, + "loss": 1.0196, + "step": 4937 + }, + { + "epoch": 0.37111077709304074, + "grad_norm": 1.3172354873839502, + "learning_rate": 2.8989032951629417e-06, + "loss": 1.042, + "step": 4938 + }, + { + "epoch": 0.3711859311588757, + "grad_norm": 0.8457620084847659, + "learning_rate": 2.898468370138036e-06, + "loss": 0.9064, + "step": 4939 + }, + { + "epoch": 0.37126108522471063, + "grad_norm": 1.6763055154708881, + "learning_rate": 2.8980333918754383e-06, + "loss": 0.9748, + "step": 4940 + }, + { + "epoch": 0.3713362392905456, + "grad_norm": 1.6351828276346532, + "learning_rate": 2.8975983604009244e-06, + "loss": 1.0191, + "step": 4941 + }, + { + "epoch": 0.37141139335638057, + "grad_norm": 2.549677798541081, + "learning_rate": 2.8971632757402694e-06, + "loss": 0.9574, + "step": 4942 + }, + { + "epoch": 0.37148654742221554, + "grad_norm": 2.3292829462914715, + "learning_rate": 2.8967281379192557e-06, + "loss": 1.0475, + "step": 4943 + }, + { + "epoch": 0.3715617014880505, + "grad_norm": 1.5664072393476067, + "learning_rate": 2.8962929469636653e-06, + "loss": 0.9048, + "step": 4944 + }, + { + "epoch": 0.3716368555538855, + "grad_norm": 1.9843687522970053, + "learning_rate": 2.8958577028992866e-06, + "loss": 0.9561, + "step": 4945 + }, + { + "epoch": 0.37171200961972045, + "grad_norm": 2.0303747785779835, + "learning_rate": 2.895422405751908e-06, + "loss": 1.0398, + "step": 4946 + }, + { + "epoch": 0.37178716368555537, + "grad_norm": 1.614378179826902, + "learning_rate": 2.8949870555473226e-06, + "loss": 0.9908, + "step": 4947 + }, + { + "epoch": 0.37186231775139034, + "grad_norm": 1.467646407406379, + "learning_rate": 2.8945516523113275e-06, + "loss": 0.9631, + "step": 4948 + }, + { + "epoch": 0.3719374718172253, + "grad_norm": 2.1480887944997002, + "learning_rate": 2.8941161960697217e-06, + "loss": 0.983, + "step": 4949 + }, + { + "epoch": 0.3720126258830603, + "grad_norm": 1.7798824749413293, + "learning_rate": 2.893680686848307e-06, + "loss": 1.0139, + "step": 4950 + }, + { + "epoch": 0.37208777994889525, + "grad_norm": 1.7994536762851978, + "learning_rate": 2.89324512467289e-06, + "loss": 0.9344, + "step": 4951 + }, + { + "epoch": 0.3721629340147302, + "grad_norm": 2.2195123786486546, + "learning_rate": 2.8928095095692783e-06, + "loss": 1.04, + "step": 4952 + }, + { + "epoch": 0.37223808808056513, + "grad_norm": 1.5172539049224962, + "learning_rate": 2.892373841563285e-06, + "loss": 0.975, + "step": 4953 + }, + { + "epoch": 0.3723132421464001, + "grad_norm": 1.9132761059929828, + "learning_rate": 2.891938120680724e-06, + "loss": 0.9856, + "step": 4954 + }, + { + "epoch": 0.3723883962122351, + "grad_norm": 0.681295728461121, + "learning_rate": 2.891502346947414e-06, + "loss": 0.8272, + "step": 4955 + }, + { + "epoch": 0.37246355027807004, + "grad_norm": 2.266981994442978, + "learning_rate": 2.8910665203891763e-06, + "loss": 1.0203, + "step": 4956 + }, + { + "epoch": 0.372538704343905, + "grad_norm": 1.9108403702861556, + "learning_rate": 2.8906306410318353e-06, + "loss": 1.0577, + "step": 4957 + }, + { + "epoch": 0.37261385840974, + "grad_norm": 1.4100295245008132, + "learning_rate": 2.890194708901218e-06, + "loss": 0.9862, + "step": 4958 + }, + { + "epoch": 0.37268901247557495, + "grad_norm": 1.4202259268148225, + "learning_rate": 2.889758724023155e-06, + "loss": 1.0298, + "step": 4959 + }, + { + "epoch": 0.37276416654140987, + "grad_norm": 2.4392859933865356, + "learning_rate": 2.8893226864234813e-06, + "loss": 0.8467, + "step": 4960 + }, + { + "epoch": 0.37283932060724484, + "grad_norm": 1.6480950456312293, + "learning_rate": 2.8888865961280325e-06, + "loss": 1.0097, + "step": 4961 + }, + { + "epoch": 0.3729144746730798, + "grad_norm": 1.480616579752433, + "learning_rate": 2.888450453162649e-06, + "loss": 0.9895, + "step": 4962 + }, + { + "epoch": 0.3729896287389148, + "grad_norm": 0.697660211635162, + "learning_rate": 2.888014257553175e-06, + "loss": 0.8167, + "step": 4963 + }, + { + "epoch": 0.37306478280474975, + "grad_norm": 0.5935341543077266, + "learning_rate": 2.8875780093254545e-06, + "loss": 0.7951, + "step": 4964 + }, + { + "epoch": 0.3731399368705847, + "grad_norm": 1.378334726650421, + "learning_rate": 2.8871417085053394e-06, + "loss": 0.9867, + "step": 4965 + }, + { + "epoch": 0.37321509093641964, + "grad_norm": 2.1597932093144316, + "learning_rate": 2.88670535511868e-06, + "loss": 1.0833, + "step": 4966 + }, + { + "epoch": 0.3732902450022546, + "grad_norm": 3.0388476925912378, + "learning_rate": 2.886268949191334e-06, + "loss": 0.9359, + "step": 4967 + }, + { + "epoch": 0.3733653990680896, + "grad_norm": 2.611156472247229, + "learning_rate": 2.885832490749158e-06, + "loss": 1.0281, + "step": 4968 + }, + { + "epoch": 0.37344055313392455, + "grad_norm": 3.866621002936114, + "learning_rate": 2.885395979818015e-06, + "loss": 0.9569, + "step": 4969 + }, + { + "epoch": 0.3735157071997595, + "grad_norm": 1.589661908822043, + "learning_rate": 2.8849594164237694e-06, + "loss": 0.9084, + "step": 4970 + }, + { + "epoch": 0.3735908612655945, + "grad_norm": 1.323329120328204, + "learning_rate": 2.8845228005922905e-06, + "loss": 0.9527, + "step": 4971 + }, + { + "epoch": 0.37366601533142946, + "grad_norm": 1.8873739743633227, + "learning_rate": 2.8840861323494482e-06, + "loss": 1.0074, + "step": 4972 + }, + { + "epoch": 0.3737411693972644, + "grad_norm": 1.739265626079094, + "learning_rate": 2.8836494117211177e-06, + "loss": 0.9785, + "step": 4973 + }, + { + "epoch": 0.37381632346309934, + "grad_norm": 2.185559092905974, + "learning_rate": 2.883212638733175e-06, + "loss": 1.0118, + "step": 4974 + }, + { + "epoch": 0.3738914775289343, + "grad_norm": 1.443137434438711, + "learning_rate": 2.8827758134115017e-06, + "loss": 0.981, + "step": 4975 + }, + { + "epoch": 0.3739666315947693, + "grad_norm": 2.1598291960046994, + "learning_rate": 2.8823389357819815e-06, + "loss": 0.9934, + "step": 4976 + }, + { + "epoch": 0.37404178566060425, + "grad_norm": 1.3430282961263946, + "learning_rate": 2.8819020058705003e-06, + "loss": 0.9471, + "step": 4977 + }, + { + "epoch": 0.3741169397264392, + "grad_norm": 1.6225423568110622, + "learning_rate": 2.881465023702948e-06, + "loss": 0.98, + "step": 4978 + }, + { + "epoch": 0.37419209379227414, + "grad_norm": 3.1360656134300315, + "learning_rate": 2.8810279893052184e-06, + "loss": 0.9431, + "step": 4979 + }, + { + "epoch": 0.3742672478581091, + "grad_norm": 1.735102389279892, + "learning_rate": 2.880590902703206e-06, + "loss": 1.0551, + "step": 4980 + }, + { + "epoch": 0.3743424019239441, + "grad_norm": 1.9950114860299626, + "learning_rate": 2.8801537639228107e-06, + "loss": 1.0269, + "step": 4981 + }, + { + "epoch": 0.37441755598977905, + "grad_norm": 1.8364307146886008, + "learning_rate": 2.8797165729899347e-06, + "loss": 1.0265, + "step": 4982 + }, + { + "epoch": 0.374492710055614, + "grad_norm": 2.033103214854321, + "learning_rate": 2.879279329930483e-06, + "loss": 0.9091, + "step": 4983 + }, + { + "epoch": 0.374567864121449, + "grad_norm": 1.9318534648957493, + "learning_rate": 2.8788420347703643e-06, + "loss": 1.0059, + "step": 4984 + }, + { + "epoch": 0.3746430181872839, + "grad_norm": 1.5551717186220484, + "learning_rate": 2.87840468753549e-06, + "loss": 1.0403, + "step": 4985 + }, + { + "epoch": 0.3747181722531189, + "grad_norm": 2.428503796893948, + "learning_rate": 2.8779672882517735e-06, + "loss": 0.9885, + "step": 4986 + }, + { + "epoch": 0.37479332631895385, + "grad_norm": 1.398803748222262, + "learning_rate": 2.877529836945134e-06, + "loss": 0.9683, + "step": 4987 + }, + { + "epoch": 0.3748684803847888, + "grad_norm": 0.7827042967430191, + "learning_rate": 2.8770923336414906e-06, + "loss": 0.9065, + "step": 4988 + }, + { + "epoch": 0.3749436344506238, + "grad_norm": 1.6567156874748805, + "learning_rate": 2.8766547783667686e-06, + "loss": 0.9549, + "step": 4989 + }, + { + "epoch": 0.37501878851645876, + "grad_norm": 1.7733360631223811, + "learning_rate": 2.8762171711468935e-06, + "loss": 1.0041, + "step": 4990 + }, + { + "epoch": 0.3750939425822937, + "grad_norm": 2.0841635257901157, + "learning_rate": 2.8757795120077955e-06, + "loss": 1.0508, + "step": 4991 + }, + { + "epoch": 0.37516909664812864, + "grad_norm": 1.9151617513725279, + "learning_rate": 2.8753418009754082e-06, + "loss": 0.9905, + "step": 4992 + }, + { + "epoch": 0.3752442507139636, + "grad_norm": 1.760524641107767, + "learning_rate": 2.874904038075668e-06, + "loss": 1.0308, + "step": 4993 + }, + { + "epoch": 0.3753194047797986, + "grad_norm": 1.8150529968258837, + "learning_rate": 2.874466223334512e-06, + "loss": 0.9682, + "step": 4994 + }, + { + "epoch": 0.37539455884563355, + "grad_norm": 2.109206920567717, + "learning_rate": 2.8740283567778844e-06, + "loss": 1.0177, + "step": 4995 + }, + { + "epoch": 0.3754697129114685, + "grad_norm": 1.769950833796178, + "learning_rate": 2.87359043843173e-06, + "loss": 0.9047, + "step": 4996 + }, + { + "epoch": 0.3755448669773035, + "grad_norm": 2.103751705030724, + "learning_rate": 2.873152468321997e-06, + "loss": 0.8805, + "step": 4997 + }, + { + "epoch": 0.3756200210431384, + "grad_norm": 1.9889082304169756, + "learning_rate": 2.872714446474636e-06, + "loss": 0.9754, + "step": 4998 + }, + { + "epoch": 0.3756951751089734, + "grad_norm": 1.9663652175544561, + "learning_rate": 2.8722763729156027e-06, + "loss": 1.0141, + "step": 4999 + }, + { + "epoch": 0.37577032917480835, + "grad_norm": 2.1547839944656113, + "learning_rate": 2.8718382476708544e-06, + "loss": 1.0468, + "step": 5000 + }, + { + "epoch": 0.3758454832406433, + "grad_norm": 2.027408194574235, + "learning_rate": 2.8714000707663507e-06, + "loss": 0.957, + "step": 5001 + }, + { + "epoch": 0.3759206373064783, + "grad_norm": 1.9279975440931705, + "learning_rate": 2.8709618422280564e-06, + "loss": 0.9133, + "step": 5002 + }, + { + "epoch": 0.37599579137231326, + "grad_norm": 1.9307768677330435, + "learning_rate": 2.8705235620819377e-06, + "loss": 1.0929, + "step": 5003 + }, + { + "epoch": 0.37607094543814823, + "grad_norm": 1.705334271639242, + "learning_rate": 2.8700852303539647e-06, + "loss": 0.9447, + "step": 5004 + }, + { + "epoch": 0.37614609950398314, + "grad_norm": 2.958197541321648, + "learning_rate": 2.8696468470701096e-06, + "loss": 0.9165, + "step": 5005 + }, + { + "epoch": 0.3762212535698181, + "grad_norm": 1.6530089425954528, + "learning_rate": 2.869208412256349e-06, + "loss": 1.0672, + "step": 5006 + }, + { + "epoch": 0.3762964076356531, + "grad_norm": 1.7624161855911107, + "learning_rate": 2.868769925938662e-06, + "loss": 0.9157, + "step": 5007 + }, + { + "epoch": 0.37637156170148806, + "grad_norm": 2.6640181999214905, + "learning_rate": 2.868331388143029e-06, + "loss": 0.9799, + "step": 5008 + }, + { + "epoch": 0.376446715767323, + "grad_norm": 2.1675458615916923, + "learning_rate": 2.867892798895437e-06, + "loss": 1.035, + "step": 5009 + }, + { + "epoch": 0.376521869833158, + "grad_norm": 1.3657021001005956, + "learning_rate": 2.867454158221873e-06, + "loss": 1.0495, + "step": 5010 + }, + { + "epoch": 0.3765970238989929, + "grad_norm": 1.5546513078780444, + "learning_rate": 2.867015466148329e-06, + "loss": 0.9722, + "step": 5011 + }, + { + "epoch": 0.3766721779648279, + "grad_norm": 1.9119298147304147, + "learning_rate": 2.8665767227007985e-06, + "loss": 0.9061, + "step": 5012 + }, + { + "epoch": 0.37674733203066285, + "grad_norm": 0.7037909427068697, + "learning_rate": 2.866137927905278e-06, + "loss": 0.8405, + "step": 5013 + }, + { + "epoch": 0.3768224860964978, + "grad_norm": 1.7696108094913077, + "learning_rate": 2.865699081787769e-06, + "loss": 1.0331, + "step": 5014 + }, + { + "epoch": 0.3768976401623328, + "grad_norm": 2.231389645606253, + "learning_rate": 2.865260184374275e-06, + "loss": 1.016, + "step": 5015 + }, + { + "epoch": 0.37697279422816776, + "grad_norm": 1.5923852500115496, + "learning_rate": 2.864821235690801e-06, + "loss": 0.9838, + "step": 5016 + }, + { + "epoch": 0.37704794829400273, + "grad_norm": 3.018512351591662, + "learning_rate": 2.8643822357633576e-06, + "loss": 0.832, + "step": 5017 + }, + { + "epoch": 0.37712310235983765, + "grad_norm": 1.8706142631449498, + "learning_rate": 2.863943184617957e-06, + "loss": 0.9699, + "step": 5018 + }, + { + "epoch": 0.3771982564256726, + "grad_norm": 2.2383277981430014, + "learning_rate": 2.8635040822806135e-06, + "loss": 0.9654, + "step": 5019 + }, + { + "epoch": 0.3772734104915076, + "grad_norm": 1.5247599021889053, + "learning_rate": 2.8630649287773475e-06, + "loss": 1.0562, + "step": 5020 + }, + { + "epoch": 0.37734856455734256, + "grad_norm": 1.7682262602282661, + "learning_rate": 2.862625724134179e-06, + "loss": 1.003, + "step": 5021 + }, + { + "epoch": 0.37742371862317753, + "grad_norm": 1.3922706467859158, + "learning_rate": 2.8621864683771337e-06, + "loss": 1.0147, + "step": 5022 + }, + { + "epoch": 0.3774988726890125, + "grad_norm": 1.952181718016871, + "learning_rate": 2.8617471615322377e-06, + "loss": 0.9134, + "step": 5023 + }, + { + "epoch": 0.3775740267548474, + "grad_norm": 1.6805123366516252, + "learning_rate": 2.8613078036255233e-06, + "loss": 0.9343, + "step": 5024 + }, + { + "epoch": 0.3776491808206824, + "grad_norm": 1.7242106303968876, + "learning_rate": 2.8608683946830236e-06, + "loss": 0.9223, + "step": 5025 + }, + { + "epoch": 0.37772433488651735, + "grad_norm": 1.5871984420672585, + "learning_rate": 2.8604289347307746e-06, + "loss": 1.0386, + "step": 5026 + }, + { + "epoch": 0.3777994889523523, + "grad_norm": 1.422167234467082, + "learning_rate": 2.859989423794816e-06, + "loss": 0.891, + "step": 5027 + }, + { + "epoch": 0.3778746430181873, + "grad_norm": 1.6132842795872355, + "learning_rate": 2.8595498619011916e-06, + "loss": 1.0156, + "step": 5028 + }, + { + "epoch": 0.37794979708402227, + "grad_norm": 1.74279425622599, + "learning_rate": 2.8591102490759468e-06, + "loss": 0.9774, + "step": 5029 + }, + { + "epoch": 0.3780249511498572, + "grad_norm": 1.6767032105626312, + "learning_rate": 2.858670585345129e-06, + "loss": 1.1123, + "step": 5030 + }, + { + "epoch": 0.37810010521569215, + "grad_norm": 1.5294928297290664, + "learning_rate": 2.8582308707347913e-06, + "loss": 0.9745, + "step": 5031 + }, + { + "epoch": 0.3781752592815271, + "grad_norm": 0.724559442902467, + "learning_rate": 2.857791105270988e-06, + "loss": 0.8671, + "step": 5032 + }, + { + "epoch": 0.3782504133473621, + "grad_norm": 1.7679944638730882, + "learning_rate": 2.8573512889797773e-06, + "loss": 1.0357, + "step": 5033 + }, + { + "epoch": 0.37832556741319706, + "grad_norm": 3.128086701881798, + "learning_rate": 2.8569114218872195e-06, + "loss": 1.0116, + "step": 5034 + }, + { + "epoch": 0.37840072147903203, + "grad_norm": 2.5121930328661257, + "learning_rate": 2.856471504019379e-06, + "loss": 0.8866, + "step": 5035 + }, + { + "epoch": 0.378475875544867, + "grad_norm": 1.5509193032261959, + "learning_rate": 2.856031535402321e-06, + "loss": 0.9091, + "step": 5036 + }, + { + "epoch": 0.3785510296107019, + "grad_norm": 1.8450670226000876, + "learning_rate": 2.8555915160621184e-06, + "loss": 1.047, + "step": 5037 + }, + { + "epoch": 0.3786261836765369, + "grad_norm": 1.9673558703202119, + "learning_rate": 2.8551514460248406e-06, + "loss": 1.1023, + "step": 5038 + }, + { + "epoch": 0.37870133774237186, + "grad_norm": 1.985875080847473, + "learning_rate": 2.8547113253165666e-06, + "loss": 1.113, + "step": 5039 + }, + { + "epoch": 0.3787764918082068, + "grad_norm": 2.022971198947538, + "learning_rate": 2.8542711539633723e-06, + "loss": 1.035, + "step": 5040 + }, + { + "epoch": 0.3788516458740418, + "grad_norm": 1.9297722554609333, + "learning_rate": 2.8538309319913413e-06, + "loss": 0.9163, + "step": 5041 + }, + { + "epoch": 0.37892679993987677, + "grad_norm": 2.1276311853219285, + "learning_rate": 2.8533906594265588e-06, + "loss": 0.9605, + "step": 5042 + }, + { + "epoch": 0.3790019540057117, + "grad_norm": 1.6849705140564764, + "learning_rate": 2.852950336295111e-06, + "loss": 0.8541, + "step": 5043 + }, + { + "epoch": 0.37907710807154665, + "grad_norm": 1.5979174561314164, + "learning_rate": 2.8525099626230894e-06, + "loss": 0.8938, + "step": 5044 + }, + { + "epoch": 0.3791522621373816, + "grad_norm": 2.1276317456129914, + "learning_rate": 2.8520695384365887e-06, + "loss": 0.8779, + "step": 5045 + }, + { + "epoch": 0.3792274162032166, + "grad_norm": 1.8978816973091872, + "learning_rate": 2.851629063761705e-06, + "loss": 1.0405, + "step": 5046 + }, + { + "epoch": 0.37930257026905156, + "grad_norm": 1.7018189628416995, + "learning_rate": 2.8511885386245373e-06, + "loss": 1.0089, + "step": 5047 + }, + { + "epoch": 0.37937772433488653, + "grad_norm": 1.8448036545274769, + "learning_rate": 2.8507479630511905e-06, + "loss": 0.9465, + "step": 5048 + }, + { + "epoch": 0.3794528784007215, + "grad_norm": 0.9617430559150082, + "learning_rate": 2.850307337067768e-06, + "loss": 0.9979, + "step": 5049 + }, + { + "epoch": 0.3795280324665564, + "grad_norm": 2.048946351748564, + "learning_rate": 2.849866660700381e-06, + "loss": 0.8539, + "step": 5050 + }, + { + "epoch": 0.3796031865323914, + "grad_norm": 1.380751199545297, + "learning_rate": 2.8494259339751396e-06, + "loss": 1.0379, + "step": 5051 + }, + { + "epoch": 0.37967834059822636, + "grad_norm": 1.7552754043155498, + "learning_rate": 2.8489851569181584e-06, + "loss": 0.9131, + "step": 5052 + }, + { + "epoch": 0.37975349466406133, + "grad_norm": 1.3749092678998749, + "learning_rate": 2.848544329555556e-06, + "loss": 1.023, + "step": 5053 + }, + { + "epoch": 0.3798286487298963, + "grad_norm": 1.3965476281839637, + "learning_rate": 2.8481034519134524e-06, + "loss": 0.9804, + "step": 5054 + }, + { + "epoch": 0.37990380279573127, + "grad_norm": 3.368790601156581, + "learning_rate": 2.8476625240179726e-06, + "loss": 1.0164, + "step": 5055 + }, + { + "epoch": 0.3799789568615662, + "grad_norm": 2.490453134539603, + "learning_rate": 2.847221545895241e-06, + "loss": 0.9402, + "step": 5056 + }, + { + "epoch": 0.38005411092740116, + "grad_norm": 2.142150001559393, + "learning_rate": 2.8467805175713897e-06, + "loss": 1.0011, + "step": 5057 + }, + { + "epoch": 0.3801292649932361, + "grad_norm": 1.426717623145252, + "learning_rate": 2.84633943907255e-06, + "loss": 0.9882, + "step": 5058 + }, + { + "epoch": 0.3802044190590711, + "grad_norm": 3.873272295310595, + "learning_rate": 2.8458983104248575e-06, + "loss": 1.0172, + "step": 5059 + }, + { + "epoch": 0.38027957312490607, + "grad_norm": 1.3941697792859093, + "learning_rate": 2.8454571316544504e-06, + "loss": 0.9631, + "step": 5060 + }, + { + "epoch": 0.38035472719074104, + "grad_norm": 1.5634529259710044, + "learning_rate": 2.845015902787472e-06, + "loss": 0.9565, + "step": 5061 + }, + { + "epoch": 0.380429881256576, + "grad_norm": 1.929478251737716, + "learning_rate": 2.8445746238500647e-06, + "loss": 0.9078, + "step": 5062 + }, + { + "epoch": 0.3805050353224109, + "grad_norm": 1.7915137543593451, + "learning_rate": 2.8441332948683768e-06, + "loss": 0.8767, + "step": 5063 + }, + { + "epoch": 0.3805801893882459, + "grad_norm": 1.1267429202043342, + "learning_rate": 2.8436919158685594e-06, + "loss": 0.9386, + "step": 5064 + }, + { + "epoch": 0.38065534345408086, + "grad_norm": 3.430301550463431, + "learning_rate": 2.8432504868767648e-06, + "loss": 1.0265, + "step": 5065 + }, + { + "epoch": 0.38073049751991583, + "grad_norm": 2.0708785524810422, + "learning_rate": 2.84280900791915e-06, + "loss": 1.0045, + "step": 5066 + }, + { + "epoch": 0.3808056515857508, + "grad_norm": 2.3113265282034052, + "learning_rate": 2.8423674790218737e-06, + "loss": 0.8955, + "step": 5067 + }, + { + "epoch": 0.3808808056515858, + "grad_norm": 2.022894591326599, + "learning_rate": 2.841925900211099e-06, + "loss": 0.9029, + "step": 5068 + }, + { + "epoch": 0.3809559597174207, + "grad_norm": 1.850347450208452, + "learning_rate": 2.841484271512991e-06, + "loss": 0.9425, + "step": 5069 + }, + { + "epoch": 0.38103111378325566, + "grad_norm": 2.1804091923451505, + "learning_rate": 2.8410425929537175e-06, + "loss": 0.9494, + "step": 5070 + }, + { + "epoch": 0.38110626784909063, + "grad_norm": 1.5756030184745686, + "learning_rate": 2.8406008645594493e-06, + "loss": 0.9454, + "step": 5071 + }, + { + "epoch": 0.3811814219149256, + "grad_norm": 2.1204857841866094, + "learning_rate": 2.840159086356362e-06, + "loss": 1.0106, + "step": 5072 + }, + { + "epoch": 0.38125657598076057, + "grad_norm": 2.0535299427272764, + "learning_rate": 2.839717258370631e-06, + "loss": 0.9985, + "step": 5073 + }, + { + "epoch": 0.38133173004659554, + "grad_norm": 1.9039983740406947, + "learning_rate": 2.8392753806284367e-06, + "loss": 0.9378, + "step": 5074 + }, + { + "epoch": 0.38140688411243046, + "grad_norm": 2.316100024716853, + "learning_rate": 2.838833453155963e-06, + "loss": 1.0534, + "step": 5075 + }, + { + "epoch": 0.3814820381782654, + "grad_norm": 1.8268300874496388, + "learning_rate": 2.8383914759793944e-06, + "loss": 0.9975, + "step": 5076 + }, + { + "epoch": 0.3815571922441004, + "grad_norm": 2.3487138190941668, + "learning_rate": 2.8379494491249214e-06, + "loss": 1.0708, + "step": 5077 + }, + { + "epoch": 0.38163234630993537, + "grad_norm": 1.756345687920255, + "learning_rate": 2.8375073726187334e-06, + "loss": 0.8973, + "step": 5078 + }, + { + "epoch": 0.38170750037577034, + "grad_norm": 1.310854106883169, + "learning_rate": 2.8370652464870277e-06, + "loss": 1.0302, + "step": 5079 + }, + { + "epoch": 0.3817826544416053, + "grad_norm": 1.2634851238098073, + "learning_rate": 2.836623070756e-06, + "loss": 0.9829, + "step": 5080 + }, + { + "epoch": 0.3818578085074403, + "grad_norm": 2.0086448517526465, + "learning_rate": 2.836180845451852e-06, + "loss": 0.9683, + "step": 5081 + }, + { + "epoch": 0.3819329625732752, + "grad_norm": 2.2850982397818482, + "learning_rate": 2.835738570600787e-06, + "loss": 0.9298, + "step": 5082 + }, + { + "epoch": 0.38200811663911016, + "grad_norm": 3.3996633138566836, + "learning_rate": 2.835296246229012e-06, + "loss": 1.0368, + "step": 5083 + }, + { + "epoch": 0.38208327070494513, + "grad_norm": 1.9008846130990946, + "learning_rate": 2.8348538723627356e-06, + "loss": 0.9263, + "step": 5084 + }, + { + "epoch": 0.3821584247707801, + "grad_norm": 1.9354167936863633, + "learning_rate": 2.83441144902817e-06, + "loss": 1.0566, + "step": 5085 + }, + { + "epoch": 0.3822335788366151, + "grad_norm": 2.2453841440216045, + "learning_rate": 2.8339689762515307e-06, + "loss": 0.9517, + "step": 5086 + }, + { + "epoch": 0.38230873290245004, + "grad_norm": 2.467274478396274, + "learning_rate": 2.8335264540590366e-06, + "loss": 0.9389, + "step": 5087 + }, + { + "epoch": 0.38238388696828496, + "grad_norm": 1.32703988120397, + "learning_rate": 2.833083882476908e-06, + "loss": 0.9985, + "step": 5088 + }, + { + "epoch": 0.38245904103411993, + "grad_norm": 1.8875050298358518, + "learning_rate": 2.8326412615313695e-06, + "loss": 0.9885, + "step": 5089 + }, + { + "epoch": 0.3825341950999549, + "grad_norm": 2.7600392977363253, + "learning_rate": 2.8321985912486476e-06, + "loss": 0.9825, + "step": 5090 + }, + { + "epoch": 0.38260934916578987, + "grad_norm": 1.5384958841084695, + "learning_rate": 2.8317558716549727e-06, + "loss": 0.9474, + "step": 5091 + }, + { + "epoch": 0.38268450323162484, + "grad_norm": 1.4072991590020427, + "learning_rate": 2.8313131027765774e-06, + "loss": 1.0178, + "step": 5092 + }, + { + "epoch": 0.3827596572974598, + "grad_norm": 1.8944448549687205, + "learning_rate": 2.830870284639697e-06, + "loss": 1.0407, + "step": 5093 + }, + { + "epoch": 0.3828348113632948, + "grad_norm": 0.7132158765468788, + "learning_rate": 2.830427417270571e-06, + "loss": 0.8758, + "step": 5094 + }, + { + "epoch": 0.3829099654291297, + "grad_norm": 1.9860361310355577, + "learning_rate": 2.829984500695441e-06, + "loss": 1.032, + "step": 5095 + }, + { + "epoch": 0.38298511949496467, + "grad_norm": 1.4267573111844603, + "learning_rate": 2.8295415349405508e-06, + "loss": 1.0218, + "step": 5096 + }, + { + "epoch": 0.38306027356079964, + "grad_norm": 5.8554808022933, + "learning_rate": 2.8290985200321477e-06, + "loss": 0.9143, + "step": 5097 + }, + { + "epoch": 0.3831354276266346, + "grad_norm": 0.8133344033760578, + "learning_rate": 2.8286554559964826e-06, + "loss": 0.8033, + "step": 5098 + }, + { + "epoch": 0.3832105816924696, + "grad_norm": 1.831007877005817, + "learning_rate": 2.8282123428598096e-06, + "loss": 0.9557, + "step": 5099 + }, + { + "epoch": 0.38328573575830455, + "grad_norm": 1.8385037282283974, + "learning_rate": 2.8277691806483824e-06, + "loss": 1.0272, + "step": 5100 + }, + { + "epoch": 0.38336088982413946, + "grad_norm": 2.1841674487789207, + "learning_rate": 2.8273259693884625e-06, + "loss": 1.0519, + "step": 5101 + }, + { + "epoch": 0.38343604388997443, + "grad_norm": 1.6985348081393739, + "learning_rate": 2.8268827091063105e-06, + "loss": 1.0551, + "step": 5102 + }, + { + "epoch": 0.3835111979558094, + "grad_norm": 1.5302372132708002, + "learning_rate": 2.8264393998281916e-06, + "loss": 0.8936, + "step": 5103 + }, + { + "epoch": 0.3835863520216444, + "grad_norm": 1.4143032480229139, + "learning_rate": 2.825996041580373e-06, + "loss": 0.9754, + "step": 5104 + }, + { + "epoch": 0.38366150608747934, + "grad_norm": 1.5825423557103258, + "learning_rate": 2.825552634389127e-06, + "loss": 0.9246, + "step": 5105 + }, + { + "epoch": 0.3837366601533143, + "grad_norm": 0.8377797513577931, + "learning_rate": 2.8251091782807265e-06, + "loss": 0.8486, + "step": 5106 + }, + { + "epoch": 0.3838118142191493, + "grad_norm": 2.2980395628044255, + "learning_rate": 2.8246656732814463e-06, + "loss": 0.9125, + "step": 5107 + }, + { + "epoch": 0.3838869682849842, + "grad_norm": 2.6447019986766374, + "learning_rate": 2.8242221194175676e-06, + "loss": 0.9354, + "step": 5108 + }, + { + "epoch": 0.38396212235081917, + "grad_norm": 1.6916610842958746, + "learning_rate": 2.8237785167153726e-06, + "loss": 1.0155, + "step": 5109 + }, + { + "epoch": 0.38403727641665414, + "grad_norm": 1.309514146207685, + "learning_rate": 2.8233348652011456e-06, + "loss": 1.0317, + "step": 5110 + }, + { + "epoch": 0.3841124304824891, + "grad_norm": 1.549143582288857, + "learning_rate": 2.8228911649011755e-06, + "loss": 1.0566, + "step": 5111 + }, + { + "epoch": 0.3841875845483241, + "grad_norm": 1.8280906347973598, + "learning_rate": 2.8224474158417526e-06, + "loss": 1.0225, + "step": 5112 + }, + { + "epoch": 0.38426273861415905, + "grad_norm": 1.8916532384365017, + "learning_rate": 2.8220036180491703e-06, + "loss": 0.9778, + "step": 5113 + }, + { + "epoch": 0.38433789267999396, + "grad_norm": 2.998665035772608, + "learning_rate": 2.8215597715497266e-06, + "loss": 0.9867, + "step": 5114 + }, + { + "epoch": 0.38441304674582893, + "grad_norm": 1.7899476239730514, + "learning_rate": 2.8211158763697205e-06, + "loss": 0.9267, + "step": 5115 + }, + { + "epoch": 0.3844882008116639, + "grad_norm": 2.3380266626435064, + "learning_rate": 2.820671932535455e-06, + "loss": 1.0077, + "step": 5116 + }, + { + "epoch": 0.3845633548774989, + "grad_norm": 1.7401697080437208, + "learning_rate": 2.8202279400732343e-06, + "loss": 0.9365, + "step": 5117 + }, + { + "epoch": 0.38463850894333385, + "grad_norm": 1.5885506885043077, + "learning_rate": 2.819783899009367e-06, + "loss": 0.9433, + "step": 5118 + }, + { + "epoch": 0.3847136630091688, + "grad_norm": 1.5585945913365131, + "learning_rate": 2.819339809370165e-06, + "loss": 0.9417, + "step": 5119 + }, + { + "epoch": 0.38478881707500373, + "grad_norm": 2.6722981078893873, + "learning_rate": 2.8188956711819413e-06, + "loss": 0.927, + "step": 5120 + }, + { + "epoch": 0.3848639711408387, + "grad_norm": 1.7074892315727468, + "learning_rate": 2.818451484471014e-06, + "loss": 1.0435, + "step": 5121 + }, + { + "epoch": 0.38493912520667367, + "grad_norm": 1.5696026920143864, + "learning_rate": 2.8180072492637016e-06, + "loss": 0.9428, + "step": 5122 + }, + { + "epoch": 0.38501427927250864, + "grad_norm": 1.8970782276366653, + "learning_rate": 2.817562965586328e-06, + "loss": 0.8606, + "step": 5123 + }, + { + "epoch": 0.3850894333383436, + "grad_norm": 1.6141292384660744, + "learning_rate": 2.8171186334652174e-06, + "loss": 1.0989, + "step": 5124 + }, + { + "epoch": 0.3851645874041786, + "grad_norm": 0.7180973904898927, + "learning_rate": 2.8166742529266988e-06, + "loss": 0.7985, + "step": 5125 + }, + { + "epoch": 0.38523974147001355, + "grad_norm": 1.6425633197348524, + "learning_rate": 2.8162298239971036e-06, + "loss": 1.0218, + "step": 5126 + }, + { + "epoch": 0.38531489553584847, + "grad_norm": 2.5127734970103917, + "learning_rate": 2.8157853467027665e-06, + "loss": 1.0855, + "step": 5127 + }, + { + "epoch": 0.38539004960168344, + "grad_norm": 0.7525440618852396, + "learning_rate": 2.815340821070023e-06, + "loss": 0.8762, + "step": 5128 + }, + { + "epoch": 0.3854652036675184, + "grad_norm": 2.131477021686096, + "learning_rate": 2.8148962471252135e-06, + "loss": 1.0274, + "step": 5129 + }, + { + "epoch": 0.3855403577333534, + "grad_norm": 2.5532876008455916, + "learning_rate": 2.8144516248946813e-06, + "loss": 0.8655, + "step": 5130 + }, + { + "epoch": 0.38561551179918835, + "grad_norm": 1.5185861663990101, + "learning_rate": 2.8140069544047717e-06, + "loss": 0.9871, + "step": 5131 + }, + { + "epoch": 0.3856906658650233, + "grad_norm": 1.8284849806876238, + "learning_rate": 2.813562235681833e-06, + "loss": 0.9874, + "step": 5132 + }, + { + "epoch": 0.38576581993085823, + "grad_norm": 0.6686359664807838, + "learning_rate": 2.813117468752216e-06, + "loss": 0.7936, + "step": 5133 + }, + { + "epoch": 0.3858409739966932, + "grad_norm": 2.60834661080228, + "learning_rate": 2.812672653642276e-06, + "loss": 1.0022, + "step": 5134 + }, + { + "epoch": 0.3859161280625282, + "grad_norm": 1.8339227538070277, + "learning_rate": 2.812227790378369e-06, + "loss": 1.0438, + "step": 5135 + }, + { + "epoch": 0.38599128212836314, + "grad_norm": 1.4375340001604942, + "learning_rate": 2.811782878986855e-06, + "loss": 0.9891, + "step": 5136 + }, + { + "epoch": 0.3860664361941981, + "grad_norm": 1.5076740576876169, + "learning_rate": 2.811337919494097e-06, + "loss": 1.1284, + "step": 5137 + }, + { + "epoch": 0.3861415902600331, + "grad_norm": 3.543877095091237, + "learning_rate": 2.8108929119264608e-06, + "loss": 0.8357, + "step": 5138 + }, + { + "epoch": 0.38621674432586806, + "grad_norm": 1.7685710668200314, + "learning_rate": 2.8104478563103145e-06, + "loss": 0.9812, + "step": 5139 + }, + { + "epoch": 0.38629189839170297, + "grad_norm": 1.7476521136832197, + "learning_rate": 2.8100027526720283e-06, + "loss": 1.0071, + "step": 5140 + }, + { + "epoch": 0.38636705245753794, + "grad_norm": 1.5853971866943144, + "learning_rate": 2.8095576010379784e-06, + "loss": 0.9644, + "step": 5141 + }, + { + "epoch": 0.3864422065233729, + "grad_norm": 1.8941376896711764, + "learning_rate": 2.80911240143454e-06, + "loss": 1.0406, + "step": 5142 + }, + { + "epoch": 0.3865173605892079, + "grad_norm": 2.387409516062901, + "learning_rate": 2.8086671538880938e-06, + "loss": 1.0417, + "step": 5143 + }, + { + "epoch": 0.38659251465504285, + "grad_norm": 3.4631806073064166, + "learning_rate": 2.808221858425022e-06, + "loss": 0.9618, + "step": 5144 + }, + { + "epoch": 0.3866676687208778, + "grad_norm": 2.2461795160255758, + "learning_rate": 2.8077765150717107e-06, + "loss": 1.0748, + "step": 5145 + }, + { + "epoch": 0.38674282278671274, + "grad_norm": 1.9692999661992987, + "learning_rate": 2.807331123854547e-06, + "loss": 1.0315, + "step": 5146 + }, + { + "epoch": 0.3868179768525477, + "grad_norm": 1.5724044454194905, + "learning_rate": 2.806885684799923e-06, + "loss": 0.986, + "step": 5147 + }, + { + "epoch": 0.3868931309183827, + "grad_norm": 1.8282746473831655, + "learning_rate": 2.8064401979342324e-06, + "loss": 1.0088, + "step": 5148 + }, + { + "epoch": 0.38696828498421765, + "grad_norm": 1.6202641949861, + "learning_rate": 2.805994663283872e-06, + "loss": 0.9394, + "step": 5149 + }, + { + "epoch": 0.3870434390500526, + "grad_norm": 2.201463871216044, + "learning_rate": 2.805549080875242e-06, + "loss": 1.0198, + "step": 5150 + }, + { + "epoch": 0.3871185931158876, + "grad_norm": 1.6375538009075303, + "learning_rate": 2.8051034507347435e-06, + "loss": 0.9489, + "step": 5151 + }, + { + "epoch": 0.38719374718172256, + "grad_norm": 1.9761803902244455, + "learning_rate": 2.804657772888783e-06, + "loss": 1.003, + "step": 5152 + }, + { + "epoch": 0.3872689012475575, + "grad_norm": 3.09610034788874, + "learning_rate": 2.804212047363768e-06, + "loss": 0.8888, + "step": 5153 + }, + { + "epoch": 0.38734405531339244, + "grad_norm": 1.619338222543919, + "learning_rate": 2.8037662741861097e-06, + "loss": 1.0595, + "step": 5154 + }, + { + "epoch": 0.3874192093792274, + "grad_norm": 1.7420024003714174, + "learning_rate": 2.803320453382222e-06, + "loss": 0.9852, + "step": 5155 + }, + { + "epoch": 0.3874943634450624, + "grad_norm": 1.4663495678278662, + "learning_rate": 2.8028745849785213e-06, + "loss": 1.0392, + "step": 5156 + }, + { + "epoch": 0.38756951751089735, + "grad_norm": 1.5578659002550541, + "learning_rate": 2.8024286690014266e-06, + "loss": 1.0275, + "step": 5157 + }, + { + "epoch": 0.3876446715767323, + "grad_norm": 1.8487691289957977, + "learning_rate": 2.801982705477361e-06, + "loss": 1.0966, + "step": 5158 + }, + { + "epoch": 0.38771982564256724, + "grad_norm": 2.2176579756519392, + "learning_rate": 2.801536694432749e-06, + "loss": 1.008, + "step": 5159 + }, + { + "epoch": 0.3877949797084022, + "grad_norm": 1.853926296583897, + "learning_rate": 2.8010906358940185e-06, + "loss": 0.9627, + "step": 5160 + }, + { + "epoch": 0.3878701337742372, + "grad_norm": 4.435482802404907, + "learning_rate": 2.8006445298876003e-06, + "loss": 0.935, + "step": 5161 + }, + { + "epoch": 0.38794528784007215, + "grad_norm": 0.9734727121416834, + "learning_rate": 2.800198376439928e-06, + "loss": 0.8878, + "step": 5162 + }, + { + "epoch": 0.3880204419059071, + "grad_norm": 1.4284949827176203, + "learning_rate": 2.7997521755774373e-06, + "loss": 1.0069, + "step": 5163 + }, + { + "epoch": 0.3880955959717421, + "grad_norm": 1.6341049700653072, + "learning_rate": 2.799305927326568e-06, + "loss": 1.0183, + "step": 5164 + }, + { + "epoch": 0.388170750037577, + "grad_norm": 1.4673023800465794, + "learning_rate": 2.7988596317137623e-06, + "loss": 0.9989, + "step": 5165 + }, + { + "epoch": 0.388245904103412, + "grad_norm": 1.6468011770767257, + "learning_rate": 2.7984132887654633e-06, + "loss": 0.9949, + "step": 5166 + }, + { + "epoch": 0.38832105816924695, + "grad_norm": 2.4561948580776662, + "learning_rate": 2.7979668985081204e-06, + "loss": 0.9615, + "step": 5167 + }, + { + "epoch": 0.3883962122350819, + "grad_norm": 1.8764090647610028, + "learning_rate": 2.797520460968183e-06, + "loss": 0.9637, + "step": 5168 + }, + { + "epoch": 0.3884713663009169, + "grad_norm": 1.590848188349229, + "learning_rate": 2.797073976172104e-06, + "loss": 0.9364, + "step": 5169 + }, + { + "epoch": 0.38854652036675186, + "grad_norm": 1.7518190058273304, + "learning_rate": 2.79662744414634e-06, + "loss": 1.0501, + "step": 5170 + }, + { + "epoch": 0.3886216744325868, + "grad_norm": 2.0245321840162065, + "learning_rate": 2.79618086491735e-06, + "loss": 1.0614, + "step": 5171 + }, + { + "epoch": 0.38869682849842174, + "grad_norm": 1.7904587007695478, + "learning_rate": 2.7957342385115944e-06, + "loss": 0.9886, + "step": 5172 + }, + { + "epoch": 0.3887719825642567, + "grad_norm": 2.1563459734359234, + "learning_rate": 2.795287564955538e-06, + "loss": 1.0498, + "step": 5173 + }, + { + "epoch": 0.3888471366300917, + "grad_norm": 1.9259705363443271, + "learning_rate": 2.7948408442756477e-06, + "loss": 1.0221, + "step": 5174 + }, + { + "epoch": 0.38892229069592665, + "grad_norm": 2.2200560688120996, + "learning_rate": 2.794394076498394e-06, + "loss": 1.1268, + "step": 5175 + }, + { + "epoch": 0.3889974447617616, + "grad_norm": 1.427613301499024, + "learning_rate": 2.79394726165025e-06, + "loss": 0.9833, + "step": 5176 + }, + { + "epoch": 0.3890725988275966, + "grad_norm": 0.7059065371845507, + "learning_rate": 2.79350039975769e-06, + "loss": 0.834, + "step": 5177 + }, + { + "epoch": 0.3891477528934315, + "grad_norm": 0.7057617761789533, + "learning_rate": 2.7930534908471927e-06, + "loss": 0.8576, + "step": 5178 + }, + { + "epoch": 0.3892229069592665, + "grad_norm": 1.4375869890512636, + "learning_rate": 2.792606534945239e-06, + "loss": 0.9983, + "step": 5179 + }, + { + "epoch": 0.38929806102510145, + "grad_norm": 1.9109371905821755, + "learning_rate": 2.7921595320783136e-06, + "loss": 0.9534, + "step": 5180 + }, + { + "epoch": 0.3893732150909364, + "grad_norm": 1.5217909302099173, + "learning_rate": 2.7917124822729022e-06, + "loss": 1.0523, + "step": 5181 + }, + { + "epoch": 0.3894483691567714, + "grad_norm": 1.7010947208887062, + "learning_rate": 2.791265385555495e-06, + "loss": 0.9721, + "step": 5182 + }, + { + "epoch": 0.38952352322260636, + "grad_norm": 1.4773802738123, + "learning_rate": 2.7908182419525834e-06, + "loss": 0.9473, + "step": 5183 + }, + { + "epoch": 0.38959867728844133, + "grad_norm": 0.7409976456330616, + "learning_rate": 2.7903710514906626e-06, + "loss": 0.8767, + "step": 5184 + }, + { + "epoch": 0.38967383135427625, + "grad_norm": 2.7908582703083886, + "learning_rate": 2.7899238141962304e-06, + "loss": 0.9699, + "step": 5185 + }, + { + "epoch": 0.3897489854201112, + "grad_norm": 2.0607296830934003, + "learning_rate": 2.7894765300957875e-06, + "loss": 1.0502, + "step": 5186 + }, + { + "epoch": 0.3898241394859462, + "grad_norm": 1.3175708362277085, + "learning_rate": 2.7890291992158376e-06, + "loss": 1.0448, + "step": 5187 + }, + { + "epoch": 0.38989929355178116, + "grad_norm": 1.8029139006754875, + "learning_rate": 2.7885818215828856e-06, + "loss": 1.0523, + "step": 5188 + }, + { + "epoch": 0.3899744476176161, + "grad_norm": 1.8077933927826957, + "learning_rate": 2.7881343972234416e-06, + "loss": 0.9544, + "step": 5189 + }, + { + "epoch": 0.3900496016834511, + "grad_norm": 1.828006707953368, + "learning_rate": 2.787686926164016e-06, + "loss": 0.783, + "step": 5190 + }, + { + "epoch": 0.390124755749286, + "grad_norm": 2.0868623090725436, + "learning_rate": 2.787239408431124e-06, + "loss": 0.9627, + "step": 5191 + }, + { + "epoch": 0.390199909815121, + "grad_norm": 1.9233179025785812, + "learning_rate": 2.786791844051282e-06, + "loss": 0.8903, + "step": 5192 + }, + { + "epoch": 0.39027506388095595, + "grad_norm": 1.7359791116336747, + "learning_rate": 2.7863442330510115e-06, + "loss": 0.9782, + "step": 5193 + }, + { + "epoch": 0.3903502179467909, + "grad_norm": 2.2198659749857454, + "learning_rate": 2.7858965754568335e-06, + "loss": 0.9469, + "step": 5194 + }, + { + "epoch": 0.3904253720126259, + "grad_norm": 1.8142328857522998, + "learning_rate": 2.7854488712952735e-06, + "loss": 0.9452, + "step": 5195 + }, + { + "epoch": 0.39050052607846086, + "grad_norm": 2.3955927078375576, + "learning_rate": 2.7850011205928607e-06, + "loss": 1.0355, + "step": 5196 + }, + { + "epoch": 0.39057568014429583, + "grad_norm": 2.463283909760844, + "learning_rate": 2.7845533233761256e-06, + "loss": 0.9674, + "step": 5197 + }, + { + "epoch": 0.39065083421013075, + "grad_norm": 4.1734099689633855, + "learning_rate": 2.784105479671602e-06, + "loss": 1.0156, + "step": 5198 + }, + { + "epoch": 0.3907259882759657, + "grad_norm": 2.197125104013599, + "learning_rate": 2.783657589505826e-06, + "loss": 0.9031, + "step": 5199 + }, + { + "epoch": 0.3908011423418007, + "grad_norm": 1.4320775467318505, + "learning_rate": 2.783209652905337e-06, + "loss": 0.9573, + "step": 5200 + }, + { + "epoch": 0.39087629640763566, + "grad_norm": 0.8159840053131433, + "learning_rate": 2.7827616698966763e-06, + "loss": 0.8925, + "step": 5201 + }, + { + "epoch": 0.39095145047347063, + "grad_norm": 2.1026456516177636, + "learning_rate": 2.78231364050639e-06, + "loss": 1.0839, + "step": 5202 + }, + { + "epoch": 0.3910266045393056, + "grad_norm": 1.4288792227498364, + "learning_rate": 2.781865564761025e-06, + "loss": 0.9336, + "step": 5203 + }, + { + "epoch": 0.3911017586051405, + "grad_norm": 1.655668696470932, + "learning_rate": 2.781417442687131e-06, + "loss": 0.9874, + "step": 5204 + }, + { + "epoch": 0.3911769126709755, + "grad_norm": 1.8190816884879277, + "learning_rate": 2.7809692743112616e-06, + "loss": 0.9515, + "step": 5205 + }, + { + "epoch": 0.39125206673681046, + "grad_norm": 5.769231727795644, + "learning_rate": 2.780521059659972e-06, + "loss": 1.1014, + "step": 5206 + }, + { + "epoch": 0.3913272208026454, + "grad_norm": 0.6209040178046387, + "learning_rate": 2.78007279875982e-06, + "loss": 0.7677, + "step": 5207 + }, + { + "epoch": 0.3914023748684804, + "grad_norm": 0.6551080941411943, + "learning_rate": 2.7796244916373686e-06, + "loss": 0.824, + "step": 5208 + }, + { + "epoch": 0.39147752893431537, + "grad_norm": 1.6576051026819922, + "learning_rate": 2.7791761383191807e-06, + "loss": 1.0003, + "step": 5209 + }, + { + "epoch": 0.3915526830001503, + "grad_norm": 1.7978312062815873, + "learning_rate": 2.778727738831822e-06, + "loss": 0.9456, + "step": 5210 + }, + { + "epoch": 0.39162783706598525, + "grad_norm": 0.7941165753432129, + "learning_rate": 2.7782792932018635e-06, + "loss": 0.8653, + "step": 5211 + }, + { + "epoch": 0.3917029911318202, + "grad_norm": 1.767390963092634, + "learning_rate": 2.7778308014558767e-06, + "loss": 0.971, + "step": 5212 + }, + { + "epoch": 0.3917781451976552, + "grad_norm": 2.2017265913857997, + "learning_rate": 2.777382263620436e-06, + "loss": 0.9474, + "step": 5213 + }, + { + "epoch": 0.39185329926349016, + "grad_norm": 2.581521311395466, + "learning_rate": 2.7769336797221197e-06, + "loss": 1.0565, + "step": 5214 + }, + { + "epoch": 0.39192845332932513, + "grad_norm": 1.6414747671418315, + "learning_rate": 2.7764850497875076e-06, + "loss": 0.9058, + "step": 5215 + }, + { + "epoch": 0.3920036073951601, + "grad_norm": 2.613314093823878, + "learning_rate": 2.776036373843183e-06, + "loss": 0.9435, + "step": 5216 + }, + { + "epoch": 0.392078761460995, + "grad_norm": 2.5212163923120112, + "learning_rate": 2.775587651915732e-06, + "loss": 1.0829, + "step": 5217 + }, + { + "epoch": 0.39215391552683, + "grad_norm": 1.6615570262268486, + "learning_rate": 2.775138884031742e-06, + "loss": 0.9412, + "step": 5218 + }, + { + "epoch": 0.39222906959266496, + "grad_norm": 2.5957805544496884, + "learning_rate": 2.7746900702178053e-06, + "loss": 1.1398, + "step": 5219 + }, + { + "epoch": 0.39230422365849993, + "grad_norm": 1.4883632687373611, + "learning_rate": 2.7742412105005154e-06, + "loss": 0.9948, + "step": 5220 + }, + { + "epoch": 0.3923793777243349, + "grad_norm": 1.81196685052911, + "learning_rate": 2.773792304906469e-06, + "loss": 1.0225, + "step": 5221 + }, + { + "epoch": 0.39245453179016987, + "grad_norm": 2.1255032560529514, + "learning_rate": 2.7733433534622655e-06, + "loss": 0.9787, + "step": 5222 + }, + { + "epoch": 0.3925296858560048, + "grad_norm": 2.0561561122434893, + "learning_rate": 2.772894356194507e-06, + "loss": 1.0559, + "step": 5223 + }, + { + "epoch": 0.39260483992183975, + "grad_norm": 0.8477139079792747, + "learning_rate": 2.7724453131297988e-06, + "loss": 0.8343, + "step": 5224 + }, + { + "epoch": 0.3926799939876747, + "grad_norm": 1.9082382339480626, + "learning_rate": 2.771996224294747e-06, + "loss": 0.9901, + "step": 5225 + }, + { + "epoch": 0.3927551480535097, + "grad_norm": 2.049538197489261, + "learning_rate": 2.7715470897159636e-06, + "loss": 1.1016, + "step": 5226 + }, + { + "epoch": 0.39283030211934467, + "grad_norm": 1.7076081233049636, + "learning_rate": 2.7710979094200593e-06, + "loss": 1.0483, + "step": 5227 + }, + { + "epoch": 0.39290545618517964, + "grad_norm": 2.2553149074072754, + "learning_rate": 2.7706486834336524e-06, + "loss": 0.9711, + "step": 5228 + }, + { + "epoch": 0.3929806102510146, + "grad_norm": 2.0347506362383485, + "learning_rate": 2.7701994117833596e-06, + "loss": 1.0396, + "step": 5229 + }, + { + "epoch": 0.3930557643168495, + "grad_norm": 3.8012329962217444, + "learning_rate": 2.7697500944958024e-06, + "loss": 1.0054, + "step": 5230 + }, + { + "epoch": 0.3931309183826845, + "grad_norm": 2.146464714814138, + "learning_rate": 2.7693007315976047e-06, + "loss": 1.0533, + "step": 5231 + }, + { + "epoch": 0.39320607244851946, + "grad_norm": 1.602242748661331, + "learning_rate": 2.7688513231153926e-06, + "loss": 0.9247, + "step": 5232 + }, + { + "epoch": 0.39328122651435443, + "grad_norm": 3.0008896462254726, + "learning_rate": 2.7684018690757954e-06, + "loss": 0.9236, + "step": 5233 + }, + { + "epoch": 0.3933563805801894, + "grad_norm": 1.9607224061486594, + "learning_rate": 2.767952369505445e-06, + "loss": 1.039, + "step": 5234 + }, + { + "epoch": 0.3934315346460244, + "grad_norm": 2.798799758203029, + "learning_rate": 2.7675028244309766e-06, + "loss": 0.9731, + "step": 5235 + }, + { + "epoch": 0.3935066887118593, + "grad_norm": 1.5046831774624765, + "learning_rate": 2.767053233879026e-06, + "loss": 0.987, + "step": 5236 + }, + { + "epoch": 0.39358184277769426, + "grad_norm": 2.1663372327318777, + "learning_rate": 2.766603597876235e-06, + "loss": 1.0223, + "step": 5237 + }, + { + "epoch": 0.3936569968435292, + "grad_norm": 1.3180413660871877, + "learning_rate": 2.7661539164492442e-06, + "loss": 0.9419, + "step": 5238 + }, + { + "epoch": 0.3937321509093642, + "grad_norm": 1.8696250805899093, + "learning_rate": 2.765704189624701e-06, + "loss": 0.9287, + "step": 5239 + }, + { + "epoch": 0.39380730497519917, + "grad_norm": 1.8257705655517833, + "learning_rate": 2.765254417429252e-06, + "loss": 0.9438, + "step": 5240 + }, + { + "epoch": 0.39388245904103414, + "grad_norm": 0.8242363950458709, + "learning_rate": 2.764804599889549e-06, + "loss": 0.8589, + "step": 5241 + }, + { + "epoch": 0.3939576131068691, + "grad_norm": 2.8594928988138264, + "learning_rate": 2.7643547370322446e-06, + "loss": 1.0046, + "step": 5242 + }, + { + "epoch": 0.394032767172704, + "grad_norm": 1.7691964681196395, + "learning_rate": 2.763904828883995e-06, + "loss": 0.9597, + "step": 5243 + }, + { + "epoch": 0.394107921238539, + "grad_norm": 1.7341047884111709, + "learning_rate": 2.763454875471459e-06, + "loss": 1.0648, + "step": 5244 + }, + { + "epoch": 0.39418307530437396, + "grad_norm": 1.7563941489146315, + "learning_rate": 2.7630048768212975e-06, + "loss": 1.0833, + "step": 5245 + }, + { + "epoch": 0.39425822937020893, + "grad_norm": 1.8737507791243977, + "learning_rate": 2.7625548329601763e-06, + "loss": 0.9606, + "step": 5246 + }, + { + "epoch": 0.3943333834360439, + "grad_norm": 1.7545244448576798, + "learning_rate": 2.7621047439147606e-06, + "loss": 0.9666, + "step": 5247 + }, + { + "epoch": 0.3944085375018789, + "grad_norm": 1.836129166358336, + "learning_rate": 2.7616546097117213e-06, + "loss": 0.9311, + "step": 5248 + }, + { + "epoch": 0.3944836915677138, + "grad_norm": 1.6080558333216648, + "learning_rate": 2.761204430377729e-06, + "loss": 1.0125, + "step": 5249 + }, + { + "epoch": 0.39455884563354876, + "grad_norm": 1.8468162772562073, + "learning_rate": 2.7607542059394604e-06, + "loss": 1.0094, + "step": 5250 + }, + { + "epoch": 0.39463399969938373, + "grad_norm": 2.508730901427813, + "learning_rate": 2.760303936423591e-06, + "loss": 0.9509, + "step": 5251 + }, + { + "epoch": 0.3947091537652187, + "grad_norm": 1.5505810663656325, + "learning_rate": 2.759853621856802e-06, + "loss": 0.9067, + "step": 5252 + }, + { + "epoch": 0.39478430783105367, + "grad_norm": 2.064769074687799, + "learning_rate": 2.759403262265777e-06, + "loss": 0.9493, + "step": 5253 + }, + { + "epoch": 0.39485946189688864, + "grad_norm": 1.5151649642116247, + "learning_rate": 2.7589528576772e-06, + "loss": 1.027, + "step": 5254 + }, + { + "epoch": 0.39493461596272356, + "grad_norm": 1.6533598244503358, + "learning_rate": 2.7585024081177602e-06, + "loss": 0.902, + "step": 5255 + }, + { + "epoch": 0.3950097700285585, + "grad_norm": 1.4819923954818774, + "learning_rate": 2.7580519136141483e-06, + "loss": 0.9969, + "step": 5256 + }, + { + "epoch": 0.3950849240943935, + "grad_norm": 1.924525348554865, + "learning_rate": 2.7576013741930576e-06, + "loss": 1.0126, + "step": 5257 + }, + { + "epoch": 0.39516007816022847, + "grad_norm": 3.2589583542987643, + "learning_rate": 2.7571507898811846e-06, + "loss": 1.1013, + "step": 5258 + }, + { + "epoch": 0.39523523222606344, + "grad_norm": 2.0859305278075775, + "learning_rate": 2.756700160705228e-06, + "loss": 0.876, + "step": 5259 + }, + { + "epoch": 0.3953103862918984, + "grad_norm": 2.2677320482319168, + "learning_rate": 2.756249486691889e-06, + "loss": 1.0741, + "step": 5260 + }, + { + "epoch": 0.3953855403577334, + "grad_norm": 1.65180625019048, + "learning_rate": 2.7557987678678723e-06, + "loss": 0.9264, + "step": 5261 + }, + { + "epoch": 0.3954606944235683, + "grad_norm": 1.7856550125094015, + "learning_rate": 2.755348004259884e-06, + "loss": 0.9854, + "step": 5262 + }, + { + "epoch": 0.39553584848940326, + "grad_norm": 3.1802618084503806, + "learning_rate": 2.7548971958946347e-06, + "loss": 1.065, + "step": 5263 + }, + { + "epoch": 0.39561100255523823, + "grad_norm": 1.8422092530290366, + "learning_rate": 2.7544463427988355e-06, + "loss": 1.1108, + "step": 5264 + }, + { + "epoch": 0.3956861566210732, + "grad_norm": 2.023867646085299, + "learning_rate": 2.7539954449992014e-06, + "loss": 0.9825, + "step": 5265 + }, + { + "epoch": 0.3957613106869082, + "grad_norm": 0.7378125999156556, + "learning_rate": 2.7535445025224506e-06, + "loss": 0.8551, + "step": 5266 + }, + { + "epoch": 0.39583646475274314, + "grad_norm": 0.6972295073229513, + "learning_rate": 2.7530935153953016e-06, + "loss": 0.8359, + "step": 5267 + }, + { + "epoch": 0.39591161881857806, + "grad_norm": 1.9008619110264848, + "learning_rate": 2.752642483644478e-06, + "loss": 1.0123, + "step": 5268 + }, + { + "epoch": 0.39598677288441303, + "grad_norm": 1.5037634050971986, + "learning_rate": 2.752191407296706e-06, + "loss": 0.8896, + "step": 5269 + }, + { + "epoch": 0.396061926950248, + "grad_norm": 4.544514361162102, + "learning_rate": 2.7517402863787123e-06, + "loss": 0.8461, + "step": 5270 + }, + { + "epoch": 0.39613708101608297, + "grad_norm": 2.1813174480514257, + "learning_rate": 2.751289120917228e-06, + "loss": 1.0634, + "step": 5271 + }, + { + "epoch": 0.39621223508191794, + "grad_norm": 1.6007885062861709, + "learning_rate": 2.750837910938987e-06, + "loss": 0.9649, + "step": 5272 + }, + { + "epoch": 0.3962873891477529, + "grad_norm": 1.8184387648611329, + "learning_rate": 2.7503866564707236e-06, + "loss": 1.0148, + "step": 5273 + }, + { + "epoch": 0.3963625432135879, + "grad_norm": 1.9787904749940055, + "learning_rate": 2.7499353575391784e-06, + "loss": 1.0678, + "step": 5274 + }, + { + "epoch": 0.3964376972794228, + "grad_norm": 1.878266413945578, + "learning_rate": 2.749484014171091e-06, + "loss": 0.8971, + "step": 5275 + }, + { + "epoch": 0.39651285134525777, + "grad_norm": 1.7612120582711865, + "learning_rate": 2.749032626393206e-06, + "loss": 1.0478, + "step": 5276 + }, + { + "epoch": 0.39658800541109274, + "grad_norm": 2.0621603483796305, + "learning_rate": 2.74858119423227e-06, + "loss": 0.9094, + "step": 5277 + }, + { + "epoch": 0.3966631594769277, + "grad_norm": 1.91124103222197, + "learning_rate": 2.748129717715031e-06, + "loss": 1.0506, + "step": 5278 + }, + { + "epoch": 0.3967383135427627, + "grad_norm": 1.468371403374184, + "learning_rate": 2.747678196868241e-06, + "loss": 0.9422, + "step": 5279 + }, + { + "epoch": 0.39681346760859765, + "grad_norm": 2.0304314944939676, + "learning_rate": 2.747226631718656e-06, + "loss": 1.0833, + "step": 5280 + }, + { + "epoch": 0.39688862167443256, + "grad_norm": 1.6025759584037236, + "learning_rate": 2.746775022293032e-06, + "loss": 1.0461, + "step": 5281 + }, + { + "epoch": 0.39696377574026753, + "grad_norm": 1.9064147127625073, + "learning_rate": 2.746323368618127e-06, + "loss": 0.9324, + "step": 5282 + }, + { + "epoch": 0.3970389298061025, + "grad_norm": 1.5162399460222562, + "learning_rate": 2.7458716707207054e-06, + "loss": 1.0318, + "step": 5283 + }, + { + "epoch": 0.3971140838719375, + "grad_norm": 2.3150388853626125, + "learning_rate": 2.74541992862753e-06, + "loss": 1.0159, + "step": 5284 + }, + { + "epoch": 0.39718923793777244, + "grad_norm": 1.6869879758188115, + "learning_rate": 2.744968142365371e-06, + "loss": 1.0208, + "step": 5285 + }, + { + "epoch": 0.3972643920036074, + "grad_norm": 1.6212332424218356, + "learning_rate": 2.744516311960996e-06, + "loss": 0.9831, + "step": 5286 + }, + { + "epoch": 0.3973395460694424, + "grad_norm": 1.7946811718026274, + "learning_rate": 2.744064437441179e-06, + "loss": 1.0005, + "step": 5287 + }, + { + "epoch": 0.3974147001352773, + "grad_norm": 1.836735069238101, + "learning_rate": 2.743612518832695e-06, + "loss": 1.0463, + "step": 5288 + }, + { + "epoch": 0.39748985420111227, + "grad_norm": 1.817323514615629, + "learning_rate": 2.743160556162321e-06, + "loss": 1.0405, + "step": 5289 + }, + { + "epoch": 0.39756500826694724, + "grad_norm": 2.1954747476048557, + "learning_rate": 2.7427085494568383e-06, + "loss": 0.9493, + "step": 5290 + }, + { + "epoch": 0.3976401623327822, + "grad_norm": 1.5045004723479372, + "learning_rate": 2.742256498743031e-06, + "loss": 0.9496, + "step": 5291 + }, + { + "epoch": 0.3977153163986172, + "grad_norm": 4.085215050629244, + "learning_rate": 2.7418044040476838e-06, + "loss": 0.8858, + "step": 5292 + }, + { + "epoch": 0.39779047046445215, + "grad_norm": 1.9493882099864699, + "learning_rate": 2.7413522653975842e-06, + "loss": 0.9331, + "step": 5293 + }, + { + "epoch": 0.39786562453028707, + "grad_norm": 1.7325252945059433, + "learning_rate": 2.7409000828195247e-06, + "loss": 0.9493, + "step": 5294 + }, + { + "epoch": 0.39794077859612204, + "grad_norm": 1.6018820095020523, + "learning_rate": 2.7404478563402976e-06, + "loss": 0.9621, + "step": 5295 + }, + { + "epoch": 0.398015932661957, + "grad_norm": 1.5419290250630568, + "learning_rate": 2.7399955859867e-06, + "loss": 0.9794, + "step": 5296 + }, + { + "epoch": 0.398091086727792, + "grad_norm": 1.9385959847896188, + "learning_rate": 2.739543271785531e-06, + "loss": 1.0656, + "step": 5297 + }, + { + "epoch": 0.39816624079362695, + "grad_norm": 2.1451785252045528, + "learning_rate": 2.7390909137635906e-06, + "loss": 0.946, + "step": 5298 + }, + { + "epoch": 0.3982413948594619, + "grad_norm": 1.9293231079517783, + "learning_rate": 2.7386385119476833e-06, + "loss": 0.9761, + "step": 5299 + }, + { + "epoch": 0.39831654892529683, + "grad_norm": 2.407766780629628, + "learning_rate": 2.738186066364616e-06, + "loss": 0.9616, + "step": 5300 + }, + { + "epoch": 0.3983917029911318, + "grad_norm": 2.0870810813635265, + "learning_rate": 2.7377335770411965e-06, + "loss": 1.0336, + "step": 5301 + }, + { + "epoch": 0.3984668570569668, + "grad_norm": 1.9322188604439072, + "learning_rate": 2.737281044004239e-06, + "loss": 0.9332, + "step": 5302 + }, + { + "epoch": 0.39854201112280174, + "grad_norm": 1.635450723216526, + "learning_rate": 2.7368284672805558e-06, + "loss": 1.0027, + "step": 5303 + }, + { + "epoch": 0.3986171651886367, + "grad_norm": 2.0177497015319386, + "learning_rate": 2.7363758468969643e-06, + "loss": 1.081, + "step": 5304 + }, + { + "epoch": 0.3986923192544717, + "grad_norm": 2.2382675090587707, + "learning_rate": 2.735923182880285e-06, + "loss": 1.0863, + "step": 5305 + }, + { + "epoch": 0.39876747332030665, + "grad_norm": 2.02151244507987, + "learning_rate": 2.7354704752573376e-06, + "loss": 1.0221, + "step": 5306 + }, + { + "epoch": 0.39884262738614157, + "grad_norm": 3.169554965913322, + "learning_rate": 2.735017724054949e-06, + "loss": 1.033, + "step": 5307 + }, + { + "epoch": 0.39891778145197654, + "grad_norm": 1.5697593663703666, + "learning_rate": 2.7345649292999456e-06, + "loss": 0.9702, + "step": 5308 + }, + { + "epoch": 0.3989929355178115, + "grad_norm": 2.199898379319712, + "learning_rate": 2.7341120910191575e-06, + "loss": 0.8703, + "step": 5309 + }, + { + "epoch": 0.3990680895836465, + "grad_norm": 1.9118809938664179, + "learning_rate": 2.733659209239417e-06, + "loss": 1.0662, + "step": 5310 + }, + { + "epoch": 0.39914324364948145, + "grad_norm": 2.378871572958978, + "learning_rate": 2.7332062839875586e-06, + "loss": 1.0597, + "step": 5311 + }, + { + "epoch": 0.3992183977153164, + "grad_norm": 0.7437929958449662, + "learning_rate": 2.73275331529042e-06, + "loss": 0.8234, + "step": 5312 + }, + { + "epoch": 0.39929355178115133, + "grad_norm": 1.8391171456177489, + "learning_rate": 2.7323003031748424e-06, + "loss": 1.0115, + "step": 5313 + }, + { + "epoch": 0.3993687058469863, + "grad_norm": 2.0533617042876915, + "learning_rate": 2.731847247667667e-06, + "loss": 1.0011, + "step": 5314 + }, + { + "epoch": 0.3994438599128213, + "grad_norm": 6.156670105827677, + "learning_rate": 2.7313941487957398e-06, + "loss": 0.9477, + "step": 5315 + }, + { + "epoch": 0.39951901397865625, + "grad_norm": 2.0011940014621366, + "learning_rate": 2.730941006585909e-06, + "loss": 1.0877, + "step": 5316 + }, + { + "epoch": 0.3995941680444912, + "grad_norm": 20.537700690637283, + "learning_rate": 2.7304878210650243e-06, + "loss": 0.9424, + "step": 5317 + }, + { + "epoch": 0.3996693221103262, + "grad_norm": 1.3685732984359418, + "learning_rate": 2.7300345922599394e-06, + "loss": 0.9307, + "step": 5318 + }, + { + "epoch": 0.39974447617616116, + "grad_norm": 1.6916628460144094, + "learning_rate": 2.7295813201975087e-06, + "loss": 0.9437, + "step": 5319 + }, + { + "epoch": 0.39981963024199607, + "grad_norm": 1.9345653212288905, + "learning_rate": 2.7291280049045916e-06, + "loss": 0.9869, + "step": 5320 + }, + { + "epoch": 0.39989478430783104, + "grad_norm": 1.5810338773530093, + "learning_rate": 2.728674646408048e-06, + "loss": 0.9634, + "step": 5321 + }, + { + "epoch": 0.399969938373666, + "grad_norm": 1.3334688723735446, + "learning_rate": 2.7282212447347413e-06, + "loss": 0.9544, + "step": 5322 + }, + { + "epoch": 0.400045092439501, + "grad_norm": 1.8150704014473056, + "learning_rate": 2.7277677999115368e-06, + "loss": 0.9641, + "step": 5323 + }, + { + "epoch": 0.40012024650533595, + "grad_norm": 1.9453591337321101, + "learning_rate": 2.7273143119653042e-06, + "loss": 0.9784, + "step": 5324 + }, + { + "epoch": 0.4001954005711709, + "grad_norm": 1.8278507776894366, + "learning_rate": 2.7268607809229137e-06, + "loss": 0.937, + "step": 5325 + }, + { + "epoch": 0.40027055463700584, + "grad_norm": 2.6471396898246735, + "learning_rate": 2.7264072068112377e-06, + "loss": 1.0681, + "step": 5326 + }, + { + "epoch": 0.4003457087028408, + "grad_norm": 1.5566734013762316, + "learning_rate": 2.725953589657154e-06, + "loss": 1.0143, + "step": 5327 + }, + { + "epoch": 0.4004208627686758, + "grad_norm": 1.5122497572265672, + "learning_rate": 2.7254999294875395e-06, + "loss": 0.9935, + "step": 5328 + }, + { + "epoch": 0.40049601683451075, + "grad_norm": 1.8894673736331542, + "learning_rate": 2.725046226329276e-06, + "loss": 0.9555, + "step": 5329 + }, + { + "epoch": 0.4005711709003457, + "grad_norm": 1.8212455355681798, + "learning_rate": 2.7245924802092476e-06, + "loss": 0.9615, + "step": 5330 + }, + { + "epoch": 0.4006463249661807, + "grad_norm": 0.646406370683474, + "learning_rate": 2.7241386911543397e-06, + "loss": 0.8374, + "step": 5331 + }, + { + "epoch": 0.40072147903201566, + "grad_norm": 2.6269047502225775, + "learning_rate": 2.7236848591914422e-06, + "loss": 0.9455, + "step": 5332 + }, + { + "epoch": 0.4007966330978506, + "grad_norm": 9.772598681904348, + "learning_rate": 2.7232309843474446e-06, + "loss": 0.8954, + "step": 5333 + }, + { + "epoch": 0.40087178716368554, + "grad_norm": 1.5139723744057774, + "learning_rate": 2.7227770666492423e-06, + "loss": 0.9267, + "step": 5334 + }, + { + "epoch": 0.4009469412295205, + "grad_norm": 0.7586314308873218, + "learning_rate": 2.722323106123731e-06, + "loss": 0.8649, + "step": 5335 + }, + { + "epoch": 0.4010220952953555, + "grad_norm": 2.283591127345184, + "learning_rate": 2.7218691027978103e-06, + "loss": 0.9894, + "step": 5336 + }, + { + "epoch": 0.40109724936119046, + "grad_norm": 2.390940028358671, + "learning_rate": 2.7214150566983807e-06, + "loss": 0.9829, + "step": 5337 + }, + { + "epoch": 0.4011724034270254, + "grad_norm": 1.751554139633712, + "learning_rate": 2.7209609678523462e-06, + "loss": 1.0395, + "step": 5338 + }, + { + "epoch": 0.40124755749286034, + "grad_norm": 1.987726160293419, + "learning_rate": 2.7205068362866134e-06, + "loss": 1.0301, + "step": 5339 + }, + { + "epoch": 0.4013227115586953, + "grad_norm": 1.3809715128342233, + "learning_rate": 2.7200526620280923e-06, + "loss": 0.9878, + "step": 5340 + }, + { + "epoch": 0.4013978656245303, + "grad_norm": 1.2434552998501895, + "learning_rate": 2.719598445103693e-06, + "loss": 0.8938, + "step": 5341 + }, + { + "epoch": 0.40147301969036525, + "grad_norm": 1.7566766078216485, + "learning_rate": 2.7191441855403304e-06, + "loss": 1.0871, + "step": 5342 + }, + { + "epoch": 0.4015481737562002, + "grad_norm": 1.6861535811517754, + "learning_rate": 2.718689883364922e-06, + "loss": 1.0128, + "step": 5343 + }, + { + "epoch": 0.4016233278220352, + "grad_norm": 1.9447613896807414, + "learning_rate": 2.7182355386043847e-06, + "loss": 0.8771, + "step": 5344 + }, + { + "epoch": 0.4016984818878701, + "grad_norm": 2.7869787472946963, + "learning_rate": 2.7177811512856415e-06, + "loss": 0.9724, + "step": 5345 + }, + { + "epoch": 0.4017736359537051, + "grad_norm": 2.001717783423192, + "learning_rate": 2.7173267214356173e-06, + "loss": 0.9679, + "step": 5346 + }, + { + "epoch": 0.40184879001954005, + "grad_norm": 1.7039883026697618, + "learning_rate": 2.716872249081238e-06, + "loss": 0.9975, + "step": 5347 + }, + { + "epoch": 0.401923944085375, + "grad_norm": 3.33266028920913, + "learning_rate": 2.7164177342494323e-06, + "loss": 1.013, + "step": 5348 + }, + { + "epoch": 0.40199909815121, + "grad_norm": 1.833303826990525, + "learning_rate": 2.7159631769671326e-06, + "loss": 1.0888, + "step": 5349 + }, + { + "epoch": 0.40207425221704496, + "grad_norm": 1.717856573768116, + "learning_rate": 2.715508577261273e-06, + "loss": 0.9828, + "step": 5350 + }, + { + "epoch": 0.40214940628287993, + "grad_norm": 1.94986605550995, + "learning_rate": 2.715053935158791e-06, + "loss": 0.8978, + "step": 5351 + }, + { + "epoch": 0.40222456034871484, + "grad_norm": 0.8759234528192643, + "learning_rate": 2.7145992506866242e-06, + "loss": 0.9384, + "step": 5352 + }, + { + "epoch": 0.4022997144145498, + "grad_norm": 5.830357182915219, + "learning_rate": 2.714144523871716e-06, + "loss": 1.0203, + "step": 5353 + }, + { + "epoch": 0.4023748684803848, + "grad_norm": 1.6846703010407165, + "learning_rate": 2.7136897547410105e-06, + "loss": 1.0552, + "step": 5354 + }, + { + "epoch": 0.40245002254621975, + "grad_norm": 2.0483054484411074, + "learning_rate": 2.7132349433214536e-06, + "loss": 0.9696, + "step": 5355 + }, + { + "epoch": 0.4025251766120547, + "grad_norm": 1.5929347178222748, + "learning_rate": 2.712780089639995e-06, + "loss": 0.9528, + "step": 5356 + }, + { + "epoch": 0.4026003306778897, + "grad_norm": 1.4519405665148302, + "learning_rate": 2.7123251937235873e-06, + "loss": 0.9426, + "step": 5357 + }, + { + "epoch": 0.4026754847437246, + "grad_norm": 3.4993976347147475, + "learning_rate": 2.7118702555991835e-06, + "loss": 1.0188, + "step": 5358 + }, + { + "epoch": 0.4027506388095596, + "grad_norm": 1.7097752076460673, + "learning_rate": 2.7114152752937417e-06, + "loss": 0.9448, + "step": 5359 + }, + { + "epoch": 0.40282579287539455, + "grad_norm": 1.7848620683607797, + "learning_rate": 2.71096025283422e-06, + "loss": 0.9868, + "step": 5360 + }, + { + "epoch": 0.4029009469412295, + "grad_norm": 1.6844394621961303, + "learning_rate": 2.7105051882475813e-06, + "loss": 0.9808, + "step": 5361 + }, + { + "epoch": 0.4029761010070645, + "grad_norm": 2.102565823680133, + "learning_rate": 2.7100500815607898e-06, + "loss": 1.0612, + "step": 5362 + }, + { + "epoch": 0.40305125507289946, + "grad_norm": 1.4405288300096988, + "learning_rate": 2.7095949328008113e-06, + "loss": 1.012, + "step": 5363 + }, + { + "epoch": 0.40312640913873443, + "grad_norm": 1.6430702515510354, + "learning_rate": 2.7091397419946162e-06, + "loss": 1.0145, + "step": 5364 + }, + { + "epoch": 0.40320156320456935, + "grad_norm": 1.6002403942952763, + "learning_rate": 2.708684509169176e-06, + "loss": 0.9837, + "step": 5365 + }, + { + "epoch": 0.4032767172704043, + "grad_norm": 1.6030140323417388, + "learning_rate": 2.7082292343514646e-06, + "loss": 0.9742, + "step": 5366 + }, + { + "epoch": 0.4033518713362393, + "grad_norm": 3.0321872481348286, + "learning_rate": 2.707773917568459e-06, + "loss": 0.9911, + "step": 5367 + }, + { + "epoch": 0.40342702540207426, + "grad_norm": 1.547096853049037, + "learning_rate": 2.707318558847139e-06, + "loss": 1.0735, + "step": 5368 + }, + { + "epoch": 0.4035021794679092, + "grad_norm": 1.5816205280409263, + "learning_rate": 2.706863158214486e-06, + "loss": 1.0355, + "step": 5369 + }, + { + "epoch": 0.4035773335337442, + "grad_norm": 1.9232491023672595, + "learning_rate": 2.7064077156974835e-06, + "loss": 0.9289, + "step": 5370 + }, + { + "epoch": 0.4036524875995791, + "grad_norm": 1.73431025203782, + "learning_rate": 2.705952231323119e-06, + "loss": 0.8887, + "step": 5371 + }, + { + "epoch": 0.4037276416654141, + "grad_norm": 1.561315163565008, + "learning_rate": 2.7054967051183813e-06, + "loss": 0.9486, + "step": 5372 + }, + { + "epoch": 0.40380279573124905, + "grad_norm": 1.6183605270510548, + "learning_rate": 2.705041137110263e-06, + "loss": 1.0322, + "step": 5373 + }, + { + "epoch": 0.403877949797084, + "grad_norm": 1.4128751881003707, + "learning_rate": 2.704585527325757e-06, + "loss": 0.9785, + "step": 5374 + }, + { + "epoch": 0.403953103862919, + "grad_norm": 1.5906128019153705, + "learning_rate": 2.704129875791861e-06, + "loss": 1.0731, + "step": 5375 + }, + { + "epoch": 0.40402825792875396, + "grad_norm": 1.6768282662788323, + "learning_rate": 2.7036741825355728e-06, + "loss": 0.9875, + "step": 5376 + }, + { + "epoch": 0.40410341199458893, + "grad_norm": 1.7426177036807702, + "learning_rate": 2.7032184475838953e-06, + "loss": 1.0171, + "step": 5377 + }, + { + "epoch": 0.40417856606042385, + "grad_norm": 1.680554401337759, + "learning_rate": 2.7027626709638317e-06, + "loss": 1.0326, + "step": 5378 + }, + { + "epoch": 0.4042537201262588, + "grad_norm": 1.944995042260135, + "learning_rate": 2.702306852702389e-06, + "loss": 0.9329, + "step": 5379 + }, + { + "epoch": 0.4043288741920938, + "grad_norm": 1.8424432293907544, + "learning_rate": 2.7018509928265763e-06, + "loss": 1.0685, + "step": 5380 + }, + { + "epoch": 0.40440402825792876, + "grad_norm": 1.5052057374010064, + "learning_rate": 2.7013950913634036e-06, + "loss": 1.0034, + "step": 5381 + }, + { + "epoch": 0.40447918232376373, + "grad_norm": 1.4525102525043916, + "learning_rate": 2.7009391483398868e-06, + "loss": 1.0594, + "step": 5382 + }, + { + "epoch": 0.4045543363895987, + "grad_norm": 2.0024181291129106, + "learning_rate": 2.7004831637830416e-06, + "loss": 0.9417, + "step": 5383 + }, + { + "epoch": 0.4046294904554336, + "grad_norm": 1.99704613705951, + "learning_rate": 2.700027137719886e-06, + "loss": 0.9703, + "step": 5384 + }, + { + "epoch": 0.4047046445212686, + "grad_norm": 1.4245025335298063, + "learning_rate": 2.699571070177442e-06, + "loss": 1.0096, + "step": 5385 + }, + { + "epoch": 0.40477979858710356, + "grad_norm": 1.8570352787781355, + "learning_rate": 2.6991149611827335e-06, + "loss": 1.042, + "step": 5386 + }, + { + "epoch": 0.4048549526529385, + "grad_norm": 2.1084006991071704, + "learning_rate": 2.6986588107627858e-06, + "loss": 1.0501, + "step": 5387 + }, + { + "epoch": 0.4049301067187735, + "grad_norm": 2.342818927203996, + "learning_rate": 2.698202618944629e-06, + "loss": 1.0193, + "step": 5388 + }, + { + "epoch": 0.40500526078460847, + "grad_norm": 1.9492181995826816, + "learning_rate": 2.697746385755293e-06, + "loss": 0.9891, + "step": 5389 + }, + { + "epoch": 0.4050804148504434, + "grad_norm": 1.722276087742339, + "learning_rate": 2.6972901112218123e-06, + "loss": 1.0313, + "step": 5390 + }, + { + "epoch": 0.40515556891627835, + "grad_norm": 1.8270251886503286, + "learning_rate": 2.696833795371222e-06, + "loss": 0.9469, + "step": 5391 + }, + { + "epoch": 0.4052307229821133, + "grad_norm": 2.9797808852062144, + "learning_rate": 2.696377438230561e-06, + "loss": 0.9591, + "step": 5392 + }, + { + "epoch": 0.4053058770479483, + "grad_norm": 3.222853775906761, + "learning_rate": 2.6959210398268703e-06, + "loss": 0.9799, + "step": 5393 + }, + { + "epoch": 0.40538103111378326, + "grad_norm": 2.591380220224401, + "learning_rate": 2.6954646001871928e-06, + "loss": 0.9404, + "step": 5394 + }, + { + "epoch": 0.40545618517961823, + "grad_norm": 1.737186112627195, + "learning_rate": 2.695008119338575e-06, + "loss": 1.0398, + "step": 5395 + }, + { + "epoch": 0.4055313392454532, + "grad_norm": 1.700455769982958, + "learning_rate": 2.6945515973080643e-06, + "loss": 0.9996, + "step": 5396 + }, + { + "epoch": 0.4056064933112881, + "grad_norm": 1.9543276936679967, + "learning_rate": 2.6940950341227124e-06, + "loss": 1.021, + "step": 5397 + }, + { + "epoch": 0.4056816473771231, + "grad_norm": 2.0166159862528965, + "learning_rate": 2.693638429809572e-06, + "loss": 0.916, + "step": 5398 + }, + { + "epoch": 0.40575680144295806, + "grad_norm": 1.9770407117048767, + "learning_rate": 2.6931817843956977e-06, + "loss": 1.0631, + "step": 5399 + }, + { + "epoch": 0.40583195550879303, + "grad_norm": 1.6316196154446612, + "learning_rate": 2.692725097908149e-06, + "loss": 0.8752, + "step": 5400 + }, + { + "epoch": 0.405907109574628, + "grad_norm": 1.7292274173344597, + "learning_rate": 2.692268370373985e-06, + "loss": 0.9395, + "step": 5401 + }, + { + "epoch": 0.40598226364046297, + "grad_norm": 1.5177722939249005, + "learning_rate": 2.69181160182027e-06, + "loss": 1.1021, + "step": 5402 + }, + { + "epoch": 0.4060574177062979, + "grad_norm": 1.65561246297661, + "learning_rate": 2.691354792274068e-06, + "loss": 1.0014, + "step": 5403 + }, + { + "epoch": 0.40613257177213286, + "grad_norm": 1.6642978842013763, + "learning_rate": 2.690897941762447e-06, + "loss": 1.0003, + "step": 5404 + }, + { + "epoch": 0.4062077258379678, + "grad_norm": 1.2620943059703504, + "learning_rate": 2.6904410503124774e-06, + "loss": 0.9726, + "step": 5405 + }, + { + "epoch": 0.4062828799038028, + "grad_norm": 1.723877627355457, + "learning_rate": 2.6899841179512324e-06, + "loss": 0.9155, + "step": 5406 + }, + { + "epoch": 0.40635803396963777, + "grad_norm": 1.7446494732511115, + "learning_rate": 2.689527144705785e-06, + "loss": 0.9417, + "step": 5407 + }, + { + "epoch": 0.40643318803547274, + "grad_norm": 1.7055337653216074, + "learning_rate": 2.6890701306032154e-06, + "loss": 0.9087, + "step": 5408 + }, + { + "epoch": 0.4065083421013077, + "grad_norm": 1.8924705388183485, + "learning_rate": 2.6886130756706003e-06, + "loss": 1.0334, + "step": 5409 + }, + { + "epoch": 0.4065834961671426, + "grad_norm": 2.027082069869659, + "learning_rate": 2.688155979935025e-06, + "loss": 0.952, + "step": 5410 + }, + { + "epoch": 0.4066586502329776, + "grad_norm": 1.6873167432956417, + "learning_rate": 2.687698843423572e-06, + "loss": 0.9317, + "step": 5411 + }, + { + "epoch": 0.40673380429881256, + "grad_norm": 1.723427182648281, + "learning_rate": 2.6872416661633296e-06, + "loss": 0.9562, + "step": 5412 + }, + { + "epoch": 0.40680895836464753, + "grad_norm": 1.470345057128045, + "learning_rate": 2.6867844481813868e-06, + "loss": 1.0167, + "step": 5413 + }, + { + "epoch": 0.4068841124304825, + "grad_norm": 2.0040515155452536, + "learning_rate": 2.6863271895048353e-06, + "loss": 0.9643, + "step": 5414 + }, + { + "epoch": 0.4069592664963175, + "grad_norm": 2.2677707376744114, + "learning_rate": 2.6858698901607696e-06, + "loss": 1.0167, + "step": 5415 + }, + { + "epoch": 0.4070344205621524, + "grad_norm": 1.590892998502516, + "learning_rate": 2.6854125501762863e-06, + "loss": 0.9435, + "step": 5416 + }, + { + "epoch": 0.40710957462798736, + "grad_norm": 1.679526884694539, + "learning_rate": 2.684955169578486e-06, + "loss": 0.9672, + "step": 5417 + }, + { + "epoch": 0.40718472869382233, + "grad_norm": 1.6154147396841978, + "learning_rate": 2.684497748394468e-06, + "loss": 1.017, + "step": 5418 + }, + { + "epoch": 0.4072598827596573, + "grad_norm": 2.4442471752136767, + "learning_rate": 2.6840402866513377e-06, + "loss": 0.9333, + "step": 5419 + }, + { + "epoch": 0.40733503682549227, + "grad_norm": 1.3902506806718862, + "learning_rate": 2.6835827843762006e-06, + "loss": 1.0663, + "step": 5420 + }, + { + "epoch": 0.40741019089132724, + "grad_norm": 1.9001801581047533, + "learning_rate": 2.6831252415961665e-06, + "loss": 1.0065, + "step": 5421 + }, + { + "epoch": 0.4074853449571622, + "grad_norm": 1.7486241245298944, + "learning_rate": 2.682667658338345e-06, + "loss": 0.9926, + "step": 5422 + }, + { + "epoch": 0.4075604990229971, + "grad_norm": 1.48050385687679, + "learning_rate": 2.6822100346298517e-06, + "loss": 0.9612, + "step": 5423 + }, + { + "epoch": 0.4076356530888321, + "grad_norm": 1.5138941841538396, + "learning_rate": 2.6817523704978014e-06, + "loss": 1.0847, + "step": 5424 + }, + { + "epoch": 0.40771080715466707, + "grad_norm": 2.0122301951507353, + "learning_rate": 2.681294665969312e-06, + "loss": 0.9302, + "step": 5425 + }, + { + "epoch": 0.40778596122050204, + "grad_norm": 2.0477298960564525, + "learning_rate": 2.6808369210715055e-06, + "loss": 0.9826, + "step": 5426 + }, + { + "epoch": 0.407861115286337, + "grad_norm": 1.9436044203036535, + "learning_rate": 2.6803791358315035e-06, + "loss": 0.9653, + "step": 5427 + }, + { + "epoch": 0.407936269352172, + "grad_norm": 1.4325294805825244, + "learning_rate": 2.679921310276432e-06, + "loss": 1.0597, + "step": 5428 + }, + { + "epoch": 0.4080114234180069, + "grad_norm": 1.5882507141150104, + "learning_rate": 2.6794634444334203e-06, + "loss": 0.831, + "step": 5429 + }, + { + "epoch": 0.40808657748384186, + "grad_norm": 2.017564535326026, + "learning_rate": 2.679005538329598e-06, + "loss": 1.015, + "step": 5430 + }, + { + "epoch": 0.40816173154967683, + "grad_norm": 1.7676332918273012, + "learning_rate": 2.678547591992096e-06, + "loss": 1.029, + "step": 5431 + }, + { + "epoch": 0.4082368856155118, + "grad_norm": 1.7304159806517538, + "learning_rate": 2.6780896054480526e-06, + "loss": 0.8419, + "step": 5432 + }, + { + "epoch": 0.4083120396813468, + "grad_norm": 2.213062464884402, + "learning_rate": 2.6776315787246024e-06, + "loss": 1.0492, + "step": 5433 + }, + { + "epoch": 0.40838719374718174, + "grad_norm": 1.3411789627674537, + "learning_rate": 2.6771735118488864e-06, + "loss": 0.9128, + "step": 5434 + }, + { + "epoch": 0.40846234781301666, + "grad_norm": 1.5861629833165174, + "learning_rate": 2.676715404848047e-06, + "loss": 1.0706, + "step": 5435 + }, + { + "epoch": 0.4085375018788516, + "grad_norm": 1.7216997670145737, + "learning_rate": 2.676257257749228e-06, + "loss": 1.018, + "step": 5436 + }, + { + "epoch": 0.4086126559446866, + "grad_norm": 1.585430346048552, + "learning_rate": 2.6757990705795777e-06, + "loss": 1.0006, + "step": 5437 + }, + { + "epoch": 0.40868781001052157, + "grad_norm": 1.7270977864812993, + "learning_rate": 2.675340843366244e-06, + "loss": 1.0606, + "step": 5438 + }, + { + "epoch": 0.40876296407635654, + "grad_norm": 1.698851166352029, + "learning_rate": 2.6748825761363794e-06, + "loss": 0.8801, + "step": 5439 + }, + { + "epoch": 0.4088381181421915, + "grad_norm": 1.9968216197918909, + "learning_rate": 2.674424268917138e-06, + "loss": 1.0447, + "step": 5440 + }, + { + "epoch": 0.4089132722080265, + "grad_norm": 1.6861487736162484, + "learning_rate": 2.6739659217356766e-06, + "loss": 1.0114, + "step": 5441 + }, + { + "epoch": 0.4089884262738614, + "grad_norm": 1.788739570891169, + "learning_rate": 2.6735075346191526e-06, + "loss": 0.9706, + "step": 5442 + }, + { + "epoch": 0.40906358033969636, + "grad_norm": 0.6753379602676995, + "learning_rate": 2.6730491075947294e-06, + "loss": 0.8429, + "step": 5443 + }, + { + "epoch": 0.40913873440553133, + "grad_norm": 21.589571999940734, + "learning_rate": 2.672590640689568e-06, + "loss": 0.967, + "step": 5444 + }, + { + "epoch": 0.4092138884713663, + "grad_norm": 1.6634741804287938, + "learning_rate": 2.6721321339308365e-06, + "loss": 0.9688, + "step": 5445 + }, + { + "epoch": 0.4092890425372013, + "grad_norm": 0.6814777431058202, + "learning_rate": 2.671673587345702e-06, + "loss": 0.8011, + "step": 5446 + }, + { + "epoch": 0.40936419660303625, + "grad_norm": 1.826314567662135, + "learning_rate": 2.671215000961335e-06, + "loss": 1.0228, + "step": 5447 + }, + { + "epoch": 0.40943935066887116, + "grad_norm": 1.9191163275617216, + "learning_rate": 2.6707563748049094e-06, + "loss": 1.0443, + "step": 5448 + }, + { + "epoch": 0.40951450473470613, + "grad_norm": 0.7875991319901511, + "learning_rate": 2.6702977089036e-06, + "loss": 0.8232, + "step": 5449 + }, + { + "epoch": 0.4095896588005411, + "grad_norm": 2.42387524112521, + "learning_rate": 2.6698390032845844e-06, + "loss": 1.085, + "step": 5450 + }, + { + "epoch": 0.40966481286637607, + "grad_norm": 1.9504230578319939, + "learning_rate": 2.6693802579750434e-06, + "loss": 0.9199, + "step": 5451 + }, + { + "epoch": 0.40973996693221104, + "grad_norm": 1.8114370978413894, + "learning_rate": 2.668921473002159e-06, + "loss": 1.0097, + "step": 5452 + }, + { + "epoch": 0.409815120998046, + "grad_norm": 1.6913356986539905, + "learning_rate": 2.668462648393115e-06, + "loss": 0.9646, + "step": 5453 + }, + { + "epoch": 0.409890275063881, + "grad_norm": 2.0569172063697323, + "learning_rate": 2.6680037841751e-06, + "loss": 1.0566, + "step": 5454 + }, + { + "epoch": 0.4099654291297159, + "grad_norm": 1.657864413585628, + "learning_rate": 2.6675448803753026e-06, + "loss": 0.9404, + "step": 5455 + }, + { + "epoch": 0.41004058319555087, + "grad_norm": 1.7265853535101972, + "learning_rate": 2.667085937020915e-06, + "loss": 1.0105, + "step": 5456 + }, + { + "epoch": 0.41011573726138584, + "grad_norm": 0.926285393176406, + "learning_rate": 2.6666269541391313e-06, + "loss": 0.9564, + "step": 5457 + }, + { + "epoch": 0.4101908913272208, + "grad_norm": 3.221897697183783, + "learning_rate": 2.6661679317571473e-06, + "loss": 1.0141, + "step": 5458 + }, + { + "epoch": 0.4102660453930558, + "grad_norm": 1.8781633395377306, + "learning_rate": 2.665708869902163e-06, + "loss": 1.0442, + "step": 5459 + }, + { + "epoch": 0.41034119945889075, + "grad_norm": 2.006726991986561, + "learning_rate": 2.6652497686013786e-06, + "loss": 0.9835, + "step": 5460 + }, + { + "epoch": 0.41041635352472566, + "grad_norm": 1.5855468870295948, + "learning_rate": 2.664790627881998e-06, + "loss": 0.9738, + "step": 5461 + }, + { + "epoch": 0.41049150759056063, + "grad_norm": 1.5365142639669909, + "learning_rate": 2.664331447771227e-06, + "loss": 1.0532, + "step": 5462 + }, + { + "epoch": 0.4105666616563956, + "grad_norm": 2.0468246803213646, + "learning_rate": 2.663872228296275e-06, + "loss": 0.9228, + "step": 5463 + }, + { + "epoch": 0.4106418157222306, + "grad_norm": 0.7194020174920324, + "learning_rate": 2.6634129694843497e-06, + "loss": 0.8256, + "step": 5464 + }, + { + "epoch": 0.41071696978806554, + "grad_norm": 2.592746586489966, + "learning_rate": 2.6629536713626664e-06, + "loss": 1.0637, + "step": 5465 + }, + { + "epoch": 0.4107921238539005, + "grad_norm": 1.4202224014627776, + "learning_rate": 2.662494333958439e-06, + "loss": 1.0141, + "step": 5466 + }, + { + "epoch": 0.4108672779197355, + "grad_norm": 1.4404594806240905, + "learning_rate": 2.662034957298886e-06, + "loss": 0.9808, + "step": 5467 + }, + { + "epoch": 0.4109424319855704, + "grad_norm": 3.707208241296399, + "learning_rate": 2.6615755414112266e-06, + "loss": 0.9045, + "step": 5468 + }, + { + "epoch": 0.41101758605140537, + "grad_norm": 1.8554340560078786, + "learning_rate": 2.6611160863226826e-06, + "loss": 0.9137, + "step": 5469 + }, + { + "epoch": 0.41109274011724034, + "grad_norm": 2.2619665665655844, + "learning_rate": 2.6606565920604793e-06, + "loss": 0.945, + "step": 5470 + }, + { + "epoch": 0.4111678941830753, + "grad_norm": 1.7714628409661402, + "learning_rate": 2.6601970586518428e-06, + "loss": 0.8993, + "step": 5471 + }, + { + "epoch": 0.4112430482489103, + "grad_norm": 1.4845086288535425, + "learning_rate": 2.6597374861240026e-06, + "loss": 0.9693, + "step": 5472 + }, + { + "epoch": 0.41131820231474525, + "grad_norm": 1.9019756861876544, + "learning_rate": 2.65927787450419e-06, + "loss": 1.0132, + "step": 5473 + }, + { + "epoch": 0.41139335638058017, + "grad_norm": 0.8820195645824492, + "learning_rate": 2.6588182238196395e-06, + "loss": 0.875, + "step": 5474 + }, + { + "epoch": 0.41146851044641514, + "grad_norm": 5.024075908050992, + "learning_rate": 2.6583585340975854e-06, + "loss": 0.9437, + "step": 5475 + }, + { + "epoch": 0.4115436645122501, + "grad_norm": 4.86657867787567, + "learning_rate": 2.657898805365268e-06, + "loss": 1.0214, + "step": 5476 + }, + { + "epoch": 0.4116188185780851, + "grad_norm": 2.1980970649148115, + "learning_rate": 2.6574390376499265e-06, + "loss": 1.0483, + "step": 5477 + }, + { + "epoch": 0.41169397264392005, + "grad_norm": 1.8655645108280976, + "learning_rate": 2.6569792309788046e-06, + "loss": 0.846, + "step": 5478 + }, + { + "epoch": 0.411769126709755, + "grad_norm": 1.8255213920240505, + "learning_rate": 2.656519385379148e-06, + "loss": 0.9642, + "step": 5479 + }, + { + "epoch": 0.41184428077558993, + "grad_norm": 3.6373973504228654, + "learning_rate": 2.6560595008782032e-06, + "loss": 0.9701, + "step": 5480 + }, + { + "epoch": 0.4119194348414249, + "grad_norm": 1.726830578183309, + "learning_rate": 2.655599577503221e-06, + "loss": 1.0005, + "step": 5481 + }, + { + "epoch": 0.4119945889072599, + "grad_norm": 1.827335611091223, + "learning_rate": 2.6551396152814534e-06, + "loss": 1.0203, + "step": 5482 + }, + { + "epoch": 0.41206974297309484, + "grad_norm": 17.756116216277068, + "learning_rate": 2.6546796142401547e-06, + "loss": 1.0115, + "step": 5483 + }, + { + "epoch": 0.4121448970389298, + "grad_norm": 0.7320795505348513, + "learning_rate": 2.6542195744065826e-06, + "loss": 0.8023, + "step": 5484 + }, + { + "epoch": 0.4122200511047648, + "grad_norm": 1.5180164622191952, + "learning_rate": 2.653759495807995e-06, + "loss": 1.0356, + "step": 5485 + }, + { + "epoch": 0.41229520517059975, + "grad_norm": 0.7481066328835941, + "learning_rate": 2.6532993784716535e-06, + "loss": 0.8833, + "step": 5486 + }, + { + "epoch": 0.41237035923643467, + "grad_norm": 1.8881598550760799, + "learning_rate": 2.652839222424823e-06, + "loss": 1.0565, + "step": 5487 + }, + { + "epoch": 0.41244551330226964, + "grad_norm": 1.3737177939280074, + "learning_rate": 2.652379027694768e-06, + "loss": 0.9756, + "step": 5488 + }, + { + "epoch": 0.4125206673681046, + "grad_norm": 1.7341775867729983, + "learning_rate": 2.651918794308758e-06, + "loss": 0.9076, + "step": 5489 + }, + { + "epoch": 0.4125958214339396, + "grad_norm": 4.091406930628763, + "learning_rate": 2.651458522294063e-06, + "loss": 1.0475, + "step": 5490 + }, + { + "epoch": 0.41267097549977455, + "grad_norm": 1.831916854312581, + "learning_rate": 2.650998211677956e-06, + "loss": 1.0006, + "step": 5491 + }, + { + "epoch": 0.4127461295656095, + "grad_norm": 2.31564292746334, + "learning_rate": 2.6505378624877116e-06, + "loss": 1.0315, + "step": 5492 + }, + { + "epoch": 0.41282128363144444, + "grad_norm": 4.9465590799633015, + "learning_rate": 2.650077474750608e-06, + "loss": 0.9662, + "step": 5493 + }, + { + "epoch": 0.4128964376972794, + "grad_norm": 0.8612027931286795, + "learning_rate": 2.649617048493925e-06, + "loss": 0.9258, + "step": 5494 + }, + { + "epoch": 0.4129715917631144, + "grad_norm": 1.674063275443676, + "learning_rate": 2.649156583744944e-06, + "loss": 0.9716, + "step": 5495 + }, + { + "epoch": 0.41304674582894935, + "grad_norm": 2.729527214248646, + "learning_rate": 2.64869608053095e-06, + "loss": 0.9318, + "step": 5496 + }, + { + "epoch": 0.4131218998947843, + "grad_norm": 1.6837697656231712, + "learning_rate": 2.648235538879229e-06, + "loss": 0.9218, + "step": 5497 + }, + { + "epoch": 0.4131970539606193, + "grad_norm": 1.44842277116087, + "learning_rate": 2.6477749588170703e-06, + "loss": 1.0087, + "step": 5498 + }, + { + "epoch": 0.41327220802645426, + "grad_norm": 1.7025333741847801, + "learning_rate": 2.647314340371764e-06, + "loss": 1.0261, + "step": 5499 + }, + { + "epoch": 0.4133473620922892, + "grad_norm": 1.540318552610146, + "learning_rate": 2.646853683570605e-06, + "loss": 0.8887, + "step": 5500 + }, + { + "epoch": 0.41342251615812414, + "grad_norm": 2.081812863231436, + "learning_rate": 2.646392988440888e-06, + "loss": 1.0409, + "step": 5501 + }, + { + "epoch": 0.4134976702239591, + "grad_norm": 1.8482668877718493, + "learning_rate": 2.6459322550099113e-06, + "loss": 0.9641, + "step": 5502 + }, + { + "epoch": 0.4135728242897941, + "grad_norm": 1.794277602701101, + "learning_rate": 2.645471483304975e-06, + "loss": 1.0049, + "step": 5503 + }, + { + "epoch": 0.41364797835562905, + "grad_norm": 1.9232231311971646, + "learning_rate": 2.645010673353382e-06, + "loss": 1.0601, + "step": 5504 + }, + { + "epoch": 0.413723132421464, + "grad_norm": 2.4990556840802465, + "learning_rate": 2.644549825182436e-06, + "loss": 1.0182, + "step": 5505 + }, + { + "epoch": 0.41379828648729894, + "grad_norm": 1.692676093943785, + "learning_rate": 2.644088938819445e-06, + "loss": 0.994, + "step": 5506 + }, + { + "epoch": 0.4138734405531339, + "grad_norm": 2.381013284767128, + "learning_rate": 2.6436280142917183e-06, + "loss": 0.9958, + "step": 5507 + }, + { + "epoch": 0.4139485946189689, + "grad_norm": 2.5136402901752537, + "learning_rate": 2.6431670516265668e-06, + "loss": 0.9473, + "step": 5508 + }, + { + "epoch": 0.41402374868480385, + "grad_norm": 1.889466301077721, + "learning_rate": 2.6427060508513052e-06, + "loss": 0.8849, + "step": 5509 + }, + { + "epoch": 0.4140989027506388, + "grad_norm": 1.3995912789727487, + "learning_rate": 2.6422450119932484e-06, + "loss": 1.0159, + "step": 5510 + }, + { + "epoch": 0.4141740568164738, + "grad_norm": 1.9361405679452575, + "learning_rate": 2.641783935079716e-06, + "loss": 0.9552, + "step": 5511 + }, + { + "epoch": 0.41424921088230876, + "grad_norm": 1.448880549449428, + "learning_rate": 2.641322820138027e-06, + "loss": 0.9367, + "step": 5512 + }, + { + "epoch": 0.4143243649481437, + "grad_norm": 1.9024740248923282, + "learning_rate": 2.6408616671955053e-06, + "loss": 0.9568, + "step": 5513 + }, + { + "epoch": 0.41439951901397865, + "grad_norm": 2.0130947110995385, + "learning_rate": 2.6404004762794766e-06, + "loss": 0.977, + "step": 5514 + }, + { + "epoch": 0.4144746730798136, + "grad_norm": 0.6742536003149625, + "learning_rate": 2.6399392474172667e-06, + "loss": 0.8091, + "step": 5515 + }, + { + "epoch": 0.4145498271456486, + "grad_norm": 0.7297303654824983, + "learning_rate": 2.6394779806362057e-06, + "loss": 0.8478, + "step": 5516 + }, + { + "epoch": 0.41462498121148356, + "grad_norm": 1.5848783517358755, + "learning_rate": 2.6390166759636263e-06, + "loss": 0.9573, + "step": 5517 + }, + { + "epoch": 0.4147001352773185, + "grad_norm": 0.770616337587502, + "learning_rate": 2.638555333426862e-06, + "loss": 0.9608, + "step": 5518 + }, + { + "epoch": 0.41477528934315344, + "grad_norm": 1.7918038500526776, + "learning_rate": 2.638093953053248e-06, + "loss": 1.0021, + "step": 5519 + }, + { + "epoch": 0.4148504434089884, + "grad_norm": 1.6550740799934345, + "learning_rate": 2.6376325348701244e-06, + "loss": 1.0276, + "step": 5520 + }, + { + "epoch": 0.4149255974748234, + "grad_norm": 1.733535219625282, + "learning_rate": 2.6371710789048313e-06, + "loss": 1.0451, + "step": 5521 + }, + { + "epoch": 0.41500075154065835, + "grad_norm": 1.7238636586295626, + "learning_rate": 2.6367095851847125e-06, + "loss": 0.9312, + "step": 5522 + }, + { + "epoch": 0.4150759056064933, + "grad_norm": 2.5123965949040468, + "learning_rate": 2.636248053737112e-06, + "loss": 0.9537, + "step": 5523 + }, + { + "epoch": 0.4151510596723283, + "grad_norm": 2.147245319083767, + "learning_rate": 2.635786484589378e-06, + "loss": 1.0282, + "step": 5524 + }, + { + "epoch": 0.4152262137381632, + "grad_norm": 1.6428586740664342, + "learning_rate": 2.6353248777688606e-06, + "loss": 0.9939, + "step": 5525 + }, + { + "epoch": 0.4153013678039982, + "grad_norm": 1.7359263037759158, + "learning_rate": 2.634863233302911e-06, + "loss": 0.894, + "step": 5526 + }, + { + "epoch": 0.41537652186983315, + "grad_norm": 1.5492496950999648, + "learning_rate": 2.634401551218884e-06, + "loss": 0.93, + "step": 5527 + }, + { + "epoch": 0.4154516759356681, + "grad_norm": 2.4057678135863783, + "learning_rate": 2.6339398315441353e-06, + "loss": 1.0693, + "step": 5528 + }, + { + "epoch": 0.4155268300015031, + "grad_norm": 1.591073874819858, + "learning_rate": 2.633478074306025e-06, + "loss": 0.997, + "step": 5529 + }, + { + "epoch": 0.41560198406733806, + "grad_norm": 1.6285602741887233, + "learning_rate": 2.633016279531912e-06, + "loss": 0.9479, + "step": 5530 + }, + { + "epoch": 0.41567713813317303, + "grad_norm": 2.0473478513646146, + "learning_rate": 2.6325544472491616e-06, + "loss": 0.8926, + "step": 5531 + }, + { + "epoch": 0.41575229219900794, + "grad_norm": 0.6315150441003476, + "learning_rate": 2.632092577485137e-06, + "loss": 0.8265, + "step": 5532 + }, + { + "epoch": 0.4158274462648429, + "grad_norm": 2.1493664761812687, + "learning_rate": 2.631630670267207e-06, + "loss": 0.9414, + "step": 5533 + }, + { + "epoch": 0.4159026003306779, + "grad_norm": 2.3029469269812606, + "learning_rate": 2.631168725622742e-06, + "loss": 1.0103, + "step": 5534 + }, + { + "epoch": 0.41597775439651286, + "grad_norm": 2.081774382643497, + "learning_rate": 2.630706743579112e-06, + "loss": 1.0817, + "step": 5535 + }, + { + "epoch": 0.4160529084623478, + "grad_norm": 1.8784130821233878, + "learning_rate": 2.6302447241636924e-06, + "loss": 0.9059, + "step": 5536 + }, + { + "epoch": 0.4161280625281828, + "grad_norm": 1.8960710383478643, + "learning_rate": 2.6297826674038595e-06, + "loss": 1.0759, + "step": 5537 + }, + { + "epoch": 0.4162032165940177, + "grad_norm": 1.9347787635953233, + "learning_rate": 2.6293205733269924e-06, + "loss": 0.8669, + "step": 5538 + }, + { + "epoch": 0.4162783706598527, + "grad_norm": 2.4923563454571416, + "learning_rate": 2.6288584419604713e-06, + "loss": 1.0553, + "step": 5539 + }, + { + "epoch": 0.41635352472568765, + "grad_norm": 1.983711131627916, + "learning_rate": 2.62839627333168e-06, + "loss": 1.0216, + "step": 5540 + }, + { + "epoch": 0.4164286787915226, + "grad_norm": 0.7703966415908435, + "learning_rate": 2.6279340674680025e-06, + "loss": 0.8675, + "step": 5541 + }, + { + "epoch": 0.4165038328573576, + "grad_norm": 1.46320530113357, + "learning_rate": 2.627471824396827e-06, + "loss": 0.9969, + "step": 5542 + }, + { + "epoch": 0.41657898692319256, + "grad_norm": 2.267200316925942, + "learning_rate": 2.6270095441455435e-06, + "loss": 0.9556, + "step": 5543 + }, + { + "epoch": 0.41665414098902753, + "grad_norm": 0.6631489247672099, + "learning_rate": 2.6265472267415432e-06, + "loss": 0.8337, + "step": 5544 + }, + { + "epoch": 0.41672929505486245, + "grad_norm": 1.9901043940243413, + "learning_rate": 2.626084872212221e-06, + "loss": 1.0129, + "step": 5545 + }, + { + "epoch": 0.4168044491206974, + "grad_norm": 1.631608363867568, + "learning_rate": 2.625622480584972e-06, + "loss": 1.0246, + "step": 5546 + }, + { + "epoch": 0.4168796031865324, + "grad_norm": 1.890235797765661, + "learning_rate": 2.6251600518871953e-06, + "loss": 1.0876, + "step": 5547 + }, + { + "epoch": 0.41695475725236736, + "grad_norm": 1.825128365274178, + "learning_rate": 2.6246975861462927e-06, + "loss": 0.9588, + "step": 5548 + }, + { + "epoch": 0.41702991131820233, + "grad_norm": 1.2956413468223535, + "learning_rate": 2.6242350833896645e-06, + "loss": 1.0128, + "step": 5549 + }, + { + "epoch": 0.4171050653840373, + "grad_norm": 1.9482539355848798, + "learning_rate": 2.623772543644718e-06, + "loss": 0.8653, + "step": 5550 + }, + { + "epoch": 0.4171802194498722, + "grad_norm": 1.2018297332316872, + "learning_rate": 2.6233099669388605e-06, + "loss": 0.924, + "step": 5551 + }, + { + "epoch": 0.4172553735157072, + "grad_norm": 2.5831064821950633, + "learning_rate": 2.6228473532995e-06, + "loss": 1.0334, + "step": 5552 + }, + { + "epoch": 0.41733052758154215, + "grad_norm": 1.6535441049438715, + "learning_rate": 2.6223847027540485e-06, + "loss": 0.8814, + "step": 5553 + }, + { + "epoch": 0.4174056816473771, + "grad_norm": 5.142506542062562, + "learning_rate": 2.62192201532992e-06, + "loss": 0.9743, + "step": 5554 + }, + { + "epoch": 0.4174808357132121, + "grad_norm": 1.8936946948949147, + "learning_rate": 2.621459291054531e-06, + "loss": 0.8538, + "step": 5555 + }, + { + "epoch": 0.41755598977904707, + "grad_norm": 3.2267025970265997, + "learning_rate": 2.6209965299552994e-06, + "loss": 1.0306, + "step": 5556 + }, + { + "epoch": 0.41763114384488204, + "grad_norm": 1.4020436850501925, + "learning_rate": 2.6205337320596452e-06, + "loss": 1.0184, + "step": 5557 + }, + { + "epoch": 0.41770629791071695, + "grad_norm": 2.542727129308191, + "learning_rate": 2.620070897394991e-06, + "loss": 1.033, + "step": 5558 + }, + { + "epoch": 0.4177814519765519, + "grad_norm": 1.807013659756514, + "learning_rate": 2.619608025988762e-06, + "loss": 0.9135, + "step": 5559 + }, + { + "epoch": 0.4178566060423869, + "grad_norm": 1.8103118052547944, + "learning_rate": 2.6191451178683842e-06, + "loss": 0.9245, + "step": 5560 + }, + { + "epoch": 0.41793176010822186, + "grad_norm": 0.7183424996929263, + "learning_rate": 2.6186821730612884e-06, + "loss": 0.8209, + "step": 5561 + }, + { + "epoch": 0.41800691417405683, + "grad_norm": 1.743202016721787, + "learning_rate": 2.6182191915949043e-06, + "loss": 1.1105, + "step": 5562 + }, + { + "epoch": 0.4180820682398918, + "grad_norm": 1.892196695571593, + "learning_rate": 2.6177561734966653e-06, + "loss": 0.9495, + "step": 5563 + }, + { + "epoch": 0.4181572223057267, + "grad_norm": 1.4640027696604143, + "learning_rate": 2.6172931187940084e-06, + "loss": 0.8413, + "step": 5564 + }, + { + "epoch": 0.4182323763715617, + "grad_norm": 2.0548014446037373, + "learning_rate": 2.6168300275143695e-06, + "loss": 0.9658, + "step": 5565 + }, + { + "epoch": 0.41830753043739666, + "grad_norm": 1.8193275504704696, + "learning_rate": 2.61636689968519e-06, + "loss": 0.9643, + "step": 5566 + }, + { + "epoch": 0.4183826845032316, + "grad_norm": 1.774842096746578, + "learning_rate": 2.6159037353339113e-06, + "loss": 0.9582, + "step": 5567 + }, + { + "epoch": 0.4184578385690666, + "grad_norm": 0.6591886844727448, + "learning_rate": 2.6154405344879776e-06, + "loss": 0.7916, + "step": 5568 + }, + { + "epoch": 0.41853299263490157, + "grad_norm": 3.1921947537048947, + "learning_rate": 2.6149772971748357e-06, + "loss": 0.8709, + "step": 5569 + }, + { + "epoch": 0.4186081467007365, + "grad_norm": 1.628487585772947, + "learning_rate": 2.614514023421934e-06, + "loss": 1.0517, + "step": 5570 + }, + { + "epoch": 0.41868330076657145, + "grad_norm": 2.139668940311398, + "learning_rate": 2.6140507132567238e-06, + "loss": 0.9672, + "step": 5571 + }, + { + "epoch": 0.4187584548324064, + "grad_norm": 2.719369138177837, + "learning_rate": 2.6135873667066567e-06, + "loss": 1.0443, + "step": 5572 + }, + { + "epoch": 0.4188336088982414, + "grad_norm": 1.4377037401526271, + "learning_rate": 2.6131239837991894e-06, + "loss": 1.0811, + "step": 5573 + }, + { + "epoch": 0.41890876296407636, + "grad_norm": 1.4310999495964156, + "learning_rate": 2.6126605645617777e-06, + "loss": 0.9074, + "step": 5574 + }, + { + "epoch": 0.41898391702991133, + "grad_norm": 2.4062947851199588, + "learning_rate": 2.6121971090218816e-06, + "loss": 0.9496, + "step": 5575 + }, + { + "epoch": 0.4190590710957463, + "grad_norm": 0.8583008381948475, + "learning_rate": 2.6117336172069625e-06, + "loss": 0.912, + "step": 5576 + }, + { + "epoch": 0.4191342251615812, + "grad_norm": 1.876641063967252, + "learning_rate": 2.6112700891444845e-06, + "loss": 0.9786, + "step": 5577 + }, + { + "epoch": 0.4192093792274162, + "grad_norm": 1.8296604066531899, + "learning_rate": 2.6108065248619128e-06, + "loss": 1.01, + "step": 5578 + }, + { + "epoch": 0.41928453329325116, + "grad_norm": 1.8677147156889484, + "learning_rate": 2.6103429243867147e-06, + "loss": 0.9641, + "step": 5579 + }, + { + "epoch": 0.41935968735908613, + "grad_norm": 1.7697897210726972, + "learning_rate": 2.609879287746362e-06, + "loss": 0.9872, + "step": 5580 + }, + { + "epoch": 0.4194348414249211, + "grad_norm": 1.5635704188675643, + "learning_rate": 2.609415614968326e-06, + "loss": 0.951, + "step": 5581 + }, + { + "epoch": 0.41950999549075607, + "grad_norm": 0.6744283904965567, + "learning_rate": 2.608951906080081e-06, + "loss": 0.8676, + "step": 5582 + }, + { + "epoch": 0.419585149556591, + "grad_norm": 2.0225722182386274, + "learning_rate": 2.608488161109104e-06, + "loss": 0.9444, + "step": 5583 + }, + { + "epoch": 0.41966030362242596, + "grad_norm": 2.4847640206797013, + "learning_rate": 2.608024380082874e-06, + "loss": 1.0336, + "step": 5584 + }, + { + "epoch": 0.4197354576882609, + "grad_norm": 1.8823389114617881, + "learning_rate": 2.60756056302887e-06, + "loss": 0.8837, + "step": 5585 + }, + { + "epoch": 0.4198106117540959, + "grad_norm": 1.8413232252965472, + "learning_rate": 2.6070967099745773e-06, + "loss": 1.0199, + "step": 5586 + }, + { + "epoch": 0.41988576581993087, + "grad_norm": 1.6191527730856474, + "learning_rate": 2.6066328209474786e-06, + "loss": 0.9924, + "step": 5587 + }, + { + "epoch": 0.41996091988576584, + "grad_norm": 1.6357687876801712, + "learning_rate": 2.6061688959750633e-06, + "loss": 0.8279, + "step": 5588 + }, + { + "epoch": 0.4200360739516008, + "grad_norm": 1.5757765717406231, + "learning_rate": 2.6057049350848194e-06, + "loss": 1.0375, + "step": 5589 + }, + { + "epoch": 0.4201112280174357, + "grad_norm": 1.9650625179139474, + "learning_rate": 2.6052409383042383e-06, + "loss": 0.8885, + "step": 5590 + }, + { + "epoch": 0.4201863820832707, + "grad_norm": 1.7217810520359151, + "learning_rate": 2.604776905660814e-06, + "loss": 1.0008, + "step": 5591 + }, + { + "epoch": 0.42026153614910566, + "grad_norm": 1.7844597524100092, + "learning_rate": 2.6043128371820427e-06, + "loss": 1.0103, + "step": 5592 + }, + { + "epoch": 0.42033669021494063, + "grad_norm": 1.6488766921318625, + "learning_rate": 2.603848732895421e-06, + "loss": 0.8548, + "step": 5593 + }, + { + "epoch": 0.4204118442807756, + "grad_norm": 2.426698471356965, + "learning_rate": 2.6033845928284503e-06, + "loss": 0.9501, + "step": 5594 + }, + { + "epoch": 0.4204869983466106, + "grad_norm": 1.9438357588809483, + "learning_rate": 2.602920417008632e-06, + "loss": 1.0221, + "step": 5595 + }, + { + "epoch": 0.4205621524124455, + "grad_norm": 1.4583705443221584, + "learning_rate": 2.60245620546347e-06, + "loss": 0.9551, + "step": 5596 + }, + { + "epoch": 0.42063730647828046, + "grad_norm": 2.726021961747813, + "learning_rate": 2.6019919582204713e-06, + "loss": 1.0211, + "step": 5597 + }, + { + "epoch": 0.42071246054411543, + "grad_norm": 1.8555440465807513, + "learning_rate": 2.601527675307143e-06, + "loss": 0.9747, + "step": 5598 + }, + { + "epoch": 0.4207876146099504, + "grad_norm": 2.988468737735655, + "learning_rate": 2.601063356750997e-06, + "loss": 0.9811, + "step": 5599 + }, + { + "epoch": 0.42086276867578537, + "grad_norm": 1.9877416331878315, + "learning_rate": 2.600599002579546e-06, + "loss": 0.9423, + "step": 5600 + }, + { + "epoch": 0.42093792274162034, + "grad_norm": 2.084745030536394, + "learning_rate": 2.6001346128203036e-06, + "loss": 0.9886, + "step": 5601 + }, + { + "epoch": 0.4210130768074553, + "grad_norm": 2.240806553081122, + "learning_rate": 2.5996701875007873e-06, + "loss": 0.9849, + "step": 5602 + }, + { + "epoch": 0.4210882308732902, + "grad_norm": 1.4121398922483692, + "learning_rate": 2.5992057266485162e-06, + "loss": 1.0004, + "step": 5603 + }, + { + "epoch": 0.4211633849391252, + "grad_norm": 1.568320533965066, + "learning_rate": 2.5987412302910114e-06, + "loss": 0.9566, + "step": 5604 + }, + { + "epoch": 0.42123853900496017, + "grad_norm": 1.768196865979677, + "learning_rate": 2.598276698455796e-06, + "loss": 0.9805, + "step": 5605 + }, + { + "epoch": 0.42131369307079514, + "grad_norm": 0.6430221693020015, + "learning_rate": 2.5978121311703955e-06, + "loss": 0.8014, + "step": 5606 + }, + { + "epoch": 0.4213888471366301, + "grad_norm": 2.2889985475948023, + "learning_rate": 2.5973475284623366e-06, + "loss": 0.9775, + "step": 5607 + }, + { + "epoch": 0.4214640012024651, + "grad_norm": 1.5545384321201186, + "learning_rate": 2.5968828903591492e-06, + "loss": 1.0647, + "step": 5608 + }, + { + "epoch": 0.4215391552683, + "grad_norm": 2.1268022411903567, + "learning_rate": 2.5964182168883654e-06, + "loss": 1.0111, + "step": 5609 + }, + { + "epoch": 0.42161430933413496, + "grad_norm": 2.333542326241501, + "learning_rate": 2.5959535080775176e-06, + "loss": 0.8116, + "step": 5610 + }, + { + "epoch": 0.42168946339996993, + "grad_norm": 1.5270323765238334, + "learning_rate": 2.595488763954143e-06, + "loss": 0.9674, + "step": 5611 + }, + { + "epoch": 0.4217646174658049, + "grad_norm": 2.2358180368046887, + "learning_rate": 2.5950239845457792e-06, + "loss": 0.9636, + "step": 5612 + }, + { + "epoch": 0.4218397715316399, + "grad_norm": 1.8751261986542442, + "learning_rate": 2.594559169879965e-06, + "loss": 0.9511, + "step": 5613 + }, + { + "epoch": 0.42191492559747484, + "grad_norm": 1.358301506176675, + "learning_rate": 2.594094319984244e-06, + "loss": 1.0524, + "step": 5614 + }, + { + "epoch": 0.42199007966330976, + "grad_norm": 0.7635672807408772, + "learning_rate": 2.593629434886159e-06, + "loss": 0.8328, + "step": 5615 + }, + { + "epoch": 0.42206523372914473, + "grad_norm": 3.212317845267787, + "learning_rate": 2.5931645146132576e-06, + "loss": 0.9015, + "step": 5616 + }, + { + "epoch": 0.4221403877949797, + "grad_norm": 1.8340641284700756, + "learning_rate": 2.592699559193086e-06, + "loss": 1.0715, + "step": 5617 + }, + { + "epoch": 0.42221554186081467, + "grad_norm": 1.4003576809507556, + "learning_rate": 2.592234568653197e-06, + "loss": 0.9804, + "step": 5618 + }, + { + "epoch": 0.42229069592664964, + "grad_norm": 1.6936539654409724, + "learning_rate": 2.5917695430211416e-06, + "loss": 0.9469, + "step": 5619 + }, + { + "epoch": 0.4223658499924846, + "grad_norm": 2.0420701802417844, + "learning_rate": 2.591304482324475e-06, + "loss": 1.049, + "step": 5620 + }, + { + "epoch": 0.4224410040583196, + "grad_norm": 1.3619863049214926, + "learning_rate": 2.590839386590754e-06, + "loss": 0.9492, + "step": 5621 + }, + { + "epoch": 0.4225161581241545, + "grad_norm": 1.7984235642039892, + "learning_rate": 2.5903742558475358e-06, + "loss": 0.9909, + "step": 5622 + }, + { + "epoch": 0.42259131218998947, + "grad_norm": 1.8629163468725445, + "learning_rate": 2.589909090122383e-06, + "loss": 1.023, + "step": 5623 + }, + { + "epoch": 0.42266646625582444, + "grad_norm": 1.6398718422150464, + "learning_rate": 2.589443889442857e-06, + "loss": 0.8962, + "step": 5624 + }, + { + "epoch": 0.4227416203216594, + "grad_norm": 1.5785089158007861, + "learning_rate": 2.5889786538365243e-06, + "loss": 0.9896, + "step": 5625 + }, + { + "epoch": 0.4228167743874944, + "grad_norm": 1.3731565255325668, + "learning_rate": 2.588513383330951e-06, + "loss": 0.9334, + "step": 5626 + }, + { + "epoch": 0.42289192845332935, + "grad_norm": 1.67100616411023, + "learning_rate": 2.588048077953705e-06, + "loss": 0.9857, + "step": 5627 + }, + { + "epoch": 0.42296708251916426, + "grad_norm": 3.0977872457622593, + "learning_rate": 2.58758273773236e-06, + "loss": 1.0373, + "step": 5628 + }, + { + "epoch": 0.42304223658499923, + "grad_norm": 2.0938028357372636, + "learning_rate": 2.5871173626944864e-06, + "loss": 0.9914, + "step": 5629 + }, + { + "epoch": 0.4231173906508342, + "grad_norm": 9.421259972282478, + "learning_rate": 2.586651952867662e-06, + "loss": 0.994, + "step": 5630 + }, + { + "epoch": 0.4231925447166692, + "grad_norm": 1.6603050973024627, + "learning_rate": 2.5861865082794625e-06, + "loss": 1.0492, + "step": 5631 + }, + { + "epoch": 0.42326769878250414, + "grad_norm": 1.501124278576376, + "learning_rate": 2.5857210289574675e-06, + "loss": 0.9181, + "step": 5632 + }, + { + "epoch": 0.4233428528483391, + "grad_norm": 2.149283613573301, + "learning_rate": 2.5852555149292593e-06, + "loss": 0.995, + "step": 5633 + }, + { + "epoch": 0.4234180069141741, + "grad_norm": 1.8760272073338542, + "learning_rate": 2.5847899662224195e-06, + "loss": 1.0573, + "step": 5634 + }, + { + "epoch": 0.423493160980009, + "grad_norm": 1.9461002822943987, + "learning_rate": 2.584324382864536e-06, + "loss": 0.8492, + "step": 5635 + }, + { + "epoch": 0.42356831504584397, + "grad_norm": 0.6593114650543535, + "learning_rate": 2.583858764883195e-06, + "loss": 0.8058, + "step": 5636 + }, + { + "epoch": 0.42364346911167894, + "grad_norm": 1.561571761022332, + "learning_rate": 2.5833931123059865e-06, + "loss": 0.9361, + "step": 5637 + }, + { + "epoch": 0.4237186231775139, + "grad_norm": 1.9567481652832788, + "learning_rate": 2.5829274251605023e-06, + "loss": 1.0485, + "step": 5638 + }, + { + "epoch": 0.4237937772433489, + "grad_norm": 2.004539345122719, + "learning_rate": 2.5824617034743354e-06, + "loss": 0.9333, + "step": 5639 + }, + { + "epoch": 0.42386893130918385, + "grad_norm": 1.4768171948563287, + "learning_rate": 2.5819959472750827e-06, + "loss": 0.9866, + "step": 5640 + }, + { + "epoch": 0.42394408537501876, + "grad_norm": 2.1205836011025685, + "learning_rate": 2.581530156590341e-06, + "loss": 1.0031, + "step": 5641 + }, + { + "epoch": 0.42401923944085373, + "grad_norm": 0.6807778918673989, + "learning_rate": 2.5810643314477116e-06, + "loss": 0.8224, + "step": 5642 + }, + { + "epoch": 0.4240943935066887, + "grad_norm": 1.8129890045320736, + "learning_rate": 2.5805984718747953e-06, + "loss": 0.9604, + "step": 5643 + }, + { + "epoch": 0.4241695475725237, + "grad_norm": 1.6922642599714686, + "learning_rate": 2.5801325778991958e-06, + "loss": 0.9673, + "step": 5644 + }, + { + "epoch": 0.42424470163835865, + "grad_norm": 2.0644769152927953, + "learning_rate": 2.5796666495485196e-06, + "loss": 0.989, + "step": 5645 + }, + { + "epoch": 0.4243198557041936, + "grad_norm": 1.8643165127528374, + "learning_rate": 2.579200686850375e-06, + "loss": 0.9951, + "step": 5646 + }, + { + "epoch": 0.4243950097700286, + "grad_norm": 1.7297059868821372, + "learning_rate": 2.5787346898323716e-06, + "loss": 0.991, + "step": 5647 + }, + { + "epoch": 0.4244701638358635, + "grad_norm": 2.0362958005808927, + "learning_rate": 2.578268658522122e-06, + "loss": 0.9527, + "step": 5648 + }, + { + "epoch": 0.42454531790169847, + "grad_norm": 1.4796973964275624, + "learning_rate": 2.5778025929472397e-06, + "loss": 0.8891, + "step": 5649 + }, + { + "epoch": 0.42462047196753344, + "grad_norm": 1.6495986392525241, + "learning_rate": 2.577336493135341e-06, + "loss": 1.0118, + "step": 5650 + }, + { + "epoch": 0.4246956260333684, + "grad_norm": 1.5752124915513552, + "learning_rate": 2.5768703591140445e-06, + "loss": 1.0173, + "step": 5651 + }, + { + "epoch": 0.4247707800992034, + "grad_norm": 2.2682219255917953, + "learning_rate": 2.5764041909109706e-06, + "loss": 0.9507, + "step": 5652 + }, + { + "epoch": 0.42484593416503835, + "grad_norm": 2.17459101284601, + "learning_rate": 2.5759379885537414e-06, + "loss": 1.1014, + "step": 5653 + }, + { + "epoch": 0.42492108823087327, + "grad_norm": 2.9260760618428807, + "learning_rate": 2.57547175206998e-06, + "loss": 0.9874, + "step": 5654 + }, + { + "epoch": 0.42499624229670824, + "grad_norm": 17.452732447185653, + "learning_rate": 2.5750054814873144e-06, + "loss": 1.0537, + "step": 5655 + }, + { + "epoch": 0.4250713963625432, + "grad_norm": 1.9026492145787628, + "learning_rate": 2.5745391768333715e-06, + "loss": 1.0713, + "step": 5656 + }, + { + "epoch": 0.4251465504283782, + "grad_norm": 1.7740146185262948, + "learning_rate": 2.574072838135783e-06, + "loss": 1.0799, + "step": 5657 + }, + { + "epoch": 0.42522170449421315, + "grad_norm": 1.6152233044984137, + "learning_rate": 2.5736064654221806e-06, + "loss": 1.0096, + "step": 5658 + }, + { + "epoch": 0.4252968585600481, + "grad_norm": 0.9454250741559285, + "learning_rate": 2.573140058720198e-06, + "loss": 0.9217, + "step": 5659 + }, + { + "epoch": 0.42537201262588303, + "grad_norm": 1.3631366117420984, + "learning_rate": 2.572673618057473e-06, + "loss": 0.9348, + "step": 5660 + }, + { + "epoch": 0.425447166691718, + "grad_norm": 4.145190676363176, + "learning_rate": 2.5722071434616426e-06, + "loss": 0.9234, + "step": 5661 + }, + { + "epoch": 0.425522320757553, + "grad_norm": 1.631018131775119, + "learning_rate": 2.5717406349603483e-06, + "loss": 0.9971, + "step": 5662 + }, + { + "epoch": 0.42559747482338794, + "grad_norm": 1.605290015450998, + "learning_rate": 2.5712740925812314e-06, + "loss": 1.0057, + "step": 5663 + }, + { + "epoch": 0.4256726288892229, + "grad_norm": 0.7195967372123748, + "learning_rate": 2.5708075163519373e-06, + "loss": 0.8608, + "step": 5664 + }, + { + "epoch": 0.4257477829550579, + "grad_norm": 1.5777331328381237, + "learning_rate": 2.5703409063001124e-06, + "loss": 0.9414, + "step": 5665 + }, + { + "epoch": 0.42582293702089286, + "grad_norm": 1.4938269591358104, + "learning_rate": 2.5698742624534046e-06, + "loss": 0.9606, + "step": 5666 + }, + { + "epoch": 0.42589809108672777, + "grad_norm": 1.3889580497476302, + "learning_rate": 2.5694075848394646e-06, + "loss": 1.0271, + "step": 5667 + }, + { + "epoch": 0.42597324515256274, + "grad_norm": 2.2689270160650246, + "learning_rate": 2.5689408734859445e-06, + "loss": 0.9594, + "step": 5668 + }, + { + "epoch": 0.4260483992183977, + "grad_norm": 1.715657729651372, + "learning_rate": 2.568474128420499e-06, + "loss": 0.8414, + "step": 5669 + }, + { + "epoch": 0.4261235532842327, + "grad_norm": 2.226161023803118, + "learning_rate": 2.5680073496707854e-06, + "loss": 0.918, + "step": 5670 + }, + { + "epoch": 0.42619870735006765, + "grad_norm": 2.8385164498891853, + "learning_rate": 2.5675405372644606e-06, + "loss": 0.9883, + "step": 5671 + }, + { + "epoch": 0.4262738614159026, + "grad_norm": 1.8157622477514772, + "learning_rate": 2.567073691229186e-06, + "loss": 0.8421, + "step": 5672 + }, + { + "epoch": 0.42634901548173754, + "grad_norm": 1.608449429026248, + "learning_rate": 2.5666068115926223e-06, + "loss": 0.9866, + "step": 5673 + }, + { + "epoch": 0.4264241695475725, + "grad_norm": 3.314848013683403, + "learning_rate": 2.5661398983824375e-06, + "loss": 0.9653, + "step": 5674 + }, + { + "epoch": 0.4264993236134075, + "grad_norm": 1.4951189416118875, + "learning_rate": 2.565672951626295e-06, + "loss": 0.966, + "step": 5675 + }, + { + "epoch": 0.42657447767924245, + "grad_norm": 1.4548588734773693, + "learning_rate": 2.5652059713518636e-06, + "loss": 0.9847, + "step": 5676 + }, + { + "epoch": 0.4266496317450774, + "grad_norm": 2.018558821468614, + "learning_rate": 2.5647389575868142e-06, + "loss": 0.891, + "step": 5677 + }, + { + "epoch": 0.4267247858109124, + "grad_norm": 2.0211511604521855, + "learning_rate": 2.564271910358819e-06, + "loss": 0.9961, + "step": 5678 + }, + { + "epoch": 0.42679993987674736, + "grad_norm": 1.6878811441131842, + "learning_rate": 2.563804829695553e-06, + "loss": 0.97, + "step": 5679 + }, + { + "epoch": 0.4268750939425823, + "grad_norm": 1.6089756294192474, + "learning_rate": 2.5633377156246917e-06, + "loss": 1.0069, + "step": 5680 + }, + { + "epoch": 0.42695024800841724, + "grad_norm": 0.6959529456012076, + "learning_rate": 2.5628705681739124e-06, + "loss": 0.8173, + "step": 5681 + }, + { + "epoch": 0.4270254020742522, + "grad_norm": 2.426212978797935, + "learning_rate": 2.5624033873708983e-06, + "loss": 0.9677, + "step": 5682 + }, + { + "epoch": 0.4271005561400872, + "grad_norm": 1.453733173381222, + "learning_rate": 2.5619361732433287e-06, + "loss": 1.0902, + "step": 5683 + }, + { + "epoch": 0.42717571020592215, + "grad_norm": 1.5383420700673776, + "learning_rate": 2.5614689258188896e-06, + "loss": 0.9874, + "step": 5684 + }, + { + "epoch": 0.4272508642717571, + "grad_norm": 1.2670750729208906, + "learning_rate": 2.561001645125266e-06, + "loss": 0.9604, + "step": 5685 + }, + { + "epoch": 0.42732601833759204, + "grad_norm": 1.5825260095131208, + "learning_rate": 2.560534331190148e-06, + "loss": 1.071, + "step": 5686 + }, + { + "epoch": 0.427401172403427, + "grad_norm": 3.043242181483211, + "learning_rate": 2.5600669840412233e-06, + "loss": 1.0031, + "step": 5687 + }, + { + "epoch": 0.427476326469262, + "grad_norm": 0.7087159712761485, + "learning_rate": 2.5595996037061853e-06, + "loss": 0.833, + "step": 5688 + }, + { + "epoch": 0.42755148053509695, + "grad_norm": 1.9919267072331064, + "learning_rate": 2.559132190212728e-06, + "loss": 1.0855, + "step": 5689 + }, + { + "epoch": 0.4276266346009319, + "grad_norm": 2.667161319472325, + "learning_rate": 2.558664743588547e-06, + "loss": 1.0071, + "step": 5690 + }, + { + "epoch": 0.4277017886667669, + "grad_norm": 1.787822043478928, + "learning_rate": 2.5581972638613417e-06, + "loss": 1.0306, + "step": 5691 + }, + { + "epoch": 0.42777694273260186, + "grad_norm": 1.1277482056042165, + "learning_rate": 2.557729751058811e-06, + "loss": 0.8686, + "step": 5692 + }, + { + "epoch": 0.4278520967984368, + "grad_norm": 1.7036461688929334, + "learning_rate": 2.557262205208656e-06, + "loss": 0.9496, + "step": 5693 + }, + { + "epoch": 0.42792725086427175, + "grad_norm": 2.2412775299883467, + "learning_rate": 2.556794626338582e-06, + "loss": 0.9716, + "step": 5694 + }, + { + "epoch": 0.4280024049301067, + "grad_norm": 2.021112232691448, + "learning_rate": 2.5563270144762933e-06, + "loss": 1.1037, + "step": 5695 + }, + { + "epoch": 0.4280775589959417, + "grad_norm": 1.6622690198099617, + "learning_rate": 2.5558593696495e-06, + "loss": 1.0151, + "step": 5696 + }, + { + "epoch": 0.42815271306177666, + "grad_norm": 3.4584520771522187, + "learning_rate": 2.5553916918859102e-06, + "loss": 1.0297, + "step": 5697 + }, + { + "epoch": 0.4282278671276116, + "grad_norm": 1.5545141228849937, + "learning_rate": 2.554923981213235e-06, + "loss": 0.9176, + "step": 5698 + }, + { + "epoch": 0.42830302119344654, + "grad_norm": 1.9545281214883041, + "learning_rate": 2.55445623765919e-06, + "loss": 1.1379, + "step": 5699 + }, + { + "epoch": 0.4283781752592815, + "grad_norm": 3.070920306509217, + "learning_rate": 2.553988461251489e-06, + "loss": 0.9457, + "step": 5700 + }, + { + "epoch": 0.4284533293251165, + "grad_norm": 0.8320635937059107, + "learning_rate": 2.553520652017851e-06, + "loss": 0.856, + "step": 5701 + }, + { + "epoch": 0.42852848339095145, + "grad_norm": 1.9026657552607777, + "learning_rate": 2.5530528099859946e-06, + "loss": 0.9636, + "step": 5702 + }, + { + "epoch": 0.4286036374567864, + "grad_norm": 1.4551591479382064, + "learning_rate": 2.5525849351836414e-06, + "loss": 1.0075, + "step": 5703 + }, + { + "epoch": 0.4286787915226214, + "grad_norm": 1.8770600763786098, + "learning_rate": 2.5521170276385147e-06, + "loss": 0.9317, + "step": 5704 + }, + { + "epoch": 0.4287539455884563, + "grad_norm": 0.7513934542113181, + "learning_rate": 2.5516490873783397e-06, + "loss": 0.8142, + "step": 5705 + }, + { + "epoch": 0.4288290996542913, + "grad_norm": 1.7568214165690195, + "learning_rate": 2.5511811144308447e-06, + "loss": 0.9569, + "step": 5706 + }, + { + "epoch": 0.42890425372012625, + "grad_norm": 1.7086585208318124, + "learning_rate": 2.550713108823757e-06, + "loss": 0.9868, + "step": 5707 + }, + { + "epoch": 0.4289794077859612, + "grad_norm": 1.8794262457946995, + "learning_rate": 2.5502450705848097e-06, + "loss": 0.9433, + "step": 5708 + }, + { + "epoch": 0.4290545618517962, + "grad_norm": 1.8853404063052102, + "learning_rate": 2.5497769997417347e-06, + "loss": 1.0085, + "step": 5709 + }, + { + "epoch": 0.42912971591763116, + "grad_norm": 1.6739236276950005, + "learning_rate": 2.5493088963222668e-06, + "loss": 0.97, + "step": 5710 + }, + { + "epoch": 0.42920486998346613, + "grad_norm": 0.7443522955749337, + "learning_rate": 2.5488407603541437e-06, + "loss": 0.8676, + "step": 5711 + }, + { + "epoch": 0.42928002404930105, + "grad_norm": 2.468385017552803, + "learning_rate": 2.5483725918651034e-06, + "loss": 1.0397, + "step": 5712 + }, + { + "epoch": 0.429355178115136, + "grad_norm": 1.8709621184044958, + "learning_rate": 2.5479043908828877e-06, + "loss": 1.0317, + "step": 5713 + }, + { + "epoch": 0.429430332180971, + "grad_norm": 1.5605896524956722, + "learning_rate": 2.547436157435239e-06, + "loss": 1.043, + "step": 5714 + }, + { + "epoch": 0.42950548624680596, + "grad_norm": 1.7919344443183112, + "learning_rate": 2.546967891549901e-06, + "loss": 0.9538, + "step": 5715 + }, + { + "epoch": 0.4295806403126409, + "grad_norm": 1.4698550348749917, + "learning_rate": 2.5464995932546217e-06, + "loss": 0.9535, + "step": 5716 + }, + { + "epoch": 0.4296557943784759, + "grad_norm": 2.002644578569785, + "learning_rate": 2.5460312625771475e-06, + "loss": 0.9731, + "step": 5717 + }, + { + "epoch": 0.4297309484443108, + "grad_norm": 2.6152163242118838, + "learning_rate": 2.5455628995452313e-06, + "loss": 1.0171, + "step": 5718 + }, + { + "epoch": 0.4298061025101458, + "grad_norm": 1.8396019908380818, + "learning_rate": 2.5450945041866246e-06, + "loss": 0.99, + "step": 5719 + }, + { + "epoch": 0.42988125657598075, + "grad_norm": 1.517009849637066, + "learning_rate": 2.54462607652908e-06, + "loss": 0.9952, + "step": 5720 + }, + { + "epoch": 0.4299564106418157, + "grad_norm": 1.5037018079637863, + "learning_rate": 2.5441576166003555e-06, + "loss": 1.0515, + "step": 5721 + }, + { + "epoch": 0.4300315647076507, + "grad_norm": 1.8047094302475708, + "learning_rate": 2.5436891244282076e-06, + "loss": 1.0394, + "step": 5722 + }, + { + "epoch": 0.43010671877348566, + "grad_norm": 1.7269118978208133, + "learning_rate": 2.5432206000403982e-06, + "loss": 1.0005, + "step": 5723 + }, + { + "epoch": 0.43018187283932063, + "grad_norm": 2.4617505864754468, + "learning_rate": 2.5427520434646884e-06, + "loss": 0.9672, + "step": 5724 + }, + { + "epoch": 0.43025702690515555, + "grad_norm": 1.6612802806703673, + "learning_rate": 2.5422834547288406e-06, + "loss": 1.0008, + "step": 5725 + }, + { + "epoch": 0.4303321809709905, + "grad_norm": 3.0055688986616245, + "learning_rate": 2.5418148338606226e-06, + "loss": 0.9977, + "step": 5726 + }, + { + "epoch": 0.4304073350368255, + "grad_norm": 1.8622055947515632, + "learning_rate": 2.5413461808878e-06, + "loss": 0.9458, + "step": 5727 + }, + { + "epoch": 0.43048248910266046, + "grad_norm": 1.3935999391796654, + "learning_rate": 2.5408774958381436e-06, + "loss": 0.9808, + "step": 5728 + }, + { + "epoch": 0.43055764316849543, + "grad_norm": 1.4471587065664107, + "learning_rate": 2.5404087787394248e-06, + "loss": 0.9303, + "step": 5729 + }, + { + "epoch": 0.4306327972343304, + "grad_norm": 1.8750516884354973, + "learning_rate": 2.5399400296194164e-06, + "loss": 0.9831, + "step": 5730 + }, + { + "epoch": 0.4307079513001653, + "grad_norm": 8.301405307423858, + "learning_rate": 2.5394712485058933e-06, + "loss": 0.9174, + "step": 5731 + }, + { + "epoch": 0.4307831053660003, + "grad_norm": 1.4949065196432598, + "learning_rate": 2.539002435426633e-06, + "loss": 0.9725, + "step": 5732 + }, + { + "epoch": 0.43085825943183526, + "grad_norm": 12.295900260168338, + "learning_rate": 2.5385335904094147e-06, + "loss": 1.0059, + "step": 5733 + }, + { + "epoch": 0.4309334134976702, + "grad_norm": 3.3782457356575293, + "learning_rate": 2.5380647134820186e-06, + "loss": 0.9718, + "step": 5734 + }, + { + "epoch": 0.4310085675635052, + "grad_norm": 1.6943092717867434, + "learning_rate": 2.5375958046722283e-06, + "loss": 1.0185, + "step": 5735 + }, + { + "epoch": 0.43108372162934017, + "grad_norm": 1.6476340844520572, + "learning_rate": 2.5371268640078277e-06, + "loss": 1.0989, + "step": 5736 + }, + { + "epoch": 0.43115887569517514, + "grad_norm": 1.7905957843733875, + "learning_rate": 2.5366578915166033e-06, + "loss": 0.9969, + "step": 5737 + }, + { + "epoch": 0.43123402976101005, + "grad_norm": 1.7811485395480304, + "learning_rate": 2.536188887226345e-06, + "loss": 0.9704, + "step": 5738 + }, + { + "epoch": 0.431309183826845, + "grad_norm": 2.036050845172688, + "learning_rate": 2.53571985116484e-06, + "loss": 0.9195, + "step": 5739 + }, + { + "epoch": 0.43138433789268, + "grad_norm": 2.119033177899607, + "learning_rate": 2.535250783359884e-06, + "loss": 1.0061, + "step": 5740 + }, + { + "epoch": 0.43145949195851496, + "grad_norm": 2.2355654019673388, + "learning_rate": 2.5347816838392695e-06, + "loss": 0.9384, + "step": 5741 + }, + { + "epoch": 0.43153464602434993, + "grad_norm": 1.7598513837403382, + "learning_rate": 2.534312552630791e-06, + "loss": 0.9839, + "step": 5742 + }, + { + "epoch": 0.4316098000901849, + "grad_norm": 1.8666748236863977, + "learning_rate": 2.533843389762249e-06, + "loss": 1.0927, + "step": 5743 + }, + { + "epoch": 0.4316849541560198, + "grad_norm": 1.9308697248949735, + "learning_rate": 2.5333741952614412e-06, + "loss": 0.997, + "step": 5744 + }, + { + "epoch": 0.4317601082218548, + "grad_norm": 2.244591889191315, + "learning_rate": 2.5329049691561705e-06, + "loss": 0.9746, + "step": 5745 + }, + { + "epoch": 0.43183526228768976, + "grad_norm": 1.5585584134563317, + "learning_rate": 2.53243571147424e-06, + "loss": 1.037, + "step": 5746 + }, + { + "epoch": 0.43191041635352473, + "grad_norm": 1.5678304918247958, + "learning_rate": 2.5319664222434534e-06, + "loss": 0.9927, + "step": 5747 + }, + { + "epoch": 0.4319855704193597, + "grad_norm": 1.3312305269921467, + "learning_rate": 2.5314971014916207e-06, + "loss": 0.9834, + "step": 5748 + }, + { + "epoch": 0.43206072448519467, + "grad_norm": 1.9671695359964938, + "learning_rate": 2.5310277492465486e-06, + "loss": 1.04, + "step": 5749 + }, + { + "epoch": 0.4321358785510296, + "grad_norm": 1.6827271392906147, + "learning_rate": 2.5305583655360495e-06, + "loss": 1.0464, + "step": 5750 + }, + { + "epoch": 0.43221103261686455, + "grad_norm": 2.1228714548474557, + "learning_rate": 2.530088950387935e-06, + "loss": 0.986, + "step": 5751 + }, + { + "epoch": 0.4322861866826995, + "grad_norm": 1.5390489451784526, + "learning_rate": 2.529619503830021e-06, + "loss": 0.9977, + "step": 5752 + }, + { + "epoch": 0.4323613407485345, + "grad_norm": 2.1942832407376844, + "learning_rate": 2.5291500258901234e-06, + "loss": 1.0117, + "step": 5753 + }, + { + "epoch": 0.43243649481436947, + "grad_norm": 1.520635914291053, + "learning_rate": 2.5286805165960597e-06, + "loss": 0.9774, + "step": 5754 + }, + { + "epoch": 0.43251164888020444, + "grad_norm": 1.5800303674749243, + "learning_rate": 2.528210975975652e-06, + "loss": 1.0318, + "step": 5755 + }, + { + "epoch": 0.4325868029460394, + "grad_norm": 2.728909419702523, + "learning_rate": 2.52774140405672e-06, + "loss": 1.0047, + "step": 5756 + }, + { + "epoch": 0.4326619570118743, + "grad_norm": 6.851444743589697, + "learning_rate": 2.5272718008670895e-06, + "loss": 1.0281, + "step": 5757 + }, + { + "epoch": 0.4327371110777093, + "grad_norm": 2.341400189243908, + "learning_rate": 2.5268021664345865e-06, + "loss": 0.9996, + "step": 5758 + }, + { + "epoch": 0.43281226514354426, + "grad_norm": 0.7614906556400851, + "learning_rate": 2.526332500787037e-06, + "loss": 0.8526, + "step": 5759 + }, + { + "epoch": 0.43288741920937923, + "grad_norm": 1.6669392680711257, + "learning_rate": 2.525862803952272e-06, + "loss": 1.0251, + "step": 5760 + }, + { + "epoch": 0.4329625732752142, + "grad_norm": 1.6181684826621887, + "learning_rate": 2.5253930759581213e-06, + "loss": 1.0982, + "step": 5761 + }, + { + "epoch": 0.4330377273410492, + "grad_norm": 1.8300212578502906, + "learning_rate": 2.5249233168324196e-06, + "loss": 0.986, + "step": 5762 + }, + { + "epoch": 0.4331128814068841, + "grad_norm": 6.61003697464336, + "learning_rate": 2.5244535266030014e-06, + "loss": 1.0647, + "step": 5763 + }, + { + "epoch": 0.43318803547271906, + "grad_norm": 1.6042157037679496, + "learning_rate": 2.5239837052977032e-06, + "loss": 1.1299, + "step": 5764 + }, + { + "epoch": 0.433263189538554, + "grad_norm": 1.995241763445056, + "learning_rate": 2.523513852944364e-06, + "loss": 0.9733, + "step": 5765 + }, + { + "epoch": 0.433338343604389, + "grad_norm": 2.856071690308754, + "learning_rate": 2.5230439695708244e-06, + "loss": 0.8932, + "step": 5766 + }, + { + "epoch": 0.43341349767022397, + "grad_norm": 1.9521571088085927, + "learning_rate": 2.5225740552049267e-06, + "loss": 1.0191, + "step": 5767 + }, + { + "epoch": 0.43348865173605894, + "grad_norm": 2.0523613936613723, + "learning_rate": 2.5221041098745157e-06, + "loss": 1.0499, + "step": 5768 + }, + { + "epoch": 0.4335638058018939, + "grad_norm": 0.6995747688122127, + "learning_rate": 2.5216341336074363e-06, + "loss": 0.8527, + "step": 5769 + }, + { + "epoch": 0.4336389598677288, + "grad_norm": 1.3956941943305972, + "learning_rate": 2.5211641264315372e-06, + "loss": 0.9749, + "step": 5770 + }, + { + "epoch": 0.4337141139335638, + "grad_norm": 1.9613122467996806, + "learning_rate": 2.520694088374668e-06, + "loss": 1.0635, + "step": 5771 + }, + { + "epoch": 0.43378926799939876, + "grad_norm": 1.2670727679033642, + "learning_rate": 2.52022401946468e-06, + "loss": 1.0257, + "step": 5772 + }, + { + "epoch": 0.43386442206523373, + "grad_norm": 1.7170866807354064, + "learning_rate": 2.519753919729427e-06, + "loss": 1.0364, + "step": 5773 + }, + { + "epoch": 0.4339395761310687, + "grad_norm": 1.9682496661245654, + "learning_rate": 2.519283789196764e-06, + "loss": 1.0704, + "step": 5774 + }, + { + "epoch": 0.4340147301969037, + "grad_norm": 1.9723765562082662, + "learning_rate": 2.518813627894548e-06, + "loss": 1.0418, + "step": 5775 + }, + { + "epoch": 0.4340898842627386, + "grad_norm": 1.4909126468042992, + "learning_rate": 2.5183434358506373e-06, + "loss": 1.0409, + "step": 5776 + }, + { + "epoch": 0.43416503832857356, + "grad_norm": 0.744458188357885, + "learning_rate": 2.5178732130928943e-06, + "loss": 0.8722, + "step": 5777 + }, + { + "epoch": 0.43424019239440853, + "grad_norm": 7.61607964700036, + "learning_rate": 2.5174029596491792e-06, + "loss": 1.0276, + "step": 5778 + }, + { + "epoch": 0.4343153464602435, + "grad_norm": 2.140696392121827, + "learning_rate": 2.5169326755473582e-06, + "loss": 0.9527, + "step": 5779 + }, + { + "epoch": 0.43439050052607847, + "grad_norm": 1.8877486740913934, + "learning_rate": 2.516462360815297e-06, + "loss": 0.9542, + "step": 5780 + }, + { + "epoch": 0.43446565459191344, + "grad_norm": 1.597383807020643, + "learning_rate": 2.5159920154808615e-06, + "loss": 0.9887, + "step": 5781 + }, + { + "epoch": 0.4345408086577484, + "grad_norm": 4.286988137166699, + "learning_rate": 2.5155216395719253e-06, + "loss": 1.0024, + "step": 5782 + }, + { + "epoch": 0.4346159627235833, + "grad_norm": 0.8323855295038024, + "learning_rate": 2.5150512331163564e-06, + "loss": 0.9268, + "step": 5783 + }, + { + "epoch": 0.4346911167894183, + "grad_norm": 6.370504552001381, + "learning_rate": 2.5145807961420303e-06, + "loss": 0.8446, + "step": 5784 + }, + { + "epoch": 0.43476627085525327, + "grad_norm": 1.4846330117303916, + "learning_rate": 2.514110328676822e-06, + "loss": 1.0324, + "step": 5785 + }, + { + "epoch": 0.43484142492108824, + "grad_norm": 1.6844632410492568, + "learning_rate": 2.5136398307486075e-06, + "loss": 1.0083, + "step": 5786 + }, + { + "epoch": 0.4349165789869232, + "grad_norm": 2.074257485697925, + "learning_rate": 2.5131693023852663e-06, + "loss": 0.8446, + "step": 5787 + }, + { + "epoch": 0.4349917330527582, + "grad_norm": 1.893534478743213, + "learning_rate": 2.5126987436146786e-06, + "loss": 0.9893, + "step": 5788 + }, + { + "epoch": 0.4350668871185931, + "grad_norm": 1.4256881056368687, + "learning_rate": 2.5122281544647273e-06, + "loss": 0.9129, + "step": 5789 + }, + { + "epoch": 0.43514204118442806, + "grad_norm": 1.7409539752510004, + "learning_rate": 2.511757534963297e-06, + "loss": 0.9317, + "step": 5790 + }, + { + "epoch": 0.43521719525026303, + "grad_norm": 1.6345239458357586, + "learning_rate": 2.5112868851382724e-06, + "loss": 0.9639, + "step": 5791 + }, + { + "epoch": 0.435292349316098, + "grad_norm": 1.6046693765569569, + "learning_rate": 2.5108162050175425e-06, + "loss": 0.9895, + "step": 5792 + }, + { + "epoch": 0.435367503381933, + "grad_norm": 1.7021541102933655, + "learning_rate": 2.510345494628996e-06, + "loss": 1.052, + "step": 5793 + }, + { + "epoch": 0.43544265744776794, + "grad_norm": 1.349175301952817, + "learning_rate": 2.509874754000525e-06, + "loss": 1.0289, + "step": 5794 + }, + { + "epoch": 0.43551781151360286, + "grad_norm": 1.5496079933616147, + "learning_rate": 2.5094039831600217e-06, + "loss": 0.9479, + "step": 5795 + }, + { + "epoch": 0.43559296557943783, + "grad_norm": 1.9347401312864607, + "learning_rate": 2.5089331821353827e-06, + "loss": 0.9601, + "step": 5796 + }, + { + "epoch": 0.4356681196452728, + "grad_norm": 2.2287741936500693, + "learning_rate": 2.5084623509545034e-06, + "loss": 1.0625, + "step": 5797 + }, + { + "epoch": 0.43574327371110777, + "grad_norm": 1.7016719954528965, + "learning_rate": 2.5079914896452823e-06, + "loss": 0.9846, + "step": 5798 + }, + { + "epoch": 0.43581842777694274, + "grad_norm": 1.6376584798984943, + "learning_rate": 2.507520598235621e-06, + "loss": 1.0385, + "step": 5799 + }, + { + "epoch": 0.4358935818427777, + "grad_norm": 7.170995188749628, + "learning_rate": 2.5070496767534202e-06, + "loss": 0.9746, + "step": 5800 + }, + { + "epoch": 0.4359687359086127, + "grad_norm": 1.616298928991632, + "learning_rate": 2.5065787252265848e-06, + "loss": 1.0226, + "step": 5801 + }, + { + "epoch": 0.4360438899744476, + "grad_norm": 1.8326238718879546, + "learning_rate": 2.50610774368302e-06, + "loss": 0.9778, + "step": 5802 + }, + { + "epoch": 0.43611904404028257, + "grad_norm": 1.7392548652636806, + "learning_rate": 2.505636732150633e-06, + "loss": 1.0152, + "step": 5803 + }, + { + "epoch": 0.43619419810611754, + "grad_norm": 1.7379996425246387, + "learning_rate": 2.505165690657334e-06, + "loss": 1.0249, + "step": 5804 + }, + { + "epoch": 0.4362693521719525, + "grad_norm": 1.7767595945082852, + "learning_rate": 2.504694619231033e-06, + "loss": 1.0386, + "step": 5805 + }, + { + "epoch": 0.4363445062377875, + "grad_norm": 1.99857899729354, + "learning_rate": 2.5042235178996436e-06, + "loss": 1.0698, + "step": 5806 + }, + { + "epoch": 0.43641966030362245, + "grad_norm": 1.9345363592781766, + "learning_rate": 2.5037523866910797e-06, + "loss": 0.9669, + "step": 5807 + }, + { + "epoch": 0.43649481436945736, + "grad_norm": 2.7971142714462127, + "learning_rate": 2.503281225633258e-06, + "loss": 0.9994, + "step": 5808 + }, + { + "epoch": 0.43656996843529233, + "grad_norm": 1.5135753970009609, + "learning_rate": 2.5028100347540967e-06, + "loss": 0.9872, + "step": 5809 + }, + { + "epoch": 0.4366451225011273, + "grad_norm": 2.493625142502048, + "learning_rate": 2.5023388140815148e-06, + "loss": 0.9832, + "step": 5810 + }, + { + "epoch": 0.4367202765669623, + "grad_norm": 1.8379466000666764, + "learning_rate": 2.5018675636434353e-06, + "loss": 1.0626, + "step": 5811 + }, + { + "epoch": 0.43679543063279724, + "grad_norm": 2.3451820576155313, + "learning_rate": 2.5013962834677804e-06, + "loss": 1.0507, + "step": 5812 + }, + { + "epoch": 0.4368705846986322, + "grad_norm": 0.7381723717226236, + "learning_rate": 2.5009249735824757e-06, + "loss": 0.8817, + "step": 5813 + }, + { + "epoch": 0.4369457387644672, + "grad_norm": 2.052404840008135, + "learning_rate": 2.500453634015449e-06, + "loss": 1.0083, + "step": 5814 + }, + { + "epoch": 0.4370208928303021, + "grad_norm": 2.8613249586693232, + "learning_rate": 2.4999822647946273e-06, + "loss": 1.025, + "step": 5815 + }, + { + "epoch": 0.43709604689613707, + "grad_norm": 2.241553452857871, + "learning_rate": 2.499510865947942e-06, + "loss": 0.9688, + "step": 5816 + }, + { + "epoch": 0.43717120096197204, + "grad_norm": 1.878802322400229, + "learning_rate": 2.4990394375033247e-06, + "loss": 1.0146, + "step": 5817 + }, + { + "epoch": 0.437246355027807, + "grad_norm": 1.869820625555545, + "learning_rate": 2.4985679794887106e-06, + "loss": 0.9631, + "step": 5818 + }, + { + "epoch": 0.437321509093642, + "grad_norm": 2.069096622067567, + "learning_rate": 2.4980964919320343e-06, + "loss": 1.0049, + "step": 5819 + }, + { + "epoch": 0.43739666315947695, + "grad_norm": 2.2803571861293546, + "learning_rate": 2.4976249748612332e-06, + "loss": 1.0248, + "step": 5820 + }, + { + "epoch": 0.43747181722531187, + "grad_norm": 1.450387735155896, + "learning_rate": 2.497153428304247e-06, + "loss": 0.9573, + "step": 5821 + }, + { + "epoch": 0.43754697129114684, + "grad_norm": 1.8082424012041227, + "learning_rate": 2.496681852289016e-06, + "loss": 1.0227, + "step": 5822 + }, + { + "epoch": 0.4376221253569818, + "grad_norm": 1.9814895316614374, + "learning_rate": 2.4962102468434843e-06, + "loss": 0.927, + "step": 5823 + }, + { + "epoch": 0.4376972794228168, + "grad_norm": 3.2965340822235145, + "learning_rate": 2.4957386119955954e-06, + "loss": 1.0684, + "step": 5824 + }, + { + "epoch": 0.43777243348865175, + "grad_norm": 1.568114531182944, + "learning_rate": 2.4952669477732938e-06, + "loss": 1.0008, + "step": 5825 + }, + { + "epoch": 0.4378475875544867, + "grad_norm": 2.332589189760176, + "learning_rate": 2.4947952542045307e-06, + "loss": 1.0819, + "step": 5826 + }, + { + "epoch": 0.43792274162032163, + "grad_norm": 1.78291417065177, + "learning_rate": 2.494323531317253e-06, + "loss": 1.0219, + "step": 5827 + }, + { + "epoch": 0.4379978956861566, + "grad_norm": 1.7245676245123336, + "learning_rate": 2.493851779139414e-06, + "loss": 0.9849, + "step": 5828 + }, + { + "epoch": 0.4380730497519916, + "grad_norm": 1.4748244779216542, + "learning_rate": 2.493379997698966e-06, + "loss": 0.9624, + "step": 5829 + }, + { + "epoch": 0.43814820381782654, + "grad_norm": 3.8928193181751345, + "learning_rate": 2.4929081870238635e-06, + "loss": 0.9746, + "step": 5830 + }, + { + "epoch": 0.4382233578836615, + "grad_norm": 0.5822372583982741, + "learning_rate": 2.4924363471420634e-06, + "loss": 0.7729, + "step": 5831 + }, + { + "epoch": 0.4382985119494965, + "grad_norm": 1.6374576330164106, + "learning_rate": 2.491964478081524e-06, + "loss": 1.0453, + "step": 5832 + }, + { + "epoch": 0.43837366601533145, + "grad_norm": 1.947670116903102, + "learning_rate": 2.4914925798702057e-06, + "loss": 0.9843, + "step": 5833 + }, + { + "epoch": 0.43844882008116637, + "grad_norm": 1.0973954620626238, + "learning_rate": 2.49102065253607e-06, + "loss": 0.8058, + "step": 5834 + }, + { + "epoch": 0.43852397414700134, + "grad_norm": 1.94704406151193, + "learning_rate": 2.49054869610708e-06, + "loss": 0.785, + "step": 5835 + }, + { + "epoch": 0.4385991282128363, + "grad_norm": 3.187293102057043, + "learning_rate": 2.490076710611202e-06, + "loss": 1.0259, + "step": 5836 + }, + { + "epoch": 0.4386742822786713, + "grad_norm": 1.873586885097545, + "learning_rate": 2.4896046960764015e-06, + "loss": 0.8858, + "step": 5837 + }, + { + "epoch": 0.43874943634450625, + "grad_norm": 4.137734840389727, + "learning_rate": 2.4891326525306487e-06, + "loss": 0.8658, + "step": 5838 + }, + { + "epoch": 0.4388245904103412, + "grad_norm": 1.719174835412494, + "learning_rate": 2.4886605800019123e-06, + "loss": 1.0412, + "step": 5839 + }, + { + "epoch": 0.43889974447617613, + "grad_norm": 1.78667493865722, + "learning_rate": 2.488188478518166e-06, + "loss": 1.0086, + "step": 5840 + }, + { + "epoch": 0.4389748985420111, + "grad_norm": 2.39781822137288, + "learning_rate": 2.487716348107383e-06, + "loss": 0.9615, + "step": 5841 + }, + { + "epoch": 0.4390500526078461, + "grad_norm": 1.5865410469051895, + "learning_rate": 2.4872441887975386e-06, + "loss": 0.9579, + "step": 5842 + }, + { + "epoch": 0.43912520667368105, + "grad_norm": 3.092807048088604, + "learning_rate": 2.48677200061661e-06, + "loss": 0.9869, + "step": 5843 + }, + { + "epoch": 0.439200360739516, + "grad_norm": 0.6789222662718871, + "learning_rate": 2.486299783592576e-06, + "loss": 0.8748, + "step": 5844 + }, + { + "epoch": 0.439275514805351, + "grad_norm": 2.2751147188878944, + "learning_rate": 2.485827537753419e-06, + "loss": 0.9941, + "step": 5845 + }, + { + "epoch": 0.43935066887118596, + "grad_norm": 2.271162959915189, + "learning_rate": 2.4853552631271193e-06, + "loss": 1.0073, + "step": 5846 + }, + { + "epoch": 0.43942582293702087, + "grad_norm": 4.284693314784363, + "learning_rate": 2.4848829597416615e-06, + "loss": 0.9572, + "step": 5847 + }, + { + "epoch": 0.43950097700285584, + "grad_norm": 2.855238460329406, + "learning_rate": 2.484410627625032e-06, + "loss": 1.0385, + "step": 5848 + }, + { + "epoch": 0.4395761310686908, + "grad_norm": 1.1703986468523069, + "learning_rate": 2.483938266805217e-06, + "loss": 0.9355, + "step": 5849 + }, + { + "epoch": 0.4396512851345258, + "grad_norm": 2.1095304078674784, + "learning_rate": 2.483465877310208e-06, + "loss": 0.9483, + "step": 5850 + }, + { + "epoch": 0.43972643920036075, + "grad_norm": 1.6412660345498509, + "learning_rate": 2.482993459167993e-06, + "loss": 1.0158, + "step": 5851 + }, + { + "epoch": 0.4398015932661957, + "grad_norm": 0.7243701303366097, + "learning_rate": 2.482521012406567e-06, + "loss": 0.8394, + "step": 5852 + }, + { + "epoch": 0.43987674733203064, + "grad_norm": 1.4619282509644642, + "learning_rate": 2.4820485370539233e-06, + "loss": 1.0195, + "step": 5853 + }, + { + "epoch": 0.4399519013978656, + "grad_norm": 1.5185947228946781, + "learning_rate": 2.481576033138057e-06, + "loss": 1.0359, + "step": 5854 + }, + { + "epoch": 0.4400270554637006, + "grad_norm": 1.778350746024251, + "learning_rate": 2.4811035006869677e-06, + "loss": 1.0496, + "step": 5855 + }, + { + "epoch": 0.44010220952953555, + "grad_norm": 0.8481115006431603, + "learning_rate": 2.4806309397286534e-06, + "loss": 0.8764, + "step": 5856 + }, + { + "epoch": 0.4401773635953705, + "grad_norm": 1.3764499042372798, + "learning_rate": 2.4801583502911154e-06, + "loss": 1.0338, + "step": 5857 + }, + { + "epoch": 0.4402525176612055, + "grad_norm": 1.8333081185889966, + "learning_rate": 2.4796857324023564e-06, + "loss": 0.8654, + "step": 5858 + }, + { + "epoch": 0.44032767172704046, + "grad_norm": 1.6900745109068394, + "learning_rate": 2.479213086090381e-06, + "loss": 1.0017, + "step": 5859 + }, + { + "epoch": 0.4404028257928754, + "grad_norm": 1.7410373740306686, + "learning_rate": 2.478740411383195e-06, + "loss": 1.0298, + "step": 5860 + }, + { + "epoch": 0.44047797985871034, + "grad_norm": 1.581834267491633, + "learning_rate": 2.478267708308807e-06, + "loss": 0.9067, + "step": 5861 + }, + { + "epoch": 0.4405531339245453, + "grad_norm": 1.597994369878865, + "learning_rate": 2.4777949768952255e-06, + "loss": 1.0125, + "step": 5862 + }, + { + "epoch": 0.4406282879903803, + "grad_norm": 2.6639561785739465, + "learning_rate": 2.477322217170462e-06, + "loss": 1.0274, + "step": 5863 + }, + { + "epoch": 0.44070344205621526, + "grad_norm": 1.7855330388866886, + "learning_rate": 2.476849429162529e-06, + "loss": 0.939, + "step": 5864 + }, + { + "epoch": 0.4407785961220502, + "grad_norm": 1.7869057127590402, + "learning_rate": 2.4763766128994423e-06, + "loss": 1.0378, + "step": 5865 + }, + { + "epoch": 0.44085375018788514, + "grad_norm": 1.864831307684914, + "learning_rate": 2.475903768409216e-06, + "loss": 0.9885, + "step": 5866 + }, + { + "epoch": 0.4409289042537201, + "grad_norm": 1.432842421557248, + "learning_rate": 2.47543089571987e-06, + "loss": 0.8944, + "step": 5867 + }, + { + "epoch": 0.4410040583195551, + "grad_norm": 2.0996352651020036, + "learning_rate": 2.4749579948594224e-06, + "loss": 0.8743, + "step": 5868 + }, + { + "epoch": 0.44107921238539005, + "grad_norm": 1.4728280936034237, + "learning_rate": 2.4744850658558943e-06, + "loss": 0.944, + "step": 5869 + }, + { + "epoch": 0.441154366451225, + "grad_norm": 1.8572564758401493, + "learning_rate": 2.47401210873731e-06, + "loss": 0.9216, + "step": 5870 + }, + { + "epoch": 0.44122952051706, + "grad_norm": 1.601880967645919, + "learning_rate": 2.473539123531693e-06, + "loss": 0.9702, + "step": 5871 + }, + { + "epoch": 0.4413046745828949, + "grad_norm": 2.979889379662751, + "learning_rate": 2.4730661102670692e-06, + "loss": 1.0805, + "step": 5872 + }, + { + "epoch": 0.4413798286487299, + "grad_norm": 1.7810253202383048, + "learning_rate": 2.4725930689714673e-06, + "loss": 0.9569, + "step": 5873 + }, + { + "epoch": 0.44145498271456485, + "grad_norm": 1.6024075395449584, + "learning_rate": 2.4721199996729167e-06, + "loss": 0.9711, + "step": 5874 + }, + { + "epoch": 0.4415301367803998, + "grad_norm": 1.8212046913056388, + "learning_rate": 2.471646902399448e-06, + "loss": 0.9401, + "step": 5875 + }, + { + "epoch": 0.4416052908462348, + "grad_norm": 1.5035703605982027, + "learning_rate": 2.471173777179094e-06, + "loss": 0.9892, + "step": 5876 + }, + { + "epoch": 0.44168044491206976, + "grad_norm": 2.7108865782602654, + "learning_rate": 2.4707006240398894e-06, + "loss": 0.9964, + "step": 5877 + }, + { + "epoch": 0.44175559897790473, + "grad_norm": 1.9614880765367277, + "learning_rate": 2.4702274430098707e-06, + "loss": 0.9787, + "step": 5878 + }, + { + "epoch": 0.44183075304373964, + "grad_norm": 2.244929534383956, + "learning_rate": 2.469754234117075e-06, + "loss": 1.0082, + "step": 5879 + }, + { + "epoch": 0.4419059071095746, + "grad_norm": 1.5144115978546502, + "learning_rate": 2.4692809973895426e-06, + "loss": 0.9857, + "step": 5880 + }, + { + "epoch": 0.4419810611754096, + "grad_norm": 1.779552922841603, + "learning_rate": 2.4688077328553136e-06, + "loss": 0.9933, + "step": 5881 + }, + { + "epoch": 0.44205621524124455, + "grad_norm": 2.5504530597697954, + "learning_rate": 2.4683344405424316e-06, + "loss": 0.9486, + "step": 5882 + }, + { + "epoch": 0.4421313693070795, + "grad_norm": 2.368362185655701, + "learning_rate": 2.4678611204789405e-06, + "loss": 0.9772, + "step": 5883 + }, + { + "epoch": 0.4422065233729145, + "grad_norm": 3.555747877985974, + "learning_rate": 2.4673877726928865e-06, + "loss": 1.0551, + "step": 5884 + }, + { + "epoch": 0.4422816774387494, + "grad_norm": 2.039265706544198, + "learning_rate": 2.4669143972123178e-06, + "loss": 0.8297, + "step": 5885 + }, + { + "epoch": 0.4423568315045844, + "grad_norm": 1.5581002661206622, + "learning_rate": 2.4664409940652817e-06, + "loss": 0.9878, + "step": 5886 + }, + { + "epoch": 0.44243198557041935, + "grad_norm": 1.9768788080845112, + "learning_rate": 2.465967563279832e-06, + "loss": 1.0516, + "step": 5887 + }, + { + "epoch": 0.4425071396362543, + "grad_norm": 1.488575743526986, + "learning_rate": 2.4654941048840184e-06, + "loss": 1.0194, + "step": 5888 + }, + { + "epoch": 0.4425822937020893, + "grad_norm": 1.9553599273602094, + "learning_rate": 2.465020618905898e-06, + "loss": 1.0686, + "step": 5889 + }, + { + "epoch": 0.44265744776792426, + "grad_norm": 1.5553415734086016, + "learning_rate": 2.464547105373525e-06, + "loss": 1.0671, + "step": 5890 + }, + { + "epoch": 0.44273260183375923, + "grad_norm": 1.5608996015344987, + "learning_rate": 2.4640735643149566e-06, + "loss": 0.9743, + "step": 5891 + }, + { + "epoch": 0.44280775589959415, + "grad_norm": 0.7210242737005328, + "learning_rate": 2.4635999957582526e-06, + "loss": 0.8652, + "step": 5892 + }, + { + "epoch": 0.4428829099654291, + "grad_norm": 1.8610082553637377, + "learning_rate": 2.4631263997314734e-06, + "loss": 0.9862, + "step": 5893 + }, + { + "epoch": 0.4429580640312641, + "grad_norm": 1.5817032085387364, + "learning_rate": 2.4626527762626822e-06, + "loss": 1.0621, + "step": 5894 + }, + { + "epoch": 0.44303321809709906, + "grad_norm": 1.6604392135920851, + "learning_rate": 2.462179125379942e-06, + "loss": 0.9715, + "step": 5895 + }, + { + "epoch": 0.443108372162934, + "grad_norm": 2.030716928812236, + "learning_rate": 2.461705447111319e-06, + "loss": 0.9804, + "step": 5896 + }, + { + "epoch": 0.443183526228769, + "grad_norm": 1.9923293717674544, + "learning_rate": 2.4612317414848803e-06, + "loss": 1.0016, + "step": 5897 + }, + { + "epoch": 0.4432586802946039, + "grad_norm": 1.514088904246028, + "learning_rate": 2.460758008528694e-06, + "loss": 1.0352, + "step": 5898 + }, + { + "epoch": 0.4433338343604389, + "grad_norm": 1.2604740014184954, + "learning_rate": 2.460284248270833e-06, + "loss": 0.9624, + "step": 5899 + }, + { + "epoch": 0.44340898842627385, + "grad_norm": 31.266173549035987, + "learning_rate": 2.4598104607393666e-06, + "loss": 1.0173, + "step": 5900 + }, + { + "epoch": 0.4434841424921088, + "grad_norm": 2.129984003642733, + "learning_rate": 2.4593366459623698e-06, + "loss": 1.0653, + "step": 5901 + }, + { + "epoch": 0.4435592965579438, + "grad_norm": 1.5899088328393935, + "learning_rate": 2.458862803967918e-06, + "loss": 0.8974, + "step": 5902 + }, + { + "epoch": 0.44363445062377876, + "grad_norm": 0.6774778072832669, + "learning_rate": 2.4583889347840873e-06, + "loss": 0.8814, + "step": 5903 + }, + { + "epoch": 0.44370960468961373, + "grad_norm": 2.5279239418986026, + "learning_rate": 2.4579150384389574e-06, + "loss": 0.9935, + "step": 5904 + }, + { + "epoch": 0.44378475875544865, + "grad_norm": 1.843062450425105, + "learning_rate": 2.4574411149606076e-06, + "loss": 0.9195, + "step": 5905 + }, + { + "epoch": 0.4438599128212836, + "grad_norm": 1.5558929370781849, + "learning_rate": 2.456967164377121e-06, + "loss": 1.076, + "step": 5906 + }, + { + "epoch": 0.4439350668871186, + "grad_norm": 1.7769116893070598, + "learning_rate": 2.4564931867165795e-06, + "loss": 0.9341, + "step": 5907 + }, + { + "epoch": 0.44401022095295356, + "grad_norm": 1.617674970641021, + "learning_rate": 2.4560191820070683e-06, + "loss": 0.9716, + "step": 5908 + }, + { + "epoch": 0.44408537501878853, + "grad_norm": 1.4914090987308686, + "learning_rate": 2.4555451502766754e-06, + "loss": 0.8637, + "step": 5909 + }, + { + "epoch": 0.4441605290846235, + "grad_norm": 0.6706602070606663, + "learning_rate": 2.4550710915534863e-06, + "loss": 0.8766, + "step": 5910 + }, + { + "epoch": 0.4442356831504584, + "grad_norm": 7.395509553367846, + "learning_rate": 2.4545970058655938e-06, + "loss": 1.0599, + "step": 5911 + }, + { + "epoch": 0.4443108372162934, + "grad_norm": 2.3294400590962026, + "learning_rate": 2.454122893241088e-06, + "loss": 0.9538, + "step": 5912 + }, + { + "epoch": 0.44438599128212836, + "grad_norm": 1.7422353284625594, + "learning_rate": 2.453648753708061e-06, + "loss": 0.972, + "step": 5913 + }, + { + "epoch": 0.4444611453479633, + "grad_norm": 1.3919992835574664, + "learning_rate": 2.4531745872946085e-06, + "loss": 0.9784, + "step": 5914 + }, + { + "epoch": 0.4445362994137983, + "grad_norm": 1.923557630393027, + "learning_rate": 2.4527003940288264e-06, + "loss": 1.0528, + "step": 5915 + }, + { + "epoch": 0.44461145347963327, + "grad_norm": 1.3669573780941382, + "learning_rate": 2.4522261739388127e-06, + "loss": 0.9538, + "step": 5916 + }, + { + "epoch": 0.4446866075454682, + "grad_norm": 1.5429544568908795, + "learning_rate": 2.451751927052666e-06, + "loss": 1.0163, + "step": 5917 + }, + { + "epoch": 0.44476176161130315, + "grad_norm": 1.5658546294584033, + "learning_rate": 2.4512776533984882e-06, + "loss": 0.8628, + "step": 5918 + }, + { + "epoch": 0.4448369156771381, + "grad_norm": 1.8365811132916137, + "learning_rate": 2.450803353004382e-06, + "loss": 0.921, + "step": 5919 + }, + { + "epoch": 0.4449120697429731, + "grad_norm": 1.4791969332038608, + "learning_rate": 2.4503290258984493e-06, + "loss": 1.052, + "step": 5920 + }, + { + "epoch": 0.44498722380880806, + "grad_norm": 1.9015112588665983, + "learning_rate": 2.4498546721087984e-06, + "loss": 1.0049, + "step": 5921 + }, + { + "epoch": 0.44506237787464303, + "grad_norm": 1.615682592785818, + "learning_rate": 2.4493802916635355e-06, + "loss": 1.0431, + "step": 5922 + }, + { + "epoch": 0.445137531940478, + "grad_norm": 1.5324118741010806, + "learning_rate": 2.448905884590769e-06, + "loss": 1.0554, + "step": 5923 + }, + { + "epoch": 0.4452126860063129, + "grad_norm": 1.4211634437957243, + "learning_rate": 2.448431450918611e-06, + "loss": 0.9826, + "step": 5924 + }, + { + "epoch": 0.4452878400721479, + "grad_norm": 2.010325243622047, + "learning_rate": 2.4479569906751714e-06, + "loss": 0.9045, + "step": 5925 + }, + { + "epoch": 0.44536299413798286, + "grad_norm": 1.9917642181107342, + "learning_rate": 2.4474825038885655e-06, + "loss": 0.8476, + "step": 5926 + }, + { + "epoch": 0.44543814820381783, + "grad_norm": 1.7595740414060277, + "learning_rate": 2.4470079905869066e-06, + "loss": 0.9289, + "step": 5927 + }, + { + "epoch": 0.4455133022696528, + "grad_norm": 6.108478294966606, + "learning_rate": 2.446533450798314e-06, + "loss": 0.9205, + "step": 5928 + }, + { + "epoch": 0.44558845633548777, + "grad_norm": 1.7832747880589135, + "learning_rate": 2.4460588845509036e-06, + "loss": 1.0023, + "step": 5929 + }, + { + "epoch": 0.4456636104013227, + "grad_norm": 2.7881995824310186, + "learning_rate": 2.4455842918727957e-06, + "loss": 0.9341, + "step": 5930 + }, + { + "epoch": 0.44573876446715766, + "grad_norm": 2.7202048081538313, + "learning_rate": 2.4451096727921135e-06, + "loss": 0.9828, + "step": 5931 + }, + { + "epoch": 0.4458139185329926, + "grad_norm": 3.850280300687193, + "learning_rate": 2.444635027336977e-06, + "loss": 1.051, + "step": 5932 + }, + { + "epoch": 0.4458890725988276, + "grad_norm": 1.664817531313008, + "learning_rate": 2.4441603555355142e-06, + "loss": 1.0005, + "step": 5933 + }, + { + "epoch": 0.44596422666466257, + "grad_norm": 2.177470621464852, + "learning_rate": 2.443685657415849e-06, + "loss": 1.042, + "step": 5934 + }, + { + "epoch": 0.44603938073049754, + "grad_norm": 2.5795934945347296, + "learning_rate": 2.4432109330061096e-06, + "loss": 0.9358, + "step": 5935 + }, + { + "epoch": 0.4461145347963325, + "grad_norm": 1.755934326473151, + "learning_rate": 2.4427361823344256e-06, + "loss": 1.0083, + "step": 5936 + }, + { + "epoch": 0.4461896888621674, + "grad_norm": 1.5148787086706454, + "learning_rate": 2.4422614054289264e-06, + "loss": 0.9237, + "step": 5937 + }, + { + "epoch": 0.4462648429280024, + "grad_norm": 1.4190936710014657, + "learning_rate": 2.4417866023177466e-06, + "loss": 0.9625, + "step": 5938 + }, + { + "epoch": 0.44633999699383736, + "grad_norm": 1.7082154380621497, + "learning_rate": 2.4413117730290186e-06, + "loss": 1.0408, + "step": 5939 + }, + { + "epoch": 0.44641515105967233, + "grad_norm": 2.190734869232913, + "learning_rate": 2.440836917590878e-06, + "loss": 1.0436, + "step": 5940 + }, + { + "epoch": 0.4464903051255073, + "grad_norm": 1.5203056029489703, + "learning_rate": 2.440362036031462e-06, + "loss": 0.9093, + "step": 5941 + }, + { + "epoch": 0.4465654591913423, + "grad_norm": 1.5992572848732673, + "learning_rate": 2.4398871283789088e-06, + "loss": 0.9897, + "step": 5942 + }, + { + "epoch": 0.4466406132571772, + "grad_norm": 2.0636383152113407, + "learning_rate": 2.439412194661359e-06, + "loss": 0.9623, + "step": 5943 + }, + { + "epoch": 0.44671576732301216, + "grad_norm": 5.161339300061693, + "learning_rate": 2.4389372349069544e-06, + "loss": 0.9006, + "step": 5944 + }, + { + "epoch": 0.44679092138884713, + "grad_norm": 1.8510105259955145, + "learning_rate": 2.4384622491438374e-06, + "loss": 0.9264, + "step": 5945 + }, + { + "epoch": 0.4468660754546821, + "grad_norm": 1.8507112888367234, + "learning_rate": 2.437987237400153e-06, + "loss": 0.9209, + "step": 5946 + }, + { + "epoch": 0.44694122952051707, + "grad_norm": 2.3439128564838505, + "learning_rate": 2.4375121997040477e-06, + "loss": 0.9255, + "step": 5947 + }, + { + "epoch": 0.44701638358635204, + "grad_norm": 1.7406880721713496, + "learning_rate": 2.4370371360836697e-06, + "loss": 1.0654, + "step": 5948 + }, + { + "epoch": 0.447091537652187, + "grad_norm": 1.9177436152632752, + "learning_rate": 2.436562046567167e-06, + "loss": 0.9921, + "step": 5949 + }, + { + "epoch": 0.4471666917180219, + "grad_norm": 1.8993671793900633, + "learning_rate": 2.4360869311826927e-06, + "loss": 0.9888, + "step": 5950 + }, + { + "epoch": 0.4472418457838569, + "grad_norm": 1.261439242272326, + "learning_rate": 2.435611789958397e-06, + "loss": 0.9817, + "step": 5951 + }, + { + "epoch": 0.44731699984969187, + "grad_norm": 1.5749821101792374, + "learning_rate": 2.435136622922434e-06, + "loss": 0.9701, + "step": 5952 + }, + { + "epoch": 0.44739215391552684, + "grad_norm": 1.6488090927952372, + "learning_rate": 2.4346614301029613e-06, + "loss": 1.0158, + "step": 5953 + }, + { + "epoch": 0.4474673079813618, + "grad_norm": 1.6776626091952302, + "learning_rate": 2.434186211528133e-06, + "loss": 1.095, + "step": 5954 + }, + { + "epoch": 0.4475424620471968, + "grad_norm": 2.3862990589404234, + "learning_rate": 2.4337109672261097e-06, + "loss": 1.0333, + "step": 5955 + }, + { + "epoch": 0.4476176161130317, + "grad_norm": 1.4662383498097296, + "learning_rate": 2.433235697225051e-06, + "loss": 0.866, + "step": 5956 + }, + { + "epoch": 0.44769277017886666, + "grad_norm": 1.639691768616692, + "learning_rate": 2.4327604015531177e-06, + "loss": 1.0752, + "step": 5957 + }, + { + "epoch": 0.44776792424470163, + "grad_norm": 1.6028223804296522, + "learning_rate": 2.432285080238474e-06, + "loss": 1.0152, + "step": 5958 + }, + { + "epoch": 0.4478430783105366, + "grad_norm": 0.6971529915073573, + "learning_rate": 2.4318097333092837e-06, + "loss": 0.8478, + "step": 5959 + }, + { + "epoch": 0.4479182323763716, + "grad_norm": 1.9719518026880458, + "learning_rate": 2.4313343607937135e-06, + "loss": 0.9774, + "step": 5960 + }, + { + "epoch": 0.44799338644220654, + "grad_norm": 2.523415391193299, + "learning_rate": 2.430858962719931e-06, + "loss": 0.8855, + "step": 5961 + }, + { + "epoch": 0.44806854050804146, + "grad_norm": 1.311442266865255, + "learning_rate": 2.4303835391161047e-06, + "loss": 1.0151, + "step": 5962 + }, + { + "epoch": 0.4481436945738764, + "grad_norm": 2.0652895471812878, + "learning_rate": 2.4299080900104055e-06, + "loss": 1.0475, + "step": 5963 + }, + { + "epoch": 0.4482188486397114, + "grad_norm": 1.9890469798451231, + "learning_rate": 2.4294326154310058e-06, + "loss": 0.9244, + "step": 5964 + }, + { + "epoch": 0.44829400270554637, + "grad_norm": 1.8943720485219993, + "learning_rate": 2.4289571154060794e-06, + "loss": 1.0746, + "step": 5965 + }, + { + "epoch": 0.44836915677138134, + "grad_norm": 0.6962217750512706, + "learning_rate": 2.4284815899638012e-06, + "loss": 0.879, + "step": 5966 + }, + { + "epoch": 0.4484443108372163, + "grad_norm": 1.7509595420167596, + "learning_rate": 2.428006039132348e-06, + "loss": 0.9295, + "step": 5967 + }, + { + "epoch": 0.4485194649030513, + "grad_norm": 1.3771010731964994, + "learning_rate": 2.4275304629398985e-06, + "loss": 0.9947, + "step": 5968 + }, + { + "epoch": 0.4485946189688862, + "grad_norm": 3.8109380305228466, + "learning_rate": 2.427054861414631e-06, + "loss": 1.0029, + "step": 5969 + }, + { + "epoch": 0.44866977303472116, + "grad_norm": 1.5018939935480915, + "learning_rate": 2.426579234584728e-06, + "loss": 1.0449, + "step": 5970 + }, + { + "epoch": 0.44874492710055613, + "grad_norm": 1.9738173891610122, + "learning_rate": 2.426103582478372e-06, + "loss": 0.9947, + "step": 5971 + }, + { + "epoch": 0.4488200811663911, + "grad_norm": 1.6107378624129136, + "learning_rate": 2.4256279051237473e-06, + "loss": 0.9656, + "step": 5972 + }, + { + "epoch": 0.4488952352322261, + "grad_norm": 1.7290614758710312, + "learning_rate": 2.4251522025490393e-06, + "loss": 0.9023, + "step": 5973 + }, + { + "epoch": 0.44897038929806105, + "grad_norm": 2.1160434010531026, + "learning_rate": 2.4246764747824347e-06, + "loss": 1.0675, + "step": 5974 + }, + { + "epoch": 0.44904554336389596, + "grad_norm": 1.8990908154009858, + "learning_rate": 2.4242007218521236e-06, + "loss": 0.9781, + "step": 5975 + }, + { + "epoch": 0.44912069742973093, + "grad_norm": 1.7543994235685334, + "learning_rate": 2.423724943786295e-06, + "loss": 0.9752, + "step": 5976 + }, + { + "epoch": 0.4491958514955659, + "grad_norm": 1.7431375282469288, + "learning_rate": 2.4232491406131408e-06, + "loss": 0.8964, + "step": 5977 + }, + { + "epoch": 0.44927100556140087, + "grad_norm": 1.7625879820379005, + "learning_rate": 2.4227733123608548e-06, + "loss": 0.9667, + "step": 5978 + }, + { + "epoch": 0.44934615962723584, + "grad_norm": 1.5717597479307879, + "learning_rate": 2.4222974590576303e-06, + "loss": 1.0351, + "step": 5979 + }, + { + "epoch": 0.4494213136930708, + "grad_norm": 1.560276356573946, + "learning_rate": 2.4218215807316647e-06, + "loss": 0.9993, + "step": 5980 + }, + { + "epoch": 0.4494964677589058, + "grad_norm": 1.9732553308874792, + "learning_rate": 2.4213456774111553e-06, + "loss": 1.0509, + "step": 5981 + }, + { + "epoch": 0.4495716218247407, + "grad_norm": 2.4478278827123017, + "learning_rate": 2.420869749124301e-06, + "loss": 0.8641, + "step": 5982 + }, + { + "epoch": 0.44964677589057567, + "grad_norm": 1.5294797356850918, + "learning_rate": 2.4203937958993027e-06, + "loss": 0.9767, + "step": 5983 + }, + { + "epoch": 0.44972192995641064, + "grad_norm": 1.6750913680479025, + "learning_rate": 2.4199178177643617e-06, + "loss": 0.979, + "step": 5984 + }, + { + "epoch": 0.4497970840222456, + "grad_norm": 1.998338605329205, + "learning_rate": 2.4194418147476827e-06, + "loss": 0.9941, + "step": 5985 + }, + { + "epoch": 0.4498722380880806, + "grad_norm": 1.9707439558262405, + "learning_rate": 2.4189657868774688e-06, + "loss": 1.0398, + "step": 5986 + }, + { + "epoch": 0.44994739215391555, + "grad_norm": 1.383391997039465, + "learning_rate": 2.418489734181929e-06, + "loss": 1.0467, + "step": 5987 + }, + { + "epoch": 0.45002254621975046, + "grad_norm": 1.8267490392870092, + "learning_rate": 2.4180136566892696e-06, + "loss": 1.0406, + "step": 5988 + }, + { + "epoch": 0.45009770028558543, + "grad_norm": 2.991986380010602, + "learning_rate": 2.4175375544276998e-06, + "loss": 0.9738, + "step": 5989 + }, + { + "epoch": 0.4501728543514204, + "grad_norm": 2.289724312891539, + "learning_rate": 2.4170614274254317e-06, + "loss": 1.0047, + "step": 5990 + }, + { + "epoch": 0.4502480084172554, + "grad_norm": 1.6776150004824557, + "learning_rate": 2.4165852757106762e-06, + "loss": 0.9076, + "step": 5991 + }, + { + "epoch": 0.45032316248309034, + "grad_norm": 1.7153947157988452, + "learning_rate": 2.416109099311649e-06, + "loss": 1.0069, + "step": 5992 + }, + { + "epoch": 0.4503983165489253, + "grad_norm": 2.181853390720241, + "learning_rate": 2.4156328982565636e-06, + "loss": 1.011, + "step": 5993 + }, + { + "epoch": 0.4504734706147603, + "grad_norm": 1.5261627860930012, + "learning_rate": 2.4151566725736375e-06, + "loss": 1.0762, + "step": 5994 + }, + { + "epoch": 0.4505486246805952, + "grad_norm": 1.8457312849207639, + "learning_rate": 2.414680422291089e-06, + "loss": 0.876, + "step": 5995 + }, + { + "epoch": 0.45062377874643017, + "grad_norm": 1.3658323274616093, + "learning_rate": 2.4142041474371368e-06, + "loss": 0.9688, + "step": 5996 + }, + { + "epoch": 0.45069893281226514, + "grad_norm": 0.7118766366673546, + "learning_rate": 2.4137278480400038e-06, + "loss": 0.9204, + "step": 5997 + }, + { + "epoch": 0.4507740868781001, + "grad_norm": 2.66660882966581, + "learning_rate": 2.4132515241279106e-06, + "loss": 0.8989, + "step": 5998 + }, + { + "epoch": 0.4508492409439351, + "grad_norm": 2.1889406092687307, + "learning_rate": 2.4127751757290826e-06, + "loss": 0.9397, + "step": 5999 + }, + { + "epoch": 0.45092439500977005, + "grad_norm": 1.7972861441568706, + "learning_rate": 2.4122988028717454e-06, + "loss": 0.9367, + "step": 6000 + }, + { + "epoch": 0.45099954907560497, + "grad_norm": 1.8121948643044843, + "learning_rate": 2.4118224055841243e-06, + "loss": 0.9695, + "step": 6001 + }, + { + "epoch": 0.45107470314143994, + "grad_norm": 2.0773186517514124, + "learning_rate": 2.4113459838944496e-06, + "loss": 0.9127, + "step": 6002 + }, + { + "epoch": 0.4511498572072749, + "grad_norm": 1.370431332454146, + "learning_rate": 2.4108695378309495e-06, + "loss": 0.962, + "step": 6003 + }, + { + "epoch": 0.4512250112731099, + "grad_norm": 2.400785210429586, + "learning_rate": 2.4103930674218565e-06, + "loss": 0.9269, + "step": 6004 + }, + { + "epoch": 0.45130016533894485, + "grad_norm": 1.6816426098819586, + "learning_rate": 2.4099165726954026e-06, + "loss": 0.979, + "step": 6005 + }, + { + "epoch": 0.4513753194047798, + "grad_norm": 1.5875113058814005, + "learning_rate": 2.409440053679822e-06, + "loss": 0.9707, + "step": 6006 + }, + { + "epoch": 0.45145047347061473, + "grad_norm": 1.9529323025058438, + "learning_rate": 2.40896351040335e-06, + "loss": 1.0015, + "step": 6007 + }, + { + "epoch": 0.4515256275364497, + "grad_norm": 3.4540705012121204, + "learning_rate": 2.4084869428942243e-06, + "loss": 1.0029, + "step": 6008 + }, + { + "epoch": 0.4516007816022847, + "grad_norm": 1.6342016999832143, + "learning_rate": 2.4080103511806836e-06, + "loss": 1.0463, + "step": 6009 + }, + { + "epoch": 0.45167593566811964, + "grad_norm": 1.745566883861147, + "learning_rate": 2.4075337352909667e-06, + "loss": 1.0567, + "step": 6010 + }, + { + "epoch": 0.4517510897339546, + "grad_norm": 1.4815109731246574, + "learning_rate": 2.4070570952533155e-06, + "loss": 0.9964, + "step": 6011 + }, + { + "epoch": 0.4518262437997896, + "grad_norm": 3.777415325192805, + "learning_rate": 2.4065804310959725e-06, + "loss": 0.9972, + "step": 6012 + }, + { + "epoch": 0.45190139786562455, + "grad_norm": 8.252020386190376, + "learning_rate": 2.406103742847182e-06, + "loss": 1.1717, + "step": 6013 + }, + { + "epoch": 0.45197655193145947, + "grad_norm": 1.2784955259445492, + "learning_rate": 2.4056270305351896e-06, + "loss": 0.9096, + "step": 6014 + }, + { + "epoch": 0.45205170599729444, + "grad_norm": 0.7656805446484157, + "learning_rate": 2.4051502941882422e-06, + "loss": 0.8421, + "step": 6015 + }, + { + "epoch": 0.4521268600631294, + "grad_norm": 2.35507872026733, + "learning_rate": 2.4046735338345897e-06, + "loss": 0.9684, + "step": 6016 + }, + { + "epoch": 0.4522020141289644, + "grad_norm": 1.6126905934857383, + "learning_rate": 2.4041967495024796e-06, + "loss": 0.9657, + "step": 6017 + }, + { + "epoch": 0.45227716819479935, + "grad_norm": 1.8161627493456074, + "learning_rate": 2.403719941220164e-06, + "loss": 1.0009, + "step": 6018 + }, + { + "epoch": 0.4523523222606343, + "grad_norm": 2.139000714944414, + "learning_rate": 2.403243109015897e-06, + "loss": 0.791, + "step": 6019 + }, + { + "epoch": 0.45242747632646924, + "grad_norm": 1.5600093827821313, + "learning_rate": 2.402766252917931e-06, + "loss": 1.039, + "step": 6020 + }, + { + "epoch": 0.4525026303923042, + "grad_norm": 1.686811801310256, + "learning_rate": 2.402289372954523e-06, + "loss": 1.049, + "step": 6021 + }, + { + "epoch": 0.4525777844581392, + "grad_norm": 1.5833427361995505, + "learning_rate": 2.4018124691539286e-06, + "loss": 0.9734, + "step": 6022 + }, + { + "epoch": 0.45265293852397415, + "grad_norm": 2.6773797154411096, + "learning_rate": 2.401335541544406e-06, + "loss": 0.9486, + "step": 6023 + }, + { + "epoch": 0.4527280925898091, + "grad_norm": 1.6928168707064604, + "learning_rate": 2.400858590154217e-06, + "loss": 0.9203, + "step": 6024 + }, + { + "epoch": 0.4528032466556441, + "grad_norm": 1.6966764527109215, + "learning_rate": 2.400381615011621e-06, + "loss": 1.0517, + "step": 6025 + }, + { + "epoch": 0.45287840072147906, + "grad_norm": 1.8147103216655402, + "learning_rate": 2.399904616144881e-06, + "loss": 1.0144, + "step": 6026 + }, + { + "epoch": 0.452953554787314, + "grad_norm": 1.5310400994130566, + "learning_rate": 2.3994275935822618e-06, + "loss": 0.9673, + "step": 6027 + }, + { + "epoch": 0.45302870885314894, + "grad_norm": 2.532210650469433, + "learning_rate": 2.398950547352028e-06, + "loss": 0.9302, + "step": 6028 + }, + { + "epoch": 0.4531038629189839, + "grad_norm": 0.8175743544009071, + "learning_rate": 2.398473477482446e-06, + "loss": 0.8641, + "step": 6029 + }, + { + "epoch": 0.4531790169848189, + "grad_norm": 2.0943308921183283, + "learning_rate": 2.397996384001785e-06, + "loss": 0.9466, + "step": 6030 + }, + { + "epoch": 0.45325417105065385, + "grad_norm": 4.253181388755855, + "learning_rate": 2.397519266938314e-06, + "loss": 0.9, + "step": 6031 + }, + { + "epoch": 0.4533293251164888, + "grad_norm": 2.6429351993949233, + "learning_rate": 2.3970421263203045e-06, + "loss": 0.9671, + "step": 6032 + }, + { + "epoch": 0.45340447918232374, + "grad_norm": 1.717064395063576, + "learning_rate": 2.396564962176028e-06, + "loss": 1.0078, + "step": 6033 + }, + { + "epoch": 0.4534796332481587, + "grad_norm": 1.5722899630540583, + "learning_rate": 2.39608777453376e-06, + "loss": 0.9894, + "step": 6034 + }, + { + "epoch": 0.4535547873139937, + "grad_norm": 1.9797705147430957, + "learning_rate": 2.395610563421774e-06, + "loss": 0.9944, + "step": 6035 + }, + { + "epoch": 0.45362994137982865, + "grad_norm": 1.8431378657686528, + "learning_rate": 2.3951333288683476e-06, + "loss": 1.056, + "step": 6036 + }, + { + "epoch": 0.4537050954456636, + "grad_norm": 1.9051109412821972, + "learning_rate": 2.394656070901757e-06, + "loss": 0.9637, + "step": 6037 + }, + { + "epoch": 0.4537802495114986, + "grad_norm": 1.9090712260391696, + "learning_rate": 2.394178789550285e-06, + "loss": 1.0184, + "step": 6038 + }, + { + "epoch": 0.45385540357733356, + "grad_norm": 1.7523157601368187, + "learning_rate": 2.3937014848422094e-06, + "loss": 0.9986, + "step": 6039 + }, + { + "epoch": 0.4539305576431685, + "grad_norm": 1.905568986158922, + "learning_rate": 2.3932241568058127e-06, + "loss": 1.0776, + "step": 6040 + }, + { + "epoch": 0.45400571170900345, + "grad_norm": 1.6124235741298463, + "learning_rate": 2.3927468054693797e-06, + "loss": 0.9001, + "step": 6041 + }, + { + "epoch": 0.4540808657748384, + "grad_norm": 1.5045128329773414, + "learning_rate": 2.392269430861194e-06, + "loss": 0.9714, + "step": 6042 + }, + { + "epoch": 0.4541560198406734, + "grad_norm": 0.7464119597103417, + "learning_rate": 2.391792033009543e-06, + "loss": 0.8756, + "step": 6043 + }, + { + "epoch": 0.45423117390650836, + "grad_norm": 1.405289852001402, + "learning_rate": 2.391314611942714e-06, + "loss": 0.8477, + "step": 6044 + }, + { + "epoch": 0.4543063279723433, + "grad_norm": 2.0644534714625418, + "learning_rate": 2.390837167688995e-06, + "loss": 0.9876, + "step": 6045 + }, + { + "epoch": 0.45438148203817824, + "grad_norm": 1.5591724822307835, + "learning_rate": 2.3903597002766777e-06, + "loss": 0.8913, + "step": 6046 + }, + { + "epoch": 0.4544566361040132, + "grad_norm": 1.2319171434196956, + "learning_rate": 2.3898822097340527e-06, + "loss": 1.0121, + "step": 6047 + }, + { + "epoch": 0.4545317901698482, + "grad_norm": 1.3191250760351234, + "learning_rate": 2.389404696089415e-06, + "loss": 1.0017, + "step": 6048 + }, + { + "epoch": 0.45460694423568315, + "grad_norm": 1.91592223457971, + "learning_rate": 2.388927159371057e-06, + "loss": 1.1626, + "step": 6049 + }, + { + "epoch": 0.4546820983015181, + "grad_norm": 2.2883377750665725, + "learning_rate": 2.3884495996072755e-06, + "loss": 0.9426, + "step": 6050 + }, + { + "epoch": 0.4547572523673531, + "grad_norm": 1.4267959955159792, + "learning_rate": 2.3879720168263683e-06, + "loss": 0.9515, + "step": 6051 + }, + { + "epoch": 0.454832406433188, + "grad_norm": 2.460856312214851, + "learning_rate": 2.387494411056633e-06, + "loss": 1.0595, + "step": 6052 + }, + { + "epoch": 0.454907560499023, + "grad_norm": 1.6160212932779565, + "learning_rate": 2.38701678232637e-06, + "loss": 1.0204, + "step": 6053 + }, + { + "epoch": 0.45498271456485795, + "grad_norm": 1.796741779198715, + "learning_rate": 2.386539130663881e-06, + "loss": 1.0908, + "step": 6054 + }, + { + "epoch": 0.4550578686306929, + "grad_norm": 1.8410924735603482, + "learning_rate": 2.386061456097468e-06, + "loss": 1.0414, + "step": 6055 + }, + { + "epoch": 0.4551330226965279, + "grad_norm": 1.7101757147784817, + "learning_rate": 2.3855837586554356e-06, + "loss": 0.9881, + "step": 6056 + }, + { + "epoch": 0.45520817676236286, + "grad_norm": 1.7291752305642207, + "learning_rate": 2.3851060383660893e-06, + "loss": 1.026, + "step": 6057 + }, + { + "epoch": 0.45528333082819783, + "grad_norm": 5.149981948672759, + "learning_rate": 2.3846282952577354e-06, + "loss": 0.8428, + "step": 6058 + }, + { + "epoch": 0.45535848489403274, + "grad_norm": 1.8858534445752946, + "learning_rate": 2.384150529358681e-06, + "loss": 1.0373, + "step": 6059 + }, + { + "epoch": 0.4554336389598677, + "grad_norm": 2.6259725676336294, + "learning_rate": 2.383672740697238e-06, + "loss": 1.0939, + "step": 6060 + }, + { + "epoch": 0.4555087930257027, + "grad_norm": 1.7903077568514731, + "learning_rate": 2.3831949293017166e-06, + "loss": 0.9161, + "step": 6061 + }, + { + "epoch": 0.45558394709153766, + "grad_norm": 1.8441292970677488, + "learning_rate": 2.3827170952004266e-06, + "loss": 0.983, + "step": 6062 + }, + { + "epoch": 0.4556591011573726, + "grad_norm": 1.5309860624828842, + "learning_rate": 2.382239238421684e-06, + "loss": 1.0395, + "step": 6063 + }, + { + "epoch": 0.4557342552232076, + "grad_norm": 2.0210113709142625, + "learning_rate": 2.3817613589938026e-06, + "loss": 0.9182, + "step": 6064 + }, + { + "epoch": 0.4558094092890425, + "grad_norm": 1.6922925780861833, + "learning_rate": 2.381283456945099e-06, + "loss": 0.8715, + "step": 6065 + }, + { + "epoch": 0.4558845633548775, + "grad_norm": 1.5511415000675508, + "learning_rate": 2.3808055323038907e-06, + "loss": 0.8893, + "step": 6066 + }, + { + "epoch": 0.45595971742071245, + "grad_norm": 4.228011300756678, + "learning_rate": 2.3803275850984963e-06, + "loss": 0.9503, + "step": 6067 + }, + { + "epoch": 0.4560348714865474, + "grad_norm": 1.2748067784222836, + "learning_rate": 2.3798496153572363e-06, + "loss": 1.0134, + "step": 6068 + }, + { + "epoch": 0.4561100255523824, + "grad_norm": 2.2770917435677176, + "learning_rate": 2.3793716231084313e-06, + "loss": 0.9372, + "step": 6069 + }, + { + "epoch": 0.45618517961821736, + "grad_norm": 1.585504106206518, + "learning_rate": 2.3788936083804058e-06, + "loss": 0.9403, + "step": 6070 + }, + { + "epoch": 0.45626033368405233, + "grad_norm": 1.7450754494737697, + "learning_rate": 2.3784155712014827e-06, + "loss": 0.886, + "step": 6071 + }, + { + "epoch": 0.45633548774988725, + "grad_norm": 1.7314527482391096, + "learning_rate": 2.3779375115999877e-06, + "loss": 1.0185, + "step": 6072 + }, + { + "epoch": 0.4564106418157222, + "grad_norm": 1.5850391568422355, + "learning_rate": 2.3774594296042485e-06, + "loss": 0.9875, + "step": 6073 + }, + { + "epoch": 0.4564857958815572, + "grad_norm": 1.7394628730533335, + "learning_rate": 2.376981325242592e-06, + "loss": 0.9381, + "step": 6074 + }, + { + "epoch": 0.45656094994739216, + "grad_norm": 3.1998008785284506, + "learning_rate": 2.376503198543349e-06, + "loss": 1.0122, + "step": 6075 + }, + { + "epoch": 0.45663610401322713, + "grad_norm": 1.772912304787326, + "learning_rate": 2.3760250495348495e-06, + "loss": 1.0028, + "step": 6076 + }, + { + "epoch": 0.4567112580790621, + "grad_norm": 4.135737460335256, + "learning_rate": 2.3755468782454265e-06, + "loss": 0.9999, + "step": 6077 + }, + { + "epoch": 0.456786412144897, + "grad_norm": 1.2667827256053865, + "learning_rate": 2.375068684703413e-06, + "loss": 0.9793, + "step": 6078 + }, + { + "epoch": 0.456861566210732, + "grad_norm": 2.199999817934896, + "learning_rate": 2.3745904689371423e-06, + "loss": 1.0504, + "step": 6079 + }, + { + "epoch": 0.45693672027656695, + "grad_norm": 1.5090629649995968, + "learning_rate": 2.374112230974953e-06, + "loss": 0.9076, + "step": 6080 + }, + { + "epoch": 0.4570118743424019, + "grad_norm": 1.666676791478239, + "learning_rate": 2.3736339708451803e-06, + "loss": 0.9918, + "step": 6081 + }, + { + "epoch": 0.4570870284082369, + "grad_norm": 1.392708835111921, + "learning_rate": 2.3731556885761656e-06, + "loss": 0.9926, + "step": 6082 + }, + { + "epoch": 0.45716218247407187, + "grad_norm": 1.5097991821191845, + "learning_rate": 2.3726773841962472e-06, + "loss": 1.0706, + "step": 6083 + }, + { + "epoch": 0.45723733653990684, + "grad_norm": 1.899845558715933, + "learning_rate": 2.372199057733766e-06, + "loss": 0.9332, + "step": 6084 + }, + { + "epoch": 0.45731249060574175, + "grad_norm": 1.3492343674171676, + "learning_rate": 2.371720709217066e-06, + "loss": 0.9151, + "step": 6085 + }, + { + "epoch": 0.4573876446715767, + "grad_norm": 0.9141644152042989, + "learning_rate": 2.3712423386744897e-06, + "loss": 0.9155, + "step": 6086 + }, + { + "epoch": 0.4574627987374117, + "grad_norm": 0.8707728799951246, + "learning_rate": 2.370763946134384e-06, + "loss": 0.9778, + "step": 6087 + }, + { + "epoch": 0.45753795280324666, + "grad_norm": 1.382899308577962, + "learning_rate": 2.3702855316250943e-06, + "loss": 0.9574, + "step": 6088 + }, + { + "epoch": 0.45761310686908163, + "grad_norm": 2.0551527533642466, + "learning_rate": 2.3698070951749692e-06, + "loss": 0.9491, + "step": 6089 + }, + { + "epoch": 0.4576882609349166, + "grad_norm": 1.6372753280379069, + "learning_rate": 2.3693286368123576e-06, + "loss": 0.982, + "step": 6090 + }, + { + "epoch": 0.4577634150007515, + "grad_norm": 1.9756786194584517, + "learning_rate": 2.3688501565656104e-06, + "loss": 1.1472, + "step": 6091 + }, + { + "epoch": 0.4578385690665865, + "grad_norm": 1.8268631060405922, + "learning_rate": 2.3683716544630784e-06, + "loss": 0.9197, + "step": 6092 + }, + { + "epoch": 0.45791372313242146, + "grad_norm": 1.8710839384773417, + "learning_rate": 2.367893130533116e-06, + "loss": 0.9639, + "step": 6093 + }, + { + "epoch": 0.4579888771982564, + "grad_norm": 1.7118509023083783, + "learning_rate": 2.367414584804076e-06, + "loss": 1.072, + "step": 6094 + }, + { + "epoch": 0.4580640312640914, + "grad_norm": 1.5356371565552254, + "learning_rate": 2.3669360173043155e-06, + "loss": 1.0572, + "step": 6095 + }, + { + "epoch": 0.45813918532992637, + "grad_norm": 1.8218241951141834, + "learning_rate": 2.3664574280621907e-06, + "loss": 0.9207, + "step": 6096 + }, + { + "epoch": 0.4582143393957613, + "grad_norm": 1.683982644890994, + "learning_rate": 2.36597881710606e-06, + "loss": 0.9906, + "step": 6097 + }, + { + "epoch": 0.45828949346159625, + "grad_norm": 1.3607365158705318, + "learning_rate": 2.3655001844642828e-06, + "loss": 1.0204, + "step": 6098 + }, + { + "epoch": 0.4583646475274312, + "grad_norm": 1.778758665855889, + "learning_rate": 2.3650215301652207e-06, + "loss": 1.0557, + "step": 6099 + }, + { + "epoch": 0.4584398015932662, + "grad_norm": 2.163222166985581, + "learning_rate": 2.3645428542372347e-06, + "loss": 1.0112, + "step": 6100 + }, + { + "epoch": 0.45851495565910116, + "grad_norm": 1.930021990616191, + "learning_rate": 2.3640641567086887e-06, + "loss": 0.9321, + "step": 6101 + }, + { + "epoch": 0.45859010972493613, + "grad_norm": 10.670808305950574, + "learning_rate": 2.363585437607947e-06, + "loss": 0.9303, + "step": 6102 + }, + { + "epoch": 0.4586652637907711, + "grad_norm": 1.8208557758941415, + "learning_rate": 2.3631066969633755e-06, + "loss": 1.081, + "step": 6103 + }, + { + "epoch": 0.458740417856606, + "grad_norm": 2.076583060800733, + "learning_rate": 2.362627934803343e-06, + "loss": 1.0093, + "step": 6104 + }, + { + "epoch": 0.458815571922441, + "grad_norm": 2.046507824295413, + "learning_rate": 2.362149151156216e-06, + "loss": 1.0152, + "step": 6105 + }, + { + "epoch": 0.45889072598827596, + "grad_norm": 1.8930125648254843, + "learning_rate": 2.3616703460503654e-06, + "loss": 1.1116, + "step": 6106 + }, + { + "epoch": 0.45896588005411093, + "grad_norm": 1.464562717182308, + "learning_rate": 2.3611915195141615e-06, + "loss": 0.864, + "step": 6107 + }, + { + "epoch": 0.4590410341199459, + "grad_norm": 2.0704102691021324, + "learning_rate": 2.3607126715759773e-06, + "loss": 0.9584, + "step": 6108 + }, + { + "epoch": 0.45911618818578087, + "grad_norm": 1.668311983836353, + "learning_rate": 2.360233802264186e-06, + "loss": 1.0436, + "step": 6109 + }, + { + "epoch": 0.4591913422516158, + "grad_norm": 1.9108439262729808, + "learning_rate": 2.359754911607163e-06, + "loss": 0.9673, + "step": 6110 + }, + { + "epoch": 0.45926649631745076, + "grad_norm": 1.2991195466503276, + "learning_rate": 2.3592759996332824e-06, + "loss": 1.0509, + "step": 6111 + }, + { + "epoch": 0.4593416503832857, + "grad_norm": 1.4578602250475718, + "learning_rate": 2.358797066370924e-06, + "loss": 0.9254, + "step": 6112 + }, + { + "epoch": 0.4594168044491207, + "grad_norm": 1.6282066837255609, + "learning_rate": 2.358318111848466e-06, + "loss": 0.9858, + "step": 6113 + }, + { + "epoch": 0.45949195851495567, + "grad_norm": 1.7616972995983127, + "learning_rate": 2.3578391360942872e-06, + "loss": 1.0463, + "step": 6114 + }, + { + "epoch": 0.45956711258079064, + "grad_norm": 1.5130701456443647, + "learning_rate": 2.3573601391367696e-06, + "loss": 1.0462, + "step": 6115 + }, + { + "epoch": 0.4596422666466256, + "grad_norm": 1.599010387985716, + "learning_rate": 2.3568811210042947e-06, + "loss": 0.9503, + "step": 6116 + }, + { + "epoch": 0.4597174207124605, + "grad_norm": 2.016997941711616, + "learning_rate": 2.3564020817252476e-06, + "loss": 0.8968, + "step": 6117 + }, + { + "epoch": 0.4597925747782955, + "grad_norm": 2.0820469381517883, + "learning_rate": 2.3559230213280115e-06, + "loss": 0.9637, + "step": 6118 + }, + { + "epoch": 0.45986772884413046, + "grad_norm": 0.6844150514074618, + "learning_rate": 2.3554439398409743e-06, + "loss": 0.8478, + "step": 6119 + }, + { + "epoch": 0.45994288290996543, + "grad_norm": 1.9898194843635397, + "learning_rate": 2.354964837292522e-06, + "loss": 1.0346, + "step": 6120 + }, + { + "epoch": 0.4600180369758004, + "grad_norm": 1.7377704684287258, + "learning_rate": 2.354485713711044e-06, + "loss": 0.8966, + "step": 6121 + }, + { + "epoch": 0.4600931910416354, + "grad_norm": 1.7595505323641585, + "learning_rate": 2.354006569124931e-06, + "loss": 0.9287, + "step": 6122 + }, + { + "epoch": 0.4601683451074703, + "grad_norm": 1.5709017886247358, + "learning_rate": 2.3535274035625713e-06, + "loss": 0.8812, + "step": 6123 + }, + { + "epoch": 0.46024349917330526, + "grad_norm": 2.2191538913238276, + "learning_rate": 2.353048217052361e-06, + "loss": 1.0147, + "step": 6124 + }, + { + "epoch": 0.46031865323914023, + "grad_norm": 1.7700356949693004, + "learning_rate": 2.3525690096226906e-06, + "loss": 0.9955, + "step": 6125 + }, + { + "epoch": 0.4603938073049752, + "grad_norm": 1.4703778924294972, + "learning_rate": 2.3520897813019566e-06, + "loss": 1.0732, + "step": 6126 + }, + { + "epoch": 0.46046896137081017, + "grad_norm": 1.5372550148899966, + "learning_rate": 2.351610532118555e-06, + "loss": 0.9982, + "step": 6127 + }, + { + "epoch": 0.46054411543664514, + "grad_norm": 1.6892355718172927, + "learning_rate": 2.3511312621008832e-06, + "loss": 1.0308, + "step": 6128 + }, + { + "epoch": 0.4606192695024801, + "grad_norm": 1.789628451588686, + "learning_rate": 2.35065197127734e-06, + "loss": 0.9709, + "step": 6129 + }, + { + "epoch": 0.460694423568315, + "grad_norm": 3.100983149999695, + "learning_rate": 2.350172659676323e-06, + "loss": 1.0128, + "step": 6130 + }, + { + "epoch": 0.46076957763415, + "grad_norm": 2.1329658648785976, + "learning_rate": 2.349693327326237e-06, + "loss": 0.9828, + "step": 6131 + }, + { + "epoch": 0.46084473169998497, + "grad_norm": 1.6210877200094929, + "learning_rate": 2.3492139742554816e-06, + "loss": 0.9839, + "step": 6132 + }, + { + "epoch": 0.46091988576581994, + "grad_norm": 1.9711130281126288, + "learning_rate": 2.3487346004924605e-06, + "loss": 0.9645, + "step": 6133 + }, + { + "epoch": 0.4609950398316549, + "grad_norm": 2.1153914938328024, + "learning_rate": 2.34825520606558e-06, + "loss": 0.8835, + "step": 6134 + }, + { + "epoch": 0.4610701938974899, + "grad_norm": 2.138028428640748, + "learning_rate": 2.3477757910032434e-06, + "loss": 0.919, + "step": 6135 + }, + { + "epoch": 0.4611453479633248, + "grad_norm": 3.1544752513418337, + "learning_rate": 2.347296355333861e-06, + "loss": 0.9156, + "step": 6136 + }, + { + "epoch": 0.46122050202915976, + "grad_norm": 1.4225201662642062, + "learning_rate": 2.346816899085839e-06, + "loss": 1.0746, + "step": 6137 + }, + { + "epoch": 0.46129565609499473, + "grad_norm": 1.838793736397326, + "learning_rate": 2.346337422287587e-06, + "loss": 0.9488, + "step": 6138 + }, + { + "epoch": 0.4613708101608297, + "grad_norm": 1.7079251739942762, + "learning_rate": 2.3458579249675176e-06, + "loss": 0.9762, + "step": 6139 + }, + { + "epoch": 0.4614459642266647, + "grad_norm": 1.6510224526748418, + "learning_rate": 2.345378407154041e-06, + "loss": 1.0292, + "step": 6140 + }, + { + "epoch": 0.46152111829249964, + "grad_norm": 1.6580509974202804, + "learning_rate": 2.344898868875572e-06, + "loss": 0.9768, + "step": 6141 + }, + { + "epoch": 0.46159627235833456, + "grad_norm": 1.3346708662085784, + "learning_rate": 2.3444193101605237e-06, + "loss": 0.9647, + "step": 6142 + }, + { + "epoch": 0.46167142642416953, + "grad_norm": 1.8481280831294165, + "learning_rate": 2.3439397310373126e-06, + "loss": 0.9828, + "step": 6143 + }, + { + "epoch": 0.4617465804900045, + "grad_norm": 0.6861642519216768, + "learning_rate": 2.343460131534356e-06, + "loss": 0.7871, + "step": 6144 + }, + { + "epoch": 0.46182173455583947, + "grad_norm": 1.753161435760103, + "learning_rate": 2.34298051168007e-06, + "loss": 0.9924, + "step": 6145 + }, + { + "epoch": 0.46189688862167444, + "grad_norm": 1.4229352582142358, + "learning_rate": 2.3425008715028766e-06, + "loss": 1.0551, + "step": 6146 + }, + { + "epoch": 0.4619720426875094, + "grad_norm": 1.4556614877690064, + "learning_rate": 2.3420212110311943e-06, + "loss": 0.8805, + "step": 6147 + }, + { + "epoch": 0.4620471967533444, + "grad_norm": 1.640629868273097, + "learning_rate": 2.3415415302934457e-06, + "loss": 1.0547, + "step": 6148 + }, + { + "epoch": 0.4621223508191793, + "grad_norm": 2.861240966302652, + "learning_rate": 2.341061829318054e-06, + "loss": 0.9685, + "step": 6149 + }, + { + "epoch": 0.46219750488501427, + "grad_norm": 1.7573622741564057, + "learning_rate": 2.340582108133442e-06, + "loss": 0.9884, + "step": 6150 + }, + { + "epoch": 0.46227265895084924, + "grad_norm": 1.7636732268301851, + "learning_rate": 2.340102366768037e-06, + "loss": 0.9584, + "step": 6151 + }, + { + "epoch": 0.4623478130166842, + "grad_norm": 1.7383460557837191, + "learning_rate": 2.339622605250264e-06, + "loss": 1.0056, + "step": 6152 + }, + { + "epoch": 0.4624229670825192, + "grad_norm": 1.5770329388823254, + "learning_rate": 2.339142823608551e-06, + "loss": 1.0175, + "step": 6153 + }, + { + "epoch": 0.46249812114835415, + "grad_norm": 1.5009009516577074, + "learning_rate": 2.3386630218713273e-06, + "loss": 0.9683, + "step": 6154 + }, + { + "epoch": 0.46257327521418906, + "grad_norm": 1.8118111189193118, + "learning_rate": 2.3381832000670223e-06, + "loss": 1.0072, + "step": 6155 + }, + { + "epoch": 0.46264842928002403, + "grad_norm": 2.0338362441858644, + "learning_rate": 2.3377033582240684e-06, + "loss": 1.0001, + "step": 6156 + }, + { + "epoch": 0.462723583345859, + "grad_norm": 2.1231107728496608, + "learning_rate": 2.3372234963708966e-06, + "loss": 0.9454, + "step": 6157 + }, + { + "epoch": 0.462798737411694, + "grad_norm": 2.533764285107068, + "learning_rate": 2.336743614535942e-06, + "loss": 0.9664, + "step": 6158 + }, + { + "epoch": 0.46287389147752894, + "grad_norm": 1.849619171535566, + "learning_rate": 2.3362637127476383e-06, + "loss": 1.0011, + "step": 6159 + }, + { + "epoch": 0.4629490455433639, + "grad_norm": 1.9169352246535558, + "learning_rate": 2.335783791034422e-06, + "loss": 0.9375, + "step": 6160 + }, + { + "epoch": 0.4630241996091989, + "grad_norm": 2.2358959863249046, + "learning_rate": 2.3353038494247305e-06, + "loss": 0.93, + "step": 6161 + }, + { + "epoch": 0.4630993536750338, + "grad_norm": 2.1688500065240937, + "learning_rate": 2.3348238879470015e-06, + "loss": 0.8911, + "step": 6162 + }, + { + "epoch": 0.46317450774086877, + "grad_norm": 1.993617303910787, + "learning_rate": 2.334343906629676e-06, + "loss": 0.9456, + "step": 6163 + }, + { + "epoch": 0.46324966180670374, + "grad_norm": 1.6651443761370175, + "learning_rate": 2.3338639055011924e-06, + "loss": 0.9193, + "step": 6164 + }, + { + "epoch": 0.4633248158725387, + "grad_norm": 1.6374732124525377, + "learning_rate": 2.333383884589995e-06, + "loss": 1.0249, + "step": 6165 + }, + { + "epoch": 0.4633999699383737, + "grad_norm": 1.6080757007240154, + "learning_rate": 2.3329038439245257e-06, + "loss": 0.9728, + "step": 6166 + }, + { + "epoch": 0.46347512400420865, + "grad_norm": 2.0900597006197073, + "learning_rate": 2.332423783533228e-06, + "loss": 1.061, + "step": 6167 + }, + { + "epoch": 0.46355027807004356, + "grad_norm": 1.5607819075835552, + "learning_rate": 2.331943703444549e-06, + "loss": 0.9346, + "step": 6168 + }, + { + "epoch": 0.46362543213587853, + "grad_norm": 2.016281021172781, + "learning_rate": 2.331463603686934e-06, + "loss": 1.065, + "step": 6169 + }, + { + "epoch": 0.4637005862017135, + "grad_norm": 1.4970976247865333, + "learning_rate": 2.330983484288832e-06, + "loss": 0.8619, + "step": 6170 + }, + { + "epoch": 0.4637757402675485, + "grad_norm": 1.831129360456533, + "learning_rate": 2.3305033452786905e-06, + "loss": 1.0585, + "step": 6171 + }, + { + "epoch": 0.46385089433338345, + "grad_norm": 2.13053252373351, + "learning_rate": 2.3300231866849606e-06, + "loss": 1.012, + "step": 6172 + }, + { + "epoch": 0.4639260483992184, + "grad_norm": 1.7229542031248752, + "learning_rate": 2.3295430085360927e-06, + "loss": 0.9335, + "step": 6173 + }, + { + "epoch": 0.4640012024650534, + "grad_norm": 1.240001939033715, + "learning_rate": 2.32906281086054e-06, + "loss": 1.0528, + "step": 6174 + }, + { + "epoch": 0.4640763565308883, + "grad_norm": 1.595959347496384, + "learning_rate": 2.3285825936867556e-06, + "loss": 1.0439, + "step": 6175 + }, + { + "epoch": 0.46415151059672327, + "grad_norm": 2.375287891808007, + "learning_rate": 2.328102357043194e-06, + "loss": 1.024, + "step": 6176 + }, + { + "epoch": 0.46422666466255824, + "grad_norm": 2.1346788449810705, + "learning_rate": 2.3276221009583116e-06, + "loss": 0.9415, + "step": 6177 + }, + { + "epoch": 0.4643018187283932, + "grad_norm": 2.161242799851717, + "learning_rate": 2.327141825460566e-06, + "loss": 1.025, + "step": 6178 + }, + { + "epoch": 0.4643769727942282, + "grad_norm": 2.213606123640264, + "learning_rate": 2.3266615305784126e-06, + "loss": 0.9509, + "step": 6179 + }, + { + "epoch": 0.46445212686006315, + "grad_norm": 1.6499279931279802, + "learning_rate": 2.3261812163403144e-06, + "loss": 1.1231, + "step": 6180 + }, + { + "epoch": 0.46452728092589807, + "grad_norm": 2.5740149826882353, + "learning_rate": 2.3257008827747294e-06, + "loss": 1.0152, + "step": 6181 + }, + { + "epoch": 0.46460243499173304, + "grad_norm": 1.7545874956936618, + "learning_rate": 2.32522052991012e-06, + "loss": 1.0049, + "step": 6182 + }, + { + "epoch": 0.464677589057568, + "grad_norm": 1.8448738293809794, + "learning_rate": 2.324740157774949e-06, + "loss": 1.0003, + "step": 6183 + }, + { + "epoch": 0.464752743123403, + "grad_norm": 1.8639782255808284, + "learning_rate": 2.3242597663976793e-06, + "loss": 0.9146, + "step": 6184 + }, + { + "epoch": 0.46482789718923795, + "grad_norm": 0.6932078236180222, + "learning_rate": 2.3237793558067776e-06, + "loss": 0.8873, + "step": 6185 + }, + { + "epoch": 0.4649030512550729, + "grad_norm": 2.2933126838257136, + "learning_rate": 2.3232989260307087e-06, + "loss": 1.0352, + "step": 6186 + }, + { + "epoch": 0.46497820532090783, + "grad_norm": 1.6034239581318188, + "learning_rate": 2.322818477097941e-06, + "loss": 1.0424, + "step": 6187 + }, + { + "epoch": 0.4650533593867428, + "grad_norm": 5.2147910811947025, + "learning_rate": 2.322338009036943e-06, + "loss": 1.0078, + "step": 6188 + }, + { + "epoch": 0.4651285134525778, + "grad_norm": 1.4438083554318994, + "learning_rate": 2.3218575218761816e-06, + "loss": 0.9975, + "step": 6189 + }, + { + "epoch": 0.46520366751841274, + "grad_norm": 1.5401471184150899, + "learning_rate": 2.3213770156441314e-06, + "loss": 0.972, + "step": 6190 + }, + { + "epoch": 0.4652788215842477, + "grad_norm": 1.7270888825048638, + "learning_rate": 2.3208964903692613e-06, + "loss": 0.9101, + "step": 6191 + }, + { + "epoch": 0.4653539756500827, + "grad_norm": 1.5456248925159262, + "learning_rate": 2.3204159460800458e-06, + "loss": 0.9617, + "step": 6192 + }, + { + "epoch": 0.46542912971591766, + "grad_norm": 2.5420131968253967, + "learning_rate": 2.319935382804959e-06, + "loss": 1.0341, + "step": 6193 + }, + { + "epoch": 0.46550428378175257, + "grad_norm": 1.8909986142577566, + "learning_rate": 2.3194548005724748e-06, + "loss": 0.9942, + "step": 6194 + }, + { + "epoch": 0.46557943784758754, + "grad_norm": 2.7882633720630707, + "learning_rate": 2.318974199411071e-06, + "loss": 0.9905, + "step": 6195 + }, + { + "epoch": 0.4656545919134225, + "grad_norm": 1.8329603364839324, + "learning_rate": 2.318493579349224e-06, + "loss": 0.9571, + "step": 6196 + }, + { + "epoch": 0.4657297459792575, + "grad_norm": 2.1855902918916965, + "learning_rate": 2.3180129404154133e-06, + "loss": 0.9603, + "step": 6197 + }, + { + "epoch": 0.46580490004509245, + "grad_norm": 1.689103318531954, + "learning_rate": 2.317532282638118e-06, + "loss": 0.9973, + "step": 6198 + }, + { + "epoch": 0.4658800541109274, + "grad_norm": 1.7684017391198872, + "learning_rate": 2.3170516060458188e-06, + "loss": 0.9477, + "step": 6199 + }, + { + "epoch": 0.46595520817676234, + "grad_norm": 1.9875818607578157, + "learning_rate": 2.3165709106669983e-06, + "loss": 0.9737, + "step": 6200 + }, + { + "epoch": 0.4660303622425973, + "grad_norm": 1.5014480912429244, + "learning_rate": 2.3160901965301386e-06, + "loss": 0.9196, + "step": 6201 + }, + { + "epoch": 0.4661055163084323, + "grad_norm": 0.7981558958210896, + "learning_rate": 2.315609463663725e-06, + "loss": 0.8059, + "step": 6202 + }, + { + "epoch": 0.46618067037426725, + "grad_norm": 2.052924266274892, + "learning_rate": 2.315128712096242e-06, + "loss": 0.8148, + "step": 6203 + }, + { + "epoch": 0.4662558244401022, + "grad_norm": 1.5568465380247325, + "learning_rate": 2.314647941856175e-06, + "loss": 0.9509, + "step": 6204 + }, + { + "epoch": 0.4663309785059372, + "grad_norm": 1.7453749028237058, + "learning_rate": 2.314167152972014e-06, + "loss": 0.9934, + "step": 6205 + }, + { + "epoch": 0.46640613257177216, + "grad_norm": 1.8504833182918037, + "learning_rate": 2.313686345472245e-06, + "loss": 1.0109, + "step": 6206 + }, + { + "epoch": 0.4664812866376071, + "grad_norm": 1.53164496485653, + "learning_rate": 2.3132055193853597e-06, + "loss": 0.9436, + "step": 6207 + }, + { + "epoch": 0.46655644070344204, + "grad_norm": 1.5388395661132435, + "learning_rate": 2.312724674739847e-06, + "loss": 0.9534, + "step": 6208 + }, + { + "epoch": 0.466631594769277, + "grad_norm": 1.473892219417634, + "learning_rate": 2.3122438115642013e-06, + "loss": 1.0801, + "step": 6209 + }, + { + "epoch": 0.466706748835112, + "grad_norm": 1.9547649974996557, + "learning_rate": 2.3117629298869135e-06, + "loss": 1.0024, + "step": 6210 + }, + { + "epoch": 0.46678190290094695, + "grad_norm": 1.8171895624937915, + "learning_rate": 2.3112820297364775e-06, + "loss": 0.9845, + "step": 6211 + }, + { + "epoch": 0.4668570569667819, + "grad_norm": 0.7325121200912839, + "learning_rate": 2.3108011111413904e-06, + "loss": 0.7933, + "step": 6212 + }, + { + "epoch": 0.46693221103261684, + "grad_norm": 4.372093761279815, + "learning_rate": 2.3103201741301465e-06, + "loss": 1.0443, + "step": 6213 + }, + { + "epoch": 0.4670073650984518, + "grad_norm": 1.8443612200882464, + "learning_rate": 2.3098392187312445e-06, + "loss": 1.0281, + "step": 6214 + }, + { + "epoch": 0.4670825191642868, + "grad_norm": 1.8015971621299602, + "learning_rate": 2.309358244973182e-06, + "loss": 1.0004, + "step": 6215 + }, + { + "epoch": 0.46715767323012175, + "grad_norm": 9.511959378373119, + "learning_rate": 2.3088772528844588e-06, + "loss": 0.9297, + "step": 6216 + }, + { + "epoch": 0.4672328272959567, + "grad_norm": 2.0981106115080204, + "learning_rate": 2.308396242493576e-06, + "loss": 1.0082, + "step": 6217 + }, + { + "epoch": 0.4673079813617917, + "grad_norm": 1.6193112052268301, + "learning_rate": 2.3079152138290347e-06, + "loss": 0.9795, + "step": 6218 + }, + { + "epoch": 0.46738313542762666, + "grad_norm": 1.8048737665136907, + "learning_rate": 2.307434166919338e-06, + "loss": 0.9373, + "step": 6219 + }, + { + "epoch": 0.4674582894934616, + "grad_norm": 1.7973490214177268, + "learning_rate": 2.30695310179299e-06, + "loss": 0.9776, + "step": 6220 + }, + { + "epoch": 0.46753344355929655, + "grad_norm": 1.4847754490423728, + "learning_rate": 2.3064720184784946e-06, + "loss": 0.9947, + "step": 6221 + }, + { + "epoch": 0.4676085976251315, + "grad_norm": 1.5301529204547486, + "learning_rate": 2.305990917004359e-06, + "loss": 0.9609, + "step": 6222 + }, + { + "epoch": 0.4676837516909665, + "grad_norm": 1.683585748852616, + "learning_rate": 2.3055097973990894e-06, + "loss": 1.0706, + "step": 6223 + }, + { + "epoch": 0.46775890575680146, + "grad_norm": 1.2472214812541103, + "learning_rate": 2.305028659691195e-06, + "loss": 1.0011, + "step": 6224 + }, + { + "epoch": 0.4678340598226364, + "grad_norm": 1.4902688518650697, + "learning_rate": 2.3045475039091846e-06, + "loss": 0.8896, + "step": 6225 + }, + { + "epoch": 0.46790921388847134, + "grad_norm": 1.6863211293414513, + "learning_rate": 2.3040663300815673e-06, + "loss": 0.8754, + "step": 6226 + }, + { + "epoch": 0.4679843679543063, + "grad_norm": 1.4663877767497748, + "learning_rate": 2.303585138236857e-06, + "loss": 0.9903, + "step": 6227 + }, + { + "epoch": 0.4680595220201413, + "grad_norm": 1.4115328423047, + "learning_rate": 2.3031039284035636e-06, + "loss": 0.9395, + "step": 6228 + }, + { + "epoch": 0.46813467608597625, + "grad_norm": 1.7666145185055593, + "learning_rate": 2.3026227006102025e-06, + "loss": 0.9317, + "step": 6229 + }, + { + "epoch": 0.4682098301518112, + "grad_norm": 1.7987802107701556, + "learning_rate": 2.3021414548852864e-06, + "loss": 0.9579, + "step": 6230 + }, + { + "epoch": 0.4682849842176462, + "grad_norm": 1.297765540952742, + "learning_rate": 2.3016601912573333e-06, + "loss": 1.0606, + "step": 6231 + }, + { + "epoch": 0.4683601382834811, + "grad_norm": 1.525180777794227, + "learning_rate": 2.301178909754859e-06, + "loss": 1.0226, + "step": 6232 + }, + { + "epoch": 0.4684352923493161, + "grad_norm": 1.709591061398528, + "learning_rate": 2.30069761040638e-06, + "loss": 1.0487, + "step": 6233 + }, + { + "epoch": 0.46851044641515105, + "grad_norm": 1.4958785975557938, + "learning_rate": 2.300216293240417e-06, + "loss": 0.9044, + "step": 6234 + }, + { + "epoch": 0.468585600480986, + "grad_norm": 1.5993310036371289, + "learning_rate": 2.299734958285488e-06, + "loss": 1.0361, + "step": 6235 + }, + { + "epoch": 0.468660754546821, + "grad_norm": 2.062896112493589, + "learning_rate": 2.2992536055701157e-06, + "loss": 1.0936, + "step": 6236 + }, + { + "epoch": 0.46873590861265596, + "grad_norm": 4.4604748487649655, + "learning_rate": 2.2987722351228216e-06, + "loss": 0.9883, + "step": 6237 + }, + { + "epoch": 0.46881106267849093, + "grad_norm": 1.7371568107604762, + "learning_rate": 2.298290846972128e-06, + "loss": 0.9327, + "step": 6238 + }, + { + "epoch": 0.46888621674432585, + "grad_norm": 1.5023768984211878, + "learning_rate": 2.29780944114656e-06, + "loss": 1.0908, + "step": 6239 + }, + { + "epoch": 0.4689613708101608, + "grad_norm": 2.1562979872519152, + "learning_rate": 2.2973280176746413e-06, + "loss": 0.9581, + "step": 6240 + }, + { + "epoch": 0.4690365248759958, + "grad_norm": 1.6288528250662546, + "learning_rate": 2.2968465765849e-06, + "loss": 0.8722, + "step": 6241 + }, + { + "epoch": 0.46911167894183076, + "grad_norm": 1.7060138105747995, + "learning_rate": 2.296365117905862e-06, + "loss": 1.0247, + "step": 6242 + }, + { + "epoch": 0.4691868330076657, + "grad_norm": 1.3558688493956466, + "learning_rate": 2.2958836416660556e-06, + "loss": 0.9875, + "step": 6243 + }, + { + "epoch": 0.4692619870735007, + "grad_norm": 1.5335154998762144, + "learning_rate": 2.295402147894011e-06, + "loss": 1.0656, + "step": 6244 + }, + { + "epoch": 0.4693371411393356, + "grad_norm": 2.1339573285342146, + "learning_rate": 2.294920636618257e-06, + "loss": 0.9942, + "step": 6245 + }, + { + "epoch": 0.4694122952051706, + "grad_norm": 2.036638009680858, + "learning_rate": 2.2944391078673267e-06, + "loss": 0.9957, + "step": 6246 + }, + { + "epoch": 0.46948744927100555, + "grad_norm": 1.5648042092015109, + "learning_rate": 2.2939575616697516e-06, + "loss": 0.9242, + "step": 6247 + }, + { + "epoch": 0.4695626033368405, + "grad_norm": 1.7221136298177036, + "learning_rate": 2.2934759980540654e-06, + "loss": 0.9794, + "step": 6248 + }, + { + "epoch": 0.4696377574026755, + "grad_norm": 1.835528843627352, + "learning_rate": 2.2929944170488025e-06, + "loss": 1.038, + "step": 6249 + }, + { + "epoch": 0.46971291146851046, + "grad_norm": 0.7316970697515022, + "learning_rate": 2.2925128186824983e-06, + "loss": 0.8341, + "step": 6250 + }, + { + "epoch": 0.46978806553434543, + "grad_norm": 1.965171346789183, + "learning_rate": 2.29203120298369e-06, + "loss": 0.9137, + "step": 6251 + }, + { + "epoch": 0.46986321960018035, + "grad_norm": 1.56416186646809, + "learning_rate": 2.2915495699809134e-06, + "loss": 0.9536, + "step": 6252 + }, + { + "epoch": 0.4699383736660153, + "grad_norm": 0.8792365915655205, + "learning_rate": 2.2910679197027093e-06, + "loss": 0.9504, + "step": 6253 + }, + { + "epoch": 0.4700135277318503, + "grad_norm": 6.139357574354278, + "learning_rate": 2.290586252177617e-06, + "loss": 0.9993, + "step": 6254 + }, + { + "epoch": 0.47008868179768526, + "grad_norm": 1.9091894281241286, + "learning_rate": 2.290104567434175e-06, + "loss": 1.0458, + "step": 6255 + }, + { + "epoch": 0.47016383586352023, + "grad_norm": 1.353478721566947, + "learning_rate": 2.2896228655009276e-06, + "loss": 0.8811, + "step": 6256 + }, + { + "epoch": 0.4702389899293552, + "grad_norm": 1.6572843956283314, + "learning_rate": 2.2891411464064155e-06, + "loss": 0.9572, + "step": 6257 + }, + { + "epoch": 0.4703141439951901, + "grad_norm": 1.5436010682702601, + "learning_rate": 2.2886594101791845e-06, + "loss": 0.9804, + "step": 6258 + }, + { + "epoch": 0.4703892980610251, + "grad_norm": 2.875464857915827, + "learning_rate": 2.2881776568477777e-06, + "loss": 0.9531, + "step": 6259 + }, + { + "epoch": 0.47046445212686006, + "grad_norm": 1.9427063424626054, + "learning_rate": 2.2876958864407407e-06, + "loss": 0.9913, + "step": 6260 + }, + { + "epoch": 0.470539606192695, + "grad_norm": 1.813066459796877, + "learning_rate": 2.287214098986621e-06, + "loss": 1.0254, + "step": 6261 + }, + { + "epoch": 0.47061476025853, + "grad_norm": 2.0407688995611473, + "learning_rate": 2.286732294513966e-06, + "loss": 0.989, + "step": 6262 + }, + { + "epoch": 0.47068991432436497, + "grad_norm": 0.741163088896328, + "learning_rate": 2.286250473051325e-06, + "loss": 0.8333, + "step": 6263 + }, + { + "epoch": 0.47076506839019994, + "grad_norm": 1.850862974501751, + "learning_rate": 2.2857686346272475e-06, + "loss": 1.0079, + "step": 6264 + }, + { + "epoch": 0.47084022245603485, + "grad_norm": 1.5854472638192443, + "learning_rate": 2.2852867792702835e-06, + "loss": 0.9421, + "step": 6265 + }, + { + "epoch": 0.4709153765218698, + "grad_norm": 1.3830666604819635, + "learning_rate": 2.284804907008986e-06, + "loss": 1.0034, + "step": 6266 + }, + { + "epoch": 0.4709905305877048, + "grad_norm": 1.7181940046422293, + "learning_rate": 2.2843230178719063e-06, + "loss": 0.9587, + "step": 6267 + }, + { + "epoch": 0.47106568465353976, + "grad_norm": 0.825850692615333, + "learning_rate": 2.2838411118875997e-06, + "loss": 0.8183, + "step": 6268 + }, + { + "epoch": 0.47114083871937473, + "grad_norm": 1.6661260284299306, + "learning_rate": 2.2833591890846204e-06, + "loss": 0.8864, + "step": 6269 + }, + { + "epoch": 0.4712159927852097, + "grad_norm": 1.7956091361003979, + "learning_rate": 2.282877249491523e-06, + "loss": 0.895, + "step": 6270 + }, + { + "epoch": 0.4712911468510446, + "grad_norm": 0.8025509515672199, + "learning_rate": 2.2823952931368667e-06, + "loss": 0.878, + "step": 6271 + }, + { + "epoch": 0.4713663009168796, + "grad_norm": 1.5427378034985815, + "learning_rate": 2.2819133200492073e-06, + "loss": 0.9173, + "step": 6272 + }, + { + "epoch": 0.47144145498271456, + "grad_norm": 1.9663080481165434, + "learning_rate": 2.281431330257105e-06, + "loss": 1.0115, + "step": 6273 + }, + { + "epoch": 0.47151660904854953, + "grad_norm": 2.438367933634925, + "learning_rate": 2.280949323789117e-06, + "loss": 0.9748, + "step": 6274 + }, + { + "epoch": 0.4715917631143845, + "grad_norm": 1.5696913216491388, + "learning_rate": 2.280467300673807e-06, + "loss": 0.944, + "step": 6275 + }, + { + "epoch": 0.47166691718021947, + "grad_norm": 0.7695340384031016, + "learning_rate": 2.2799852609397353e-06, + "loss": 0.8267, + "step": 6276 + }, + { + "epoch": 0.4717420712460544, + "grad_norm": 1.8810174386383742, + "learning_rate": 2.2795032046154644e-06, + "loss": 0.9661, + "step": 6277 + }, + { + "epoch": 0.47181722531188935, + "grad_norm": 2.4055758436001855, + "learning_rate": 2.279021131729559e-06, + "loss": 0.9514, + "step": 6278 + }, + { + "epoch": 0.4718923793777243, + "grad_norm": 1.7464538475991351, + "learning_rate": 2.2785390423105822e-06, + "loss": 1.0777, + "step": 6279 + }, + { + "epoch": 0.4719675334435593, + "grad_norm": 2.3846918418023924, + "learning_rate": 2.2780569363871016e-06, + "loss": 1.0476, + "step": 6280 + }, + { + "epoch": 0.47204268750939427, + "grad_norm": 0.7786235912467643, + "learning_rate": 2.277574813987682e-06, + "loss": 0.8633, + "step": 6281 + }, + { + "epoch": 0.47211784157522924, + "grad_norm": 1.6362926866341478, + "learning_rate": 2.2770926751408916e-06, + "loss": 0.9878, + "step": 6282 + }, + { + "epoch": 0.4721929956410642, + "grad_norm": 1.3715413115701498, + "learning_rate": 2.2766105198753e-06, + "loss": 1.0608, + "step": 6283 + }, + { + "epoch": 0.4722681497068991, + "grad_norm": 1.5081305291983103, + "learning_rate": 2.2761283482194747e-06, + "loss": 0.9773, + "step": 6284 + }, + { + "epoch": 0.4723433037727341, + "grad_norm": 1.697547178104414, + "learning_rate": 2.2756461602019886e-06, + "loss": 0.8924, + "step": 6285 + }, + { + "epoch": 0.47241845783856906, + "grad_norm": 1.1400169946842658, + "learning_rate": 2.2751639558514117e-06, + "loss": 0.9358, + "step": 6286 + }, + { + "epoch": 0.47249361190440403, + "grad_norm": 2.71931022046102, + "learning_rate": 2.2746817351963163e-06, + "loss": 0.9453, + "step": 6287 + }, + { + "epoch": 0.472568765970239, + "grad_norm": 1.340703571569136, + "learning_rate": 2.274199498265276e-06, + "loss": 0.9129, + "step": 6288 + }, + { + "epoch": 0.472643920036074, + "grad_norm": 2.0535809107970255, + "learning_rate": 2.2737172450868663e-06, + "loss": 1.0045, + "step": 6289 + }, + { + "epoch": 0.4727190741019089, + "grad_norm": 1.86392418347424, + "learning_rate": 2.2732349756896615e-06, + "loss": 1.0562, + "step": 6290 + }, + { + "epoch": 0.47279422816774386, + "grad_norm": 1.8733028996643846, + "learning_rate": 2.272752690102238e-06, + "loss": 0.9814, + "step": 6291 + }, + { + "epoch": 0.4728693822335788, + "grad_norm": 1.521910934493983, + "learning_rate": 2.272270388353173e-06, + "loss": 0.96, + "step": 6292 + }, + { + "epoch": 0.4729445362994138, + "grad_norm": 1.495179616669966, + "learning_rate": 2.2717880704710453e-06, + "loss": 0.9959, + "step": 6293 + }, + { + "epoch": 0.47301969036524877, + "grad_norm": 1.522007119085678, + "learning_rate": 2.2713057364844323e-06, + "loss": 0.8756, + "step": 6294 + }, + { + "epoch": 0.47309484443108374, + "grad_norm": 1.6292170312743808, + "learning_rate": 2.2708233864219175e-06, + "loss": 0.9646, + "step": 6295 + }, + { + "epoch": 0.4731699984969187, + "grad_norm": 1.6756874894097367, + "learning_rate": 2.270341020312078e-06, + "loss": 0.9893, + "step": 6296 + }, + { + "epoch": 0.4732451525627536, + "grad_norm": 1.913804577946816, + "learning_rate": 2.2698586381834993e-06, + "loss": 0.9323, + "step": 6297 + }, + { + "epoch": 0.4733203066285886, + "grad_norm": 1.9104869844841394, + "learning_rate": 2.269376240064763e-06, + "loss": 0.9712, + "step": 6298 + }, + { + "epoch": 0.47339546069442356, + "grad_norm": 3.0791957174194557, + "learning_rate": 2.2688938259844525e-06, + "loss": 1.0463, + "step": 6299 + }, + { + "epoch": 0.47347061476025853, + "grad_norm": 1.497539170633335, + "learning_rate": 2.268411395971153e-06, + "loss": 0.9189, + "step": 6300 + }, + { + "epoch": 0.4735457688260935, + "grad_norm": 1.7223753408161326, + "learning_rate": 2.2679289500534504e-06, + "loss": 0.9649, + "step": 6301 + }, + { + "epoch": 0.4736209228919285, + "grad_norm": 1.5225574005280826, + "learning_rate": 2.267446488259932e-06, + "loss": 0.8174, + "step": 6302 + }, + { + "epoch": 0.4736960769577634, + "grad_norm": 1.5985777792432507, + "learning_rate": 2.266964010619185e-06, + "loss": 1.042, + "step": 6303 + }, + { + "epoch": 0.47377123102359836, + "grad_norm": 1.6213427984286604, + "learning_rate": 2.2664815171597983e-06, + "loss": 1.0173, + "step": 6304 + }, + { + "epoch": 0.47384638508943333, + "grad_norm": 1.6368167820526014, + "learning_rate": 2.2659990079103604e-06, + "loss": 1.023, + "step": 6305 + }, + { + "epoch": 0.4739215391552683, + "grad_norm": 1.5916768221839086, + "learning_rate": 2.2655164828994635e-06, + "loss": 1.062, + "step": 6306 + }, + { + "epoch": 0.47399669322110327, + "grad_norm": 2.63629879468145, + "learning_rate": 2.2650339421556982e-06, + "loss": 1.0505, + "step": 6307 + }, + { + "epoch": 0.47407184728693824, + "grad_norm": 1.9283079088403872, + "learning_rate": 2.2645513857076567e-06, + "loss": 0.9382, + "step": 6308 + }, + { + "epoch": 0.4741470013527732, + "grad_norm": 1.4121915125932305, + "learning_rate": 2.2640688135839326e-06, + "loss": 0.8407, + "step": 6309 + }, + { + "epoch": 0.4742221554186081, + "grad_norm": 2.024766169104665, + "learning_rate": 2.26358622581312e-06, + "loss": 1.0091, + "step": 6310 + }, + { + "epoch": 0.4742973094844431, + "grad_norm": 1.6866766722511783, + "learning_rate": 2.2631036224238144e-06, + "loss": 0.9757, + "step": 6311 + }, + { + "epoch": 0.47437246355027807, + "grad_norm": 1.9353844567757525, + "learning_rate": 2.262621003444611e-06, + "loss": 0.9871, + "step": 6312 + }, + { + "epoch": 0.47444761761611304, + "grad_norm": 1.454337076004975, + "learning_rate": 2.2621383689041087e-06, + "loss": 0.9423, + "step": 6313 + }, + { + "epoch": 0.474522771681948, + "grad_norm": 4.636507718844464, + "learning_rate": 2.2616557188309033e-06, + "loss": 0.8974, + "step": 6314 + }, + { + "epoch": 0.474597925747783, + "grad_norm": 2.5628926976455286, + "learning_rate": 2.261173053253595e-06, + "loss": 0.9954, + "step": 6315 + }, + { + "epoch": 0.4746730798136179, + "grad_norm": 1.58170139971356, + "learning_rate": 2.260690372200783e-06, + "loss": 0.9597, + "step": 6316 + }, + { + "epoch": 0.47474823387945286, + "grad_norm": 1.367636034822762, + "learning_rate": 2.260207675701069e-06, + "loss": 0.9571, + "step": 6317 + }, + { + "epoch": 0.47482338794528783, + "grad_norm": 1.8031728093570685, + "learning_rate": 2.259724963783052e-06, + "loss": 0.8991, + "step": 6318 + }, + { + "epoch": 0.4748985420111228, + "grad_norm": 1.5705093336341114, + "learning_rate": 2.2592422364753377e-06, + "loss": 1.0354, + "step": 6319 + }, + { + "epoch": 0.4749736960769578, + "grad_norm": 1.5334385393511163, + "learning_rate": 2.2587594938065285e-06, + "loss": 1.0067, + "step": 6320 + }, + { + "epoch": 0.47504885014279274, + "grad_norm": 1.613969706760387, + "learning_rate": 2.2582767358052272e-06, + "loss": 1.0219, + "step": 6321 + }, + { + "epoch": 0.47512400420862766, + "grad_norm": 1.4862631584335741, + "learning_rate": 2.2577939625000414e-06, + "loss": 1.0069, + "step": 6322 + }, + { + "epoch": 0.47519915827446263, + "grad_norm": 0.6528084293455869, + "learning_rate": 2.2573111739195756e-06, + "loss": 0.803, + "step": 6323 + }, + { + "epoch": 0.4752743123402976, + "grad_norm": 1.7306800179434287, + "learning_rate": 2.2568283700924375e-06, + "loss": 1.0687, + "step": 6324 + }, + { + "epoch": 0.47534946640613257, + "grad_norm": 1.8274905416136906, + "learning_rate": 2.2563455510472353e-06, + "loss": 0.9804, + "step": 6325 + }, + { + "epoch": 0.47542462047196754, + "grad_norm": 1.3973430421203166, + "learning_rate": 2.255862716812577e-06, + "loss": 0.883, + "step": 6326 + }, + { + "epoch": 0.4754997745378025, + "grad_norm": 1.742923119399513, + "learning_rate": 2.2553798674170735e-06, + "loss": 0.9778, + "step": 6327 + }, + { + "epoch": 0.4755749286036375, + "grad_norm": 1.696753667213408, + "learning_rate": 2.2548970028893348e-06, + "loss": 0.9163, + "step": 6328 + }, + { + "epoch": 0.4756500826694724, + "grad_norm": 2.0750521549886045, + "learning_rate": 2.254414123257973e-06, + "loss": 1.0227, + "step": 6329 + }, + { + "epoch": 0.47572523673530737, + "grad_norm": 6.647879194940845, + "learning_rate": 2.2539312285516e-06, + "loss": 1.0405, + "step": 6330 + }, + { + "epoch": 0.47580039080114234, + "grad_norm": 1.7145522441185306, + "learning_rate": 2.2534483187988288e-06, + "loss": 0.9387, + "step": 6331 + }, + { + "epoch": 0.4758755448669773, + "grad_norm": 1.876237524449846, + "learning_rate": 2.2529653940282743e-06, + "loss": 1.0347, + "step": 6332 + }, + { + "epoch": 0.4759506989328123, + "grad_norm": 2.3812292623743296, + "learning_rate": 2.2524824542685515e-06, + "loss": 1.0955, + "step": 6333 + }, + { + "epoch": 0.47602585299864725, + "grad_norm": 1.3484893877250455, + "learning_rate": 2.2519994995482774e-06, + "loss": 0.868, + "step": 6334 + }, + { + "epoch": 0.47610100706448216, + "grad_norm": 0.6761695099276341, + "learning_rate": 2.2515165298960674e-06, + "loss": 0.8419, + "step": 6335 + }, + { + "epoch": 0.47617616113031713, + "grad_norm": 2.073044381608285, + "learning_rate": 2.251033545340539e-06, + "loss": 1.0483, + "step": 6336 + }, + { + "epoch": 0.4762513151961521, + "grad_norm": 1.6473294552184004, + "learning_rate": 2.2505505459103133e-06, + "loss": 1.0056, + "step": 6337 + }, + { + "epoch": 0.4763264692619871, + "grad_norm": 1.6178643473377856, + "learning_rate": 2.250067531634007e-06, + "loss": 0.8952, + "step": 6338 + }, + { + "epoch": 0.47640162332782204, + "grad_norm": 1.601989837947623, + "learning_rate": 2.249584502540242e-06, + "loss": 1.0184, + "step": 6339 + }, + { + "epoch": 0.476476777393657, + "grad_norm": 2.3974660092550732, + "learning_rate": 2.2491014586576404e-06, + "loss": 0.8994, + "step": 6340 + }, + { + "epoch": 0.476551931459492, + "grad_norm": 0.7560209747185094, + "learning_rate": 2.248618400014823e-06, + "loss": 0.8656, + "step": 6341 + }, + { + "epoch": 0.4766270855253269, + "grad_norm": 2.3088786664219807, + "learning_rate": 2.248135326640414e-06, + "loss": 0.9391, + "step": 6342 + }, + { + "epoch": 0.47670223959116187, + "grad_norm": 1.7510066543095373, + "learning_rate": 2.2476522385630354e-06, + "loss": 0.94, + "step": 6343 + }, + { + "epoch": 0.47677739365699684, + "grad_norm": 0.677576711823186, + "learning_rate": 2.2471691358113146e-06, + "loss": 0.8226, + "step": 6344 + }, + { + "epoch": 0.4768525477228318, + "grad_norm": 1.471234594116966, + "learning_rate": 2.246686018413875e-06, + "loss": 1.0093, + "step": 6345 + }, + { + "epoch": 0.4769277017886668, + "grad_norm": 2.3850045549201258, + "learning_rate": 2.246202886399345e-06, + "loss": 0.9008, + "step": 6346 + }, + { + "epoch": 0.47700285585450175, + "grad_norm": 1.7488278823317274, + "learning_rate": 2.245719739796351e-06, + "loss": 0.9078, + "step": 6347 + }, + { + "epoch": 0.47707800992033667, + "grad_norm": 1.727954976166394, + "learning_rate": 2.2452365786335214e-06, + "loss": 1.1029, + "step": 6348 + }, + { + "epoch": 0.47715316398617164, + "grad_norm": 0.7466581915760433, + "learning_rate": 2.2447534029394856e-06, + "loss": 0.8481, + "step": 6349 + }, + { + "epoch": 0.4772283180520066, + "grad_norm": 1.656418557856756, + "learning_rate": 2.244270212742873e-06, + "loss": 0.8437, + "step": 6350 + }, + { + "epoch": 0.4773034721178416, + "grad_norm": 0.7755670226727086, + "learning_rate": 2.2437870080723153e-06, + "loss": 0.8172, + "step": 6351 + }, + { + "epoch": 0.47737862618367655, + "grad_norm": 1.8618052652597428, + "learning_rate": 2.243303788956444e-06, + "loss": 0.94, + "step": 6352 + }, + { + "epoch": 0.4774537802495115, + "grad_norm": 1.6659056436029207, + "learning_rate": 2.2428205554238914e-06, + "loss": 0.8536, + "step": 6353 + }, + { + "epoch": 0.4775289343153465, + "grad_norm": 1.4030228779423601, + "learning_rate": 2.2423373075032913e-06, + "loss": 1.0025, + "step": 6354 + }, + { + "epoch": 0.4776040883811814, + "grad_norm": 2.2721180446073594, + "learning_rate": 2.241854045223277e-06, + "loss": 1.064, + "step": 6355 + }, + { + "epoch": 0.4776792424470164, + "grad_norm": 1.7894511904203279, + "learning_rate": 2.241370768612485e-06, + "loss": 1.0688, + "step": 6356 + }, + { + "epoch": 0.47775439651285134, + "grad_norm": 1.885562202639463, + "learning_rate": 2.2408874776995508e-06, + "loss": 1.013, + "step": 6357 + }, + { + "epoch": 0.4778295505786863, + "grad_norm": 2.945786819918122, + "learning_rate": 2.2404041725131106e-06, + "loss": 1.0514, + "step": 6358 + }, + { + "epoch": 0.4779047046445213, + "grad_norm": 1.3178758879069432, + "learning_rate": 2.239920853081803e-06, + "loss": 0.9952, + "step": 6359 + }, + { + "epoch": 0.47797985871035625, + "grad_norm": 1.5628352759184672, + "learning_rate": 2.2394375194342653e-06, + "loss": 0.995, + "step": 6360 + }, + { + "epoch": 0.47805501277619117, + "grad_norm": 1.4491132096266928, + "learning_rate": 2.2389541715991385e-06, + "loss": 0.992, + "step": 6361 + }, + { + "epoch": 0.47813016684202614, + "grad_norm": 4.849171612969709, + "learning_rate": 2.238470809605062e-06, + "loss": 0.9424, + "step": 6362 + }, + { + "epoch": 0.4782053209078611, + "grad_norm": 1.9118321717557232, + "learning_rate": 2.2379874334806764e-06, + "loss": 0.9216, + "step": 6363 + }, + { + "epoch": 0.4782804749736961, + "grad_norm": 1.5563004910654628, + "learning_rate": 2.237504043254625e-06, + "loss": 1.0771, + "step": 6364 + }, + { + "epoch": 0.47835562903953105, + "grad_norm": 1.5894842467749877, + "learning_rate": 2.2370206389555485e-06, + "loss": 0.7828, + "step": 6365 + }, + { + "epoch": 0.478430783105366, + "grad_norm": 1.8453408829382076, + "learning_rate": 2.2365372206120923e-06, + "loss": 0.9991, + "step": 6366 + }, + { + "epoch": 0.47850593717120093, + "grad_norm": 1.4883597445876726, + "learning_rate": 2.2360537882528996e-06, + "loss": 1.0276, + "step": 6367 + }, + { + "epoch": 0.4785810912370359, + "grad_norm": 1.669028238253044, + "learning_rate": 2.2355703419066163e-06, + "loss": 1.1188, + "step": 6368 + }, + { + "epoch": 0.4786562453028709, + "grad_norm": 1.8950299737215506, + "learning_rate": 2.2350868816018886e-06, + "loss": 0.8968, + "step": 6369 + }, + { + "epoch": 0.47873139936870585, + "grad_norm": 0.87862646100016, + "learning_rate": 2.234603407367362e-06, + "loss": 0.8971, + "step": 6370 + }, + { + "epoch": 0.4788065534345408, + "grad_norm": 3.242774413885898, + "learning_rate": 2.234119919231686e-06, + "loss": 1.0122, + "step": 6371 + }, + { + "epoch": 0.4788817075003758, + "grad_norm": 1.6415151451915728, + "learning_rate": 2.2336364172235074e-06, + "loss": 0.9316, + "step": 6372 + }, + { + "epoch": 0.47895686156621076, + "grad_norm": 2.0375731425818433, + "learning_rate": 2.2331529013714775e-06, + "loss": 1.032, + "step": 6373 + }, + { + "epoch": 0.47903201563204567, + "grad_norm": 1.452604221012538, + "learning_rate": 2.2326693717042446e-06, + "loss": 0.9336, + "step": 6374 + }, + { + "epoch": 0.47910716969788064, + "grad_norm": 1.6057437559098757, + "learning_rate": 2.2321858282504603e-06, + "loss": 0.9022, + "step": 6375 + }, + { + "epoch": 0.4791823237637156, + "grad_norm": 0.7265275516103304, + "learning_rate": 2.231702271038777e-06, + "loss": 0.8344, + "step": 6376 + }, + { + "epoch": 0.4792574778295506, + "grad_norm": 1.7342096198336814, + "learning_rate": 2.2312187000978467e-06, + "loss": 1.0162, + "step": 6377 + }, + { + "epoch": 0.47933263189538555, + "grad_norm": 1.6659612434339592, + "learning_rate": 2.230735115456324e-06, + "loss": 0.9533, + "step": 6378 + }, + { + "epoch": 0.4794077859612205, + "grad_norm": 1.8552928319154856, + "learning_rate": 2.2302515171428613e-06, + "loss": 0.9564, + "step": 6379 + }, + { + "epoch": 0.47948294002705544, + "grad_norm": 1.4686388825934067, + "learning_rate": 2.229767905186114e-06, + "loss": 0.966, + "step": 6380 + }, + { + "epoch": 0.4795580940928904, + "grad_norm": 2.7711338845834605, + "learning_rate": 2.2292842796147395e-06, + "loss": 0.8771, + "step": 6381 + }, + { + "epoch": 0.4796332481587254, + "grad_norm": 1.4456098096076655, + "learning_rate": 2.2288006404573922e-06, + "loss": 0.9706, + "step": 6382 + }, + { + "epoch": 0.47970840222456035, + "grad_norm": 1.4240191714785744, + "learning_rate": 2.228316987742732e-06, + "loss": 1.0487, + "step": 6383 + }, + { + "epoch": 0.4797835562903953, + "grad_norm": 1.53446871282925, + "learning_rate": 2.227833321499415e-06, + "loss": 0.9986, + "step": 6384 + }, + { + "epoch": 0.4798587103562303, + "grad_norm": 1.98981654879097, + "learning_rate": 2.227349641756102e-06, + "loss": 0.7955, + "step": 6385 + }, + { + "epoch": 0.47993386442206526, + "grad_norm": 1.6248613445108164, + "learning_rate": 2.2268659485414526e-06, + "loss": 1.0022, + "step": 6386 + }, + { + "epoch": 0.4800090184879002, + "grad_norm": 1.617271458017437, + "learning_rate": 2.226382241884126e-06, + "loss": 1.0101, + "step": 6387 + }, + { + "epoch": 0.48008417255373514, + "grad_norm": 1.8481925847245229, + "learning_rate": 2.225898521812785e-06, + "loss": 1.0449, + "step": 6388 + }, + { + "epoch": 0.4801593266195701, + "grad_norm": 1.8935104294120257, + "learning_rate": 2.2254147883560916e-06, + "loss": 1.0018, + "step": 6389 + }, + { + "epoch": 0.4802344806854051, + "grad_norm": 1.4347247825436409, + "learning_rate": 2.2249310415427087e-06, + "loss": 0.9856, + "step": 6390 + }, + { + "epoch": 0.48030963475124006, + "grad_norm": 2.002063640242257, + "learning_rate": 2.224447281401301e-06, + "loss": 0.854, + "step": 6391 + }, + { + "epoch": 0.480384788817075, + "grad_norm": 1.4794993586098897, + "learning_rate": 2.2239635079605316e-06, + "loss": 1.0088, + "step": 6392 + }, + { + "epoch": 0.48045994288290994, + "grad_norm": 1.6448842301836282, + "learning_rate": 2.223479721249067e-06, + "loss": 0.8516, + "step": 6393 + }, + { + "epoch": 0.4805350969487449, + "grad_norm": 1.8175528240267025, + "learning_rate": 2.222995921295573e-06, + "loss": 0.9024, + "step": 6394 + }, + { + "epoch": 0.4806102510145799, + "grad_norm": 1.8365115303157413, + "learning_rate": 2.2225121081287174e-06, + "loss": 0.9073, + "step": 6395 + }, + { + "epoch": 0.48068540508041485, + "grad_norm": 2.115337844811762, + "learning_rate": 2.2220282817771668e-06, + "loss": 0.9492, + "step": 6396 + }, + { + "epoch": 0.4807605591462498, + "grad_norm": 1.6572343312465228, + "learning_rate": 2.2215444422695906e-06, + "loss": 0.8553, + "step": 6397 + }, + { + "epoch": 0.4808357132120848, + "grad_norm": 2.3916661606986365, + "learning_rate": 2.2210605896346575e-06, + "loss": 0.9964, + "step": 6398 + }, + { + "epoch": 0.48091086727791976, + "grad_norm": 1.7613983201222592, + "learning_rate": 2.2205767239010376e-06, + "loss": 1.0558, + "step": 6399 + }, + { + "epoch": 0.4809860213437547, + "grad_norm": 1.9776276142403755, + "learning_rate": 2.2200928450974028e-06, + "loss": 1.0683, + "step": 6400 + }, + { + "epoch": 0.48106117540958965, + "grad_norm": 2.1512538137185993, + "learning_rate": 2.2196089532524244e-06, + "loss": 1.0193, + "step": 6401 + }, + { + "epoch": 0.4811363294754246, + "grad_norm": 1.7595914527994192, + "learning_rate": 2.2191250483947736e-06, + "loss": 1.0435, + "step": 6402 + }, + { + "epoch": 0.4812114835412596, + "grad_norm": 1.6907291598854004, + "learning_rate": 2.2186411305531254e-06, + "loss": 1.0264, + "step": 6403 + }, + { + "epoch": 0.48128663760709456, + "grad_norm": 1.2431033135349785, + "learning_rate": 2.2181571997561523e-06, + "loss": 0.9916, + "step": 6404 + }, + { + "epoch": 0.48136179167292953, + "grad_norm": 1.6596776653925693, + "learning_rate": 2.2176732560325302e-06, + "loss": 1.0821, + "step": 6405 + }, + { + "epoch": 0.48143694573876444, + "grad_norm": 1.4357798692192423, + "learning_rate": 2.217189299410934e-06, + "loss": 0.9332, + "step": 6406 + }, + { + "epoch": 0.4815120998045994, + "grad_norm": 2.6332677792787607, + "learning_rate": 2.21670532992004e-06, + "loss": 0.9791, + "step": 6407 + }, + { + "epoch": 0.4815872538704344, + "grad_norm": 4.076673227358446, + "learning_rate": 2.2162213475885262e-06, + "loss": 1.0862, + "step": 6408 + }, + { + "epoch": 0.48166240793626935, + "grad_norm": 1.7871989576889933, + "learning_rate": 2.215737352445069e-06, + "loss": 0.8562, + "step": 6409 + }, + { + "epoch": 0.4817375620021043, + "grad_norm": 1.786062398766686, + "learning_rate": 2.2152533445183477e-06, + "loss": 1.0027, + "step": 6410 + }, + { + "epoch": 0.4818127160679393, + "grad_norm": 1.4803088260320774, + "learning_rate": 2.214769323837041e-06, + "loss": 0.9986, + "step": 6411 + }, + { + "epoch": 0.4818878701337742, + "grad_norm": 0.7633512170090422, + "learning_rate": 2.21428529042983e-06, + "loss": 0.8468, + "step": 6412 + }, + { + "epoch": 0.4819630241996092, + "grad_norm": 1.587387399359637, + "learning_rate": 2.213801244325395e-06, + "loss": 0.9875, + "step": 6413 + }, + { + "epoch": 0.48203817826544415, + "grad_norm": 1.9379868049644793, + "learning_rate": 2.2133171855524167e-06, + "loss": 1.0378, + "step": 6414 + }, + { + "epoch": 0.4821133323312791, + "grad_norm": 0.7271873494448504, + "learning_rate": 2.21283311413958e-06, + "loss": 0.8325, + "step": 6415 + }, + { + "epoch": 0.4821884863971141, + "grad_norm": 1.4274382695147088, + "learning_rate": 2.2123490301155647e-06, + "loss": 0.8606, + "step": 6416 + }, + { + "epoch": 0.48226364046294906, + "grad_norm": 1.6957235936579573, + "learning_rate": 2.2118649335090568e-06, + "loss": 1.0334, + "step": 6417 + }, + { + "epoch": 0.48233879452878403, + "grad_norm": 1.8138824978154864, + "learning_rate": 2.2113808243487404e-06, + "loss": 1.0128, + "step": 6418 + }, + { + "epoch": 0.48241394859461895, + "grad_norm": 2.1476556638492745, + "learning_rate": 2.210896702663301e-06, + "loss": 1.0053, + "step": 6419 + }, + { + "epoch": 0.4824891026604539, + "grad_norm": 1.712485810381165, + "learning_rate": 2.2104125684814238e-06, + "loss": 0.9279, + "step": 6420 + }, + { + "epoch": 0.4825642567262889, + "grad_norm": 1.5606416714060056, + "learning_rate": 2.209928421831796e-06, + "loss": 1.0253, + "step": 6421 + }, + { + "epoch": 0.48263941079212386, + "grad_norm": 1.5251016770533539, + "learning_rate": 2.209444262743106e-06, + "loss": 1.0044, + "step": 6422 + }, + { + "epoch": 0.4827145648579588, + "grad_norm": 1.3952618563048664, + "learning_rate": 2.2089600912440413e-06, + "loss": 0.8717, + "step": 6423 + }, + { + "epoch": 0.4827897189237938, + "grad_norm": 2.4932716427105337, + "learning_rate": 2.2084759073632904e-06, + "loss": 0.9544, + "step": 6424 + }, + { + "epoch": 0.4828648729896287, + "grad_norm": 1.746459171708, + "learning_rate": 2.2079917111295448e-06, + "loss": 0.9765, + "step": 6425 + }, + { + "epoch": 0.4829400270554637, + "grad_norm": 1.7246347427662534, + "learning_rate": 2.2075075025714922e-06, + "loss": 0.8314, + "step": 6426 + }, + { + "epoch": 0.48301518112129865, + "grad_norm": 1.9360958056192303, + "learning_rate": 2.2070232817178272e-06, + "loss": 1.0678, + "step": 6427 + }, + { + "epoch": 0.4830903351871336, + "grad_norm": 1.6280161669300974, + "learning_rate": 2.206539048597239e-06, + "loss": 0.9588, + "step": 6428 + }, + { + "epoch": 0.4831654892529686, + "grad_norm": 1.7240495302294323, + "learning_rate": 2.206054803238422e-06, + "loss": 0.9988, + "step": 6429 + }, + { + "epoch": 0.48324064331880356, + "grad_norm": 0.8008072267366321, + "learning_rate": 2.2055705456700686e-06, + "loss": 0.8986, + "step": 6430 + }, + { + "epoch": 0.48331579738463853, + "grad_norm": 0.8998061660148342, + "learning_rate": 2.2050862759208728e-06, + "loss": 0.8202, + "step": 6431 + }, + { + "epoch": 0.48339095145047345, + "grad_norm": 1.8215977138767967, + "learning_rate": 2.2046019940195303e-06, + "loss": 1.0611, + "step": 6432 + }, + { + "epoch": 0.4834661055163084, + "grad_norm": 1.6712942808300926, + "learning_rate": 2.2041176999947353e-06, + "loss": 0.9046, + "step": 6433 + }, + { + "epoch": 0.4835412595821434, + "grad_norm": 1.3976923912838515, + "learning_rate": 2.203633393875186e-06, + "loss": 1.0474, + "step": 6434 + }, + { + "epoch": 0.48361641364797836, + "grad_norm": 1.6964614420409516, + "learning_rate": 2.2031490756895784e-06, + "loss": 1.0027, + "step": 6435 + }, + { + "epoch": 0.48369156771381333, + "grad_norm": 3.276134608920861, + "learning_rate": 2.2026647454666097e-06, + "loss": 1.0391, + "step": 6436 + }, + { + "epoch": 0.4837667217796483, + "grad_norm": 1.6043032385748899, + "learning_rate": 2.202180403234979e-06, + "loss": 0.944, + "step": 6437 + }, + { + "epoch": 0.4838418758454832, + "grad_norm": 1.3870108283403377, + "learning_rate": 2.2016960490233845e-06, + "loss": 0.9838, + "step": 6438 + }, + { + "epoch": 0.4839170299113182, + "grad_norm": 1.7724970538030116, + "learning_rate": 2.2012116828605275e-06, + "loss": 0.9873, + "step": 6439 + }, + { + "epoch": 0.48399218397715316, + "grad_norm": 1.4872190835303964, + "learning_rate": 2.200727304775108e-06, + "loss": 0.9586, + "step": 6440 + }, + { + "epoch": 0.4840673380429881, + "grad_norm": 1.6894210194971455, + "learning_rate": 2.200242914795826e-06, + "loss": 0.9196, + "step": 6441 + }, + { + "epoch": 0.4841424921088231, + "grad_norm": 2.152341866881833, + "learning_rate": 2.1997585129513852e-06, + "loss": 0.8764, + "step": 6442 + }, + { + "epoch": 0.48421764617465807, + "grad_norm": 1.8086143805560988, + "learning_rate": 2.1992740992704877e-06, + "loss": 0.9648, + "step": 6443 + }, + { + "epoch": 0.48429280024049304, + "grad_norm": 1.8436381419243186, + "learning_rate": 2.1987896737818365e-06, + "loss": 1.1255, + "step": 6444 + }, + { + "epoch": 0.48436795430632795, + "grad_norm": 2.150343171166914, + "learning_rate": 2.198305236514136e-06, + "loss": 1.111, + "step": 6445 + }, + { + "epoch": 0.4844431083721629, + "grad_norm": 0.6838497227574973, + "learning_rate": 2.1978207874960908e-06, + "loss": 0.8017, + "step": 6446 + }, + { + "epoch": 0.4845182624379979, + "grad_norm": 1.3067657912537036, + "learning_rate": 2.1973363267564063e-06, + "loss": 1.0187, + "step": 6447 + }, + { + "epoch": 0.48459341650383286, + "grad_norm": 2.390470611972484, + "learning_rate": 2.196851854323789e-06, + "loss": 1.0694, + "step": 6448 + }, + { + "epoch": 0.48466857056966783, + "grad_norm": 2.150759465238554, + "learning_rate": 2.1963673702269454e-06, + "loss": 1.0088, + "step": 6449 + }, + { + "epoch": 0.4847437246355028, + "grad_norm": 1.7333910788796272, + "learning_rate": 2.195882874494583e-06, + "loss": 0.9514, + "step": 6450 + }, + { + "epoch": 0.4848188787013377, + "grad_norm": 0.8893931385798988, + "learning_rate": 2.195398367155411e-06, + "loss": 0.8844, + "step": 6451 + }, + { + "epoch": 0.4848940327671727, + "grad_norm": 1.3593929333435666, + "learning_rate": 2.194913848238137e-06, + "loss": 0.934, + "step": 6452 + }, + { + "epoch": 0.48496918683300766, + "grad_norm": 1.7099186201975707, + "learning_rate": 2.194429317771471e-06, + "loss": 0.9263, + "step": 6453 + }, + { + "epoch": 0.48504434089884263, + "grad_norm": 1.6461930022803632, + "learning_rate": 2.1939447757841236e-06, + "loss": 1.0444, + "step": 6454 + }, + { + "epoch": 0.4851194949646776, + "grad_norm": 1.3866087963946037, + "learning_rate": 2.1934602223048054e-06, + "loss": 0.9497, + "step": 6455 + }, + { + "epoch": 0.48519464903051257, + "grad_norm": 1.6069152701163967, + "learning_rate": 2.1929756573622282e-06, + "loss": 0.9771, + "step": 6456 + }, + { + "epoch": 0.4852698030963475, + "grad_norm": 1.3597603175085167, + "learning_rate": 2.192491080985105e-06, + "loss": 0.9825, + "step": 6457 + }, + { + "epoch": 0.48534495716218246, + "grad_norm": 1.607220438946592, + "learning_rate": 2.192006493202147e-06, + "loss": 1.0041, + "step": 6458 + }, + { + "epoch": 0.4854201112280174, + "grad_norm": 1.9274907136883903, + "learning_rate": 2.19152189404207e-06, + "loss": 0.9275, + "step": 6459 + }, + { + "epoch": 0.4854952652938524, + "grad_norm": 1.4794077430521433, + "learning_rate": 2.191037283533587e-06, + "loss": 1.0156, + "step": 6460 + }, + { + "epoch": 0.48557041935968737, + "grad_norm": 2.4400348687416433, + "learning_rate": 2.1905526617054136e-06, + "loss": 0.891, + "step": 6461 + }, + { + "epoch": 0.48564557342552234, + "grad_norm": 2.466267556902828, + "learning_rate": 2.1900680285862655e-06, + "loss": 0.9815, + "step": 6462 + }, + { + "epoch": 0.4857207274913573, + "grad_norm": 1.9840197508005741, + "learning_rate": 2.1895833842048583e-06, + "loss": 1.026, + "step": 6463 + }, + { + "epoch": 0.4857958815571922, + "grad_norm": 1.653315193181863, + "learning_rate": 2.18909872858991e-06, + "loss": 1.0162, + "step": 6464 + }, + { + "epoch": 0.4858710356230272, + "grad_norm": 8.719462703985345, + "learning_rate": 2.1886140617701374e-06, + "loss": 0.9588, + "step": 6465 + }, + { + "epoch": 0.48594618968886216, + "grad_norm": 10.779930498398075, + "learning_rate": 2.1881293837742604e-06, + "loss": 0.8524, + "step": 6466 + }, + { + "epoch": 0.48602134375469713, + "grad_norm": 1.9379563563248208, + "learning_rate": 2.1876446946309965e-06, + "loss": 1.0398, + "step": 6467 + }, + { + "epoch": 0.4860964978205321, + "grad_norm": 1.3295728084889222, + "learning_rate": 2.187159994369065e-06, + "loss": 0.9289, + "step": 6468 + }, + { + "epoch": 0.4861716518863671, + "grad_norm": 2.3427680438076197, + "learning_rate": 2.1866752830171884e-06, + "loss": 0.8907, + "step": 6469 + }, + { + "epoch": 0.486246805952202, + "grad_norm": 1.5987302718984668, + "learning_rate": 2.1861905606040857e-06, + "loss": 1.0081, + "step": 6470 + }, + { + "epoch": 0.48632196001803696, + "grad_norm": 1.7918184201342562, + "learning_rate": 2.18570582715848e-06, + "loss": 1.0111, + "step": 6471 + }, + { + "epoch": 0.48639711408387193, + "grad_norm": 2.1578852768985692, + "learning_rate": 2.1852210827090922e-06, + "loss": 1.0125, + "step": 6472 + }, + { + "epoch": 0.4864722681497069, + "grad_norm": 1.9705024065043175, + "learning_rate": 2.1847363272846465e-06, + "loss": 0.8809, + "step": 6473 + }, + { + "epoch": 0.48654742221554187, + "grad_norm": 1.9306592463663375, + "learning_rate": 2.184251560913866e-06, + "loss": 0.9549, + "step": 6474 + }, + { + "epoch": 0.48662257628137684, + "grad_norm": 4.8533708086418414, + "learning_rate": 2.1837667836254746e-06, + "loss": 1.0061, + "step": 6475 + }, + { + "epoch": 0.4866977303472118, + "grad_norm": 1.7229968921414163, + "learning_rate": 2.183281995448198e-06, + "loss": 0.9833, + "step": 6476 + }, + { + "epoch": 0.4867728844130467, + "grad_norm": 2.714303073970953, + "learning_rate": 2.1827971964107607e-06, + "loss": 0.898, + "step": 6477 + }, + { + "epoch": 0.4868480384788817, + "grad_norm": 2.378512946628686, + "learning_rate": 2.1823123865418903e-06, + "loss": 0.9942, + "step": 6478 + }, + { + "epoch": 0.48692319254471667, + "grad_norm": 1.585762126979077, + "learning_rate": 2.1818275658703128e-06, + "loss": 1.068, + "step": 6479 + }, + { + "epoch": 0.48699834661055164, + "grad_norm": 2.046698293265523, + "learning_rate": 2.1813427344247557e-06, + "loss": 1.0511, + "step": 6480 + }, + { + "epoch": 0.4870735006763866, + "grad_norm": 0.8744603604689023, + "learning_rate": 2.1808578922339466e-06, + "loss": 0.8878, + "step": 6481 + }, + { + "epoch": 0.4871486547422216, + "grad_norm": 1.3210027120794832, + "learning_rate": 2.180373039326615e-06, + "loss": 0.9837, + "step": 6482 + }, + { + "epoch": 0.4872238088080565, + "grad_norm": 1.3310991980876317, + "learning_rate": 2.1798881757314905e-06, + "loss": 0.909, + "step": 6483 + }, + { + "epoch": 0.48729896287389146, + "grad_norm": 1.5460688484426417, + "learning_rate": 2.1794033014773033e-06, + "loss": 0.9998, + "step": 6484 + }, + { + "epoch": 0.48737411693972643, + "grad_norm": 2.0156214514412802, + "learning_rate": 2.1789184165927824e-06, + "loss": 0.8402, + "step": 6485 + }, + { + "epoch": 0.4874492710055614, + "grad_norm": 2.142754936051461, + "learning_rate": 2.178433521106661e-06, + "loss": 1.0635, + "step": 6486 + }, + { + "epoch": 0.4875244250713964, + "grad_norm": 1.7396500245366835, + "learning_rate": 2.177948615047669e-06, + "loss": 1.0555, + "step": 6487 + }, + { + "epoch": 0.48759957913723134, + "grad_norm": 1.8398663626737048, + "learning_rate": 2.1774636984445417e-06, + "loss": 0.9936, + "step": 6488 + }, + { + "epoch": 0.4876747332030663, + "grad_norm": 2.027063956826399, + "learning_rate": 2.1769787713260097e-06, + "loss": 1.0778, + "step": 6489 + }, + { + "epoch": 0.4877498872689012, + "grad_norm": 1.6505051793176373, + "learning_rate": 2.1764938337208075e-06, + "loss": 1.0347, + "step": 6490 + }, + { + "epoch": 0.4878250413347362, + "grad_norm": 1.6892199052131533, + "learning_rate": 2.1760088856576706e-06, + "loss": 1.0604, + "step": 6491 + }, + { + "epoch": 0.48790019540057117, + "grad_norm": 1.2653135693583983, + "learning_rate": 2.175523927165333e-06, + "loss": 0.8999, + "step": 6492 + }, + { + "epoch": 0.48797534946640614, + "grad_norm": 1.6544928676697044, + "learning_rate": 2.1750389582725307e-06, + "loss": 0.9203, + "step": 6493 + }, + { + "epoch": 0.4880505035322411, + "grad_norm": 0.7376752515857252, + "learning_rate": 2.1745539790079993e-06, + "loss": 0.8123, + "step": 6494 + }, + { + "epoch": 0.4881256575980761, + "grad_norm": 2.214214255678535, + "learning_rate": 2.1740689894004773e-06, + "loss": 1.0144, + "step": 6495 + }, + { + "epoch": 0.488200811663911, + "grad_norm": 1.441020634870747, + "learning_rate": 2.1735839894787003e-06, + "loss": 0.9911, + "step": 6496 + }, + { + "epoch": 0.48827596572974596, + "grad_norm": 1.8439586973881328, + "learning_rate": 2.1730989792714074e-06, + "loss": 1.0378, + "step": 6497 + }, + { + "epoch": 0.48835111979558093, + "grad_norm": 1.3409524228775391, + "learning_rate": 2.1726139588073374e-06, + "loss": 0.9265, + "step": 6498 + }, + { + "epoch": 0.4884262738614159, + "grad_norm": 1.9207201954034747, + "learning_rate": 2.172128928115229e-06, + "loss": 0.9433, + "step": 6499 + }, + { + "epoch": 0.4885014279272509, + "grad_norm": 0.663231138531299, + "learning_rate": 2.1716438872238227e-06, + "loss": 0.7854, + "step": 6500 + }, + { + "epoch": 0.48857658199308585, + "grad_norm": 2.1916551635595027, + "learning_rate": 2.171158836161859e-06, + "loss": 0.9731, + "step": 6501 + }, + { + "epoch": 0.48865173605892076, + "grad_norm": 0.7768078974671027, + "learning_rate": 2.1706737749580783e-06, + "loss": 0.9014, + "step": 6502 + }, + { + "epoch": 0.48872689012475573, + "grad_norm": 3.949980921638867, + "learning_rate": 2.1701887036412236e-06, + "loss": 1.0605, + "step": 6503 + }, + { + "epoch": 0.4888020441905907, + "grad_norm": 1.887005707142361, + "learning_rate": 2.1697036222400355e-06, + "loss": 1.0229, + "step": 6504 + }, + { + "epoch": 0.48887719825642567, + "grad_norm": 2.013312380768, + "learning_rate": 2.169218530783259e-06, + "loss": 1.0123, + "step": 6505 + }, + { + "epoch": 0.48895235232226064, + "grad_norm": 1.541120441000256, + "learning_rate": 2.1687334292996363e-06, + "loss": 0.913, + "step": 6506 + }, + { + "epoch": 0.4890275063880956, + "grad_norm": 1.5158031398261005, + "learning_rate": 2.168248317817911e-06, + "loss": 0.9511, + "step": 6507 + }, + { + "epoch": 0.4891026604539306, + "grad_norm": 0.8202564401998692, + "learning_rate": 2.1677631963668298e-06, + "loss": 0.9618, + "step": 6508 + }, + { + "epoch": 0.4891778145197655, + "grad_norm": 2.2983450823680207, + "learning_rate": 2.1672780649751353e-06, + "loss": 0.9729, + "step": 6509 + }, + { + "epoch": 0.48925296858560047, + "grad_norm": 1.7337062336761382, + "learning_rate": 2.166792923671576e-06, + "loss": 0.946, + "step": 6510 + }, + { + "epoch": 0.48932812265143544, + "grad_norm": 1.723779982208573, + "learning_rate": 2.166307772484898e-06, + "loss": 0.9944, + "step": 6511 + }, + { + "epoch": 0.4894032767172704, + "grad_norm": 1.5416212161126046, + "learning_rate": 2.1658226114438457e-06, + "loss": 1.0305, + "step": 6512 + }, + { + "epoch": 0.4894784307831054, + "grad_norm": 1.8798084908197972, + "learning_rate": 2.1653374405771696e-06, + "loss": 1.0659, + "step": 6513 + }, + { + "epoch": 0.48955358484894035, + "grad_norm": 1.7496220316751685, + "learning_rate": 2.1648522599136173e-06, + "loss": 0.8521, + "step": 6514 + }, + { + "epoch": 0.48962873891477526, + "grad_norm": 1.6064767017022632, + "learning_rate": 2.1643670694819375e-06, + "loss": 0.96, + "step": 6515 + }, + { + "epoch": 0.48970389298061023, + "grad_norm": 1.447420634464118, + "learning_rate": 2.163881869310879e-06, + "loss": 1.0279, + "step": 6516 + }, + { + "epoch": 0.4897790470464452, + "grad_norm": 1.5401797039904488, + "learning_rate": 2.163396659429192e-06, + "loss": 0.9826, + "step": 6517 + }, + { + "epoch": 0.4898542011122802, + "grad_norm": 1.8023166689183772, + "learning_rate": 2.162911439865628e-06, + "loss": 0.9975, + "step": 6518 + }, + { + "epoch": 0.48992935517811514, + "grad_norm": 2.2754340040795267, + "learning_rate": 2.1624262106489368e-06, + "loss": 0.9019, + "step": 6519 + }, + { + "epoch": 0.4900045092439501, + "grad_norm": 1.7502221920102115, + "learning_rate": 2.161940971807871e-06, + "loss": 1.0103, + "step": 6520 + }, + { + "epoch": 0.4900796633097851, + "grad_norm": 2.0602904535355417, + "learning_rate": 2.1614557233711817e-06, + "loss": 0.9288, + "step": 6521 + }, + { + "epoch": 0.49015481737562, + "grad_norm": 3.3877648450633018, + "learning_rate": 2.1609704653676234e-06, + "loss": 0.9724, + "step": 6522 + }, + { + "epoch": 0.49022997144145497, + "grad_norm": 1.4733593625722465, + "learning_rate": 2.1604851978259485e-06, + "loss": 0.8662, + "step": 6523 + }, + { + "epoch": 0.49030512550728994, + "grad_norm": 2.1519355179084907, + "learning_rate": 2.159999920774911e-06, + "loss": 0.9579, + "step": 6524 + }, + { + "epoch": 0.4903802795731249, + "grad_norm": 1.8468630743529215, + "learning_rate": 2.1595146342432655e-06, + "loss": 1.0096, + "step": 6525 + }, + { + "epoch": 0.4904554336389599, + "grad_norm": 1.78760539190999, + "learning_rate": 2.1590293382597667e-06, + "loss": 1.0256, + "step": 6526 + }, + { + "epoch": 0.49053058770479485, + "grad_norm": 2.1705733316226983, + "learning_rate": 2.158544032853171e-06, + "loss": 1.0226, + "step": 6527 + }, + { + "epoch": 0.49060574177062977, + "grad_norm": 2.3126114483345486, + "learning_rate": 2.1580587180522345e-06, + "loss": 0.8795, + "step": 6528 + }, + { + "epoch": 0.49068089583646474, + "grad_norm": 1.4874869402143838, + "learning_rate": 2.1575733938857134e-06, + "loss": 1.0679, + "step": 6529 + }, + { + "epoch": 0.4907560499022997, + "grad_norm": 1.4002014628278057, + "learning_rate": 2.1570880603823654e-06, + "loss": 0.9909, + "step": 6530 + }, + { + "epoch": 0.4908312039681347, + "grad_norm": 2.431553752527018, + "learning_rate": 2.156602717570948e-06, + "loss": 0.8724, + "step": 6531 + }, + { + "epoch": 0.49090635803396965, + "grad_norm": 2.9062135755655945, + "learning_rate": 2.15611736548022e-06, + "loss": 0.946, + "step": 6532 + }, + { + "epoch": 0.4909815120998046, + "grad_norm": 2.3672442035996872, + "learning_rate": 2.1556320041389407e-06, + "loss": 1.0372, + "step": 6533 + }, + { + "epoch": 0.4910566661656396, + "grad_norm": 1.5018898661651998, + "learning_rate": 2.155146633575869e-06, + "loss": 0.9648, + "step": 6534 + }, + { + "epoch": 0.4911318202314745, + "grad_norm": 1.9772881544143126, + "learning_rate": 2.154661253819765e-06, + "loss": 0.8152, + "step": 6535 + }, + { + "epoch": 0.4912069742973095, + "grad_norm": 1.6179675740928927, + "learning_rate": 2.1541758648993895e-06, + "loss": 1.0443, + "step": 6536 + }, + { + "epoch": 0.49128212836314444, + "grad_norm": 1.6069663828886567, + "learning_rate": 2.1536904668435035e-06, + "loss": 0.9545, + "step": 6537 + }, + { + "epoch": 0.4913572824289794, + "grad_norm": 3.110533685307729, + "learning_rate": 2.153205059680869e-06, + "loss": 1.0721, + "step": 6538 + }, + { + "epoch": 0.4914324364948144, + "grad_norm": 1.4631669276315071, + "learning_rate": 2.1527196434402483e-06, + "loss": 0.9912, + "step": 6539 + }, + { + "epoch": 0.49150759056064935, + "grad_norm": 1.8189269592611, + "learning_rate": 2.152234218150404e-06, + "loss": 0.8633, + "step": 6540 + }, + { + "epoch": 0.49158274462648427, + "grad_norm": 9.308965024855576, + "learning_rate": 2.1517487838400984e-06, + "loss": 1.0101, + "step": 6541 + }, + { + "epoch": 0.49165789869231924, + "grad_norm": 1.6446735383951852, + "learning_rate": 2.1512633405380972e-06, + "loss": 1.0181, + "step": 6542 + }, + { + "epoch": 0.4917330527581542, + "grad_norm": 3.8160616388201087, + "learning_rate": 2.1507778882731635e-06, + "loss": 0.9672, + "step": 6543 + }, + { + "epoch": 0.4918082068239892, + "grad_norm": 2.0316737979996375, + "learning_rate": 2.1502924270740626e-06, + "loss": 0.9365, + "step": 6544 + }, + { + "epoch": 0.49188336088982415, + "grad_norm": 2.46732163454196, + "learning_rate": 2.1498069569695605e-06, + "loss": 0.9241, + "step": 6545 + }, + { + "epoch": 0.4919585149556591, + "grad_norm": 0.7475472716627098, + "learning_rate": 2.149321477988421e-06, + "loss": 0.8858, + "step": 6546 + }, + { + "epoch": 0.49203366902149404, + "grad_norm": 2.023745598037202, + "learning_rate": 2.1488359901594137e-06, + "loss": 0.9729, + "step": 6547 + }, + { + "epoch": 0.492108823087329, + "grad_norm": 1.5839514111570765, + "learning_rate": 2.148350493511303e-06, + "loss": 0.9582, + "step": 6548 + }, + { + "epoch": 0.492183977153164, + "grad_norm": 4.7949359963514055, + "learning_rate": 2.1478649880728582e-06, + "loss": 1.0454, + "step": 6549 + }, + { + "epoch": 0.49225913121899895, + "grad_norm": 1.792589334472402, + "learning_rate": 2.1473794738728466e-06, + "loss": 1.0364, + "step": 6550 + }, + { + "epoch": 0.4923342852848339, + "grad_norm": 1.3737122834766595, + "learning_rate": 2.1468939509400363e-06, + "loss": 1.0263, + "step": 6551 + }, + { + "epoch": 0.4924094393506689, + "grad_norm": 1.5589196956620932, + "learning_rate": 2.1464084193031976e-06, + "loss": 0.9021, + "step": 6552 + }, + { + "epoch": 0.49248459341650386, + "grad_norm": 1.814974181356776, + "learning_rate": 2.1459228789910983e-06, + "loss": 0.9457, + "step": 6553 + }, + { + "epoch": 0.4925597474823388, + "grad_norm": 1.43713523549077, + "learning_rate": 2.1454373300325103e-06, + "loss": 0.9842, + "step": 6554 + }, + { + "epoch": 0.49263490154817374, + "grad_norm": 1.4701446242779432, + "learning_rate": 2.1449517724562037e-06, + "loss": 1.0106, + "step": 6555 + }, + { + "epoch": 0.4927100556140087, + "grad_norm": 1.7262476293062785, + "learning_rate": 2.1444662062909495e-06, + "loss": 0.9534, + "step": 6556 + }, + { + "epoch": 0.4927852096798437, + "grad_norm": 4.732751549082393, + "learning_rate": 2.1439806315655197e-06, + "loss": 1.0586, + "step": 6557 + }, + { + "epoch": 0.49286036374567865, + "grad_norm": 1.631702392519514, + "learning_rate": 2.1434950483086856e-06, + "loss": 1.025, + "step": 6558 + }, + { + "epoch": 0.4929355178115136, + "grad_norm": 1.8636755056351857, + "learning_rate": 2.143009456549221e-06, + "loss": 0.9489, + "step": 6559 + }, + { + "epoch": 0.49301067187734854, + "grad_norm": 1.563031983412577, + "learning_rate": 2.1425238563158975e-06, + "loss": 0.9846, + "step": 6560 + }, + { + "epoch": 0.4930858259431835, + "grad_norm": 1.6580420102409947, + "learning_rate": 2.1420382476374905e-06, + "loss": 1.0078, + "step": 6561 + }, + { + "epoch": 0.4931609800090185, + "grad_norm": 1.892942600171959, + "learning_rate": 2.1415526305427735e-06, + "loss": 1.0126, + "step": 6562 + }, + { + "epoch": 0.49323613407485345, + "grad_norm": 1.7614633582291837, + "learning_rate": 2.1410670050605203e-06, + "loss": 0.8628, + "step": 6563 + }, + { + "epoch": 0.4933112881406884, + "grad_norm": 0.7004819998678228, + "learning_rate": 2.140581371219508e-06, + "loss": 0.8137, + "step": 6564 + }, + { + "epoch": 0.4933864422065234, + "grad_norm": 1.7086832882095904, + "learning_rate": 2.14009572904851e-06, + "loss": 1.0269, + "step": 6565 + }, + { + "epoch": 0.49346159627235836, + "grad_norm": 2.324087238598144, + "learning_rate": 2.139610078576305e-06, + "loss": 0.903, + "step": 6566 + }, + { + "epoch": 0.4935367503381933, + "grad_norm": 1.6109895477618863, + "learning_rate": 2.139124419831667e-06, + "loss": 0.9718, + "step": 6567 + }, + { + "epoch": 0.49361190440402825, + "grad_norm": 1.6161727301469526, + "learning_rate": 2.1386387528433743e-06, + "loss": 0.9604, + "step": 6568 + }, + { + "epoch": 0.4936870584698632, + "grad_norm": 2.0435738770543637, + "learning_rate": 2.1381530776402058e-06, + "loss": 1.0289, + "step": 6569 + }, + { + "epoch": 0.4937622125356982, + "grad_norm": 2.4009585969557357, + "learning_rate": 2.1376673942509373e-06, + "loss": 0.9451, + "step": 6570 + }, + { + "epoch": 0.49383736660153316, + "grad_norm": 2.3752786573265823, + "learning_rate": 2.1371817027043494e-06, + "loss": 1.0765, + "step": 6571 + }, + { + "epoch": 0.4939125206673681, + "grad_norm": 1.8301053528607527, + "learning_rate": 2.1366960030292195e-06, + "loss": 0.906, + "step": 6572 + }, + { + "epoch": 0.49398767473320304, + "grad_norm": 2.226512065115466, + "learning_rate": 2.1362102952543277e-06, + "loss": 0.805, + "step": 6573 + }, + { + "epoch": 0.494062828799038, + "grad_norm": 1.3947587882799415, + "learning_rate": 2.135724579408456e-06, + "loss": 0.9986, + "step": 6574 + }, + { + "epoch": 0.494137982864873, + "grad_norm": 1.5806824008448688, + "learning_rate": 2.1352388555203814e-06, + "loss": 1.0365, + "step": 6575 + }, + { + "epoch": 0.49421313693070795, + "grad_norm": 1.44800807014872, + "learning_rate": 2.1347531236188878e-06, + "loss": 0.9794, + "step": 6576 + }, + { + "epoch": 0.4942882909965429, + "grad_norm": 1.5470302003009866, + "learning_rate": 2.134267383732755e-06, + "loss": 0.9197, + "step": 6577 + }, + { + "epoch": 0.4943634450623779, + "grad_norm": 1.839115330691624, + "learning_rate": 2.1337816358907663e-06, + "loss": 0.9464, + "step": 6578 + }, + { + "epoch": 0.49443859912821286, + "grad_norm": 1.435930058207857, + "learning_rate": 2.133295880121703e-06, + "loss": 1.0151, + "step": 6579 + }, + { + "epoch": 0.4945137531940478, + "grad_norm": 1.8940386259496809, + "learning_rate": 2.132810116454348e-06, + "loss": 1.0374, + "step": 6580 + }, + { + "epoch": 0.49458890725988275, + "grad_norm": 1.8089901714195018, + "learning_rate": 2.132324344917486e-06, + "loss": 1.0051, + "step": 6581 + }, + { + "epoch": 0.4946640613257177, + "grad_norm": 3.6174768451616606, + "learning_rate": 2.131838565539899e-06, + "loss": 1.0343, + "step": 6582 + }, + { + "epoch": 0.4947392153915527, + "grad_norm": 1.9558662403764069, + "learning_rate": 2.1313527783503727e-06, + "loss": 0.9126, + "step": 6583 + }, + { + "epoch": 0.49481436945738766, + "grad_norm": 0.5932952996942049, + "learning_rate": 2.1308669833776907e-06, + "loss": 0.8083, + "step": 6584 + }, + { + "epoch": 0.49488952352322263, + "grad_norm": 1.6522226852904334, + "learning_rate": 2.130381180650639e-06, + "loss": 0.9317, + "step": 6585 + }, + { + "epoch": 0.49496467758905754, + "grad_norm": 1.4325085100117567, + "learning_rate": 2.1298953701980038e-06, + "loss": 1.1341, + "step": 6586 + }, + { + "epoch": 0.4950398316548925, + "grad_norm": 0.7488434934755339, + "learning_rate": 2.129409552048569e-06, + "loss": 0.8884, + "step": 6587 + }, + { + "epoch": 0.4951149857207275, + "grad_norm": 2.3751768749027415, + "learning_rate": 2.1289237262311243e-06, + "loss": 1.0084, + "step": 6588 + }, + { + "epoch": 0.49519013978656246, + "grad_norm": 1.7715871964027081, + "learning_rate": 2.1284378927744546e-06, + "loss": 0.9801, + "step": 6589 + }, + { + "epoch": 0.4952652938523974, + "grad_norm": 2.126120664507922, + "learning_rate": 2.1279520517073475e-06, + "loss": 0.971, + "step": 6590 + }, + { + "epoch": 0.4953404479182324, + "grad_norm": 1.409712322661787, + "learning_rate": 2.127466203058592e-06, + "loss": 1.0693, + "step": 6591 + }, + { + "epoch": 0.4954156019840673, + "grad_norm": 1.4520462297746273, + "learning_rate": 2.1269803468569756e-06, + "loss": 1.0243, + "step": 6592 + }, + { + "epoch": 0.4954907560499023, + "grad_norm": 0.6902103534311597, + "learning_rate": 2.1264944831312874e-06, + "loss": 0.8613, + "step": 6593 + }, + { + "epoch": 0.49556591011573725, + "grad_norm": 1.519403055074035, + "learning_rate": 2.1260086119103165e-06, + "loss": 0.9334, + "step": 6594 + }, + { + "epoch": 0.4956410641815722, + "grad_norm": 1.6860152176275442, + "learning_rate": 2.1255227332228527e-06, + "loss": 1.0305, + "step": 6595 + }, + { + "epoch": 0.4957162182474072, + "grad_norm": 1.303196540533688, + "learning_rate": 2.125036847097687e-06, + "loss": 0.9901, + "step": 6596 + }, + { + "epoch": 0.49579137231324216, + "grad_norm": 1.5707394603064178, + "learning_rate": 2.1245509535636088e-06, + "loss": 1.0319, + "step": 6597 + }, + { + "epoch": 0.49586652637907713, + "grad_norm": 1.5900561591493105, + "learning_rate": 2.1240650526494096e-06, + "loss": 0.9555, + "step": 6598 + }, + { + "epoch": 0.49594168044491205, + "grad_norm": 0.7875876287116422, + "learning_rate": 2.1235791443838804e-06, + "loss": 0.8367, + "step": 6599 + }, + { + "epoch": 0.496016834510747, + "grad_norm": 1.777983900213994, + "learning_rate": 2.123093228795815e-06, + "loss": 0.913, + "step": 6600 + }, + { + "epoch": 0.496091988576582, + "grad_norm": 2.031085667931935, + "learning_rate": 2.122607305914004e-06, + "loss": 0.8825, + "step": 6601 + }, + { + "epoch": 0.49616714264241696, + "grad_norm": 1.5860922601666714, + "learning_rate": 2.1221213757672404e-06, + "loss": 0.886, + "step": 6602 + }, + { + "epoch": 0.49624229670825193, + "grad_norm": 0.8534556701692604, + "learning_rate": 2.1216354383843176e-06, + "loss": 0.9155, + "step": 6603 + }, + { + "epoch": 0.4963174507740869, + "grad_norm": 2.551522537170783, + "learning_rate": 2.1211494937940296e-06, + "loss": 0.9117, + "step": 6604 + }, + { + "epoch": 0.4963926048399218, + "grad_norm": 1.6972109102093118, + "learning_rate": 2.12066354202517e-06, + "loss": 0.9023, + "step": 6605 + }, + { + "epoch": 0.4964677589057568, + "grad_norm": 1.2763049366603927, + "learning_rate": 2.1201775831065336e-06, + "loss": 0.9406, + "step": 6606 + }, + { + "epoch": 0.49654291297159175, + "grad_norm": 1.6235099710217613, + "learning_rate": 2.119691617066915e-06, + "loss": 0.826, + "step": 6607 + }, + { + "epoch": 0.4966180670374267, + "grad_norm": 2.393037962029358, + "learning_rate": 2.11920564393511e-06, + "loss": 1.0208, + "step": 6608 + }, + { + "epoch": 0.4966932211032617, + "grad_norm": 1.6988846373659046, + "learning_rate": 2.1187196637399138e-06, + "loss": 0.9492, + "step": 6609 + }, + { + "epoch": 0.49676837516909667, + "grad_norm": 1.627042293967195, + "learning_rate": 2.118233676510123e-06, + "loss": 0.9471, + "step": 6610 + }, + { + "epoch": 0.49684352923493164, + "grad_norm": 2.1784373427497568, + "learning_rate": 2.1177476822745344e-06, + "loss": 1.0524, + "step": 6611 + }, + { + "epoch": 0.49691868330076655, + "grad_norm": 1.568322662265863, + "learning_rate": 2.1172616810619446e-06, + "loss": 0.9515, + "step": 6612 + }, + { + "epoch": 0.4969938373666015, + "grad_norm": 2.5350797903230666, + "learning_rate": 2.116775672901151e-06, + "loss": 0.9254, + "step": 6613 + }, + { + "epoch": 0.4970689914324365, + "grad_norm": 1.2835267470291982, + "learning_rate": 2.1162896578209516e-06, + "loss": 0.8385, + "step": 6614 + }, + { + "epoch": 0.49714414549827146, + "grad_norm": 2.118496648444126, + "learning_rate": 2.115803635850145e-06, + "loss": 1.0782, + "step": 6615 + }, + { + "epoch": 0.49721929956410643, + "grad_norm": 1.638116499184882, + "learning_rate": 2.1153176070175297e-06, + "loss": 0.8779, + "step": 6616 + }, + { + "epoch": 0.4972944536299414, + "grad_norm": 32.14867766465578, + "learning_rate": 2.1148315713519036e-06, + "loss": 0.9735, + "step": 6617 + }, + { + "epoch": 0.4973696076957763, + "grad_norm": 2.4126334790872903, + "learning_rate": 2.114345528882068e-06, + "loss": 0.8683, + "step": 6618 + }, + { + "epoch": 0.4974447617616113, + "grad_norm": 2.1281714614972285, + "learning_rate": 2.1138594796368213e-06, + "loss": 0.9134, + "step": 6619 + }, + { + "epoch": 0.49751991582744626, + "grad_norm": 5.823436989053674, + "learning_rate": 2.1133734236449654e-06, + "loss": 0.9549, + "step": 6620 + }, + { + "epoch": 0.4975950698932812, + "grad_norm": 1.7051877470488541, + "learning_rate": 2.112887360935299e-06, + "loss": 0.9658, + "step": 6621 + }, + { + "epoch": 0.4976702239591162, + "grad_norm": 1.7049049385619954, + "learning_rate": 2.112401291536625e-06, + "loss": 0.9854, + "step": 6622 + }, + { + "epoch": 0.49774537802495117, + "grad_norm": 0.7277971258050616, + "learning_rate": 2.1119152154777442e-06, + "loss": 0.926, + "step": 6623 + }, + { + "epoch": 0.49782053209078614, + "grad_norm": 1.6772387730689946, + "learning_rate": 2.1114291327874578e-06, + "loss": 1.0843, + "step": 6624 + }, + { + "epoch": 0.49789568615662105, + "grad_norm": 1.8418081370198496, + "learning_rate": 2.1109430434945685e-06, + "loss": 1.0078, + "step": 6625 + }, + { + "epoch": 0.497970840222456, + "grad_norm": 2.0246560687701654, + "learning_rate": 2.1104569476278794e-06, + "loss": 1.0481, + "step": 6626 + }, + { + "epoch": 0.498045994288291, + "grad_norm": 1.5904997051182377, + "learning_rate": 2.109970845216193e-06, + "loss": 0.9376, + "step": 6627 + }, + { + "epoch": 0.49812114835412596, + "grad_norm": 1.4348616229604434, + "learning_rate": 2.109484736288313e-06, + "loss": 1.0371, + "step": 6628 + }, + { + "epoch": 0.49819630241996093, + "grad_norm": 1.8163154825481627, + "learning_rate": 2.108998620873043e-06, + "loss": 1.0412, + "step": 6629 + }, + { + "epoch": 0.4982714564857959, + "grad_norm": 0.6902750535429788, + "learning_rate": 2.1085124989991876e-06, + "loss": 0.838, + "step": 6630 + }, + { + "epoch": 0.4983466105516308, + "grad_norm": 1.6762738693096721, + "learning_rate": 2.108026370695551e-06, + "loss": 1.0207, + "step": 6631 + }, + { + "epoch": 0.4984217646174658, + "grad_norm": 1.7278875729359306, + "learning_rate": 2.107540235990938e-06, + "loss": 1.0608, + "step": 6632 + }, + { + "epoch": 0.49849691868330076, + "grad_norm": 1.576685182544682, + "learning_rate": 2.107054094914155e-06, + "loss": 0.9044, + "step": 6633 + }, + { + "epoch": 0.49857207274913573, + "grad_norm": 2.1841639557314276, + "learning_rate": 2.106567947494006e-06, + "loss": 1.1051, + "step": 6634 + }, + { + "epoch": 0.4986472268149707, + "grad_norm": 2.1413721047775383, + "learning_rate": 2.106081793759298e-06, + "loss": 1.0064, + "step": 6635 + }, + { + "epoch": 0.49872238088080567, + "grad_norm": 1.1868839673772715, + "learning_rate": 2.1055956337388376e-06, + "loss": 1.0165, + "step": 6636 + }, + { + "epoch": 0.4987975349466406, + "grad_norm": 1.9118511271038474, + "learning_rate": 2.1051094674614327e-06, + "loss": 1.0645, + "step": 6637 + }, + { + "epoch": 0.49887268901247556, + "grad_norm": 2.4919079471320726, + "learning_rate": 2.1046232949558887e-06, + "loss": 0.9709, + "step": 6638 + }, + { + "epoch": 0.4989478430783105, + "grad_norm": 1.641410857407042, + "learning_rate": 2.104137116251013e-06, + "loss": 0.9318, + "step": 6639 + }, + { + "epoch": 0.4990229971441455, + "grad_norm": 1.2763759668189838, + "learning_rate": 2.103650931375615e-06, + "loss": 1.0386, + "step": 6640 + }, + { + "epoch": 0.49909815120998047, + "grad_norm": 0.715443385704093, + "learning_rate": 2.103164740358502e-06, + "loss": 0.8689, + "step": 6641 + }, + { + "epoch": 0.49917330527581544, + "grad_norm": 1.9087602940090072, + "learning_rate": 2.1026785432284837e-06, + "loss": 0.9477, + "step": 6642 + }, + { + "epoch": 0.4992484593416504, + "grad_norm": 1.8423207449186454, + "learning_rate": 2.1021923400143683e-06, + "loss": 1.0599, + "step": 6643 + }, + { + "epoch": 0.4993236134074853, + "grad_norm": 1.5833229349029134, + "learning_rate": 2.101706130744966e-06, + "loss": 1.0704, + "step": 6644 + }, + { + "epoch": 0.4993987674733203, + "grad_norm": 1.6236782567338828, + "learning_rate": 2.1012199154490852e-06, + "loss": 0.929, + "step": 6645 + }, + { + "epoch": 0.49947392153915526, + "grad_norm": 1.5007669554411396, + "learning_rate": 2.1007336941555374e-06, + "loss": 0.9963, + "step": 6646 + }, + { + "epoch": 0.49954907560499023, + "grad_norm": 1.6660403823354673, + "learning_rate": 2.100247466893132e-06, + "loss": 0.9498, + "step": 6647 + }, + { + "epoch": 0.4996242296708252, + "grad_norm": 2.1718215386925683, + "learning_rate": 2.0997612336906805e-06, + "loss": 0.9611, + "step": 6648 + }, + { + "epoch": 0.4996993837366602, + "grad_norm": 1.3640882841384674, + "learning_rate": 2.099274994576994e-06, + "loss": 0.9336, + "step": 6649 + }, + { + "epoch": 0.4997745378024951, + "grad_norm": 1.432087619002967, + "learning_rate": 2.098788749580884e-06, + "loss": 0.875, + "step": 6650 + }, + { + "epoch": 0.49984969186833006, + "grad_norm": 1.4323235063192814, + "learning_rate": 2.098302498731162e-06, + "loss": 0.9303, + "step": 6651 + }, + { + "epoch": 0.49992484593416503, + "grad_norm": 1.590898843212068, + "learning_rate": 2.0978162420566406e-06, + "loss": 1.0315, + "step": 6652 + }, + { + "epoch": 0.5, + "grad_norm": 0.663147846191505, + "learning_rate": 2.0973299795861322e-06, + "loss": 0.8574, + "step": 6653 + }, + { + "epoch": 0.5000751540658349, + "grad_norm": 1.4692734941983348, + "learning_rate": 2.09684371134845e-06, + "loss": 0.9448, + "step": 6654 + }, + { + "epoch": 0.5001503081316699, + "grad_norm": 1.72245050356415, + "learning_rate": 2.0963574373724074e-06, + "loss": 1.0433, + "step": 6655 + }, + { + "epoch": 0.5002254621975049, + "grad_norm": 4.060125038213446, + "learning_rate": 2.095871157686817e-06, + "loss": 1.0152, + "step": 6656 + }, + { + "epoch": 0.5003006162633399, + "grad_norm": 1.8967283742126106, + "learning_rate": 2.095384872320494e-06, + "loss": 0.9177, + "step": 6657 + }, + { + "epoch": 0.5003757703291748, + "grad_norm": 1.6869867745306737, + "learning_rate": 2.094898581302251e-06, + "loss": 0.9558, + "step": 6658 + }, + { + "epoch": 0.5004509243950098, + "grad_norm": 1.5006578115667748, + "learning_rate": 2.094412284660905e-06, + "loss": 0.9572, + "step": 6659 + }, + { + "epoch": 0.5005260784608447, + "grad_norm": 1.3924753544405388, + "learning_rate": 2.093925982425269e-06, + "loss": 0.9958, + "step": 6660 + }, + { + "epoch": 0.5006012325266797, + "grad_norm": 3.3209052779677974, + "learning_rate": 2.093439674624158e-06, + "loss": 0.9157, + "step": 6661 + }, + { + "epoch": 0.5006763865925147, + "grad_norm": 1.5668565668848522, + "learning_rate": 2.09295336128639e-06, + "loss": 0.9719, + "step": 6662 + }, + { + "epoch": 0.5007515406583496, + "grad_norm": 1.3575343682310017, + "learning_rate": 2.0924670424407785e-06, + "loss": 0.9177, + "step": 6663 + }, + { + "epoch": 0.5008266947241846, + "grad_norm": 1.5406459659906273, + "learning_rate": 2.091980718116141e-06, + "loss": 0.9659, + "step": 6664 + }, + { + "epoch": 0.5009018487900195, + "grad_norm": 1.5846196604572012, + "learning_rate": 2.0914943883412935e-06, + "loss": 0.9125, + "step": 6665 + }, + { + "epoch": 0.5009770028558544, + "grad_norm": 1.4869890986532315, + "learning_rate": 2.0910080531450534e-06, + "loss": 1.0243, + "step": 6666 + }, + { + "epoch": 0.5010521569216895, + "grad_norm": 1.625397193490205, + "learning_rate": 2.0905217125562378e-06, + "loss": 0.9955, + "step": 6667 + }, + { + "epoch": 0.5011273109875244, + "grad_norm": 0.9284815299354503, + "learning_rate": 2.0900353666036635e-06, + "loss": 0.9315, + "step": 6668 + }, + { + "epoch": 0.5012024650533594, + "grad_norm": 1.7283994129145481, + "learning_rate": 2.0895490153161496e-06, + "loss": 1.0812, + "step": 6669 + }, + { + "epoch": 0.5012776191191943, + "grad_norm": 41.441519471814885, + "learning_rate": 2.089062658722513e-06, + "loss": 1.0293, + "step": 6670 + }, + { + "epoch": 0.5013527731850294, + "grad_norm": 2.5269832678279784, + "learning_rate": 2.0885762968515737e-06, + "loss": 1.1782, + "step": 6671 + }, + { + "epoch": 0.5014279272508643, + "grad_norm": 5.389258877789325, + "learning_rate": 2.088089929732149e-06, + "loss": 0.9529, + "step": 6672 + }, + { + "epoch": 0.5015030813166992, + "grad_norm": 1.748326455033951, + "learning_rate": 2.0876035573930587e-06, + "loss": 1.0256, + "step": 6673 + }, + { + "epoch": 0.5015782353825342, + "grad_norm": 1.8340778428575186, + "learning_rate": 2.0871171798631224e-06, + "loss": 0.968, + "step": 6674 + }, + { + "epoch": 0.5016533894483691, + "grad_norm": 1.5363178857093076, + "learning_rate": 2.0866307971711594e-06, + "loss": 1.0156, + "step": 6675 + }, + { + "epoch": 0.5017285435142042, + "grad_norm": 1.793879790711108, + "learning_rate": 2.08614440934599e-06, + "loss": 0.9163, + "step": 6676 + }, + { + "epoch": 0.5018036975800391, + "grad_norm": 1.6575470649861026, + "learning_rate": 2.0856580164164344e-06, + "loss": 0.9911, + "step": 6677 + }, + { + "epoch": 0.5018788516458741, + "grad_norm": 1.7523953529273655, + "learning_rate": 2.085171618411313e-06, + "loss": 0.9703, + "step": 6678 + }, + { + "epoch": 0.501954005711709, + "grad_norm": 1.9355697861440417, + "learning_rate": 2.0846852153594477e-06, + "loss": 1.0322, + "step": 6679 + }, + { + "epoch": 0.5020291597775439, + "grad_norm": 1.6161370297961646, + "learning_rate": 2.0841988072896585e-06, + "loss": 0.9843, + "step": 6680 + }, + { + "epoch": 0.502104313843379, + "grad_norm": 1.9889303470649933, + "learning_rate": 2.0837123942307677e-06, + "loss": 0.9837, + "step": 6681 + }, + { + "epoch": 0.5021794679092139, + "grad_norm": 2.0695079463742587, + "learning_rate": 2.0832259762115977e-06, + "loss": 0.9615, + "step": 6682 + }, + { + "epoch": 0.5022546219750489, + "grad_norm": 1.7999511924060456, + "learning_rate": 2.0827395532609685e-06, + "loss": 1.0338, + "step": 6683 + }, + { + "epoch": 0.5023297760408838, + "grad_norm": 1.6072308228654495, + "learning_rate": 2.082253125407705e-06, + "loss": 1.0244, + "step": 6684 + }, + { + "epoch": 0.5024049301067188, + "grad_norm": 1.5031246383571388, + "learning_rate": 2.0817666926806287e-06, + "loss": 1.0216, + "step": 6685 + }, + { + "epoch": 0.5024800841725537, + "grad_norm": 1.5463299031686588, + "learning_rate": 2.0812802551085633e-06, + "loss": 1.0028, + "step": 6686 + }, + { + "epoch": 0.5025552382383887, + "grad_norm": 1.9624813127539138, + "learning_rate": 2.0807938127203304e-06, + "loss": 1.0009, + "step": 6687 + }, + { + "epoch": 0.5026303923042237, + "grad_norm": 1.3255128641644678, + "learning_rate": 2.080307365544755e-06, + "loss": 0.9817, + "step": 6688 + }, + { + "epoch": 0.5027055463700586, + "grad_norm": 2.599677425327857, + "learning_rate": 2.0798209136106615e-06, + "loss": 1.0958, + "step": 6689 + }, + { + "epoch": 0.5027807004358936, + "grad_norm": 1.579354288971532, + "learning_rate": 2.0793344569468725e-06, + "loss": 0.9769, + "step": 6690 + }, + { + "epoch": 0.5028558545017285, + "grad_norm": 11.33933308680522, + "learning_rate": 2.0788479955822136e-06, + "loss": 0.9718, + "step": 6691 + }, + { + "epoch": 0.5029310085675635, + "grad_norm": 1.9049117597032834, + "learning_rate": 2.0783615295455082e-06, + "loss": 0.993, + "step": 6692 + }, + { + "epoch": 0.5030061626333985, + "grad_norm": 1.446989993310585, + "learning_rate": 2.077875058865583e-06, + "loss": 0.9602, + "step": 6693 + }, + { + "epoch": 0.5030813166992334, + "grad_norm": 1.8440473930904995, + "learning_rate": 2.077388583571262e-06, + "loss": 0.9645, + "step": 6694 + }, + { + "epoch": 0.5031564707650684, + "grad_norm": 1.6894515726600527, + "learning_rate": 2.076902103691371e-06, + "loss": 0.9534, + "step": 6695 + }, + { + "epoch": 0.5032316248309033, + "grad_norm": 1.6507340474563184, + "learning_rate": 2.076415619254736e-06, + "loss": 1.1025, + "step": 6696 + }, + { + "epoch": 0.5033067788967384, + "grad_norm": 3.0581883811034936, + "learning_rate": 2.075929130290183e-06, + "loss": 0.9823, + "step": 6697 + }, + { + "epoch": 0.5033819329625733, + "grad_norm": 0.6106778912786305, + "learning_rate": 2.075442636826538e-06, + "loss": 0.8405, + "step": 6698 + }, + { + "epoch": 0.5034570870284082, + "grad_norm": 1.6683179860484127, + "learning_rate": 2.0749561388926283e-06, + "loss": 1.0009, + "step": 6699 + }, + { + "epoch": 0.5035322410942432, + "grad_norm": 1.5639800881730503, + "learning_rate": 2.07446963651728e-06, + "loss": 0.9436, + "step": 6700 + }, + { + "epoch": 0.5036073951600781, + "grad_norm": 2.170782129876097, + "learning_rate": 2.073983129729321e-06, + "loss": 0.8823, + "step": 6701 + }, + { + "epoch": 0.5036825492259132, + "grad_norm": 2.0536518460278517, + "learning_rate": 2.073496618557577e-06, + "loss": 0.9378, + "step": 6702 + }, + { + "epoch": 0.5037577032917481, + "grad_norm": 2.2977711492849076, + "learning_rate": 2.073010103030878e-06, + "loss": 0.9768, + "step": 6703 + }, + { + "epoch": 0.5038328573575831, + "grad_norm": 2.0090719940820403, + "learning_rate": 2.0725235831780516e-06, + "loss": 0.8294, + "step": 6704 + }, + { + "epoch": 0.503908011423418, + "grad_norm": 1.5543622007253333, + "learning_rate": 2.0720370590279234e-06, + "loss": 1.0076, + "step": 6705 + }, + { + "epoch": 0.5039831654892529, + "grad_norm": 1.518056197698065, + "learning_rate": 2.0715505306093247e-06, + "loss": 1.0362, + "step": 6706 + }, + { + "epoch": 0.504058319555088, + "grad_norm": 0.6371344397109193, + "learning_rate": 2.071063997951082e-06, + "loss": 0.8559, + "step": 6707 + }, + { + "epoch": 0.5041334736209229, + "grad_norm": 2.080725398117937, + "learning_rate": 2.0705774610820267e-06, + "loss": 0.906, + "step": 6708 + }, + { + "epoch": 0.5042086276867579, + "grad_norm": 1.5642724474563936, + "learning_rate": 2.070090920030986e-06, + "loss": 0.9399, + "step": 6709 + }, + { + "epoch": 0.5042837817525928, + "grad_norm": 1.464060011691622, + "learning_rate": 2.0696043748267897e-06, + "loss": 1.0802, + "step": 6710 + }, + { + "epoch": 0.5043589358184277, + "grad_norm": 1.783813572628233, + "learning_rate": 2.0691178254982684e-06, + "loss": 0.9331, + "step": 6711 + }, + { + "epoch": 0.5044340898842627, + "grad_norm": 0.7522022973932034, + "learning_rate": 2.0686312720742504e-06, + "loss": 0.8249, + "step": 6712 + }, + { + "epoch": 0.5045092439500977, + "grad_norm": 1.8636595144123806, + "learning_rate": 2.068144714583567e-06, + "loss": 1.0529, + "step": 6713 + }, + { + "epoch": 0.5045843980159327, + "grad_norm": 2.054009387305927, + "learning_rate": 2.0676581530550485e-06, + "loss": 1.0045, + "step": 6714 + }, + { + "epoch": 0.5046595520817676, + "grad_norm": 1.6353388320316182, + "learning_rate": 2.067171587517525e-06, + "loss": 1.0386, + "step": 6715 + }, + { + "epoch": 0.5047347061476026, + "grad_norm": 1.8014404679032316, + "learning_rate": 2.066685017999828e-06, + "loss": 0.9849, + "step": 6716 + }, + { + "epoch": 0.5048098602134375, + "grad_norm": 1.764289283197256, + "learning_rate": 2.0661984445307886e-06, + "loss": 0.9519, + "step": 6717 + }, + { + "epoch": 0.5048850142792725, + "grad_norm": 1.7182805200422429, + "learning_rate": 2.0657118671392373e-06, + "loss": 0.9703, + "step": 6718 + }, + { + "epoch": 0.5049601683451075, + "grad_norm": 1.886171439262074, + "learning_rate": 2.0652252858540064e-06, + "loss": 0.9704, + "step": 6719 + }, + { + "epoch": 0.5050353224109424, + "grad_norm": 1.4399163213894617, + "learning_rate": 2.0647387007039277e-06, + "loss": 0.9823, + "step": 6720 + }, + { + "epoch": 0.5051104764767774, + "grad_norm": 1.9551049534560339, + "learning_rate": 2.0642521117178332e-06, + "loss": 1.0489, + "step": 6721 + }, + { + "epoch": 0.5051856305426123, + "grad_norm": 1.5982646203544129, + "learning_rate": 2.0637655189245548e-06, + "loss": 0.9678, + "step": 6722 + }, + { + "epoch": 0.5052607846084474, + "grad_norm": 1.3284591254513611, + "learning_rate": 2.0632789223529254e-06, + "loss": 1.0298, + "step": 6723 + }, + { + "epoch": 0.5053359386742823, + "grad_norm": 1.7332267740702612, + "learning_rate": 2.0627923220317767e-06, + "loss": 1.072, + "step": 6724 + }, + { + "epoch": 0.5054110927401172, + "grad_norm": 1.4989610888906693, + "learning_rate": 2.062305717989943e-06, + "loss": 0.9886, + "step": 6725 + }, + { + "epoch": 0.5054862468059522, + "grad_norm": 2.5647999746100125, + "learning_rate": 2.0618191102562575e-06, + "loss": 0.9541, + "step": 6726 + }, + { + "epoch": 0.5055614008717871, + "grad_norm": 2.0860624240228685, + "learning_rate": 2.061332498859553e-06, + "loss": 0.9302, + "step": 6727 + }, + { + "epoch": 0.5056365549376222, + "grad_norm": 1.2869075912194776, + "learning_rate": 2.060845883828663e-06, + "loss": 0.9185, + "step": 6728 + }, + { + "epoch": 0.5057117090034571, + "grad_norm": 0.804794637714234, + "learning_rate": 2.0603592651924206e-06, + "loss": 0.7833, + "step": 6729 + }, + { + "epoch": 0.5057868630692921, + "grad_norm": 1.5218430220046573, + "learning_rate": 2.059872642979661e-06, + "loss": 0.8556, + "step": 6730 + }, + { + "epoch": 0.505862017135127, + "grad_norm": 2.2122606826922624, + "learning_rate": 2.0593860172192178e-06, + "loss": 0.8804, + "step": 6731 + }, + { + "epoch": 0.5059371712009619, + "grad_norm": 1.669627309774892, + "learning_rate": 2.0588993879399265e-06, + "loss": 1.0012, + "step": 6732 + }, + { + "epoch": 0.506012325266797, + "grad_norm": 1.4485921404774793, + "learning_rate": 2.0584127551706202e-06, + "loss": 0.9704, + "step": 6733 + }, + { + "epoch": 0.5060874793326319, + "grad_norm": 2.0184422403065225, + "learning_rate": 2.0579261189401345e-06, + "loss": 0.9817, + "step": 6734 + }, + { + "epoch": 0.5061626333984669, + "grad_norm": 1.7482437449152026, + "learning_rate": 2.0574394792773048e-06, + "loss": 1.0461, + "step": 6735 + }, + { + "epoch": 0.5062377874643018, + "grad_norm": 2.16702475400833, + "learning_rate": 2.0569528362109662e-06, + "loss": 1.0442, + "step": 6736 + }, + { + "epoch": 0.5063129415301367, + "grad_norm": 1.6215596092670321, + "learning_rate": 2.056466189769953e-06, + "loss": 0.8851, + "step": 6737 + }, + { + "epoch": 0.5063880955959718, + "grad_norm": 1.947570103569854, + "learning_rate": 2.055979539983103e-06, + "loss": 0.8984, + "step": 6738 + }, + { + "epoch": 0.5064632496618067, + "grad_norm": 0.823949833250188, + "learning_rate": 2.05549288687925e-06, + "loss": 0.8244, + "step": 6739 + }, + { + "epoch": 0.5065384037276417, + "grad_norm": 1.3815118308484038, + "learning_rate": 2.0550062304872317e-06, + "loss": 0.9976, + "step": 6740 + }, + { + "epoch": 0.5066135577934766, + "grad_norm": 1.5152362442890153, + "learning_rate": 2.054519570835883e-06, + "loss": 0.8591, + "step": 6741 + }, + { + "epoch": 0.5066887118593116, + "grad_norm": 1.6213996322862927, + "learning_rate": 2.0540329079540414e-06, + "loss": 1.0387, + "step": 6742 + }, + { + "epoch": 0.5067638659251466, + "grad_norm": 2.346283725664818, + "learning_rate": 2.053546241870543e-06, + "loss": 0.9794, + "step": 6743 + }, + { + "epoch": 0.5068390199909815, + "grad_norm": 1.3617895759026688, + "learning_rate": 2.053059572614224e-06, + "loss": 1.0697, + "step": 6744 + }, + { + "epoch": 0.5069141740568165, + "grad_norm": 2.5467561097274474, + "learning_rate": 2.0525729002139233e-06, + "loss": 1.0233, + "step": 6745 + }, + { + "epoch": 0.5069893281226514, + "grad_norm": 1.8444926818925842, + "learning_rate": 2.052086224698476e-06, + "loss": 0.9088, + "step": 6746 + }, + { + "epoch": 0.5070644821884864, + "grad_norm": 2.8670143678037037, + "learning_rate": 2.0515995460967204e-06, + "loss": 0.9865, + "step": 6747 + }, + { + "epoch": 0.5071396362543213, + "grad_norm": 1.520304818834776, + "learning_rate": 2.0511128644374953e-06, + "loss": 1.0017, + "step": 6748 + }, + { + "epoch": 0.5072147903201564, + "grad_norm": 1.8163320218578152, + "learning_rate": 2.0506261797496357e-06, + "loss": 0.881, + "step": 6749 + }, + { + "epoch": 0.5072899443859913, + "grad_norm": 1.7699608018274002, + "learning_rate": 2.0501394920619822e-06, + "loss": 1.0047, + "step": 6750 + }, + { + "epoch": 0.5073650984518262, + "grad_norm": 1.708627473893434, + "learning_rate": 2.0496528014033717e-06, + "loss": 1.0333, + "step": 6751 + }, + { + "epoch": 0.5074402525176612, + "grad_norm": 1.6363838963988349, + "learning_rate": 2.0491661078026423e-06, + "loss": 0.9273, + "step": 6752 + }, + { + "epoch": 0.5075154065834961, + "grad_norm": 0.7687661099490423, + "learning_rate": 2.0486794112886328e-06, + "loss": 0.8652, + "step": 6753 + }, + { + "epoch": 0.5075905606493312, + "grad_norm": 1.7093058430910542, + "learning_rate": 2.0481927118901817e-06, + "loss": 0.9702, + "step": 6754 + }, + { + "epoch": 0.5076657147151661, + "grad_norm": 3.7708269965749084, + "learning_rate": 2.047706009636128e-06, + "loss": 1.0054, + "step": 6755 + }, + { + "epoch": 0.507740868781001, + "grad_norm": 1.5048433629137346, + "learning_rate": 2.0472193045553104e-06, + "loss": 1.0364, + "step": 6756 + }, + { + "epoch": 0.507816022846836, + "grad_norm": 2.0165851287541647, + "learning_rate": 2.0467325966765683e-06, + "loss": 0.9937, + "step": 6757 + }, + { + "epoch": 0.5078911769126709, + "grad_norm": 1.677464135890934, + "learning_rate": 2.046245886028741e-06, + "loss": 1.0583, + "step": 6758 + }, + { + "epoch": 0.507966330978506, + "grad_norm": 0.751556529697489, + "learning_rate": 2.045759172640668e-06, + "loss": 0.827, + "step": 6759 + }, + { + "epoch": 0.5080414850443409, + "grad_norm": 2.0055266552877278, + "learning_rate": 2.0452724565411886e-06, + "loss": 0.9538, + "step": 6760 + }, + { + "epoch": 0.5081166391101759, + "grad_norm": 1.5304173911231687, + "learning_rate": 2.044785737759143e-06, + "loss": 0.9936, + "step": 6761 + }, + { + "epoch": 0.5081917931760108, + "grad_norm": 1.5464340506254899, + "learning_rate": 2.0442990163233704e-06, + "loss": 0.9387, + "step": 6762 + }, + { + "epoch": 0.5082669472418457, + "grad_norm": 1.8293792464946936, + "learning_rate": 2.0438122922627114e-06, + "loss": 1.0192, + "step": 6763 + }, + { + "epoch": 0.5083421013076808, + "grad_norm": 5.016308128919384, + "learning_rate": 2.0433255656060066e-06, + "loss": 0.9761, + "step": 6764 + }, + { + "epoch": 0.5084172553735157, + "grad_norm": 1.9463391639703584, + "learning_rate": 2.0428388363820966e-06, + "loss": 0.8634, + "step": 6765 + }, + { + "epoch": 0.5084924094393507, + "grad_norm": 1.3667855680117387, + "learning_rate": 2.0423521046198206e-06, + "loss": 1.0154, + "step": 6766 + }, + { + "epoch": 0.5085675635051856, + "grad_norm": 0.7794714710576712, + "learning_rate": 2.041865370348021e-06, + "loss": 0.8498, + "step": 6767 + }, + { + "epoch": 0.5086427175710206, + "grad_norm": 1.745808281487357, + "learning_rate": 2.0413786335955374e-06, + "loss": 0.9636, + "step": 6768 + }, + { + "epoch": 0.5087178716368556, + "grad_norm": 1.7922138295322356, + "learning_rate": 2.0408918943912113e-06, + "loss": 0.9034, + "step": 6769 + }, + { + "epoch": 0.5087930257026905, + "grad_norm": 1.687970413762284, + "learning_rate": 2.0404051527638844e-06, + "loss": 0.9907, + "step": 6770 + }, + { + "epoch": 0.5088681797685255, + "grad_norm": 1.4366019387647528, + "learning_rate": 2.039918408742397e-06, + "loss": 0.9361, + "step": 6771 + }, + { + "epoch": 0.5089433338343604, + "grad_norm": 5.393482987567738, + "learning_rate": 2.039431662355591e-06, + "loss": 0.9899, + "step": 6772 + }, + { + "epoch": 0.5090184879001954, + "grad_norm": 1.6834939807617693, + "learning_rate": 2.0389449136323082e-06, + "loss": 0.9586, + "step": 6773 + }, + { + "epoch": 0.5090936419660304, + "grad_norm": 1.708026517282514, + "learning_rate": 2.0384581626013905e-06, + "loss": 0.9667, + "step": 6774 + }, + { + "epoch": 0.5091687960318654, + "grad_norm": 1.7860772826592213, + "learning_rate": 2.037971409291679e-06, + "loss": 1.0398, + "step": 6775 + }, + { + "epoch": 0.5092439500977003, + "grad_norm": 1.7863785380197033, + "learning_rate": 2.037484653732016e-06, + "loss": 1.0097, + "step": 6776 + }, + { + "epoch": 0.5093191041635352, + "grad_norm": 2.0298080020928353, + "learning_rate": 2.036997895951244e-06, + "loss": 1.016, + "step": 6777 + }, + { + "epoch": 0.5093942582293702, + "grad_norm": 2.160925840179475, + "learning_rate": 2.0365111359782046e-06, + "loss": 0.9632, + "step": 6778 + }, + { + "epoch": 0.5094694122952051, + "grad_norm": 2.336184893619881, + "learning_rate": 2.0360243738417414e-06, + "loss": 0.9111, + "step": 6779 + }, + { + "epoch": 0.5095445663610402, + "grad_norm": 1.7357873063647549, + "learning_rate": 2.035537609570695e-06, + "loss": 1.0141, + "step": 6780 + }, + { + "epoch": 0.5096197204268751, + "grad_norm": 0.7322532357938518, + "learning_rate": 2.03505084319391e-06, + "loss": 0.8255, + "step": 6781 + }, + { + "epoch": 0.50969487449271, + "grad_norm": 3.1984239273267834, + "learning_rate": 2.0345640747402283e-06, + "loss": 0.9828, + "step": 6782 + }, + { + "epoch": 0.509770028558545, + "grad_norm": 1.4891037157725162, + "learning_rate": 2.034077304238492e-06, + "loss": 0.9925, + "step": 6783 + }, + { + "epoch": 0.50984518262438, + "grad_norm": 1.4964294853688884, + "learning_rate": 2.0335905317175457e-06, + "loss": 1.0257, + "step": 6784 + }, + { + "epoch": 0.509920336690215, + "grad_norm": 2.2852566167194994, + "learning_rate": 2.0331037572062314e-06, + "loss": 1.0255, + "step": 6785 + }, + { + "epoch": 0.5099954907560499, + "grad_norm": 1.3777092771566817, + "learning_rate": 2.032616980733393e-06, + "loss": 0.9177, + "step": 6786 + }, + { + "epoch": 0.5100706448218849, + "grad_norm": 1.7453953927350074, + "learning_rate": 2.0321302023278734e-06, + "loss": 0.9982, + "step": 6787 + }, + { + "epoch": 0.5101457988877198, + "grad_norm": 1.6844065534081452, + "learning_rate": 2.031643422018516e-06, + "loss": 0.9416, + "step": 6788 + }, + { + "epoch": 0.5102209529535547, + "grad_norm": 1.4617039103241494, + "learning_rate": 2.0311566398341653e-06, + "loss": 0.907, + "step": 6789 + }, + { + "epoch": 0.5102961070193898, + "grad_norm": 1.9818313997080823, + "learning_rate": 2.0306698558036635e-06, + "loss": 1.0301, + "step": 6790 + }, + { + "epoch": 0.5103712610852247, + "grad_norm": 1.426692723559164, + "learning_rate": 2.0301830699558563e-06, + "loss": 0.9667, + "step": 6791 + }, + { + "epoch": 0.5104464151510597, + "grad_norm": 1.600539983586617, + "learning_rate": 2.029696282319586e-06, + "loss": 1.0619, + "step": 6792 + }, + { + "epoch": 0.5105215692168946, + "grad_norm": 1.7009822420048855, + "learning_rate": 2.0292094929236976e-06, + "loss": 1.0098, + "step": 6793 + }, + { + "epoch": 0.5105967232827296, + "grad_norm": 1.3058593877803197, + "learning_rate": 2.028722701797035e-06, + "loss": 0.9054, + "step": 6794 + }, + { + "epoch": 0.5106718773485646, + "grad_norm": 0.7687299787814004, + "learning_rate": 2.0282359089684417e-06, + "loss": 0.8133, + "step": 6795 + }, + { + "epoch": 0.5107470314143995, + "grad_norm": 2.9319771326378103, + "learning_rate": 2.027749114466763e-06, + "loss": 0.9551, + "step": 6796 + }, + { + "epoch": 0.5108221854802345, + "grad_norm": 2.1039686078315136, + "learning_rate": 2.0272623183208433e-06, + "loss": 0.9585, + "step": 6797 + }, + { + "epoch": 0.5108973395460694, + "grad_norm": 0.80300252848948, + "learning_rate": 2.0267755205595266e-06, + "loss": 0.8902, + "step": 6798 + }, + { + "epoch": 0.5109724936119044, + "grad_norm": 1.818316105738544, + "learning_rate": 2.026288721211658e-06, + "loss": 0.9467, + "step": 6799 + }, + { + "epoch": 0.5110476476777394, + "grad_norm": 1.5184826211295956, + "learning_rate": 2.0258019203060816e-06, + "loss": 0.988, + "step": 6800 + }, + { + "epoch": 0.5111228017435743, + "grad_norm": 4.182896831176757, + "learning_rate": 2.025315117871643e-06, + "loss": 0.9656, + "step": 6801 + }, + { + "epoch": 0.5111979558094093, + "grad_norm": 1.9049904210808488, + "learning_rate": 2.0248283139371862e-06, + "loss": 0.9509, + "step": 6802 + }, + { + "epoch": 0.5112731098752442, + "grad_norm": 1.4271740942539028, + "learning_rate": 2.0243415085315573e-06, + "loss": 1.0769, + "step": 6803 + }, + { + "epoch": 0.5113482639410792, + "grad_norm": 1.6791874340783042, + "learning_rate": 2.023854701683601e-06, + "loss": 0.8696, + "step": 6804 + }, + { + "epoch": 0.5114234180069142, + "grad_norm": 2.008566154607939, + "learning_rate": 2.0233678934221615e-06, + "loss": 0.9639, + "step": 6805 + }, + { + "epoch": 0.5114985720727492, + "grad_norm": 1.7967616833325155, + "learning_rate": 2.0228810837760853e-06, + "loss": 0.8793, + "step": 6806 + }, + { + "epoch": 0.5115737261385841, + "grad_norm": 2.9624964653670074, + "learning_rate": 2.0223942727742168e-06, + "loss": 0.9613, + "step": 6807 + }, + { + "epoch": 0.511648880204419, + "grad_norm": 1.9016797044503793, + "learning_rate": 2.0219074604454026e-06, + "loss": 0.9742, + "step": 6808 + }, + { + "epoch": 0.511724034270254, + "grad_norm": 1.7039925002036018, + "learning_rate": 2.021420646818487e-06, + "loss": 1.0161, + "step": 6809 + }, + { + "epoch": 0.511799188336089, + "grad_norm": 1.5907001110397014, + "learning_rate": 2.0209338319223155e-06, + "loss": 1.0562, + "step": 6810 + }, + { + "epoch": 0.511874342401924, + "grad_norm": 1.9271327106920115, + "learning_rate": 2.0204470157857354e-06, + "loss": 0.8965, + "step": 6811 + }, + { + "epoch": 0.5119494964677589, + "grad_norm": 0.7002953493809003, + "learning_rate": 2.0199601984375907e-06, + "loss": 0.8615, + "step": 6812 + }, + { + "epoch": 0.5120246505335939, + "grad_norm": 1.5311680985949208, + "learning_rate": 2.0194733799067284e-06, + "loss": 0.952, + "step": 6813 + }, + { + "epoch": 0.5120998045994288, + "grad_norm": 2.0018380778680482, + "learning_rate": 2.0189865602219934e-06, + "loss": 0.9787, + "step": 6814 + }, + { + "epoch": 0.5121749586652637, + "grad_norm": 1.5438064039791126, + "learning_rate": 2.0184997394122317e-06, + "loss": 1.0729, + "step": 6815 + }, + { + "epoch": 0.5122501127310988, + "grad_norm": 1.8083741155616821, + "learning_rate": 2.01801291750629e-06, + "loss": 1.0626, + "step": 6816 + }, + { + "epoch": 0.5123252667969337, + "grad_norm": 2.075254135195729, + "learning_rate": 2.0175260945330134e-06, + "loss": 0.9296, + "step": 6817 + }, + { + "epoch": 0.5124004208627687, + "grad_norm": 1.4974736555581123, + "learning_rate": 2.0170392705212495e-06, + "loss": 0.9753, + "step": 6818 + }, + { + "epoch": 0.5124755749286036, + "grad_norm": 1.559759865398363, + "learning_rate": 2.016552445499843e-06, + "loss": 0.9301, + "step": 6819 + }, + { + "epoch": 0.5125507289944387, + "grad_norm": 1.4962912645760702, + "learning_rate": 2.0160656194976407e-06, + "loss": 1.0187, + "step": 6820 + }, + { + "epoch": 0.5126258830602736, + "grad_norm": 1.4630830890269795, + "learning_rate": 2.0155787925434893e-06, + "loss": 1.1034, + "step": 6821 + }, + { + "epoch": 0.5127010371261085, + "grad_norm": 1.4773590522832976, + "learning_rate": 2.0150919646662342e-06, + "loss": 0.9453, + "step": 6822 + }, + { + "epoch": 0.5127761911919435, + "grad_norm": 1.525494873548388, + "learning_rate": 2.014605135894723e-06, + "loss": 0.9917, + "step": 6823 + }, + { + "epoch": 0.5128513452577784, + "grad_norm": 1.5548760381272395, + "learning_rate": 2.0141183062578013e-06, + "loss": 0.9195, + "step": 6824 + }, + { + "epoch": 0.5129264993236134, + "grad_norm": 1.9675052280186915, + "learning_rate": 2.013631475784316e-06, + "loss": 0.8642, + "step": 6825 + }, + { + "epoch": 0.5130016533894484, + "grad_norm": 1.6415075925383604, + "learning_rate": 2.0131446445031134e-06, + "loss": 0.9543, + "step": 6826 + }, + { + "epoch": 0.5130768074552833, + "grad_norm": 0.6062916731521676, + "learning_rate": 2.0126578124430402e-06, + "loss": 0.7667, + "step": 6827 + }, + { + "epoch": 0.5131519615211183, + "grad_norm": 1.4586536691430618, + "learning_rate": 2.012170979632944e-06, + "loss": 1.071, + "step": 6828 + }, + { + "epoch": 0.5132271155869532, + "grad_norm": 1.6970477386845273, + "learning_rate": 2.0116841461016685e-06, + "loss": 1.0134, + "step": 6829 + }, + { + "epoch": 0.5133022696527882, + "grad_norm": 1.8084846611986094, + "learning_rate": 2.0111973118780653e-06, + "loss": 0.9931, + "step": 6830 + }, + { + "epoch": 0.5133774237186232, + "grad_norm": 1.7608581606331288, + "learning_rate": 2.0107104769909773e-06, + "loss": 1.0141, + "step": 6831 + }, + { + "epoch": 0.5134525777844582, + "grad_norm": 2.8949559115147605, + "learning_rate": 2.0102236414692515e-06, + "loss": 0.8631, + "step": 6832 + }, + { + "epoch": 0.5135277318502931, + "grad_norm": 1.3689429647910207, + "learning_rate": 2.009736805341737e-06, + "loss": 0.9608, + "step": 6833 + }, + { + "epoch": 0.513602885916128, + "grad_norm": 1.8374388704057687, + "learning_rate": 2.0092499686372794e-06, + "loss": 0.8929, + "step": 6834 + }, + { + "epoch": 0.513678039981963, + "grad_norm": 1.4653155978335426, + "learning_rate": 2.0087631313847252e-06, + "loss": 0.9046, + "step": 6835 + }, + { + "epoch": 0.513753194047798, + "grad_norm": 1.5484938194587596, + "learning_rate": 2.0082762936129226e-06, + "loss": 0.9675, + "step": 6836 + }, + { + "epoch": 0.513828348113633, + "grad_norm": 1.3948114364776543, + "learning_rate": 2.0077894553507174e-06, + "loss": 0.8967, + "step": 6837 + }, + { + "epoch": 0.5139035021794679, + "grad_norm": 1.702856550145709, + "learning_rate": 2.0073026166269577e-06, + "loss": 0.9222, + "step": 6838 + }, + { + "epoch": 0.5139786562453029, + "grad_norm": 2.0632674205566444, + "learning_rate": 2.006815777470489e-06, + "loss": 1.071, + "step": 6839 + }, + { + "epoch": 0.5140538103111378, + "grad_norm": 1.572127853896881, + "learning_rate": 2.0063289379101606e-06, + "loss": 0.9533, + "step": 6840 + }, + { + "epoch": 0.5141289643769728, + "grad_norm": 1.489388041821366, + "learning_rate": 2.0058420979748172e-06, + "loss": 1.0861, + "step": 6841 + }, + { + "epoch": 0.5142041184428078, + "grad_norm": 1.4442122111739433, + "learning_rate": 2.005355257693308e-06, + "loss": 0.9413, + "step": 6842 + }, + { + "epoch": 0.5142792725086427, + "grad_norm": 2.6574189250704814, + "learning_rate": 2.0048684170944795e-06, + "loss": 1.1076, + "step": 6843 + }, + { + "epoch": 0.5143544265744777, + "grad_norm": 1.837119318655147, + "learning_rate": 2.004381576207178e-06, + "loss": 1.0729, + "step": 6844 + }, + { + "epoch": 0.5144295806403126, + "grad_norm": 1.9828962688621983, + "learning_rate": 2.0038947350602516e-06, + "loss": 0.9675, + "step": 6845 + }, + { + "epoch": 0.5145047347061475, + "grad_norm": 1.6491420740674816, + "learning_rate": 2.0034078936825467e-06, + "loss": 0.9724, + "step": 6846 + }, + { + "epoch": 0.5145798887719826, + "grad_norm": 1.6134579910375024, + "learning_rate": 2.002921052102912e-06, + "loss": 0.9498, + "step": 6847 + }, + { + "epoch": 0.5146550428378175, + "grad_norm": 1.5757382918002614, + "learning_rate": 2.0024342103501934e-06, + "loss": 0.9772, + "step": 6848 + }, + { + "epoch": 0.5147301969036525, + "grad_norm": 1.545875766031244, + "learning_rate": 2.001947368453238e-06, + "loss": 1.0222, + "step": 6849 + }, + { + "epoch": 0.5148053509694874, + "grad_norm": 0.751430815582509, + "learning_rate": 2.001460526440894e-06, + "loss": 0.9269, + "step": 6850 + }, + { + "epoch": 0.5148805050353225, + "grad_norm": 2.4850992550590645, + "learning_rate": 2.0009736843420076e-06, + "loss": 1.0582, + "step": 6851 + }, + { + "epoch": 0.5149556591011574, + "grad_norm": 1.5200603123042857, + "learning_rate": 2.0004868421854274e-06, + "loss": 0.9425, + "step": 6852 + }, + { + "epoch": 0.5150308131669923, + "grad_norm": 1.860736252467049, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9947, + "step": 6853 + }, + { + "epoch": 0.5151059672328273, + "grad_norm": 1.9861753211797852, + "learning_rate": 1.999513157814572e-06, + "loss": 1.0043, + "step": 6854 + }, + { + "epoch": 0.5151811212986622, + "grad_norm": 1.5227829529611971, + "learning_rate": 1.9990263156579922e-06, + "loss": 1.0603, + "step": 6855 + }, + { + "epoch": 0.5152562753644973, + "grad_norm": 1.816377963594436, + "learning_rate": 1.998539473559106e-06, + "loss": 0.8729, + "step": 6856 + }, + { + "epoch": 0.5153314294303322, + "grad_norm": 38.46395177224384, + "learning_rate": 1.998052631546762e-06, + "loss": 0.8067, + "step": 6857 + }, + { + "epoch": 0.5154065834961672, + "grad_norm": 1.6358839286769822, + "learning_rate": 1.9975657896498073e-06, + "loss": 1.0174, + "step": 6858 + }, + { + "epoch": 0.5154817375620021, + "grad_norm": 1.63330859081186, + "learning_rate": 1.9970789478970882e-06, + "loss": 1.0632, + "step": 6859 + }, + { + "epoch": 0.515556891627837, + "grad_norm": 1.4723743604028259, + "learning_rate": 1.996592106317453e-06, + "loss": 0.9802, + "step": 6860 + }, + { + "epoch": 0.515632045693672, + "grad_norm": 1.5569084827278987, + "learning_rate": 1.9961052649397486e-06, + "loss": 0.8854, + "step": 6861 + }, + { + "epoch": 0.515707199759507, + "grad_norm": 3.702779890630421, + "learning_rate": 1.995618423792822e-06, + "loss": 0.9441, + "step": 6862 + }, + { + "epoch": 0.515782353825342, + "grad_norm": 1.8585489185084727, + "learning_rate": 1.9951315829055208e-06, + "loss": 1.0165, + "step": 6863 + }, + { + "epoch": 0.5158575078911769, + "grad_norm": 1.4361233961382103, + "learning_rate": 1.994644742306692e-06, + "loss": 0.9774, + "step": 6864 + }, + { + "epoch": 0.5159326619570119, + "grad_norm": 1.612012535164799, + "learning_rate": 1.9941579020251826e-06, + "loss": 0.9389, + "step": 6865 + }, + { + "epoch": 0.5160078160228468, + "grad_norm": 3.728679319677966, + "learning_rate": 1.9936710620898396e-06, + "loss": 0.9442, + "step": 6866 + }, + { + "epoch": 0.5160829700886818, + "grad_norm": 1.801273369726867, + "learning_rate": 1.9931842225295108e-06, + "loss": 0.9791, + "step": 6867 + }, + { + "epoch": 0.5161581241545168, + "grad_norm": 2.084159178762955, + "learning_rate": 1.9926973833730426e-06, + "loss": 1.0806, + "step": 6868 + }, + { + "epoch": 0.5162332782203517, + "grad_norm": 2.044517382354992, + "learning_rate": 1.9922105446492824e-06, + "loss": 0.983, + "step": 6869 + }, + { + "epoch": 0.5163084322861867, + "grad_norm": 1.5723520575039218, + "learning_rate": 1.9917237063870777e-06, + "loss": 1.0662, + "step": 6870 + }, + { + "epoch": 0.5163835863520216, + "grad_norm": 1.3841364067225657, + "learning_rate": 1.9912368686152746e-06, + "loss": 1.0022, + "step": 6871 + }, + { + "epoch": 0.5164587404178566, + "grad_norm": 1.7999096662318497, + "learning_rate": 1.990750031362721e-06, + "loss": 1.0345, + "step": 6872 + }, + { + "epoch": 0.5165338944836916, + "grad_norm": 1.442057640701378, + "learning_rate": 1.9902631946582627e-06, + "loss": 0.9773, + "step": 6873 + }, + { + "epoch": 0.5166090485495265, + "grad_norm": 1.8925780620122217, + "learning_rate": 1.9897763585307483e-06, + "loss": 0.9842, + "step": 6874 + }, + { + "epoch": 0.5166842026153615, + "grad_norm": 1.7273026342860063, + "learning_rate": 1.989289523009024e-06, + "loss": 1.0453, + "step": 6875 + }, + { + "epoch": 0.5167593566811964, + "grad_norm": 1.7855698922225354, + "learning_rate": 1.988802688121935e-06, + "loss": 0.9504, + "step": 6876 + }, + { + "epoch": 0.5168345107470315, + "grad_norm": 1.4773307295396447, + "learning_rate": 1.988315853898331e-06, + "loss": 0.9373, + "step": 6877 + }, + { + "epoch": 0.5169096648128664, + "grad_norm": 1.7170693243180837, + "learning_rate": 1.9878290203670563e-06, + "loss": 1.0532, + "step": 6878 + }, + { + "epoch": 0.5169848188787013, + "grad_norm": 2.161858713839157, + "learning_rate": 1.98734218755696e-06, + "loss": 1.0125, + "step": 6879 + }, + { + "epoch": 0.5170599729445363, + "grad_norm": 1.4195810614723423, + "learning_rate": 1.986855355496887e-06, + "loss": 1.0841, + "step": 6880 + }, + { + "epoch": 0.5171351270103712, + "grad_norm": 1.8410528466152598, + "learning_rate": 1.986368524215684e-06, + "loss": 1.0041, + "step": 6881 + }, + { + "epoch": 0.5172102810762063, + "grad_norm": 1.441420144114946, + "learning_rate": 1.985881693742199e-06, + "loss": 0.983, + "step": 6882 + }, + { + "epoch": 0.5172854351420412, + "grad_norm": 1.59263324880743, + "learning_rate": 1.985394864105277e-06, + "loss": 0.9408, + "step": 6883 + }, + { + "epoch": 0.5173605892078762, + "grad_norm": 0.8295325068013235, + "learning_rate": 1.9849080353337656e-06, + "loss": 0.858, + "step": 6884 + }, + { + "epoch": 0.5174357432737111, + "grad_norm": 7.137665583999692, + "learning_rate": 1.984421207456511e-06, + "loss": 0.8251, + "step": 6885 + }, + { + "epoch": 0.517510897339546, + "grad_norm": 1.772417825642186, + "learning_rate": 1.983934380502359e-06, + "loss": 0.9586, + "step": 6886 + }, + { + "epoch": 0.517586051405381, + "grad_norm": 1.6149337470339022, + "learning_rate": 1.983447554500157e-06, + "loss": 0.879, + "step": 6887 + }, + { + "epoch": 0.517661205471216, + "grad_norm": 1.5989523109877637, + "learning_rate": 1.9829607294787503e-06, + "loss": 0.9522, + "step": 6888 + }, + { + "epoch": 0.517736359537051, + "grad_norm": 1.8052550905412368, + "learning_rate": 1.9824739054669864e-06, + "loss": 0.8708, + "step": 6889 + }, + { + "epoch": 0.5178115136028859, + "grad_norm": 2.686431583677142, + "learning_rate": 1.98198708249371e-06, + "loss": 0.9939, + "step": 6890 + }, + { + "epoch": 0.5178866676687208, + "grad_norm": 1.339220930984959, + "learning_rate": 1.9815002605877685e-06, + "loss": 0.9305, + "step": 6891 + }, + { + "epoch": 0.5179618217345558, + "grad_norm": 2.011310543384266, + "learning_rate": 1.9810134397780073e-06, + "loss": 0.7537, + "step": 6892 + }, + { + "epoch": 0.5180369758003908, + "grad_norm": 1.860305553277847, + "learning_rate": 1.980526620093272e-06, + "loss": 1.0527, + "step": 6893 + }, + { + "epoch": 0.5181121298662258, + "grad_norm": 1.8960680833736774, + "learning_rate": 1.9800398015624095e-06, + "loss": 0.9811, + "step": 6894 + }, + { + "epoch": 0.5181872839320607, + "grad_norm": 1.429730023861705, + "learning_rate": 1.9795529842142644e-06, + "loss": 1.056, + "step": 6895 + }, + { + "epoch": 0.5182624379978957, + "grad_norm": 1.6502509966591408, + "learning_rate": 1.979066168077684e-06, + "loss": 0.9311, + "step": 6896 + }, + { + "epoch": 0.5183375920637306, + "grad_norm": 1.850074910889643, + "learning_rate": 1.978579353181513e-06, + "loss": 0.9267, + "step": 6897 + }, + { + "epoch": 0.5184127461295656, + "grad_norm": 1.9269531893387, + "learning_rate": 1.9780925395545977e-06, + "loss": 0.9877, + "step": 6898 + }, + { + "epoch": 0.5184879001954006, + "grad_norm": 3.1550607990385675, + "learning_rate": 1.977605727225783e-06, + "loss": 1.0328, + "step": 6899 + }, + { + "epoch": 0.5185630542612355, + "grad_norm": 1.9652201176404664, + "learning_rate": 1.977118916223915e-06, + "loss": 1.0213, + "step": 6900 + }, + { + "epoch": 0.5186382083270705, + "grad_norm": 1.4491811577787344, + "learning_rate": 1.9766321065778387e-06, + "loss": 1.0215, + "step": 6901 + }, + { + "epoch": 0.5187133623929054, + "grad_norm": 2.2534210899389273, + "learning_rate": 1.9761452983163996e-06, + "loss": 1.0059, + "step": 6902 + }, + { + "epoch": 0.5187885164587405, + "grad_norm": 1.652893768464783, + "learning_rate": 1.9756584914684425e-06, + "loss": 1.029, + "step": 6903 + }, + { + "epoch": 0.5188636705245754, + "grad_norm": 1.7270707983184055, + "learning_rate": 1.975171686062814e-06, + "loss": 1.016, + "step": 6904 + }, + { + "epoch": 0.5189388245904103, + "grad_norm": 1.5623103980422497, + "learning_rate": 1.974684882128357e-06, + "loss": 1.0036, + "step": 6905 + }, + { + "epoch": 0.5190139786562453, + "grad_norm": 1.5351741100563336, + "learning_rate": 1.974198079693918e-06, + "loss": 1.0652, + "step": 6906 + }, + { + "epoch": 0.5190891327220802, + "grad_norm": 1.6385221537959822, + "learning_rate": 1.973711278788342e-06, + "loss": 0.9459, + "step": 6907 + }, + { + "epoch": 0.5191642867879153, + "grad_norm": 1.6099374029181064, + "learning_rate": 1.973224479440473e-06, + "loss": 0.8948, + "step": 6908 + }, + { + "epoch": 0.5192394408537502, + "grad_norm": 1.4828058883792932, + "learning_rate": 1.972737681679157e-06, + "loss": 0.9097, + "step": 6909 + }, + { + "epoch": 0.5193145949195852, + "grad_norm": 3.026052364433319, + "learning_rate": 1.9722508855332367e-06, + "loss": 0.9718, + "step": 6910 + }, + { + "epoch": 0.5193897489854201, + "grad_norm": 1.6571765683061932, + "learning_rate": 1.971764091031558e-06, + "loss": 0.9381, + "step": 6911 + }, + { + "epoch": 0.519464903051255, + "grad_norm": 1.4489301615247279, + "learning_rate": 1.971277298202965e-06, + "loss": 0.9755, + "step": 6912 + }, + { + "epoch": 0.5195400571170901, + "grad_norm": 1.52808379759366, + "learning_rate": 1.9707905070763027e-06, + "loss": 0.908, + "step": 6913 + }, + { + "epoch": 0.519615211182925, + "grad_norm": 1.6976386780757642, + "learning_rate": 1.970303717680414e-06, + "loss": 0.9939, + "step": 6914 + }, + { + "epoch": 0.51969036524876, + "grad_norm": 1.6649061759655006, + "learning_rate": 1.9698169300441435e-06, + "loss": 1.0142, + "step": 6915 + }, + { + "epoch": 0.5197655193145949, + "grad_norm": 1.9893939608845714, + "learning_rate": 1.9693301441963363e-06, + "loss": 0.993, + "step": 6916 + }, + { + "epoch": 0.5198406733804298, + "grad_norm": 0.7109574535210691, + "learning_rate": 1.9688433601658345e-06, + "loss": 0.8583, + "step": 6917 + }, + { + "epoch": 0.5199158274462649, + "grad_norm": 5.979663717453359, + "learning_rate": 1.9683565779814838e-06, + "loss": 1.0131, + "step": 6918 + }, + { + "epoch": 0.5199909815120998, + "grad_norm": 1.8886660050120454, + "learning_rate": 1.967869797672127e-06, + "loss": 1.0323, + "step": 6919 + }, + { + "epoch": 0.5200661355779348, + "grad_norm": 1.9645204069456825, + "learning_rate": 1.967383019266607e-06, + "loss": 1.0028, + "step": 6920 + }, + { + "epoch": 0.5201412896437697, + "grad_norm": 1.6289973610844322, + "learning_rate": 1.966896242793769e-06, + "loss": 0.9823, + "step": 6921 + }, + { + "epoch": 0.5202164437096047, + "grad_norm": 1.8634649222194024, + "learning_rate": 1.9664094682824545e-06, + "loss": 1.0038, + "step": 6922 + }, + { + "epoch": 0.5202915977754397, + "grad_norm": 1.8484679812113265, + "learning_rate": 1.965922695761508e-06, + "loss": 0.9726, + "step": 6923 + }, + { + "epoch": 0.5203667518412746, + "grad_norm": 2.164951073179594, + "learning_rate": 1.9654359252597723e-06, + "loss": 1.0451, + "step": 6924 + }, + { + "epoch": 0.5204419059071096, + "grad_norm": 1.5367858623827388, + "learning_rate": 1.96494915680609e-06, + "loss": 0.8854, + "step": 6925 + }, + { + "epoch": 0.5205170599729445, + "grad_norm": 1.677352630918647, + "learning_rate": 1.964462390429305e-06, + "loss": 0.8949, + "step": 6926 + }, + { + "epoch": 0.5205922140387795, + "grad_norm": 1.7690885886981085, + "learning_rate": 1.963975626158259e-06, + "loss": 1.0148, + "step": 6927 + }, + { + "epoch": 0.5206673681046144, + "grad_norm": 2.0073392436458795, + "learning_rate": 1.963488864021795e-06, + "loss": 1.0813, + "step": 6928 + }, + { + "epoch": 0.5207425221704495, + "grad_norm": 2.103644831947516, + "learning_rate": 1.9630021040487557e-06, + "loss": 0.9523, + "step": 6929 + }, + { + "epoch": 0.5208176762362844, + "grad_norm": 11.964835460534184, + "learning_rate": 1.962515346267984e-06, + "loss": 1.0314, + "step": 6930 + }, + { + "epoch": 0.5208928303021193, + "grad_norm": 1.3639734909888015, + "learning_rate": 1.9620285907083213e-06, + "loss": 1.0299, + "step": 6931 + }, + { + "epoch": 0.5209679843679543, + "grad_norm": 1.6154463235715448, + "learning_rate": 1.9615418373986097e-06, + "loss": 1.0082, + "step": 6932 + }, + { + "epoch": 0.5210431384337892, + "grad_norm": 1.6565652223430458, + "learning_rate": 1.961055086367692e-06, + "loss": 1.0246, + "step": 6933 + }, + { + "epoch": 0.5211182924996243, + "grad_norm": 1.507124986200395, + "learning_rate": 1.960568337644409e-06, + "loss": 0.934, + "step": 6934 + }, + { + "epoch": 0.5211934465654592, + "grad_norm": 1.6886348440578152, + "learning_rate": 1.9600815912576034e-06, + "loss": 0.9182, + "step": 6935 + }, + { + "epoch": 0.5212686006312941, + "grad_norm": 1.6773064348452524, + "learning_rate": 1.9595948472361163e-06, + "loss": 0.9712, + "step": 6936 + }, + { + "epoch": 0.5213437546971291, + "grad_norm": 1.7726977986106303, + "learning_rate": 1.959108105608788e-06, + "loss": 0.9901, + "step": 6937 + }, + { + "epoch": 0.521418908762964, + "grad_norm": 0.7059586962269843, + "learning_rate": 1.958621366404463e-06, + "loss": 0.8389, + "step": 6938 + }, + { + "epoch": 0.5214940628287991, + "grad_norm": 2.7316136831653917, + "learning_rate": 1.958134629651979e-06, + "loss": 1.0691, + "step": 6939 + }, + { + "epoch": 0.521569216894634, + "grad_norm": 1.7574878986316436, + "learning_rate": 1.957647895380179e-06, + "loss": 0.9651, + "step": 6940 + }, + { + "epoch": 0.521644370960469, + "grad_norm": 1.4969757906243688, + "learning_rate": 1.9571611636179037e-06, + "loss": 0.9891, + "step": 6941 + }, + { + "epoch": 0.5217195250263039, + "grad_norm": 1.6601687442084996, + "learning_rate": 1.956674434393993e-06, + "loss": 1.0472, + "step": 6942 + }, + { + "epoch": 0.5217946790921388, + "grad_norm": 1.6293542918128339, + "learning_rate": 1.9561877077372884e-06, + "loss": 0.9675, + "step": 6943 + }, + { + "epoch": 0.5218698331579739, + "grad_norm": 1.4511139602707808, + "learning_rate": 1.9557009836766294e-06, + "loss": 0.8927, + "step": 6944 + }, + { + "epoch": 0.5219449872238088, + "grad_norm": 1.6479909572260634, + "learning_rate": 1.9552142622408574e-06, + "loss": 1.0628, + "step": 6945 + }, + { + "epoch": 0.5220201412896438, + "grad_norm": 1.8797087987642809, + "learning_rate": 1.954727543458812e-06, + "loss": 0.9125, + "step": 6946 + }, + { + "epoch": 0.5220952953554787, + "grad_norm": 2.021625075304241, + "learning_rate": 1.9542408273593324e-06, + "loss": 0.9945, + "step": 6947 + }, + { + "epoch": 0.5221704494213137, + "grad_norm": 2.0283896627678413, + "learning_rate": 1.9537541139712594e-06, + "loss": 1.0971, + "step": 6948 + }, + { + "epoch": 0.5222456034871487, + "grad_norm": 1.587918176916374, + "learning_rate": 1.9532674033234315e-06, + "loss": 1.036, + "step": 6949 + }, + { + "epoch": 0.5223207575529836, + "grad_norm": 1.76324160239317, + "learning_rate": 1.95278069544469e-06, + "loss": 0.9809, + "step": 6950 + }, + { + "epoch": 0.5223959116188186, + "grad_norm": 2.809747238873312, + "learning_rate": 1.952293990363872e-06, + "loss": 1.0742, + "step": 6951 + }, + { + "epoch": 0.5224710656846535, + "grad_norm": 1.673393771813931, + "learning_rate": 1.9518072881098185e-06, + "loss": 0.9759, + "step": 6952 + }, + { + "epoch": 0.5225462197504885, + "grad_norm": 0.8992503925393684, + "learning_rate": 1.9513205887113675e-06, + "loss": 0.9437, + "step": 6953 + }, + { + "epoch": 0.5226213738163235, + "grad_norm": 1.57257645656132, + "learning_rate": 1.9508338921973576e-06, + "loss": 0.8931, + "step": 6954 + }, + { + "epoch": 0.5226965278821585, + "grad_norm": 3.149401647262351, + "learning_rate": 1.9503471985966285e-06, + "loss": 1.0331, + "step": 6955 + }, + { + "epoch": 0.5227716819479934, + "grad_norm": 1.6605754726893436, + "learning_rate": 1.9498605079380176e-06, + "loss": 0.9354, + "step": 6956 + }, + { + "epoch": 0.5228468360138283, + "grad_norm": 1.4731884710518999, + "learning_rate": 1.949373820250364e-06, + "loss": 0.9621, + "step": 6957 + }, + { + "epoch": 0.5229219900796633, + "grad_norm": 1.7874050541984174, + "learning_rate": 1.9488871355625054e-06, + "loss": 0.999, + "step": 6958 + }, + { + "epoch": 0.5229971441454982, + "grad_norm": 1.4903877149412426, + "learning_rate": 1.9484004539032786e-06, + "loss": 1.001, + "step": 6959 + }, + { + "epoch": 0.5230722982113333, + "grad_norm": 1.6155460891784441, + "learning_rate": 1.947913775301524e-06, + "loss": 0.9768, + "step": 6960 + }, + { + "epoch": 0.5231474522771682, + "grad_norm": 1.5933779861997994, + "learning_rate": 1.9474270997860766e-06, + "loss": 0.9693, + "step": 6961 + }, + { + "epoch": 0.5232226063430031, + "grad_norm": 2.0489117920759212, + "learning_rate": 1.946940427385776e-06, + "loss": 1.0347, + "step": 6962 + }, + { + "epoch": 0.5232977604088381, + "grad_norm": 2.0625907415600384, + "learning_rate": 1.9464537581294576e-06, + "loss": 0.9448, + "step": 6963 + }, + { + "epoch": 0.523372914474673, + "grad_norm": 0.6499084655794388, + "learning_rate": 1.945967092045959e-06, + "loss": 0.8011, + "step": 6964 + }, + { + "epoch": 0.5234480685405081, + "grad_norm": 2.2075606259826515, + "learning_rate": 1.945480429164117e-06, + "loss": 0.936, + "step": 6965 + }, + { + "epoch": 0.523523222606343, + "grad_norm": 1.4569506584562661, + "learning_rate": 1.944993769512768e-06, + "loss": 0.9558, + "step": 6966 + }, + { + "epoch": 0.523598376672178, + "grad_norm": 3.106400643884679, + "learning_rate": 1.9445071131207497e-06, + "loss": 0.9892, + "step": 6967 + }, + { + "epoch": 0.5236735307380129, + "grad_norm": 2.0922477229184926, + "learning_rate": 1.9440204600168975e-06, + "loss": 1.0058, + "step": 6968 + }, + { + "epoch": 0.5237486848038478, + "grad_norm": 1.5795189772308444, + "learning_rate": 1.9435338102300467e-06, + "loss": 1.0083, + "step": 6969 + }, + { + "epoch": 0.5238238388696829, + "grad_norm": 2.8945557957482366, + "learning_rate": 1.943047163789034e-06, + "loss": 1.002, + "step": 6970 + }, + { + "epoch": 0.5238989929355178, + "grad_norm": 1.7910285375164983, + "learning_rate": 1.942560520722695e-06, + "loss": 0.8933, + "step": 6971 + }, + { + "epoch": 0.5239741470013528, + "grad_norm": 1.440376223848911, + "learning_rate": 1.9420738810598653e-06, + "loss": 0.9679, + "step": 6972 + }, + { + "epoch": 0.5240493010671877, + "grad_norm": 1.5378096074939696, + "learning_rate": 1.9415872448293796e-06, + "loss": 1.0688, + "step": 6973 + }, + { + "epoch": 0.5241244551330227, + "grad_norm": 1.596931871569762, + "learning_rate": 1.9411006120600737e-06, + "loss": 0.9541, + "step": 6974 + }, + { + "epoch": 0.5241996091988577, + "grad_norm": 2.6629785922454516, + "learning_rate": 1.940613982780782e-06, + "loss": 0.9551, + "step": 6975 + }, + { + "epoch": 0.5242747632646926, + "grad_norm": 1.81859012700741, + "learning_rate": 1.940127357020339e-06, + "loss": 0.9361, + "step": 6976 + }, + { + "epoch": 0.5243499173305276, + "grad_norm": 2.3311079743922516, + "learning_rate": 1.9396407348075796e-06, + "loss": 1.0854, + "step": 6977 + }, + { + "epoch": 0.5244250713963625, + "grad_norm": 1.813319908693118, + "learning_rate": 1.939154116171337e-06, + "loss": 1.028, + "step": 6978 + }, + { + "epoch": 0.5245002254621975, + "grad_norm": 2.282121988103442, + "learning_rate": 1.9386675011404473e-06, + "loss": 0.9693, + "step": 6979 + }, + { + "epoch": 0.5245753795280325, + "grad_norm": 4.030661843146709, + "learning_rate": 1.9381808897437427e-06, + "loss": 1.0074, + "step": 6980 + }, + { + "epoch": 0.5246505335938674, + "grad_norm": 1.6354518165779586, + "learning_rate": 1.9376942820100563e-06, + "loss": 0.9123, + "step": 6981 + }, + { + "epoch": 0.5247256876597024, + "grad_norm": 1.582709800348292, + "learning_rate": 1.937207677968223e-06, + "loss": 0.9603, + "step": 6982 + }, + { + "epoch": 0.5248008417255373, + "grad_norm": 4.204481168438259, + "learning_rate": 1.9367210776470744e-06, + "loss": 1.0158, + "step": 6983 + }, + { + "epoch": 0.5248759957913723, + "grad_norm": 1.8012251234487646, + "learning_rate": 1.9362344810754455e-06, + "loss": 0.9621, + "step": 6984 + }, + { + "epoch": 0.5249511498572073, + "grad_norm": 1.4382735948913186, + "learning_rate": 1.935747888282167e-06, + "loss": 0.97, + "step": 6985 + }, + { + "epoch": 0.5250263039230423, + "grad_norm": 1.4478414319295185, + "learning_rate": 1.935261299296072e-06, + "loss": 0.9871, + "step": 6986 + }, + { + "epoch": 0.5251014579888772, + "grad_norm": 1.6913378836049684, + "learning_rate": 1.9347747141459934e-06, + "loss": 0.9902, + "step": 6987 + }, + { + "epoch": 0.5251766120547121, + "grad_norm": 1.7124379863787749, + "learning_rate": 1.9342881328607625e-06, + "loss": 0.9516, + "step": 6988 + }, + { + "epoch": 0.5252517661205471, + "grad_norm": 2.0170881298596264, + "learning_rate": 1.9338015554692116e-06, + "loss": 0.9306, + "step": 6989 + }, + { + "epoch": 0.525326920186382, + "grad_norm": 1.9120420418749848, + "learning_rate": 1.933314982000172e-06, + "loss": 0.9908, + "step": 6990 + }, + { + "epoch": 0.5254020742522171, + "grad_norm": 1.195219048730245, + "learning_rate": 1.932828412482475e-06, + "loss": 1.0159, + "step": 6991 + }, + { + "epoch": 0.525477228318052, + "grad_norm": 1.785963147397248, + "learning_rate": 1.9323418469449517e-06, + "loss": 0.9937, + "step": 6992 + }, + { + "epoch": 0.525552382383887, + "grad_norm": 1.369398672298848, + "learning_rate": 1.931855285416433e-06, + "loss": 1.0334, + "step": 6993 + }, + { + "epoch": 0.5256275364497219, + "grad_norm": 4.817896010033942, + "learning_rate": 1.93136872792575e-06, + "loss": 0.9451, + "step": 6994 + }, + { + "epoch": 0.5257026905155568, + "grad_norm": 1.626658840056388, + "learning_rate": 1.930882174501732e-06, + "loss": 0.9092, + "step": 6995 + }, + { + "epoch": 0.5257778445813919, + "grad_norm": 3.861307749021542, + "learning_rate": 1.93039562517321e-06, + "loss": 0.9602, + "step": 6996 + }, + { + "epoch": 0.5258529986472268, + "grad_norm": 1.7992296133239336, + "learning_rate": 1.929909079969014e-06, + "loss": 0.9526, + "step": 6997 + }, + { + "epoch": 0.5259281527130618, + "grad_norm": 0.662582925788787, + "learning_rate": 1.929422538917973e-06, + "loss": 0.8234, + "step": 6998 + }, + { + "epoch": 0.5260033067788967, + "grad_norm": 1.6161298011167349, + "learning_rate": 1.9289360020489177e-06, + "loss": 0.9291, + "step": 6999 + }, + { + "epoch": 0.5260784608447318, + "grad_norm": 1.5408318897169717, + "learning_rate": 1.928449469390675e-06, + "loss": 1.0481, + "step": 7000 + }, + { + "epoch": 0.5261536149105667, + "grad_norm": 1.3841002766493917, + "learning_rate": 1.927962940972077e-06, + "loss": 0.961, + "step": 7001 + }, + { + "epoch": 0.5262287689764016, + "grad_norm": 1.5898106826979994, + "learning_rate": 1.9274764168219495e-06, + "loss": 0.9397, + "step": 7002 + }, + { + "epoch": 0.5263039230422366, + "grad_norm": 1.200439130384655, + "learning_rate": 1.9269898969691214e-06, + "loss": 0.9492, + "step": 7003 + }, + { + "epoch": 0.5263790771080715, + "grad_norm": 2.130571018921618, + "learning_rate": 1.9265033814424227e-06, + "loss": 0.9793, + "step": 7004 + }, + { + "epoch": 0.5264542311739066, + "grad_norm": 1.888360361723405, + "learning_rate": 1.9260168702706794e-06, + "loss": 0.9467, + "step": 7005 + }, + { + "epoch": 0.5265293852397415, + "grad_norm": 1.4448696798409304, + "learning_rate": 1.9255303634827204e-06, + "loss": 1.0269, + "step": 7006 + }, + { + "epoch": 0.5266045393055764, + "grad_norm": 1.950153317904295, + "learning_rate": 1.9250438611073724e-06, + "loss": 0.8464, + "step": 7007 + }, + { + "epoch": 0.5266796933714114, + "grad_norm": 1.568051508690014, + "learning_rate": 1.924557363173462e-06, + "loss": 0.9705, + "step": 7008 + }, + { + "epoch": 0.5267548474372463, + "grad_norm": 5.748159362749985, + "learning_rate": 1.9240708697098174e-06, + "loss": 1.0438, + "step": 7009 + }, + { + "epoch": 0.5268300015030813, + "grad_norm": 1.5748379154666854, + "learning_rate": 1.9235843807452642e-06, + "loss": 1.0283, + "step": 7010 + }, + { + "epoch": 0.5269051555689163, + "grad_norm": 1.4755251030591416, + "learning_rate": 1.923097896308629e-06, + "loss": 0.924, + "step": 7011 + }, + { + "epoch": 0.5269803096347513, + "grad_norm": 1.3910333269680637, + "learning_rate": 1.9226114164287384e-06, + "loss": 1.0049, + "step": 7012 + }, + { + "epoch": 0.5270554637005862, + "grad_norm": 0.7063977264247825, + "learning_rate": 1.9221249411344173e-06, + "loss": 0.7941, + "step": 7013 + }, + { + "epoch": 0.5271306177664211, + "grad_norm": 1.864770386191271, + "learning_rate": 1.921638470454492e-06, + "loss": 0.8502, + "step": 7014 + }, + { + "epoch": 0.5272057718322561, + "grad_norm": 2.1462959852496644, + "learning_rate": 1.9211520044177866e-06, + "loss": 1.0858, + "step": 7015 + }, + { + "epoch": 0.5272809258980911, + "grad_norm": 1.880854874998673, + "learning_rate": 1.9206655430531277e-06, + "loss": 0.9027, + "step": 7016 + }, + { + "epoch": 0.5273560799639261, + "grad_norm": 1.597143188769479, + "learning_rate": 1.9201790863893387e-06, + "loss": 0.9812, + "step": 7017 + }, + { + "epoch": 0.527431234029761, + "grad_norm": 1.5289160399203923, + "learning_rate": 1.919692634455245e-06, + "loss": 0.869, + "step": 7018 + }, + { + "epoch": 0.527506388095596, + "grad_norm": 1.6716376875967323, + "learning_rate": 1.91920618727967e-06, + "loss": 0.9732, + "step": 7019 + }, + { + "epoch": 0.5275815421614309, + "grad_norm": 1.8529409436173208, + "learning_rate": 1.9187197448914374e-06, + "loss": 0.9389, + "step": 7020 + }, + { + "epoch": 0.5276566962272659, + "grad_norm": 1.4843210310912676, + "learning_rate": 1.918233307319371e-06, + "loss": 0.9785, + "step": 7021 + }, + { + "epoch": 0.5277318502931009, + "grad_norm": 1.533423224523224, + "learning_rate": 1.9177468745922944e-06, + "loss": 1.0199, + "step": 7022 + }, + { + "epoch": 0.5278070043589358, + "grad_norm": 1.7738044795997514, + "learning_rate": 1.917260446739031e-06, + "loss": 0.9803, + "step": 7023 + }, + { + "epoch": 0.5278821584247708, + "grad_norm": 1.8998693396163158, + "learning_rate": 1.916774023788403e-06, + "loss": 1.0207, + "step": 7024 + }, + { + "epoch": 0.5279573124906057, + "grad_norm": 1.6363641541412768, + "learning_rate": 1.9162876057692317e-06, + "loss": 0.9641, + "step": 7025 + }, + { + "epoch": 0.5280324665564406, + "grad_norm": 1.5420255844625947, + "learning_rate": 1.9158011927103413e-06, + "loss": 0.9054, + "step": 7026 + }, + { + "epoch": 0.5281076206222757, + "grad_norm": 1.6075492799125668, + "learning_rate": 1.9153147846405525e-06, + "loss": 0.9846, + "step": 7027 + }, + { + "epoch": 0.5281827746881106, + "grad_norm": 1.5228566162683073, + "learning_rate": 1.914828381588687e-06, + "loss": 0.8982, + "step": 7028 + }, + { + "epoch": 0.5282579287539456, + "grad_norm": 2.132710772313556, + "learning_rate": 1.9143419835835663e-06, + "loss": 1.0181, + "step": 7029 + }, + { + "epoch": 0.5283330828197805, + "grad_norm": 1.5080204163138133, + "learning_rate": 1.9138555906540103e-06, + "loss": 1.0271, + "step": 7030 + }, + { + "epoch": 0.5284082368856156, + "grad_norm": 5.732098452332822, + "learning_rate": 1.9133692028288413e-06, + "loss": 0.9584, + "step": 7031 + }, + { + "epoch": 0.5284833909514505, + "grad_norm": 5.210676698578074, + "learning_rate": 1.912882820136878e-06, + "loss": 0.88, + "step": 7032 + }, + { + "epoch": 0.5285585450172854, + "grad_norm": 1.417817331011157, + "learning_rate": 1.9123964426069416e-06, + "loss": 0.8978, + "step": 7033 + }, + { + "epoch": 0.5286336990831204, + "grad_norm": 1.7632835188729796, + "learning_rate": 1.9119100702678515e-06, + "loss": 0.9556, + "step": 7034 + }, + { + "epoch": 0.5287088531489553, + "grad_norm": 1.7883153296373495, + "learning_rate": 1.9114237031484266e-06, + "loss": 0.9898, + "step": 7035 + }, + { + "epoch": 0.5287840072147904, + "grad_norm": 1.3072233397522326, + "learning_rate": 1.9109373412774867e-06, + "loss": 1.0411, + "step": 7036 + }, + { + "epoch": 0.5288591612806253, + "grad_norm": 1.4900945395361231, + "learning_rate": 1.91045098468385e-06, + "loss": 1.0762, + "step": 7037 + }, + { + "epoch": 0.5289343153464603, + "grad_norm": 2.532364023216556, + "learning_rate": 1.9099646333963363e-06, + "loss": 0.951, + "step": 7038 + }, + { + "epoch": 0.5290094694122952, + "grad_norm": 0.7355575779537789, + "learning_rate": 1.9094782874437625e-06, + "loss": 0.8376, + "step": 7039 + }, + { + "epoch": 0.5290846234781301, + "grad_norm": 1.6514047249577533, + "learning_rate": 1.9089919468549464e-06, + "loss": 0.9604, + "step": 7040 + }, + { + "epoch": 0.5291597775439651, + "grad_norm": 0.7638758706418873, + "learning_rate": 1.9085056116587068e-06, + "loss": 0.8667, + "step": 7041 + }, + { + "epoch": 0.5292349316098001, + "grad_norm": 0.7959328486018716, + "learning_rate": 1.908019281883859e-06, + "loss": 0.8661, + "step": 7042 + }, + { + "epoch": 0.5293100856756351, + "grad_norm": 2.725342048385979, + "learning_rate": 1.9075329575592217e-06, + "loss": 0.9402, + "step": 7043 + }, + { + "epoch": 0.52938523974147, + "grad_norm": 1.4992332088057503, + "learning_rate": 1.9070466387136095e-06, + "loss": 0.9928, + "step": 7044 + }, + { + "epoch": 0.529460393807305, + "grad_norm": 1.6134932334859682, + "learning_rate": 1.906560325375841e-06, + "loss": 0.958, + "step": 7045 + }, + { + "epoch": 0.52953554787314, + "grad_norm": 1.67305891953149, + "learning_rate": 1.9060740175747317e-06, + "loss": 1.0231, + "step": 7046 + }, + { + "epoch": 0.5296107019389749, + "grad_norm": 1.6530426927927653, + "learning_rate": 1.9055877153390948e-06, + "loss": 1.0611, + "step": 7047 + }, + { + "epoch": 0.5296858560048099, + "grad_norm": 2.167590848961348, + "learning_rate": 1.9051014186977485e-06, + "loss": 0.9759, + "step": 7048 + }, + { + "epoch": 0.5297610100706448, + "grad_norm": 1.8420469532966346, + "learning_rate": 1.9046151276795062e-06, + "loss": 1.0142, + "step": 7049 + }, + { + "epoch": 0.5298361641364798, + "grad_norm": 2.421169547426713, + "learning_rate": 1.904128842313183e-06, + "loss": 0.8635, + "step": 7050 + }, + { + "epoch": 0.5299113182023147, + "grad_norm": 2.2416569417674004, + "learning_rate": 1.9036425626275929e-06, + "loss": 1.0301, + "step": 7051 + }, + { + "epoch": 0.5299864722681497, + "grad_norm": 1.9141831340532622, + "learning_rate": 1.9031562886515497e-06, + "loss": 0.9615, + "step": 7052 + }, + { + "epoch": 0.5300616263339847, + "grad_norm": 2.7471107564010904, + "learning_rate": 1.9026700204138676e-06, + "loss": 0.9875, + "step": 7053 + }, + { + "epoch": 0.5301367803998196, + "grad_norm": 3.2757624197340673, + "learning_rate": 1.9021837579433593e-06, + "loss": 1.0249, + "step": 7054 + }, + { + "epoch": 0.5302119344656546, + "grad_norm": 1.3792347287135893, + "learning_rate": 1.9016975012688382e-06, + "loss": 1.0176, + "step": 7055 + }, + { + "epoch": 0.5302870885314895, + "grad_norm": 0.7794104471841191, + "learning_rate": 1.901211250419116e-06, + "loss": 0.8328, + "step": 7056 + }, + { + "epoch": 0.5303622425973246, + "grad_norm": 1.8906324402213492, + "learning_rate": 1.900725005423006e-06, + "loss": 1.0514, + "step": 7057 + }, + { + "epoch": 0.5304373966631595, + "grad_norm": 1.9023682516349172, + "learning_rate": 1.9002387663093195e-06, + "loss": 0.9823, + "step": 7058 + }, + { + "epoch": 0.5305125507289944, + "grad_norm": 1.5058030094428256, + "learning_rate": 1.899752533106868e-06, + "loss": 1.1099, + "step": 7059 + }, + { + "epoch": 0.5305877047948294, + "grad_norm": 2.5657094999609167, + "learning_rate": 1.8992663058444629e-06, + "loss": 0.8991, + "step": 7060 + }, + { + "epoch": 0.5306628588606643, + "grad_norm": 0.740588745002409, + "learning_rate": 1.8987800845509146e-06, + "loss": 0.8462, + "step": 7061 + }, + { + "epoch": 0.5307380129264994, + "grad_norm": 2.266390125384298, + "learning_rate": 1.8982938692550344e-06, + "loss": 0.8774, + "step": 7062 + }, + { + "epoch": 0.5308131669923343, + "grad_norm": 1.3761002733377656, + "learning_rate": 1.8978076599856317e-06, + "loss": 1.0376, + "step": 7063 + }, + { + "epoch": 0.5308883210581693, + "grad_norm": 1.6424619291123232, + "learning_rate": 1.897321456771516e-06, + "loss": 0.982, + "step": 7064 + }, + { + "epoch": 0.5309634751240042, + "grad_norm": 1.4645754148758487, + "learning_rate": 1.8968352596414977e-06, + "loss": 0.9644, + "step": 7065 + }, + { + "epoch": 0.5310386291898391, + "grad_norm": 1.9279433178741083, + "learning_rate": 1.8963490686243847e-06, + "loss": 0.9758, + "step": 7066 + }, + { + "epoch": 0.5311137832556742, + "grad_norm": 2.1863105810026178, + "learning_rate": 1.895862883748987e-06, + "loss": 1.0497, + "step": 7067 + }, + { + "epoch": 0.5311889373215091, + "grad_norm": 1.7062668125316378, + "learning_rate": 1.895376705044112e-06, + "loss": 1.0116, + "step": 7068 + }, + { + "epoch": 0.5312640913873441, + "grad_norm": 3.0316572407356963, + "learning_rate": 1.8948905325385675e-06, + "loss": 0.9854, + "step": 7069 + }, + { + "epoch": 0.531339245453179, + "grad_norm": 1.2723921715518052, + "learning_rate": 1.894404366261162e-06, + "loss": 0.906, + "step": 7070 + }, + { + "epoch": 0.5314143995190139, + "grad_norm": 1.5453647986531005, + "learning_rate": 1.8939182062407017e-06, + "loss": 0.9684, + "step": 7071 + }, + { + "epoch": 0.531489553584849, + "grad_norm": 0.6601813497794053, + "learning_rate": 1.8934320525059944e-06, + "loss": 0.7853, + "step": 7072 + }, + { + "epoch": 0.5315647076506839, + "grad_norm": 1.615648799993945, + "learning_rate": 1.8929459050858458e-06, + "loss": 0.9646, + "step": 7073 + }, + { + "epoch": 0.5316398617165189, + "grad_norm": 1.8862489861851541, + "learning_rate": 1.892459764009062e-06, + "loss": 0.9452, + "step": 7074 + }, + { + "epoch": 0.5317150157823538, + "grad_norm": 1.3640779719398353, + "learning_rate": 1.8919736293044495e-06, + "loss": 0.9135, + "step": 7075 + }, + { + "epoch": 0.5317901698481888, + "grad_norm": 1.2713959600175895, + "learning_rate": 1.8914875010008124e-06, + "loss": 1.005, + "step": 7076 + }, + { + "epoch": 0.5318653239140237, + "grad_norm": 1.846046375469124, + "learning_rate": 1.891001379126957e-06, + "loss": 0.9964, + "step": 7077 + }, + { + "epoch": 0.5319404779798587, + "grad_norm": 1.5972129001437536, + "learning_rate": 1.8905152637116868e-06, + "loss": 0.9324, + "step": 7078 + }, + { + "epoch": 0.5320156320456937, + "grad_norm": 2.252855290580518, + "learning_rate": 1.890029154783807e-06, + "loss": 0.8531, + "step": 7079 + }, + { + "epoch": 0.5320907861115286, + "grad_norm": 1.7788848567408375, + "learning_rate": 1.889543052372121e-06, + "loss": 1.056, + "step": 7080 + }, + { + "epoch": 0.5321659401773636, + "grad_norm": 1.8808102546845566, + "learning_rate": 1.8890569565054313e-06, + "loss": 1.0332, + "step": 7081 + }, + { + "epoch": 0.5322410942431985, + "grad_norm": 2.777507964911659, + "learning_rate": 1.8885708672125425e-06, + "loss": 0.9135, + "step": 7082 + }, + { + "epoch": 0.5323162483090336, + "grad_norm": 1.5574922039149144, + "learning_rate": 1.888084784522256e-06, + "loss": 1.0473, + "step": 7083 + }, + { + "epoch": 0.5323914023748685, + "grad_norm": 1.6252988393803953, + "learning_rate": 1.8875987084633748e-06, + "loss": 0.9907, + "step": 7084 + }, + { + "epoch": 0.5324665564407034, + "grad_norm": 4.478120382962771, + "learning_rate": 1.887112639064701e-06, + "loss": 0.9166, + "step": 7085 + }, + { + "epoch": 0.5325417105065384, + "grad_norm": 1.3899597548414993, + "learning_rate": 1.8866265763550344e-06, + "loss": 0.885, + "step": 7086 + }, + { + "epoch": 0.5326168645723733, + "grad_norm": 1.7120206705030925, + "learning_rate": 1.8861405203631786e-06, + "loss": 1.1131, + "step": 7087 + }, + { + "epoch": 0.5326920186382084, + "grad_norm": 1.5968446045612368, + "learning_rate": 1.8856544711179317e-06, + "loss": 0.9562, + "step": 7088 + }, + { + "epoch": 0.5327671727040433, + "grad_norm": 1.6867167456483791, + "learning_rate": 1.8851684286480962e-06, + "loss": 0.907, + "step": 7089 + }, + { + "epoch": 0.5328423267698783, + "grad_norm": 1.63846795105635, + "learning_rate": 1.884682392982471e-06, + "loss": 0.8981, + "step": 7090 + }, + { + "epoch": 0.5329174808357132, + "grad_norm": 1.824451374521438, + "learning_rate": 1.884196364149855e-06, + "loss": 0.9429, + "step": 7091 + }, + { + "epoch": 0.5329926349015481, + "grad_norm": 1.7061854870795263, + "learning_rate": 1.8837103421790483e-06, + "loss": 0.968, + "step": 7092 + }, + { + "epoch": 0.5330677889673832, + "grad_norm": 2.115332547460667, + "learning_rate": 1.8832243270988488e-06, + "loss": 0.9062, + "step": 7093 + }, + { + "epoch": 0.5331429430332181, + "grad_norm": 1.540543207005291, + "learning_rate": 1.8827383189380556e-06, + "loss": 0.9803, + "step": 7094 + }, + { + "epoch": 0.5332180970990531, + "grad_norm": 1.5974810441255107, + "learning_rate": 1.8822523177254658e-06, + "loss": 0.9303, + "step": 7095 + }, + { + "epoch": 0.533293251164888, + "grad_norm": 3.0098971190248776, + "learning_rate": 1.881766323489877e-06, + "loss": 1.0037, + "step": 7096 + }, + { + "epoch": 0.5333684052307229, + "grad_norm": 2.296822994966904, + "learning_rate": 1.8812803362600865e-06, + "loss": 0.9515, + "step": 7097 + }, + { + "epoch": 0.533443559296558, + "grad_norm": 2.477745567800472, + "learning_rate": 1.8807943560648903e-06, + "loss": 0.9326, + "step": 7098 + }, + { + "epoch": 0.5335187133623929, + "grad_norm": 1.6336142291952227, + "learning_rate": 1.8803083829330853e-06, + "loss": 0.9207, + "step": 7099 + }, + { + "epoch": 0.5335938674282279, + "grad_norm": 1.7929078467835722, + "learning_rate": 1.8798224168934664e-06, + "loss": 1.0072, + "step": 7100 + }, + { + "epoch": 0.5336690214940628, + "grad_norm": 1.5469129153142132, + "learning_rate": 1.87933645797483e-06, + "loss": 0.8924, + "step": 7101 + }, + { + "epoch": 0.5337441755598978, + "grad_norm": 1.4581784393158541, + "learning_rate": 1.8788505062059708e-06, + "loss": 0.9975, + "step": 7102 + }, + { + "epoch": 0.5338193296257328, + "grad_norm": 1.682234071231827, + "learning_rate": 1.8783645616156822e-06, + "loss": 0.9722, + "step": 7103 + }, + { + "epoch": 0.5338944836915677, + "grad_norm": 2.81104597017038, + "learning_rate": 1.8778786242327598e-06, + "loss": 1.0223, + "step": 7104 + }, + { + "epoch": 0.5339696377574027, + "grad_norm": 1.3902747752618438, + "learning_rate": 1.877392694085996e-06, + "loss": 0.9974, + "step": 7105 + }, + { + "epoch": 0.5340447918232376, + "grad_norm": 1.7053169354169115, + "learning_rate": 1.876906771204185e-06, + "loss": 1.0806, + "step": 7106 + }, + { + "epoch": 0.5341199458890726, + "grad_norm": 1.6021715448283362, + "learning_rate": 1.8764208556161192e-06, + "loss": 0.972, + "step": 7107 + }, + { + "epoch": 0.5341950999549075, + "grad_norm": 1.8170729859368098, + "learning_rate": 1.87593494735059e-06, + "loss": 0.913, + "step": 7108 + }, + { + "epoch": 0.5342702540207426, + "grad_norm": 2.4297959282038613, + "learning_rate": 1.8754490464363917e-06, + "loss": 0.888, + "step": 7109 + }, + { + "epoch": 0.5343454080865775, + "grad_norm": 1.5775084093598208, + "learning_rate": 1.8749631529023129e-06, + "loss": 0.9796, + "step": 7110 + }, + { + "epoch": 0.5344205621524124, + "grad_norm": 2.9462125096223915, + "learning_rate": 1.874477266777147e-06, + "loss": 0.9593, + "step": 7111 + }, + { + "epoch": 0.5344957162182474, + "grad_norm": 1.6168534675641886, + "learning_rate": 1.8739913880896835e-06, + "loss": 0.9816, + "step": 7112 + }, + { + "epoch": 0.5345708702840823, + "grad_norm": 1.8951730801091906, + "learning_rate": 1.8735055168687126e-06, + "loss": 1.0082, + "step": 7113 + }, + { + "epoch": 0.5346460243499174, + "grad_norm": 1.2940897619843663, + "learning_rate": 1.8730196531430246e-06, + "loss": 0.9971, + "step": 7114 + }, + { + "epoch": 0.5347211784157523, + "grad_norm": 1.6162507665976042, + "learning_rate": 1.872533796941408e-06, + "loss": 0.9447, + "step": 7115 + }, + { + "epoch": 0.5347963324815872, + "grad_norm": 1.6772403367122812, + "learning_rate": 1.8720479482926523e-06, + "loss": 1.0235, + "step": 7116 + }, + { + "epoch": 0.5348714865474222, + "grad_norm": 1.623450347291487, + "learning_rate": 1.8715621072255457e-06, + "loss": 0.9847, + "step": 7117 + }, + { + "epoch": 0.5349466406132571, + "grad_norm": 1.611952190348056, + "learning_rate": 1.8710762737688757e-06, + "loss": 0.9958, + "step": 7118 + }, + { + "epoch": 0.5350217946790922, + "grad_norm": 1.234070704170365, + "learning_rate": 1.8705904479514305e-06, + "loss": 0.9821, + "step": 7119 + }, + { + "epoch": 0.5350969487449271, + "grad_norm": 2.0386265548109757, + "learning_rate": 1.8701046298019965e-06, + "loss": 1.0374, + "step": 7120 + }, + { + "epoch": 0.5351721028107621, + "grad_norm": 2.8232448898508338, + "learning_rate": 1.869618819349361e-06, + "loss": 0.9752, + "step": 7121 + }, + { + "epoch": 0.535247256876597, + "grad_norm": 2.2506958627178126, + "learning_rate": 1.8691330166223091e-06, + "loss": 0.9481, + "step": 7122 + }, + { + "epoch": 0.5353224109424319, + "grad_norm": 2.929671712197044, + "learning_rate": 1.8686472216496275e-06, + "loss": 0.9355, + "step": 7123 + }, + { + "epoch": 0.535397565008267, + "grad_norm": 1.5632167698012382, + "learning_rate": 1.8681614344601013e-06, + "loss": 1.0193, + "step": 7124 + }, + { + "epoch": 0.5354727190741019, + "grad_norm": 1.762027890979447, + "learning_rate": 1.8676756550825144e-06, + "loss": 0.9653, + "step": 7125 + }, + { + "epoch": 0.5355478731399369, + "grad_norm": 1.614322427978376, + "learning_rate": 1.8671898835456518e-06, + "loss": 0.9709, + "step": 7126 + }, + { + "epoch": 0.5356230272057718, + "grad_norm": 1.803003822102571, + "learning_rate": 1.8667041198782972e-06, + "loss": 0.9832, + "step": 7127 + }, + { + "epoch": 0.5356981812716068, + "grad_norm": 1.5168154259273527, + "learning_rate": 1.866218364109234e-06, + "loss": 1.0244, + "step": 7128 + }, + { + "epoch": 0.5357733353374418, + "grad_norm": 1.3628904115781129, + "learning_rate": 1.8657326162672452e-06, + "loss": 1.0088, + "step": 7129 + }, + { + "epoch": 0.5358484894032767, + "grad_norm": 1.671073792897062, + "learning_rate": 1.865246876381112e-06, + "loss": 0.9702, + "step": 7130 + }, + { + "epoch": 0.5359236434691117, + "grad_norm": 1.8595920724540203, + "learning_rate": 1.8647611444796182e-06, + "loss": 0.8173, + "step": 7131 + }, + { + "epoch": 0.5359987975349466, + "grad_norm": 1.9420115335163242, + "learning_rate": 1.8642754205915444e-06, + "loss": 1.0635, + "step": 7132 + }, + { + "epoch": 0.5360739516007816, + "grad_norm": 1.794544001705449, + "learning_rate": 1.8637897047456717e-06, + "loss": 0.9854, + "step": 7133 + }, + { + "epoch": 0.5361491056666166, + "grad_norm": 1.766446690389609, + "learning_rate": 1.8633039969707808e-06, + "loss": 0.9801, + "step": 7134 + }, + { + "epoch": 0.5362242597324516, + "grad_norm": 1.7428500707447123, + "learning_rate": 1.8628182972956509e-06, + "loss": 0.9626, + "step": 7135 + }, + { + "epoch": 0.5362994137982865, + "grad_norm": 1.4872719052885455, + "learning_rate": 1.8623326057490627e-06, + "loss": 0.9706, + "step": 7136 + }, + { + "epoch": 0.5363745678641214, + "grad_norm": 2.1820181688134457, + "learning_rate": 1.8618469223597943e-06, + "loss": 1.107, + "step": 7137 + }, + { + "epoch": 0.5364497219299564, + "grad_norm": 2.190687092160927, + "learning_rate": 1.8613612471566253e-06, + "loss": 1.0418, + "step": 7138 + }, + { + "epoch": 0.5365248759957914, + "grad_norm": 1.6398990296473606, + "learning_rate": 1.8608755801683334e-06, + "loss": 0.972, + "step": 7139 + }, + { + "epoch": 0.5366000300616264, + "grad_norm": 1.7417991444824654, + "learning_rate": 1.8603899214236956e-06, + "loss": 0.9997, + "step": 7140 + }, + { + "epoch": 0.5366751841274613, + "grad_norm": 1.537769374665201, + "learning_rate": 1.85990427095149e-06, + "loss": 0.9803, + "step": 7141 + }, + { + "epoch": 0.5367503381932962, + "grad_norm": 1.6522271586377686, + "learning_rate": 1.8594186287804923e-06, + "loss": 0.9977, + "step": 7142 + }, + { + "epoch": 0.5368254922591312, + "grad_norm": 1.8352735252145702, + "learning_rate": 1.8589329949394793e-06, + "loss": 0.9814, + "step": 7143 + }, + { + "epoch": 0.5369006463249661, + "grad_norm": 0.7395210553306195, + "learning_rate": 1.8584473694572268e-06, + "loss": 0.8345, + "step": 7144 + }, + { + "epoch": 0.5369758003908012, + "grad_norm": 2.6769566091584105, + "learning_rate": 1.8579617523625096e-06, + "loss": 0.9651, + "step": 7145 + }, + { + "epoch": 0.5370509544566361, + "grad_norm": 4.478519457995525, + "learning_rate": 1.8574761436841027e-06, + "loss": 0.9604, + "step": 7146 + }, + { + "epoch": 0.5371261085224711, + "grad_norm": 2.613976629822567, + "learning_rate": 1.8569905434507796e-06, + "loss": 1.0621, + "step": 7147 + }, + { + "epoch": 0.537201262588306, + "grad_norm": 1.6913338661122839, + "learning_rate": 1.8565049516913146e-06, + "loss": 0.8681, + "step": 7148 + }, + { + "epoch": 0.5372764166541409, + "grad_norm": 1.4676275631312257, + "learning_rate": 1.85601936843448e-06, + "loss": 0.8304, + "step": 7149 + }, + { + "epoch": 0.537351570719976, + "grad_norm": 1.7439754873530422, + "learning_rate": 1.8555337937090506e-06, + "loss": 1.0354, + "step": 7150 + }, + { + "epoch": 0.5374267247858109, + "grad_norm": 2.3135843569890198, + "learning_rate": 1.8550482275437964e-06, + "loss": 0.9621, + "step": 7151 + }, + { + "epoch": 0.5375018788516459, + "grad_norm": 1.7182004570236726, + "learning_rate": 1.854562669967489e-06, + "loss": 1.0415, + "step": 7152 + }, + { + "epoch": 0.5375770329174808, + "grad_norm": 1.7720145680192534, + "learning_rate": 1.8540771210089016e-06, + "loss": 0.833, + "step": 7153 + }, + { + "epoch": 0.5376521869833158, + "grad_norm": 1.45255407787511, + "learning_rate": 1.8535915806968026e-06, + "loss": 1.0262, + "step": 7154 + }, + { + "epoch": 0.5377273410491508, + "grad_norm": 1.8346569384000004, + "learning_rate": 1.8531060490599637e-06, + "loss": 1.1432, + "step": 7155 + }, + { + "epoch": 0.5378024951149857, + "grad_norm": 1.6898144991490704, + "learning_rate": 1.8526205261271538e-06, + "loss": 0.9592, + "step": 7156 + }, + { + "epoch": 0.5378776491808207, + "grad_norm": 2.1465596818426462, + "learning_rate": 1.8521350119271418e-06, + "loss": 0.9192, + "step": 7157 + }, + { + "epoch": 0.5379528032466556, + "grad_norm": 1.860999543687538, + "learning_rate": 1.8516495064886967e-06, + "loss": 0.9272, + "step": 7158 + }, + { + "epoch": 0.5380279573124906, + "grad_norm": 1.7861213329148595, + "learning_rate": 1.8511640098405863e-06, + "loss": 0.9567, + "step": 7159 + }, + { + "epoch": 0.5381031113783256, + "grad_norm": 2.0734115725450697, + "learning_rate": 1.8506785220115787e-06, + "loss": 0.9325, + "step": 7160 + }, + { + "epoch": 0.5381782654441605, + "grad_norm": 3.1452043085493013, + "learning_rate": 1.8501930430304402e-06, + "loss": 1.0317, + "step": 7161 + }, + { + "epoch": 0.5382534195099955, + "grad_norm": 1.9682137500744312, + "learning_rate": 1.8497075729259372e-06, + "loss": 1.0824, + "step": 7162 + }, + { + "epoch": 0.5383285735758304, + "grad_norm": 0.7378313419403347, + "learning_rate": 1.8492221117268367e-06, + "loss": 0.8595, + "step": 7163 + }, + { + "epoch": 0.5384037276416654, + "grad_norm": 1.8166484056074106, + "learning_rate": 1.8487366594619028e-06, + "loss": 0.9746, + "step": 7164 + }, + { + "epoch": 0.5384788817075004, + "grad_norm": 1.4074164321374902, + "learning_rate": 1.8482512161599016e-06, + "loss": 0.927, + "step": 7165 + }, + { + "epoch": 0.5385540357733354, + "grad_norm": 1.6489139248074625, + "learning_rate": 1.8477657818495963e-06, + "loss": 0.987, + "step": 7166 + }, + { + "epoch": 0.5386291898391703, + "grad_norm": 1.7604933577165485, + "learning_rate": 1.847280356559752e-06, + "loss": 0.9438, + "step": 7167 + }, + { + "epoch": 0.5387043439050052, + "grad_norm": 2.0479620456680405, + "learning_rate": 1.8467949403191312e-06, + "loss": 0.9805, + "step": 7168 + }, + { + "epoch": 0.5387794979708402, + "grad_norm": 1.8393317480053948, + "learning_rate": 1.8463095331564965e-06, + "loss": 1.0528, + "step": 7169 + }, + { + "epoch": 0.5388546520366752, + "grad_norm": 2.126795178758725, + "learning_rate": 1.8458241351006107e-06, + "loss": 0.8936, + "step": 7170 + }, + { + "epoch": 0.5389298061025102, + "grad_norm": 2.337686859658667, + "learning_rate": 1.8453387461802347e-06, + "loss": 1.0186, + "step": 7171 + }, + { + "epoch": 0.5390049601683451, + "grad_norm": 1.7425131044483408, + "learning_rate": 1.8448533664241316e-06, + "loss": 0.9183, + "step": 7172 + }, + { + "epoch": 0.5390801142341801, + "grad_norm": 1.5097707573136003, + "learning_rate": 1.84436799586106e-06, + "loss": 0.9687, + "step": 7173 + }, + { + "epoch": 0.539155268300015, + "grad_norm": 1.7851571848927392, + "learning_rate": 1.8438826345197796e-06, + "loss": 1.0612, + "step": 7174 + }, + { + "epoch": 0.53923042236585, + "grad_norm": 1.5753153349055, + "learning_rate": 1.843397282429052e-06, + "loss": 1.0252, + "step": 7175 + }, + { + "epoch": 0.539305576431685, + "grad_norm": 1.4940438432452177, + "learning_rate": 1.8429119396176348e-06, + "loss": 1.0377, + "step": 7176 + }, + { + "epoch": 0.5393807304975199, + "grad_norm": 1.93819599571456, + "learning_rate": 1.8424266061142869e-06, + "loss": 0.9894, + "step": 7177 + }, + { + "epoch": 0.5394558845633549, + "grad_norm": 2.7078205405686577, + "learning_rate": 1.841941281947766e-06, + "loss": 1.0111, + "step": 7178 + }, + { + "epoch": 0.5395310386291898, + "grad_norm": 4.445645445571922, + "learning_rate": 1.8414559671468288e-06, + "loss": 0.9769, + "step": 7179 + }, + { + "epoch": 0.5396061926950249, + "grad_norm": 2.207365243671678, + "learning_rate": 1.8409706617402333e-06, + "loss": 1.0129, + "step": 7180 + }, + { + "epoch": 0.5396813467608598, + "grad_norm": 1.9953821037178936, + "learning_rate": 1.8404853657567347e-06, + "loss": 0.993, + "step": 7181 + }, + { + "epoch": 0.5397565008266947, + "grad_norm": 1.5537743114923308, + "learning_rate": 1.8400000792250894e-06, + "loss": 1.0198, + "step": 7182 + }, + { + "epoch": 0.5398316548925297, + "grad_norm": 1.5889389875304716, + "learning_rate": 1.8395148021740518e-06, + "loss": 1.0056, + "step": 7183 + }, + { + "epoch": 0.5399068089583646, + "grad_norm": 1.7893171504, + "learning_rate": 1.8390295346323765e-06, + "loss": 0.8983, + "step": 7184 + }, + { + "epoch": 0.5399819630241997, + "grad_norm": 1.7107243513999497, + "learning_rate": 1.8385442766288181e-06, + "loss": 0.9834, + "step": 7185 + }, + { + "epoch": 0.5400571170900346, + "grad_norm": 1.5049722435758461, + "learning_rate": 1.8380590281921294e-06, + "loss": 1.0268, + "step": 7186 + }, + { + "epoch": 0.5401322711558695, + "grad_norm": 1.9728517341066094, + "learning_rate": 1.8375737893510635e-06, + "loss": 1.057, + "step": 7187 + }, + { + "epoch": 0.5402074252217045, + "grad_norm": 1.6523011114151722, + "learning_rate": 1.837088560134372e-06, + "loss": 0.9791, + "step": 7188 + }, + { + "epoch": 0.5402825792875394, + "grad_norm": 1.4944352401787147, + "learning_rate": 1.8366033405708076e-06, + "loss": 0.8924, + "step": 7189 + }, + { + "epoch": 0.5403577333533744, + "grad_norm": 0.7577042649832638, + "learning_rate": 1.8361181306891214e-06, + "loss": 0.8395, + "step": 7190 + }, + { + "epoch": 0.5404328874192094, + "grad_norm": 1.532423076099214, + "learning_rate": 1.8356329305180626e-06, + "loss": 1.0179, + "step": 7191 + }, + { + "epoch": 0.5405080414850444, + "grad_norm": 1.5191289766413874, + "learning_rate": 1.835147740086383e-06, + "loss": 0.9983, + "step": 7192 + }, + { + "epoch": 0.5405831955508793, + "grad_norm": 1.877989928201271, + "learning_rate": 1.8346625594228295e-06, + "loss": 0.9871, + "step": 7193 + }, + { + "epoch": 0.5406583496167142, + "grad_norm": 0.6699100415488675, + "learning_rate": 1.8341773885561539e-06, + "loss": 0.8021, + "step": 7194 + }, + { + "epoch": 0.5407335036825492, + "grad_norm": 1.4022307709048596, + "learning_rate": 1.8336922275151032e-06, + "loss": 1.0104, + "step": 7195 + }, + { + "epoch": 0.5408086577483842, + "grad_norm": 1.5988739521571032, + "learning_rate": 1.8332070763284236e-06, + "loss": 1.0515, + "step": 7196 + }, + { + "epoch": 0.5408838118142192, + "grad_norm": 1.3678089582272317, + "learning_rate": 1.8327219350248643e-06, + "loss": 1.058, + "step": 7197 + }, + { + "epoch": 0.5409589658800541, + "grad_norm": 4.384700375349699, + "learning_rate": 1.8322368036331705e-06, + "loss": 0.9763, + "step": 7198 + }, + { + "epoch": 0.5410341199458891, + "grad_norm": 1.8333267804230997, + "learning_rate": 1.8317516821820888e-06, + "loss": 1.017, + "step": 7199 + }, + { + "epoch": 0.541109274011724, + "grad_norm": 1.6169579384451926, + "learning_rate": 1.8312665707003643e-06, + "loss": 0.9145, + "step": 7200 + }, + { + "epoch": 0.541184428077559, + "grad_norm": 1.6445922115380418, + "learning_rate": 1.8307814692167412e-06, + "loss": 0.8035, + "step": 7201 + }, + { + "epoch": 0.541259582143394, + "grad_norm": 6.624915932175927, + "learning_rate": 1.8302963777599645e-06, + "loss": 1.048, + "step": 7202 + }, + { + "epoch": 0.5413347362092289, + "grad_norm": 0.6426077374994832, + "learning_rate": 1.8298112963587766e-06, + "loss": 0.8361, + "step": 7203 + }, + { + "epoch": 0.5414098902750639, + "grad_norm": 1.6012461280031463, + "learning_rate": 1.8293262250419217e-06, + "loss": 0.9471, + "step": 7204 + }, + { + "epoch": 0.5414850443408988, + "grad_norm": 1.4770968645961924, + "learning_rate": 1.8288411638381415e-06, + "loss": 0.9754, + "step": 7205 + }, + { + "epoch": 0.5415601984067338, + "grad_norm": 1.7285338320015675, + "learning_rate": 1.8283561127761773e-06, + "loss": 1.0133, + "step": 7206 + }, + { + "epoch": 0.5416353524725688, + "grad_norm": 1.3301343191770567, + "learning_rate": 1.8278710718847711e-06, + "loss": 0.9446, + "step": 7207 + }, + { + "epoch": 0.5417105065384037, + "grad_norm": 1.419319203337045, + "learning_rate": 1.8273860411926627e-06, + "loss": 0.974, + "step": 7208 + }, + { + "epoch": 0.5417856606042387, + "grad_norm": 2.5629609786570877, + "learning_rate": 1.8269010207285927e-06, + "loss": 0.9447, + "step": 7209 + }, + { + "epoch": 0.5418608146700736, + "grad_norm": 3.1315301948710568, + "learning_rate": 1.8264160105212995e-06, + "loss": 0.9539, + "step": 7210 + }, + { + "epoch": 0.5419359687359087, + "grad_norm": 1.3651404614543579, + "learning_rate": 1.825931010599523e-06, + "loss": 1.0649, + "step": 7211 + }, + { + "epoch": 0.5420111228017436, + "grad_norm": 1.6050938820639924, + "learning_rate": 1.8254460209920007e-06, + "loss": 0.9949, + "step": 7212 + }, + { + "epoch": 0.5420862768675785, + "grad_norm": 1.8807951697366625, + "learning_rate": 1.8249610417274695e-06, + "loss": 0.9237, + "step": 7213 + }, + { + "epoch": 0.5421614309334135, + "grad_norm": 1.659095048008527, + "learning_rate": 1.8244760728346674e-06, + "loss": 1.015, + "step": 7214 + }, + { + "epoch": 0.5422365849992484, + "grad_norm": 1.8718961456581946, + "learning_rate": 1.823991114342329e-06, + "loss": 0.9725, + "step": 7215 + }, + { + "epoch": 0.5423117390650835, + "grad_norm": 2.0376807899070934, + "learning_rate": 1.823506166279192e-06, + "loss": 0.8186, + "step": 7216 + }, + { + "epoch": 0.5423868931309184, + "grad_norm": 1.4608652025581161, + "learning_rate": 1.823021228673991e-06, + "loss": 0.9436, + "step": 7217 + }, + { + "epoch": 0.5424620471967534, + "grad_norm": 2.160932129076105, + "learning_rate": 1.8225363015554586e-06, + "loss": 0.9035, + "step": 7218 + }, + { + "epoch": 0.5425372012625883, + "grad_norm": 2.005818010923308, + "learning_rate": 1.822051384952331e-06, + "loss": 0.9777, + "step": 7219 + }, + { + "epoch": 0.5426123553284232, + "grad_norm": 1.8455482377554049, + "learning_rate": 1.8215664788933394e-06, + "loss": 1.016, + "step": 7220 + }, + { + "epoch": 0.5426875093942582, + "grad_norm": 0.7403507191561588, + "learning_rate": 1.8210815834072177e-06, + "loss": 0.8489, + "step": 7221 + }, + { + "epoch": 0.5427626634600932, + "grad_norm": 1.7749979153473123, + "learning_rate": 1.8205966985226975e-06, + "loss": 0.9852, + "step": 7222 + }, + { + "epoch": 0.5428378175259282, + "grad_norm": 1.5800225963666776, + "learning_rate": 1.8201118242685093e-06, + "loss": 1.0123, + "step": 7223 + }, + { + "epoch": 0.5429129715917631, + "grad_norm": 1.6904289822987408, + "learning_rate": 1.819626960673385e-06, + "loss": 0.9253, + "step": 7224 + }, + { + "epoch": 0.5429881256575981, + "grad_norm": 0.7529288166996018, + "learning_rate": 1.8191421077660535e-06, + "loss": 0.8345, + "step": 7225 + }, + { + "epoch": 0.543063279723433, + "grad_norm": 1.5124766103654184, + "learning_rate": 1.8186572655752448e-06, + "loss": 0.9918, + "step": 7226 + }, + { + "epoch": 0.543138433789268, + "grad_norm": 1.4409390651253429, + "learning_rate": 1.8181724341296877e-06, + "loss": 0.9084, + "step": 7227 + }, + { + "epoch": 0.543213587855103, + "grad_norm": 1.673641448143894, + "learning_rate": 1.8176876134581098e-06, + "loss": 1.05, + "step": 7228 + }, + { + "epoch": 0.5432887419209379, + "grad_norm": 1.6800169865566892, + "learning_rate": 1.8172028035892394e-06, + "loss": 0.9581, + "step": 7229 + }, + { + "epoch": 0.5433638959867729, + "grad_norm": 1.466922840952234, + "learning_rate": 1.816718004551802e-06, + "loss": 0.9779, + "step": 7230 + }, + { + "epoch": 0.5434390500526078, + "grad_norm": 1.7120722661140717, + "learning_rate": 1.8162332163745254e-06, + "loss": 0.8789, + "step": 7231 + }, + { + "epoch": 0.5435142041184428, + "grad_norm": 0.7558277527653247, + "learning_rate": 1.8157484390861342e-06, + "loss": 0.8866, + "step": 7232 + }, + { + "epoch": 0.5435893581842778, + "grad_norm": 1.690168319762028, + "learning_rate": 1.8152636727153536e-06, + "loss": 0.9843, + "step": 7233 + }, + { + "epoch": 0.5436645122501127, + "grad_norm": 1.789842327355461, + "learning_rate": 1.814778917290908e-06, + "loss": 0.9813, + "step": 7234 + }, + { + "epoch": 0.5437396663159477, + "grad_norm": 0.7660686997813767, + "learning_rate": 1.8142941728415204e-06, + "loss": 0.8926, + "step": 7235 + }, + { + "epoch": 0.5438148203817826, + "grad_norm": 1.482704266375571, + "learning_rate": 1.8138094393959144e-06, + "loss": 0.9448, + "step": 7236 + }, + { + "epoch": 0.5438899744476177, + "grad_norm": 1.559445791223918, + "learning_rate": 1.8133247169828114e-06, + "loss": 0.9835, + "step": 7237 + }, + { + "epoch": 0.5439651285134526, + "grad_norm": 1.667307809852215, + "learning_rate": 1.8128400056309345e-06, + "loss": 0.9336, + "step": 7238 + }, + { + "epoch": 0.5440402825792875, + "grad_norm": 1.339743251757204, + "learning_rate": 1.8123553053690046e-06, + "loss": 1.0108, + "step": 7239 + }, + { + "epoch": 0.5441154366451225, + "grad_norm": 1.9152752829315605, + "learning_rate": 1.81187061622574e-06, + "loss": 0.8904, + "step": 7240 + }, + { + "epoch": 0.5441905907109574, + "grad_norm": 1.4016843217621577, + "learning_rate": 1.8113859382298627e-06, + "loss": 0.9791, + "step": 7241 + }, + { + "epoch": 0.5442657447767925, + "grad_norm": 1.7702210284015782, + "learning_rate": 1.81090127141009e-06, + "loss": 0.907, + "step": 7242 + }, + { + "epoch": 0.5443408988426274, + "grad_norm": 1.8385374449194256, + "learning_rate": 1.8104166157951419e-06, + "loss": 0.9992, + "step": 7243 + }, + { + "epoch": 0.5444160529084624, + "grad_norm": 1.3492595036928066, + "learning_rate": 1.809931971413735e-06, + "loss": 1.0777, + "step": 7244 + }, + { + "epoch": 0.5444912069742973, + "grad_norm": 2.359012500123415, + "learning_rate": 1.8094473382945866e-06, + "loss": 1.0237, + "step": 7245 + }, + { + "epoch": 0.5445663610401322, + "grad_norm": 1.493451846066409, + "learning_rate": 1.8089627164664132e-06, + "loss": 1.0001, + "step": 7246 + }, + { + "epoch": 0.5446415151059673, + "grad_norm": 1.848360536316087, + "learning_rate": 1.80847810595793e-06, + "loss": 0.8755, + "step": 7247 + }, + { + "epoch": 0.5447166691718022, + "grad_norm": 1.6149261438825344, + "learning_rate": 1.8079935067978528e-06, + "loss": 0.9378, + "step": 7248 + }, + { + "epoch": 0.5447918232376372, + "grad_norm": 1.5452758537861238, + "learning_rate": 1.8075089190148956e-06, + "loss": 0.8898, + "step": 7249 + }, + { + "epoch": 0.5448669773034721, + "grad_norm": 1.7561407659369925, + "learning_rate": 1.8070243426377716e-06, + "loss": 0.9467, + "step": 7250 + }, + { + "epoch": 0.544942131369307, + "grad_norm": 1.344916170336934, + "learning_rate": 1.8065397776951946e-06, + "loss": 0.935, + "step": 7251 + }, + { + "epoch": 0.545017285435142, + "grad_norm": 1.492454544857627, + "learning_rate": 1.8060552242158765e-06, + "loss": 0.9949, + "step": 7252 + }, + { + "epoch": 0.545092439500977, + "grad_norm": 2.358355876783408, + "learning_rate": 1.8055706822285291e-06, + "loss": 1.0211, + "step": 7253 + }, + { + "epoch": 0.545167593566812, + "grad_norm": 2.2893715086797597, + "learning_rate": 1.8050861517618629e-06, + "loss": 0.9701, + "step": 7254 + }, + { + "epoch": 0.5452427476326469, + "grad_norm": 0.8662090107961943, + "learning_rate": 1.8046016328445893e-06, + "loss": 0.8667, + "step": 7255 + }, + { + "epoch": 0.5453179016984819, + "grad_norm": 1.6807843552607098, + "learning_rate": 1.804117125505417e-06, + "loss": 0.9219, + "step": 7256 + }, + { + "epoch": 0.5453930557643168, + "grad_norm": 1.5841546020101915, + "learning_rate": 1.803632629773054e-06, + "loss": 0.9555, + "step": 7257 + }, + { + "epoch": 0.5454682098301518, + "grad_norm": 2.016346528781002, + "learning_rate": 1.8031481456762112e-06, + "loss": 1.0453, + "step": 7258 + }, + { + "epoch": 0.5455433638959868, + "grad_norm": 1.6915367023298606, + "learning_rate": 1.802663673243593e-06, + "loss": 0.988, + "step": 7259 + }, + { + "epoch": 0.5456185179618217, + "grad_norm": 1.5992021240205003, + "learning_rate": 1.802179212503909e-06, + "loss": 0.9856, + "step": 7260 + }, + { + "epoch": 0.5456936720276567, + "grad_norm": 2.107129406366529, + "learning_rate": 1.801694763485864e-06, + "loss": 0.9195, + "step": 7261 + }, + { + "epoch": 0.5457688260934916, + "grad_norm": 1.57101773799447, + "learning_rate": 1.8012103262181635e-06, + "loss": 1.0381, + "step": 7262 + }, + { + "epoch": 0.5458439801593267, + "grad_norm": 1.4348967656607348, + "learning_rate": 1.8007259007295125e-06, + "loss": 0.974, + "step": 7263 + }, + { + "epoch": 0.5459191342251616, + "grad_norm": 2.581925336595949, + "learning_rate": 1.8002414870486144e-06, + "loss": 0.9479, + "step": 7264 + }, + { + "epoch": 0.5459942882909965, + "grad_norm": 1.9666368557669556, + "learning_rate": 1.7997570852041739e-06, + "loss": 0.9753, + "step": 7265 + }, + { + "epoch": 0.5460694423568315, + "grad_norm": 1.6055268140256704, + "learning_rate": 1.7992726952248926e-06, + "loss": 0.9429, + "step": 7266 + }, + { + "epoch": 0.5461445964226664, + "grad_norm": 1.9046270622196688, + "learning_rate": 1.7987883171394724e-06, + "loss": 0.9987, + "step": 7267 + }, + { + "epoch": 0.5462197504885015, + "grad_norm": 1.8031451748094491, + "learning_rate": 1.7983039509766156e-06, + "loss": 0.9549, + "step": 7268 + }, + { + "epoch": 0.5462949045543364, + "grad_norm": 15.497339820237093, + "learning_rate": 1.7978195967650214e-06, + "loss": 1.0191, + "step": 7269 + }, + { + "epoch": 0.5463700586201714, + "grad_norm": 2.5156445235922322, + "learning_rate": 1.7973352545333905e-06, + "loss": 0.8722, + "step": 7270 + }, + { + "epoch": 0.5464452126860063, + "grad_norm": 2.2269106944026125, + "learning_rate": 1.796850924310422e-06, + "loss": 0.9986, + "step": 7271 + }, + { + "epoch": 0.5465203667518412, + "grad_norm": 1.6759196048700888, + "learning_rate": 1.796366606124814e-06, + "loss": 1.0955, + "step": 7272 + }, + { + "epoch": 0.5465955208176763, + "grad_norm": 1.3659450880410222, + "learning_rate": 1.7958823000052643e-06, + "loss": 1.0055, + "step": 7273 + }, + { + "epoch": 0.5466706748835112, + "grad_norm": 1.5486629441296333, + "learning_rate": 1.79539800598047e-06, + "loss": 1.0014, + "step": 7274 + }, + { + "epoch": 0.5467458289493462, + "grad_norm": 1.7103192331469763, + "learning_rate": 1.7949137240791275e-06, + "loss": 1.0686, + "step": 7275 + }, + { + "epoch": 0.5468209830151811, + "grad_norm": 1.6851596144287593, + "learning_rate": 1.7944294543299317e-06, + "loss": 0.9781, + "step": 7276 + }, + { + "epoch": 0.546896137081016, + "grad_norm": 1.8402760639622544, + "learning_rate": 1.7939451967615783e-06, + "loss": 0.9153, + "step": 7277 + }, + { + "epoch": 0.5469712911468511, + "grad_norm": 4.272069513319715, + "learning_rate": 1.793460951402761e-06, + "loss": 0.9104, + "step": 7278 + }, + { + "epoch": 0.547046445212686, + "grad_norm": 1.525074944452534, + "learning_rate": 1.7929767182821724e-06, + "loss": 0.9115, + "step": 7279 + }, + { + "epoch": 0.547121599278521, + "grad_norm": 1.57055085304317, + "learning_rate": 1.7924924974285074e-06, + "loss": 1.0093, + "step": 7280 + }, + { + "epoch": 0.5471967533443559, + "grad_norm": 2.0439779731922307, + "learning_rate": 1.7920082888704553e-06, + "loss": 0.925, + "step": 7281 + }, + { + "epoch": 0.5472719074101909, + "grad_norm": 1.8046574445591492, + "learning_rate": 1.7915240926367092e-06, + "loss": 0.9987, + "step": 7282 + }, + { + "epoch": 0.5473470614760259, + "grad_norm": 1.546712770528514, + "learning_rate": 1.791039908755959e-06, + "loss": 0.9651, + "step": 7283 + }, + { + "epoch": 0.5474222155418608, + "grad_norm": 1.6573761764005472, + "learning_rate": 1.790555737256894e-06, + "loss": 0.8921, + "step": 7284 + }, + { + "epoch": 0.5474973696076958, + "grad_norm": 0.7187012780303016, + "learning_rate": 1.7900715781682039e-06, + "loss": 0.8094, + "step": 7285 + }, + { + "epoch": 0.5475725236735307, + "grad_norm": 1.9478648649729386, + "learning_rate": 1.7895874315185763e-06, + "loss": 0.9945, + "step": 7286 + }, + { + "epoch": 0.5476476777393657, + "grad_norm": 2.5724926082997945, + "learning_rate": 1.7891032973366996e-06, + "loss": 0.982, + "step": 7287 + }, + { + "epoch": 0.5477228318052006, + "grad_norm": 1.790495119826956, + "learning_rate": 1.7886191756512598e-06, + "loss": 0.9487, + "step": 7288 + }, + { + "epoch": 0.5477979858710357, + "grad_norm": 1.7914504726589542, + "learning_rate": 1.788135066490943e-06, + "loss": 0.9329, + "step": 7289 + }, + { + "epoch": 0.5478731399368706, + "grad_norm": 1.817927488507132, + "learning_rate": 1.7876509698844356e-06, + "loss": 0.9122, + "step": 7290 + }, + { + "epoch": 0.5479482940027055, + "grad_norm": 2.6287406427105324, + "learning_rate": 1.7871668858604206e-06, + "loss": 0.959, + "step": 7291 + }, + { + "epoch": 0.5480234480685405, + "grad_norm": 1.4087880436956925, + "learning_rate": 1.786682814447583e-06, + "loss": 0.9964, + "step": 7292 + }, + { + "epoch": 0.5480986021343754, + "grad_norm": 1.6095781661005084, + "learning_rate": 1.7861987556746056e-06, + "loss": 0.9927, + "step": 7293 + }, + { + "epoch": 0.5481737562002105, + "grad_norm": 1.510644812179777, + "learning_rate": 1.78571470957017e-06, + "loss": 0.9407, + "step": 7294 + }, + { + "epoch": 0.5482489102660454, + "grad_norm": 1.4277237710415467, + "learning_rate": 1.7852306761629592e-06, + "loss": 0.976, + "step": 7295 + }, + { + "epoch": 0.5483240643318803, + "grad_norm": 1.4632599673783049, + "learning_rate": 1.7847466554816526e-06, + "loss": 0.9789, + "step": 7296 + }, + { + "epoch": 0.5483992183977153, + "grad_norm": 1.8976183721437794, + "learning_rate": 1.7842626475549314e-06, + "loss": 1.0608, + "step": 7297 + }, + { + "epoch": 0.5484743724635502, + "grad_norm": 1.867434688789552, + "learning_rate": 1.783778652411474e-06, + "loss": 0.935, + "step": 7298 + }, + { + "epoch": 0.5485495265293853, + "grad_norm": 1.9180947940221778, + "learning_rate": 1.7832946700799596e-06, + "loss": 1.0787, + "step": 7299 + }, + { + "epoch": 0.5486246805952202, + "grad_norm": 3.636050514264778, + "learning_rate": 1.7828107005890663e-06, + "loss": 0.8909, + "step": 7300 + }, + { + "epoch": 0.5486998346610552, + "grad_norm": 1.6514050858907054, + "learning_rate": 1.7823267439674694e-06, + "loss": 1.0773, + "step": 7301 + }, + { + "epoch": 0.5487749887268901, + "grad_norm": 3.052807475732251, + "learning_rate": 1.7818428002438475e-06, + "loss": 1.0563, + "step": 7302 + }, + { + "epoch": 0.548850142792725, + "grad_norm": 0.7185715163495466, + "learning_rate": 1.7813588694468745e-06, + "loss": 0.8718, + "step": 7303 + }, + { + "epoch": 0.5489252968585601, + "grad_norm": 1.3386016515607917, + "learning_rate": 1.780874951605226e-06, + "loss": 0.9435, + "step": 7304 + }, + { + "epoch": 0.549000450924395, + "grad_norm": 1.5820478273629925, + "learning_rate": 1.7803910467475763e-06, + "loss": 0.9732, + "step": 7305 + }, + { + "epoch": 0.54907560499023, + "grad_norm": 2.4528496460149793, + "learning_rate": 1.779907154902597e-06, + "loss": 0.8864, + "step": 7306 + }, + { + "epoch": 0.5491507590560649, + "grad_norm": 1.844965001009491, + "learning_rate": 1.7794232760989623e-06, + "loss": 0.9955, + "step": 7307 + }, + { + "epoch": 0.5492259131219, + "grad_norm": 1.7623891295062732, + "learning_rate": 1.7789394103653425e-06, + "loss": 1.0627, + "step": 7308 + }, + { + "epoch": 0.5493010671877349, + "grad_norm": 0.7358598105209297, + "learning_rate": 1.7784555577304099e-06, + "loss": 0.814, + "step": 7309 + }, + { + "epoch": 0.5493762212535698, + "grad_norm": 0.791891559647472, + "learning_rate": 1.7779717182228335e-06, + "loss": 0.8705, + "step": 7310 + }, + { + "epoch": 0.5494513753194048, + "grad_norm": 1.461847440109989, + "learning_rate": 1.7774878918712828e-06, + "loss": 0.997, + "step": 7311 + }, + { + "epoch": 0.5495265293852397, + "grad_norm": 1.3355220572572106, + "learning_rate": 1.777004078704427e-06, + "loss": 0.9546, + "step": 7312 + }, + { + "epoch": 0.5496016834510747, + "grad_norm": 1.9182223834054997, + "learning_rate": 1.7765202787509327e-06, + "loss": 0.8716, + "step": 7313 + }, + { + "epoch": 0.5496768375169097, + "grad_norm": 1.4796646068054788, + "learning_rate": 1.7760364920394684e-06, + "loss": 0.8645, + "step": 7314 + }, + { + "epoch": 0.5497519915827447, + "grad_norm": 1.6211638286104666, + "learning_rate": 1.7755527185986996e-06, + "loss": 0.9904, + "step": 7315 + }, + { + "epoch": 0.5498271456485796, + "grad_norm": 1.4090521127500262, + "learning_rate": 1.775068958457291e-06, + "loss": 1.0623, + "step": 7316 + }, + { + "epoch": 0.5499022997144145, + "grad_norm": 1.2857188932396815, + "learning_rate": 1.7745852116439087e-06, + "loss": 1.0967, + "step": 7317 + }, + { + "epoch": 0.5499774537802495, + "grad_norm": 1.5589402657321325, + "learning_rate": 1.774101478187215e-06, + "loss": 1.0209, + "step": 7318 + }, + { + "epoch": 0.5500526078460845, + "grad_norm": 0.6208195350914292, + "learning_rate": 1.7736177581158742e-06, + "loss": 0.7923, + "step": 7319 + }, + { + "epoch": 0.5501277619119195, + "grad_norm": 1.75120707163178, + "learning_rate": 1.7731340514585474e-06, + "loss": 0.9258, + "step": 7320 + }, + { + "epoch": 0.5502029159777544, + "grad_norm": 1.6795891311584554, + "learning_rate": 1.7726503582438982e-06, + "loss": 0.8844, + "step": 7321 + }, + { + "epoch": 0.5502780700435893, + "grad_norm": 3.4231161018048377, + "learning_rate": 1.772166678500585e-06, + "loss": 0.9746, + "step": 7322 + }, + { + "epoch": 0.5503532241094243, + "grad_norm": 2.062092133854804, + "learning_rate": 1.771683012257268e-06, + "loss": 1.0435, + "step": 7323 + }, + { + "epoch": 0.5504283781752592, + "grad_norm": 1.869288328068891, + "learning_rate": 1.7711993595426076e-06, + "loss": 0.8762, + "step": 7324 + }, + { + "epoch": 0.5505035322410943, + "grad_norm": 1.395810819795397, + "learning_rate": 1.7707157203852608e-06, + "loss": 0.9981, + "step": 7325 + }, + { + "epoch": 0.5505786863069292, + "grad_norm": 1.6373329921907407, + "learning_rate": 1.770232094813886e-06, + "loss": 0.9708, + "step": 7326 + }, + { + "epoch": 0.5506538403727642, + "grad_norm": 1.9310313499153966, + "learning_rate": 1.7697484828571394e-06, + "loss": 0.9627, + "step": 7327 + }, + { + "epoch": 0.5507289944385991, + "grad_norm": 1.8941038927830447, + "learning_rate": 1.7692648845436764e-06, + "loss": 0.9708, + "step": 7328 + }, + { + "epoch": 0.550804148504434, + "grad_norm": 1.3700825149162141, + "learning_rate": 1.7687812999021531e-06, + "loss": 0.877, + "step": 7329 + }, + { + "epoch": 0.5508793025702691, + "grad_norm": 1.9232747632744047, + "learning_rate": 1.7682977289612226e-06, + "loss": 0.9695, + "step": 7330 + }, + { + "epoch": 0.550954456636104, + "grad_norm": 1.8060615230021302, + "learning_rate": 1.7678141717495395e-06, + "loss": 0.9747, + "step": 7331 + }, + { + "epoch": 0.551029610701939, + "grad_norm": 1.4965595685508377, + "learning_rate": 1.7673306282957559e-06, + "loss": 0.9507, + "step": 7332 + }, + { + "epoch": 0.5511047647677739, + "grad_norm": 1.474358584064785, + "learning_rate": 1.766847098628523e-06, + "loss": 1.0959, + "step": 7333 + }, + { + "epoch": 0.551179918833609, + "grad_norm": 1.263218133466709, + "learning_rate": 1.7663635827764924e-06, + "loss": 0.893, + "step": 7334 + }, + { + "epoch": 0.5512550728994439, + "grad_norm": 1.5587682028826457, + "learning_rate": 1.7658800807683142e-06, + "loss": 1.0008, + "step": 7335 + }, + { + "epoch": 0.5513302269652788, + "grad_norm": 2.6869422866194084, + "learning_rate": 1.7653965926326379e-06, + "loss": 0.9178, + "step": 7336 + }, + { + "epoch": 0.5514053810311138, + "grad_norm": 1.5340701233454306, + "learning_rate": 1.764913118398112e-06, + "loss": 1.0509, + "step": 7337 + }, + { + "epoch": 0.5514805350969487, + "grad_norm": 1.5048173794888111, + "learning_rate": 1.7644296580933835e-06, + "loss": 0.919, + "step": 7338 + }, + { + "epoch": 0.5515556891627837, + "grad_norm": 1.6676932114013434, + "learning_rate": 1.7639462117471004e-06, + "loss": 0.9277, + "step": 7339 + }, + { + "epoch": 0.5516308432286187, + "grad_norm": 6.35943228581001, + "learning_rate": 1.7634627793879075e-06, + "loss": 0.9857, + "step": 7340 + }, + { + "epoch": 0.5517059972944536, + "grad_norm": 1.49268847938355, + "learning_rate": 1.7629793610444513e-06, + "loss": 0.8878, + "step": 7341 + }, + { + "epoch": 0.5517811513602886, + "grad_norm": 1.517179812442306, + "learning_rate": 1.7624959567453746e-06, + "loss": 0.9988, + "step": 7342 + }, + { + "epoch": 0.5518563054261235, + "grad_norm": 1.8738962739596055, + "learning_rate": 1.7620125665193232e-06, + "loss": 0.9168, + "step": 7343 + }, + { + "epoch": 0.5519314594919585, + "grad_norm": 3.687699781275429, + "learning_rate": 1.7615291903949382e-06, + "loss": 0.9289, + "step": 7344 + }, + { + "epoch": 0.5520066135577935, + "grad_norm": 0.6837593967831148, + "learning_rate": 1.761045828400861e-06, + "loss": 0.87, + "step": 7345 + }, + { + "epoch": 0.5520817676236285, + "grad_norm": 0.8244727227028859, + "learning_rate": 1.7605624805657343e-06, + "loss": 0.9023, + "step": 7346 + }, + { + "epoch": 0.5521569216894634, + "grad_norm": 1.5955906131444364, + "learning_rate": 1.760079146918197e-06, + "loss": 0.9954, + "step": 7347 + }, + { + "epoch": 0.5522320757552983, + "grad_norm": 1.5880676396087203, + "learning_rate": 1.7595958274868896e-06, + "loss": 0.9586, + "step": 7348 + }, + { + "epoch": 0.5523072298211333, + "grad_norm": 1.6424657032490777, + "learning_rate": 1.75911252230045e-06, + "loss": 1.0602, + "step": 7349 + }, + { + "epoch": 0.5523823838869683, + "grad_norm": 2.15398188179068, + "learning_rate": 1.758629231387515e-06, + "loss": 0.8624, + "step": 7350 + }, + { + "epoch": 0.5524575379528033, + "grad_norm": 1.6092963338347583, + "learning_rate": 1.7581459547767233e-06, + "loss": 1.0079, + "step": 7351 + }, + { + "epoch": 0.5525326920186382, + "grad_norm": 2.6346953137335856, + "learning_rate": 1.7576626924967091e-06, + "loss": 1.015, + "step": 7352 + }, + { + "epoch": 0.5526078460844732, + "grad_norm": 1.5837710762578558, + "learning_rate": 1.7571794445761089e-06, + "loss": 0.9318, + "step": 7353 + }, + { + "epoch": 0.5526830001503081, + "grad_norm": 1.9785228552126035, + "learning_rate": 1.7566962110435563e-06, + "loss": 0.9418, + "step": 7354 + }, + { + "epoch": 0.552758154216143, + "grad_norm": 1.9316363050144438, + "learning_rate": 1.7562129919276845e-06, + "loss": 0.9552, + "step": 7355 + }, + { + "epoch": 0.5528333082819781, + "grad_norm": 1.6539719227549168, + "learning_rate": 1.7557297872571272e-06, + "loss": 0.9901, + "step": 7356 + }, + { + "epoch": 0.552908462347813, + "grad_norm": 2.90621390371569, + "learning_rate": 1.7552465970605145e-06, + "loss": 1.0613, + "step": 7357 + }, + { + "epoch": 0.552983616413648, + "grad_norm": 6.619962108307697, + "learning_rate": 1.7547634213664786e-06, + "loss": 0.9265, + "step": 7358 + }, + { + "epoch": 0.5530587704794829, + "grad_norm": 1.3805450124783625, + "learning_rate": 1.7542802602036492e-06, + "loss": 0.9527, + "step": 7359 + }, + { + "epoch": 0.553133924545318, + "grad_norm": 2.320221664757141, + "learning_rate": 1.753797113600655e-06, + "loss": 1.0219, + "step": 7360 + }, + { + "epoch": 0.5532090786111529, + "grad_norm": 1.4955818754770642, + "learning_rate": 1.7533139815861248e-06, + "loss": 1.0324, + "step": 7361 + }, + { + "epoch": 0.5532842326769878, + "grad_norm": 2.83034297738435, + "learning_rate": 1.7528308641886856e-06, + "loss": 1.0215, + "step": 7362 + }, + { + "epoch": 0.5533593867428228, + "grad_norm": 1.459400041189738, + "learning_rate": 1.7523477614369645e-06, + "loss": 0.8993, + "step": 7363 + }, + { + "epoch": 0.5534345408086577, + "grad_norm": 1.6829628879854794, + "learning_rate": 1.751864673359586e-06, + "loss": 0.9773, + "step": 7364 + }, + { + "epoch": 0.5535096948744928, + "grad_norm": 1.9548898886381885, + "learning_rate": 1.7513815999851767e-06, + "loss": 0.8614, + "step": 7365 + }, + { + "epoch": 0.5535848489403277, + "grad_norm": 0.6826834585085384, + "learning_rate": 1.7508985413423599e-06, + "loss": 0.8638, + "step": 7366 + }, + { + "epoch": 0.5536600030061626, + "grad_norm": 2.0308954956575747, + "learning_rate": 1.7504154974597572e-06, + "loss": 0.9029, + "step": 7367 + }, + { + "epoch": 0.5537351570719976, + "grad_norm": 0.7088984077394834, + "learning_rate": 1.7499324683659928e-06, + "loss": 0.8521, + "step": 7368 + }, + { + "epoch": 0.5538103111378325, + "grad_norm": 1.5390765968877724, + "learning_rate": 1.749449454089687e-06, + "loss": 0.9321, + "step": 7369 + }, + { + "epoch": 0.5538854652036675, + "grad_norm": 1.3803194493124658, + "learning_rate": 1.7489664546594606e-06, + "loss": 0.9654, + "step": 7370 + }, + { + "epoch": 0.5539606192695025, + "grad_norm": 2.151681012205037, + "learning_rate": 1.7484834701039333e-06, + "loss": 0.9018, + "step": 7371 + }, + { + "epoch": 0.5540357733353375, + "grad_norm": 1.4797556425394016, + "learning_rate": 1.7480005004517228e-06, + "loss": 1.0054, + "step": 7372 + }, + { + "epoch": 0.5541109274011724, + "grad_norm": 6.981102366581194, + "learning_rate": 1.7475175457314481e-06, + "loss": 0.9141, + "step": 7373 + }, + { + "epoch": 0.5541860814670073, + "grad_norm": 1.5759784716912388, + "learning_rate": 1.7470346059717253e-06, + "loss": 0.9698, + "step": 7374 + }, + { + "epoch": 0.5542612355328423, + "grad_norm": 1.656184645029103, + "learning_rate": 1.7465516812011713e-06, + "loss": 0.9578, + "step": 7375 + }, + { + "epoch": 0.5543363895986773, + "grad_norm": 1.715810238369508, + "learning_rate": 1.7460687714484008e-06, + "loss": 0.9575, + "step": 7376 + }, + { + "epoch": 0.5544115436645123, + "grad_norm": 1.6205042491081483, + "learning_rate": 1.7455858767420272e-06, + "loss": 0.9372, + "step": 7377 + }, + { + "epoch": 0.5544866977303472, + "grad_norm": 1.6982628249087481, + "learning_rate": 1.7451029971106653e-06, + "loss": 1.0979, + "step": 7378 + }, + { + "epoch": 0.5545618517961822, + "grad_norm": 1.5691770937278247, + "learning_rate": 1.7446201325829261e-06, + "loss": 0.9917, + "step": 7379 + }, + { + "epoch": 0.5546370058620171, + "grad_norm": 1.8845256435221382, + "learning_rate": 1.7441372831874228e-06, + "loss": 0.9727, + "step": 7380 + }, + { + "epoch": 0.554712159927852, + "grad_norm": 0.7759185070305127, + "learning_rate": 1.7436544489527652e-06, + "loss": 0.8508, + "step": 7381 + }, + { + "epoch": 0.5547873139936871, + "grad_norm": 0.6716160607934927, + "learning_rate": 1.7431716299075625e-06, + "loss": 0.9036, + "step": 7382 + }, + { + "epoch": 0.554862468059522, + "grad_norm": 1.696107036550835, + "learning_rate": 1.7426888260804247e-06, + "loss": 0.9985, + "step": 7383 + }, + { + "epoch": 0.554937622125357, + "grad_norm": 0.814990554708032, + "learning_rate": 1.7422060374999587e-06, + "loss": 0.8567, + "step": 7384 + }, + { + "epoch": 0.5550127761911919, + "grad_norm": 2.445314060003496, + "learning_rate": 1.7417232641947728e-06, + "loss": 0.8555, + "step": 7385 + }, + { + "epoch": 0.5550879302570269, + "grad_norm": 1.966386920893063, + "learning_rate": 1.7412405061934714e-06, + "loss": 1.0238, + "step": 7386 + }, + { + "epoch": 0.5551630843228619, + "grad_norm": 1.5059098013859522, + "learning_rate": 1.740757763524662e-06, + "loss": 0.9635, + "step": 7387 + }, + { + "epoch": 0.5552382383886968, + "grad_norm": 1.370408802715474, + "learning_rate": 1.740275036216948e-06, + "loss": 1.0311, + "step": 7388 + }, + { + "epoch": 0.5553133924545318, + "grad_norm": 2.1565265201783785, + "learning_rate": 1.7397923242989314e-06, + "loss": 0.9521, + "step": 7389 + }, + { + "epoch": 0.5553885465203667, + "grad_norm": 2.25944064349661, + "learning_rate": 1.739309627799217e-06, + "loss": 1.0287, + "step": 7390 + }, + { + "epoch": 0.5554637005862018, + "grad_norm": 2.1777339370795525, + "learning_rate": 1.7388269467464047e-06, + "loss": 1.0975, + "step": 7391 + }, + { + "epoch": 0.5555388546520367, + "grad_norm": 1.522457413978243, + "learning_rate": 1.7383442811690967e-06, + "loss": 0.9058, + "step": 7392 + }, + { + "epoch": 0.5556140087178716, + "grad_norm": 1.7236202941958658, + "learning_rate": 1.7378616310958917e-06, + "loss": 0.9909, + "step": 7393 + }, + { + "epoch": 0.5556891627837066, + "grad_norm": 2.2450834174684027, + "learning_rate": 1.7373789965553886e-06, + "loss": 0.964, + "step": 7394 + }, + { + "epoch": 0.5557643168495415, + "grad_norm": 4.567894868365776, + "learning_rate": 1.736896377576186e-06, + "loss": 0.8689, + "step": 7395 + }, + { + "epoch": 0.5558394709153766, + "grad_norm": 1.3606968295719444, + "learning_rate": 1.73641377418688e-06, + "loss": 1.0009, + "step": 7396 + }, + { + "epoch": 0.5559146249812115, + "grad_norm": 1.4498307819939664, + "learning_rate": 1.7359311864160677e-06, + "loss": 0.9817, + "step": 7397 + }, + { + "epoch": 0.5559897790470465, + "grad_norm": 1.5499508265416893, + "learning_rate": 1.7354486142923438e-06, + "loss": 0.988, + "step": 7398 + }, + { + "epoch": 0.5560649331128814, + "grad_norm": 1.609709695623394, + "learning_rate": 1.7349660578443022e-06, + "loss": 0.9341, + "step": 7399 + }, + { + "epoch": 0.5561400871787163, + "grad_norm": 1.4042324428075388, + "learning_rate": 1.7344835171005368e-06, + "loss": 1.0014, + "step": 7400 + }, + { + "epoch": 0.5562152412445514, + "grad_norm": 1.6863846801860007, + "learning_rate": 1.7340009920896392e-06, + "loss": 0.9388, + "step": 7401 + }, + { + "epoch": 0.5562903953103863, + "grad_norm": 1.4905843061414574, + "learning_rate": 1.7335184828402022e-06, + "loss": 0.9457, + "step": 7402 + }, + { + "epoch": 0.5563655493762213, + "grad_norm": 1.6319684489042654, + "learning_rate": 1.7330359893808154e-06, + "loss": 1.0285, + "step": 7403 + }, + { + "epoch": 0.5564407034420562, + "grad_norm": 2.586799132291484, + "learning_rate": 1.732553511740068e-06, + "loss": 0.9909, + "step": 7404 + }, + { + "epoch": 0.5565158575078912, + "grad_norm": 1.4127630933979418, + "learning_rate": 1.7320710499465494e-06, + "loss": 1.0151, + "step": 7405 + }, + { + "epoch": 0.5565910115737261, + "grad_norm": 1.6960938230959939, + "learning_rate": 1.7315886040288468e-06, + "loss": 0.9599, + "step": 7406 + }, + { + "epoch": 0.5566661656395611, + "grad_norm": 1.641948493093918, + "learning_rate": 1.7311061740155477e-06, + "loss": 0.9102, + "step": 7407 + }, + { + "epoch": 0.5567413197053961, + "grad_norm": 1.5125238999929644, + "learning_rate": 1.7306237599352365e-06, + "loss": 0.8662, + "step": 7408 + }, + { + "epoch": 0.556816473771231, + "grad_norm": 22.32719010464035, + "learning_rate": 1.7301413618165e-06, + "loss": 0.9734, + "step": 7409 + }, + { + "epoch": 0.556891627837066, + "grad_norm": 0.6454332455779008, + "learning_rate": 1.7296589796879215e-06, + "loss": 0.8555, + "step": 7410 + }, + { + "epoch": 0.5569667819029009, + "grad_norm": 1.6455780829302096, + "learning_rate": 1.7291766135780825e-06, + "loss": 0.9508, + "step": 7411 + }, + { + "epoch": 0.5570419359687359, + "grad_norm": 1.49746752581149, + "learning_rate": 1.728694263515567e-06, + "loss": 0.9992, + "step": 7412 + }, + { + "epoch": 0.5571170900345709, + "grad_norm": 1.7505949235298417, + "learning_rate": 1.728211929528955e-06, + "loss": 0.9617, + "step": 7413 + }, + { + "epoch": 0.5571922441004058, + "grad_norm": 1.8907086062270115, + "learning_rate": 1.727729611646827e-06, + "loss": 0.9194, + "step": 7414 + }, + { + "epoch": 0.5572673981662408, + "grad_norm": 1.6481271605320158, + "learning_rate": 1.7272473098977623e-06, + "loss": 0.9743, + "step": 7415 + }, + { + "epoch": 0.5573425522320757, + "grad_norm": 1.9331097854523664, + "learning_rate": 1.7267650243103384e-06, + "loss": 0.9871, + "step": 7416 + }, + { + "epoch": 0.5574177062979108, + "grad_norm": 0.7550448975255432, + "learning_rate": 1.7262827549131337e-06, + "loss": 0.8461, + "step": 7417 + }, + { + "epoch": 0.5574928603637457, + "grad_norm": 1.4385573603358472, + "learning_rate": 1.7258005017347234e-06, + "loss": 1.0179, + "step": 7418 + }, + { + "epoch": 0.5575680144295806, + "grad_norm": 1.5886131228859097, + "learning_rate": 1.725318264803684e-06, + "loss": 0.8905, + "step": 7419 + }, + { + "epoch": 0.5576431684954156, + "grad_norm": 1.4503442552933095, + "learning_rate": 1.724836044148589e-06, + "loss": 0.9529, + "step": 7420 + }, + { + "epoch": 0.5577183225612505, + "grad_norm": 1.7242362114420113, + "learning_rate": 1.7243538397980115e-06, + "loss": 1.0251, + "step": 7421 + }, + { + "epoch": 0.5577934766270856, + "grad_norm": 1.5654206254855065, + "learning_rate": 1.7238716517805249e-06, + "loss": 0.997, + "step": 7422 + }, + { + "epoch": 0.5578686306929205, + "grad_norm": 1.6470459040467669, + "learning_rate": 1.7233894801247002e-06, + "loss": 0.9662, + "step": 7423 + }, + { + "epoch": 0.5579437847587555, + "grad_norm": 2.143630430929725, + "learning_rate": 1.7229073248591084e-06, + "loss": 1.0191, + "step": 7424 + }, + { + "epoch": 0.5580189388245904, + "grad_norm": 1.5030606357173169, + "learning_rate": 1.7224251860123185e-06, + "loss": 0.9889, + "step": 7425 + }, + { + "epoch": 0.5580940928904253, + "grad_norm": 1.461156410502989, + "learning_rate": 1.7219430636128989e-06, + "loss": 0.9856, + "step": 7426 + }, + { + "epoch": 0.5581692469562604, + "grad_norm": 1.5240749492688244, + "learning_rate": 1.721460957689418e-06, + "loss": 0.97, + "step": 7427 + }, + { + "epoch": 0.5582444010220953, + "grad_norm": 1.9689428446946953, + "learning_rate": 1.720978868270441e-06, + "loss": 1.0831, + "step": 7428 + }, + { + "epoch": 0.5583195550879303, + "grad_norm": 1.5660636698469594, + "learning_rate": 1.7204967953845358e-06, + "loss": 1.0014, + "step": 7429 + }, + { + "epoch": 0.5583947091537652, + "grad_norm": 2.09819629046947, + "learning_rate": 1.7200147390602643e-06, + "loss": 0.9953, + "step": 7430 + }, + { + "epoch": 0.5584698632196001, + "grad_norm": 1.974978026738643, + "learning_rate": 1.7195326993261927e-06, + "loss": 0.9411, + "step": 7431 + }, + { + "epoch": 0.5585450172854352, + "grad_norm": 1.5834278864819966, + "learning_rate": 1.7190506762108828e-06, + "loss": 0.9158, + "step": 7432 + }, + { + "epoch": 0.5586201713512701, + "grad_norm": 2.3407857842220348, + "learning_rate": 1.7185686697428954e-06, + "loss": 0.9906, + "step": 7433 + }, + { + "epoch": 0.5586953254171051, + "grad_norm": 1.46465745885736, + "learning_rate": 1.7180866799507925e-06, + "loss": 0.9324, + "step": 7434 + }, + { + "epoch": 0.55877047948294, + "grad_norm": 1.4001485916028715, + "learning_rate": 1.717604706863133e-06, + "loss": 0.9487, + "step": 7435 + }, + { + "epoch": 0.558845633548775, + "grad_norm": 1.4815325374989132, + "learning_rate": 1.7171227505084764e-06, + "loss": 0.9748, + "step": 7436 + }, + { + "epoch": 0.55892078761461, + "grad_norm": 1.7165203544461054, + "learning_rate": 1.71664081091538e-06, + "loss": 1.0237, + "step": 7437 + }, + { + "epoch": 0.5589959416804449, + "grad_norm": 1.7892475282364972, + "learning_rate": 1.7161588881124003e-06, + "loss": 1.0626, + "step": 7438 + }, + { + "epoch": 0.5590710957462799, + "grad_norm": 1.8263606499187879, + "learning_rate": 1.7156769821280937e-06, + "loss": 0.9627, + "step": 7439 + }, + { + "epoch": 0.5591462498121148, + "grad_norm": 1.9836649187443876, + "learning_rate": 1.7151950929910145e-06, + "loss": 0.9049, + "step": 7440 + }, + { + "epoch": 0.5592214038779498, + "grad_norm": 1.808612205460643, + "learning_rate": 1.7147132207297165e-06, + "loss": 0.9559, + "step": 7441 + }, + { + "epoch": 0.5592965579437847, + "grad_norm": 0.7338108980424357, + "learning_rate": 1.7142313653727531e-06, + "loss": 0.8105, + "step": 7442 + }, + { + "epoch": 0.5593717120096198, + "grad_norm": 6.894589622185617, + "learning_rate": 1.7137495269486749e-06, + "loss": 1.0971, + "step": 7443 + }, + { + "epoch": 0.5594468660754547, + "grad_norm": 1.3390205901986543, + "learning_rate": 1.7132677054860339e-06, + "loss": 0.9217, + "step": 7444 + }, + { + "epoch": 0.5595220201412896, + "grad_norm": 2.852203380732348, + "learning_rate": 1.7127859010133788e-06, + "loss": 0.9109, + "step": 7445 + }, + { + "epoch": 0.5595971742071246, + "grad_norm": 1.4327084668911831, + "learning_rate": 1.7123041135592593e-06, + "loss": 0.8294, + "step": 7446 + }, + { + "epoch": 0.5596723282729595, + "grad_norm": 1.6249464466367056, + "learning_rate": 1.7118223431522227e-06, + "loss": 0.9006, + "step": 7447 + }, + { + "epoch": 0.5597474823387946, + "grad_norm": 1.7146275412266818, + "learning_rate": 1.7113405898208156e-06, + "loss": 0.9928, + "step": 7448 + }, + { + "epoch": 0.5598226364046295, + "grad_norm": 6.948015371495594, + "learning_rate": 1.710858853593584e-06, + "loss": 0.9995, + "step": 7449 + }, + { + "epoch": 0.5598977904704645, + "grad_norm": 2.272276067160552, + "learning_rate": 1.710377134499072e-06, + "loss": 0.9414, + "step": 7450 + }, + { + "epoch": 0.5599729445362994, + "grad_norm": 1.8810855654207237, + "learning_rate": 1.7098954325658249e-06, + "loss": 0.996, + "step": 7451 + }, + { + "epoch": 0.5600480986021343, + "grad_norm": 3.0482827089116435, + "learning_rate": 1.7094137478223831e-06, + "loss": 0.9952, + "step": 7452 + }, + { + "epoch": 0.5601232526679694, + "grad_norm": 2.028554095706536, + "learning_rate": 1.7089320802972901e-06, + "loss": 0.7878, + "step": 7453 + }, + { + "epoch": 0.5601984067338043, + "grad_norm": 1.7646525582971047, + "learning_rate": 1.7084504300190862e-06, + "loss": 0.9592, + "step": 7454 + }, + { + "epoch": 0.5602735607996393, + "grad_norm": 1.884465801524851, + "learning_rate": 1.7079687970163105e-06, + "loss": 1.0658, + "step": 7455 + }, + { + "epoch": 0.5603487148654742, + "grad_norm": 1.8149449530837014, + "learning_rate": 1.7074871813175018e-06, + "loss": 0.9885, + "step": 7456 + }, + { + "epoch": 0.5604238689313091, + "grad_norm": 1.6319786753570245, + "learning_rate": 1.7070055829511973e-06, + "loss": 0.9207, + "step": 7457 + }, + { + "epoch": 0.5604990229971442, + "grad_norm": 1.9509605316166454, + "learning_rate": 1.7065240019459347e-06, + "loss": 0.9758, + "step": 7458 + }, + { + "epoch": 0.5605741770629791, + "grad_norm": 1.9686879950553364, + "learning_rate": 1.7060424383302485e-06, + "loss": 0.9829, + "step": 7459 + }, + { + "epoch": 0.5606493311288141, + "grad_norm": 1.4357591952096382, + "learning_rate": 1.7055608921326731e-06, + "loss": 0.9568, + "step": 7460 + }, + { + "epoch": 0.560724485194649, + "grad_norm": 1.4850607643154223, + "learning_rate": 1.7050793633817431e-06, + "loss": 0.8504, + "step": 7461 + }, + { + "epoch": 0.560799639260484, + "grad_norm": 1.7971491728881124, + "learning_rate": 1.7045978521059894e-06, + "loss": 0.951, + "step": 7462 + }, + { + "epoch": 0.560874793326319, + "grad_norm": 0.72478242437514, + "learning_rate": 1.7041163583339446e-06, + "loss": 0.817, + "step": 7463 + }, + { + "epoch": 0.5609499473921539, + "grad_norm": 0.7574272356242016, + "learning_rate": 1.7036348820941386e-06, + "loss": 0.9391, + "step": 7464 + }, + { + "epoch": 0.5610251014579889, + "grad_norm": 1.8367344202097649, + "learning_rate": 1.7031534234151001e-06, + "loss": 0.9684, + "step": 7465 + }, + { + "epoch": 0.5611002555238238, + "grad_norm": 1.8274270706132254, + "learning_rate": 1.7026719823253585e-06, + "loss": 0.9566, + "step": 7466 + }, + { + "epoch": 0.5611754095896588, + "grad_norm": 1.7017740613778298, + "learning_rate": 1.7021905588534402e-06, + "loss": 1.059, + "step": 7467 + }, + { + "epoch": 0.5612505636554938, + "grad_norm": 1.7594513437839616, + "learning_rate": 1.701709153027872e-06, + "loss": 0.9177, + "step": 7468 + }, + { + "epoch": 0.5613257177213288, + "grad_norm": 1.635621132997308, + "learning_rate": 1.7012277648771787e-06, + "loss": 0.8893, + "step": 7469 + }, + { + "epoch": 0.5614008717871637, + "grad_norm": 1.9563762611861353, + "learning_rate": 1.700746394429884e-06, + "loss": 1.0007, + "step": 7470 + }, + { + "epoch": 0.5614760258529986, + "grad_norm": 1.4629901195148154, + "learning_rate": 1.7002650417145119e-06, + "loss": 0.8796, + "step": 7471 + }, + { + "epoch": 0.5615511799188336, + "grad_norm": 3.7485919852183254, + "learning_rate": 1.699783706759583e-06, + "loss": 0.9825, + "step": 7472 + }, + { + "epoch": 0.5616263339846685, + "grad_norm": 1.421218552858002, + "learning_rate": 1.6993023895936196e-06, + "loss": 1.0382, + "step": 7473 + }, + { + "epoch": 0.5617014880505036, + "grad_norm": 1.596410587510624, + "learning_rate": 1.6988210902451407e-06, + "loss": 1.0478, + "step": 7474 + }, + { + "epoch": 0.5617766421163385, + "grad_norm": 1.6158653419921902, + "learning_rate": 1.698339808742666e-06, + "loss": 0.9408, + "step": 7475 + }, + { + "epoch": 0.5618517961821734, + "grad_norm": 2.910535872104169, + "learning_rate": 1.697858545114713e-06, + "loss": 1.0679, + "step": 7476 + }, + { + "epoch": 0.5619269502480084, + "grad_norm": 1.4911991061457628, + "learning_rate": 1.6973772993897978e-06, + "loss": 0.9968, + "step": 7477 + }, + { + "epoch": 0.5620021043138433, + "grad_norm": 1.8161417450643227, + "learning_rate": 1.6968960715964364e-06, + "loss": 1.0164, + "step": 7478 + }, + { + "epoch": 0.5620772583796784, + "grad_norm": 1.6619782625835673, + "learning_rate": 1.6964148617631432e-06, + "loss": 0.9393, + "step": 7479 + }, + { + "epoch": 0.5621524124455133, + "grad_norm": 1.46243274892745, + "learning_rate": 1.6959336699184323e-06, + "loss": 0.9646, + "step": 7480 + }, + { + "epoch": 0.5622275665113483, + "grad_norm": 1.6119507852334987, + "learning_rate": 1.695452496090816e-06, + "loss": 1.0027, + "step": 7481 + }, + { + "epoch": 0.5623027205771832, + "grad_norm": 1.5857786653403458, + "learning_rate": 1.694971340308805e-06, + "loss": 0.9391, + "step": 7482 + }, + { + "epoch": 0.5623778746430181, + "grad_norm": 1.5291016747628967, + "learning_rate": 1.6944902026009107e-06, + "loss": 0.9658, + "step": 7483 + }, + { + "epoch": 0.5624530287088532, + "grad_norm": 2.2427543960237353, + "learning_rate": 1.694009082995641e-06, + "loss": 0.9868, + "step": 7484 + }, + { + "epoch": 0.5625281827746881, + "grad_norm": 2.36602583478816, + "learning_rate": 1.6935279815215056e-06, + "loss": 0.947, + "step": 7485 + }, + { + "epoch": 0.5626033368405231, + "grad_norm": 1.4347722253561939, + "learning_rate": 1.6930468982070106e-06, + "loss": 0.8982, + "step": 7486 + }, + { + "epoch": 0.562678490906358, + "grad_norm": 1.847958239661502, + "learning_rate": 1.6925658330806618e-06, + "loss": 0.9825, + "step": 7487 + }, + { + "epoch": 0.562753644972193, + "grad_norm": 1.695723031258237, + "learning_rate": 1.6920847861709653e-06, + "loss": 1.0172, + "step": 7488 + }, + { + "epoch": 0.562828799038028, + "grad_norm": 2.193113590100999, + "learning_rate": 1.6916037575064238e-06, + "loss": 0.9288, + "step": 7489 + }, + { + "epoch": 0.5629039531038629, + "grad_norm": 1.4169831857323048, + "learning_rate": 1.6911227471155408e-06, + "loss": 0.9614, + "step": 7490 + }, + { + "epoch": 0.5629791071696979, + "grad_norm": 1.9290910795273981, + "learning_rate": 1.6906417550268182e-06, + "loss": 0.9518, + "step": 7491 + }, + { + "epoch": 0.5630542612355328, + "grad_norm": 1.9564874620706114, + "learning_rate": 1.6901607812687558e-06, + "loss": 0.9841, + "step": 7492 + }, + { + "epoch": 0.5631294153013678, + "grad_norm": 1.6784218921550045, + "learning_rate": 1.6896798258698538e-06, + "loss": 0.9163, + "step": 7493 + }, + { + "epoch": 0.5632045693672028, + "grad_norm": 2.020974563959524, + "learning_rate": 1.6891988888586094e-06, + "loss": 0.9893, + "step": 7494 + }, + { + "epoch": 0.5632797234330378, + "grad_norm": 1.3477064040429625, + "learning_rate": 1.6887179702635219e-06, + "loss": 0.9948, + "step": 7495 + }, + { + "epoch": 0.5633548774988727, + "grad_norm": 1.4721130664513749, + "learning_rate": 1.6882370701130863e-06, + "loss": 0.9459, + "step": 7496 + }, + { + "epoch": 0.5634300315647076, + "grad_norm": 1.9151253376872004, + "learning_rate": 1.6877561884357987e-06, + "loss": 0.9219, + "step": 7497 + }, + { + "epoch": 0.5635051856305426, + "grad_norm": 1.503862652827896, + "learning_rate": 1.6872753252601525e-06, + "loss": 0.9187, + "step": 7498 + }, + { + "epoch": 0.5635803396963776, + "grad_norm": 1.5430576733424801, + "learning_rate": 1.6867944806146403e-06, + "loss": 0.9151, + "step": 7499 + }, + { + "epoch": 0.5636554937622126, + "grad_norm": 1.895917750844086, + "learning_rate": 1.6863136545277547e-06, + "loss": 0.8759, + "step": 7500 + }, + { + "epoch": 0.5637306478280475, + "grad_norm": 1.4721771189340431, + "learning_rate": 1.685832847027986e-06, + "loss": 0.9129, + "step": 7501 + }, + { + "epoch": 0.5638058018938824, + "grad_norm": 1.7367888827283946, + "learning_rate": 1.6853520581438246e-06, + "loss": 0.9792, + "step": 7502 + }, + { + "epoch": 0.5638809559597174, + "grad_norm": 1.8083628430866605, + "learning_rate": 1.6848712879037588e-06, + "loss": 0.9631, + "step": 7503 + }, + { + "epoch": 0.5639561100255523, + "grad_norm": 1.49443013496749, + "learning_rate": 1.6843905363362754e-06, + "loss": 0.8783, + "step": 7504 + }, + { + "epoch": 0.5640312640913874, + "grad_norm": 1.6420046863372872, + "learning_rate": 1.6839098034698616e-06, + "loss": 0.9242, + "step": 7505 + }, + { + "epoch": 0.5641064181572223, + "grad_norm": 1.9128922602479226, + "learning_rate": 1.6834290893330017e-06, + "loss": 0.9664, + "step": 7506 + }, + { + "epoch": 0.5641815722230573, + "grad_norm": 0.8232977182070108, + "learning_rate": 1.682948393954181e-06, + "loss": 0.9023, + "step": 7507 + }, + { + "epoch": 0.5642567262888922, + "grad_norm": 1.5115542457833118, + "learning_rate": 1.6824677173618822e-06, + "loss": 1.0618, + "step": 7508 + }, + { + "epoch": 0.5643318803547271, + "grad_norm": 9.224675747101367, + "learning_rate": 1.6819870595845867e-06, + "loss": 0.9901, + "step": 7509 + }, + { + "epoch": 0.5644070344205622, + "grad_norm": 1.7651696082041348, + "learning_rate": 1.681506420650776e-06, + "loss": 1.0764, + "step": 7510 + }, + { + "epoch": 0.5644821884863971, + "grad_norm": 1.4563277031465227, + "learning_rate": 1.6810258005889287e-06, + "loss": 0.8774, + "step": 7511 + }, + { + "epoch": 0.5645573425522321, + "grad_norm": 1.5572829315382846, + "learning_rate": 1.680545199427525e-06, + "loss": 1.032, + "step": 7512 + }, + { + "epoch": 0.564632496618067, + "grad_norm": 1.701674447348526, + "learning_rate": 1.6800646171950415e-06, + "loss": 1.0269, + "step": 7513 + }, + { + "epoch": 0.564707650683902, + "grad_norm": 1.6288724388071392, + "learning_rate": 1.6795840539199538e-06, + "loss": 0.9735, + "step": 7514 + }, + { + "epoch": 0.564782804749737, + "grad_norm": 2.523063419356898, + "learning_rate": 1.6791035096307387e-06, + "loss": 0.8627, + "step": 7515 + }, + { + "epoch": 0.5648579588155719, + "grad_norm": 1.7332196210583866, + "learning_rate": 1.6786229843558685e-06, + "loss": 1.0689, + "step": 7516 + }, + { + "epoch": 0.5649331128814069, + "grad_norm": 1.4744513218535724, + "learning_rate": 1.6781424781238178e-06, + "loss": 1.0017, + "step": 7517 + }, + { + "epoch": 0.5650082669472418, + "grad_norm": 0.6813824883516464, + "learning_rate": 1.6776619909630574e-06, + "loss": 0.8078, + "step": 7518 + }, + { + "epoch": 0.5650834210130768, + "grad_norm": 1.7263052217903176, + "learning_rate": 1.6771815229020586e-06, + "loss": 0.9101, + "step": 7519 + }, + { + "epoch": 0.5651585750789118, + "grad_norm": 2.1767723798371352, + "learning_rate": 1.676701073969291e-06, + "loss": 1.0126, + "step": 7520 + }, + { + "epoch": 0.5652337291447467, + "grad_norm": 0.8178651177486898, + "learning_rate": 1.676220644193222e-06, + "loss": 0.8816, + "step": 7521 + }, + { + "epoch": 0.5653088832105817, + "grad_norm": 2.4475130654786046, + "learning_rate": 1.6757402336023205e-06, + "loss": 1.0401, + "step": 7522 + }, + { + "epoch": 0.5653840372764166, + "grad_norm": 2.327791126044014, + "learning_rate": 1.6752598422250512e-06, + "loss": 0.9968, + "step": 7523 + }, + { + "epoch": 0.5654591913422516, + "grad_norm": 1.7433642189664598, + "learning_rate": 1.6747794700898803e-06, + "loss": 1.1114, + "step": 7524 + }, + { + "epoch": 0.5655343454080866, + "grad_norm": 1.7003747274618302, + "learning_rate": 1.674299117225271e-06, + "loss": 0.9695, + "step": 7525 + }, + { + "epoch": 0.5656094994739216, + "grad_norm": 3.9173822087518855, + "learning_rate": 1.6738187836596858e-06, + "loss": 0.8599, + "step": 7526 + }, + { + "epoch": 0.5656846535397565, + "grad_norm": 1.7997256016858585, + "learning_rate": 1.6733384694215872e-06, + "loss": 1.0257, + "step": 7527 + }, + { + "epoch": 0.5657598076055914, + "grad_norm": 0.672238140885405, + "learning_rate": 1.6728581745394346e-06, + "loss": 0.8002, + "step": 7528 + }, + { + "epoch": 0.5658349616714264, + "grad_norm": 2.254693434318193, + "learning_rate": 1.6723778990416883e-06, + "loss": 1.1329, + "step": 7529 + }, + { + "epoch": 0.5659101157372614, + "grad_norm": 1.2167899191276403, + "learning_rate": 1.671897642956806e-06, + "loss": 0.9826, + "step": 7530 + }, + { + "epoch": 0.5659852698030964, + "grad_norm": 2.2643955446470008, + "learning_rate": 1.6714174063132447e-06, + "loss": 0.8668, + "step": 7531 + }, + { + "epoch": 0.5660604238689313, + "grad_norm": 1.5324714615670916, + "learning_rate": 1.6709371891394605e-06, + "loss": 1.0266, + "step": 7532 + }, + { + "epoch": 0.5661355779347663, + "grad_norm": 1.4946624044249237, + "learning_rate": 1.6704569914639073e-06, + "loss": 0.9372, + "step": 7533 + }, + { + "epoch": 0.5662107320006012, + "grad_norm": 3.754582593248406, + "learning_rate": 1.6699768133150399e-06, + "loss": 0.9995, + "step": 7534 + }, + { + "epoch": 0.5662858860664362, + "grad_norm": 1.2573585875915518, + "learning_rate": 1.6694966547213098e-06, + "loss": 0.9307, + "step": 7535 + }, + { + "epoch": 0.5663610401322712, + "grad_norm": 1.9597350305393573, + "learning_rate": 1.669016515711168e-06, + "loss": 0.9497, + "step": 7536 + }, + { + "epoch": 0.5664361941981061, + "grad_norm": 1.568872275279523, + "learning_rate": 1.668536396313066e-06, + "loss": 1.0641, + "step": 7537 + }, + { + "epoch": 0.5665113482639411, + "grad_norm": 1.9256461755763863, + "learning_rate": 1.6680562965554508e-06, + "loss": 0.9626, + "step": 7538 + }, + { + "epoch": 0.566586502329776, + "grad_norm": 1.889701365436893, + "learning_rate": 1.6675762164667717e-06, + "loss": 1.0124, + "step": 7539 + }, + { + "epoch": 0.5666616563956111, + "grad_norm": 1.5945176912261139, + "learning_rate": 1.6670961560754744e-06, + "loss": 0.9478, + "step": 7540 + }, + { + "epoch": 0.566736810461446, + "grad_norm": 1.3805914245225301, + "learning_rate": 1.6666161154100052e-06, + "loss": 0.9973, + "step": 7541 + }, + { + "epoch": 0.5668119645272809, + "grad_norm": 0.7621855259267996, + "learning_rate": 1.6661360944988076e-06, + "loss": 0.8297, + "step": 7542 + }, + { + "epoch": 0.5668871185931159, + "grad_norm": 2.2543953244292787, + "learning_rate": 1.6656560933703244e-06, + "loss": 1.102, + "step": 7543 + }, + { + "epoch": 0.5669622726589508, + "grad_norm": 2.0304294983085027, + "learning_rate": 1.6651761120529983e-06, + "loss": 0.9005, + "step": 7544 + }, + { + "epoch": 0.5670374267247859, + "grad_norm": 2.304399996136435, + "learning_rate": 1.6646961505752696e-06, + "loss": 0.9776, + "step": 7545 + }, + { + "epoch": 0.5671125807906208, + "grad_norm": 1.9039263085962816, + "learning_rate": 1.6642162089655782e-06, + "loss": 1.0584, + "step": 7546 + }, + { + "epoch": 0.5671877348564557, + "grad_norm": 1.8864917184145948, + "learning_rate": 1.6637362872523621e-06, + "loss": 0.8749, + "step": 7547 + }, + { + "epoch": 0.5672628889222907, + "grad_norm": 2.0273890260554266, + "learning_rate": 1.6632563854640583e-06, + "loss": 1.0435, + "step": 7548 + }, + { + "epoch": 0.5673380429881256, + "grad_norm": 1.4982808911343706, + "learning_rate": 1.6627765036291034e-06, + "loss": 0.9223, + "step": 7549 + }, + { + "epoch": 0.5674131970539606, + "grad_norm": 1.3945121443265145, + "learning_rate": 1.6622966417759319e-06, + "loss": 1.0088, + "step": 7550 + }, + { + "epoch": 0.5674883511197956, + "grad_norm": 1.7352821108648933, + "learning_rate": 1.6618167999329778e-06, + "loss": 0.9402, + "step": 7551 + }, + { + "epoch": 0.5675635051856306, + "grad_norm": 1.7924441565376674, + "learning_rate": 1.6613369781286732e-06, + "loss": 0.9526, + "step": 7552 + }, + { + "epoch": 0.5676386592514655, + "grad_norm": 1.4046292183637779, + "learning_rate": 1.660857176391449e-06, + "loss": 0.9633, + "step": 7553 + }, + { + "epoch": 0.5677138133173004, + "grad_norm": 3.563572370851666, + "learning_rate": 1.6603773947497364e-06, + "loss": 0.9465, + "step": 7554 + }, + { + "epoch": 0.5677889673831354, + "grad_norm": 1.6295509434778424, + "learning_rate": 1.6598976332319631e-06, + "loss": 0.9485, + "step": 7555 + }, + { + "epoch": 0.5678641214489704, + "grad_norm": 1.6185219828727562, + "learning_rate": 1.6594178918665578e-06, + "loss": 1.0077, + "step": 7556 + }, + { + "epoch": 0.5679392755148054, + "grad_norm": 1.6409093701187776, + "learning_rate": 1.6589381706819467e-06, + "loss": 1.0151, + "step": 7557 + }, + { + "epoch": 0.5680144295806403, + "grad_norm": 1.46491251466202, + "learning_rate": 1.658458469706554e-06, + "loss": 1.0019, + "step": 7558 + }, + { + "epoch": 0.5680895836464753, + "grad_norm": 2.2454158921086167, + "learning_rate": 1.6579787889688062e-06, + "loss": 1.0884, + "step": 7559 + }, + { + "epoch": 0.5681647377123102, + "grad_norm": 1.729222315923072, + "learning_rate": 1.6574991284971235e-06, + "loss": 0.9458, + "step": 7560 + }, + { + "epoch": 0.5682398917781452, + "grad_norm": 1.6681564189743368, + "learning_rate": 1.6570194883199298e-06, + "loss": 0.9384, + "step": 7561 + }, + { + "epoch": 0.5683150458439802, + "grad_norm": 1.5723752569987934, + "learning_rate": 1.6565398684656442e-06, + "loss": 0.9375, + "step": 7562 + }, + { + "epoch": 0.5683901999098151, + "grad_norm": 3.491421267702724, + "learning_rate": 1.6560602689626872e-06, + "loss": 0.9051, + "step": 7563 + }, + { + "epoch": 0.5684653539756501, + "grad_norm": 4.307704525972163, + "learning_rate": 1.6555806898394764e-06, + "loss": 0.97, + "step": 7564 + }, + { + "epoch": 0.568540508041485, + "grad_norm": 1.8145828118599086, + "learning_rate": 1.655101131124428e-06, + "loss": 0.9481, + "step": 7565 + }, + { + "epoch": 0.56861566210732, + "grad_norm": 2.02203473693863, + "learning_rate": 1.6546215928459589e-06, + "loss": 0.8817, + "step": 7566 + }, + { + "epoch": 0.568690816173155, + "grad_norm": 0.6767986555008915, + "learning_rate": 1.6541420750324825e-06, + "loss": 0.8302, + "step": 7567 + }, + { + "epoch": 0.5687659702389899, + "grad_norm": 1.688939857794368, + "learning_rate": 1.6536625777124128e-06, + "loss": 0.8332, + "step": 7568 + }, + { + "epoch": 0.5688411243048249, + "grad_norm": 0.7853947367418699, + "learning_rate": 1.6531831009141616e-06, + "loss": 0.8745, + "step": 7569 + }, + { + "epoch": 0.5689162783706598, + "grad_norm": 1.9597417217497812, + "learning_rate": 1.6527036446661393e-06, + "loss": 0.9061, + "step": 7570 + }, + { + "epoch": 0.5689914324364949, + "grad_norm": 1.7552733668684612, + "learning_rate": 1.6522242089967564e-06, + "loss": 0.878, + "step": 7571 + }, + { + "epoch": 0.5690665865023298, + "grad_norm": 3.4420495184395787, + "learning_rate": 1.6517447939344205e-06, + "loss": 0.862, + "step": 7572 + }, + { + "epoch": 0.5691417405681647, + "grad_norm": 1.2669385992495823, + "learning_rate": 1.6512653995075393e-06, + "loss": 0.9382, + "step": 7573 + }, + { + "epoch": 0.5692168946339997, + "grad_norm": 1.5857788908622619, + "learning_rate": 1.650786025744519e-06, + "loss": 0.9913, + "step": 7574 + }, + { + "epoch": 0.5692920486998346, + "grad_norm": 1.622895785757685, + "learning_rate": 1.6503066726737632e-06, + "loss": 1.0193, + "step": 7575 + }, + { + "epoch": 0.5693672027656697, + "grad_norm": 1.8720356714648405, + "learning_rate": 1.6498273403236764e-06, + "loss": 0.8902, + "step": 7576 + }, + { + "epoch": 0.5694423568315046, + "grad_norm": 1.6702748421972873, + "learning_rate": 1.6493480287226605e-06, + "loss": 0.865, + "step": 7577 + }, + { + "epoch": 0.5695175108973396, + "grad_norm": 1.6203335001953976, + "learning_rate": 1.6488687378991168e-06, + "loss": 1.007, + "step": 7578 + }, + { + "epoch": 0.5695926649631745, + "grad_norm": 1.8568570695266238, + "learning_rate": 1.648389467881444e-06, + "loss": 0.9582, + "step": 7579 + }, + { + "epoch": 0.5696678190290094, + "grad_norm": 15.879349240438183, + "learning_rate": 1.6479102186980428e-06, + "loss": 0.9233, + "step": 7580 + }, + { + "epoch": 0.5697429730948445, + "grad_norm": 0.6650324973792028, + "learning_rate": 1.6474309903773098e-06, + "loss": 0.7793, + "step": 7581 + }, + { + "epoch": 0.5698181271606794, + "grad_norm": 1.6792089445914546, + "learning_rate": 1.6469517829476391e-06, + "loss": 1.0218, + "step": 7582 + }, + { + "epoch": 0.5698932812265144, + "grad_norm": 1.5302907313457228, + "learning_rate": 1.6464725964374285e-06, + "loss": 0.9989, + "step": 7583 + }, + { + "epoch": 0.5699684352923493, + "grad_norm": 1.7597349378003802, + "learning_rate": 1.6459934308750694e-06, + "loss": 1.1461, + "step": 7584 + }, + { + "epoch": 0.5700435893581843, + "grad_norm": 1.270053134167145, + "learning_rate": 1.6455142862889557e-06, + "loss": 1.0354, + "step": 7585 + }, + { + "epoch": 0.5701187434240192, + "grad_norm": 0.6933913532458446, + "learning_rate": 1.6450351627074781e-06, + "loss": 0.8307, + "step": 7586 + }, + { + "epoch": 0.5701938974898542, + "grad_norm": 2.074634230214682, + "learning_rate": 1.6445560601590257e-06, + "loss": 0.9767, + "step": 7587 + }, + { + "epoch": 0.5702690515556892, + "grad_norm": 2.059630970538714, + "learning_rate": 1.6440769786719883e-06, + "loss": 0.8595, + "step": 7588 + }, + { + "epoch": 0.5703442056215241, + "grad_norm": 1.5597202752043537, + "learning_rate": 1.6435979182747526e-06, + "loss": 1.0314, + "step": 7589 + }, + { + "epoch": 0.5704193596873591, + "grad_norm": 15.135218675670053, + "learning_rate": 1.6431188789957053e-06, + "loss": 0.9291, + "step": 7590 + }, + { + "epoch": 0.570494513753194, + "grad_norm": 1.5965736657269447, + "learning_rate": 1.642639860863231e-06, + "loss": 0.9251, + "step": 7591 + }, + { + "epoch": 0.570569667819029, + "grad_norm": 1.6296615496111848, + "learning_rate": 1.642160863905713e-06, + "loss": 0.992, + "step": 7592 + }, + { + "epoch": 0.570644821884864, + "grad_norm": 2.868033385685008, + "learning_rate": 1.6416818881515344e-06, + "loss": 0.8936, + "step": 7593 + }, + { + "epoch": 0.5707199759506989, + "grad_norm": 2.240654291575992, + "learning_rate": 1.6412029336290755e-06, + "loss": 0.9239, + "step": 7594 + }, + { + "epoch": 0.5707951300165339, + "grad_norm": 1.4778757865133825, + "learning_rate": 1.6407240003667172e-06, + "loss": 0.9884, + "step": 7595 + }, + { + "epoch": 0.5708702840823688, + "grad_norm": 1.427981333111054, + "learning_rate": 1.640245088392838e-06, + "loss": 0.9676, + "step": 7596 + }, + { + "epoch": 0.5709454381482039, + "grad_norm": 1.4342801895887372, + "learning_rate": 1.6397661977358142e-06, + "loss": 0.95, + "step": 7597 + }, + { + "epoch": 0.5710205922140388, + "grad_norm": 1.7299018430702116, + "learning_rate": 1.639287328424023e-06, + "loss": 1.0473, + "step": 7598 + }, + { + "epoch": 0.5710957462798737, + "grad_norm": 2.7451532307921473, + "learning_rate": 1.638808480485838e-06, + "loss": 1.0066, + "step": 7599 + }, + { + "epoch": 0.5711709003457087, + "grad_norm": 1.6809601686377798, + "learning_rate": 1.638329653949635e-06, + "loss": 0.9372, + "step": 7600 + }, + { + "epoch": 0.5712460544115436, + "grad_norm": 0.7337650038242828, + "learning_rate": 1.6378508488437835e-06, + "loss": 0.839, + "step": 7601 + }, + { + "epoch": 0.5713212084773787, + "grad_norm": 1.75314178458762, + "learning_rate": 1.6373720651966569e-06, + "loss": 1.0032, + "step": 7602 + }, + { + "epoch": 0.5713963625432136, + "grad_norm": 0.7587921653104795, + "learning_rate": 1.6368933030366241e-06, + "loss": 0.83, + "step": 7603 + }, + { + "epoch": 0.5714715166090486, + "grad_norm": 2.854439733951863, + "learning_rate": 1.6364145623920528e-06, + "loss": 0.9883, + "step": 7604 + }, + { + "epoch": 0.5715466706748835, + "grad_norm": 1.648818491799366, + "learning_rate": 1.6359358432913118e-06, + "loss": 1.0068, + "step": 7605 + }, + { + "epoch": 0.5716218247407184, + "grad_norm": 0.7022922670974209, + "learning_rate": 1.6354571457627656e-06, + "loss": 0.8489, + "step": 7606 + }, + { + "epoch": 0.5716969788065535, + "grad_norm": 1.6904233406780766, + "learning_rate": 1.6349784698347797e-06, + "loss": 0.859, + "step": 7607 + }, + { + "epoch": 0.5717721328723884, + "grad_norm": 0.6739038618343075, + "learning_rate": 1.6344998155357175e-06, + "loss": 0.7532, + "step": 7608 + }, + { + "epoch": 0.5718472869382234, + "grad_norm": 4.280915852893021, + "learning_rate": 1.63402118289394e-06, + "loss": 0.9317, + "step": 7609 + }, + { + "epoch": 0.5719224410040583, + "grad_norm": 1.554907241704133, + "learning_rate": 1.6335425719378097e-06, + "loss": 0.9712, + "step": 7610 + }, + { + "epoch": 0.5719975950698932, + "grad_norm": 0.67644375646652, + "learning_rate": 1.6330639826956848e-06, + "loss": 0.854, + "step": 7611 + }, + { + "epoch": 0.5720727491357283, + "grad_norm": 2.6680291589290226, + "learning_rate": 1.632585415195924e-06, + "loss": 0.9471, + "step": 7612 + }, + { + "epoch": 0.5721479032015632, + "grad_norm": 2.288707093649701, + "learning_rate": 1.6321068694668846e-06, + "loss": 0.9276, + "step": 7613 + }, + { + "epoch": 0.5722230572673982, + "grad_norm": 2.0576109304455175, + "learning_rate": 1.6316283455369215e-06, + "loss": 0.9982, + "step": 7614 + }, + { + "epoch": 0.5722982113332331, + "grad_norm": 2.2473616496138917, + "learning_rate": 1.63114984343439e-06, + "loss": 0.9537, + "step": 7615 + }, + { + "epoch": 0.5723733653990681, + "grad_norm": 2.252629333237909, + "learning_rate": 1.630671363187642e-06, + "loss": 0.9341, + "step": 7616 + }, + { + "epoch": 0.572448519464903, + "grad_norm": 1.6710404068760136, + "learning_rate": 1.6301929048250306e-06, + "loss": 0.9183, + "step": 7617 + }, + { + "epoch": 0.572523673530738, + "grad_norm": 1.678169664751093, + "learning_rate": 1.6297144683749057e-06, + "loss": 1.1063, + "step": 7618 + }, + { + "epoch": 0.572598827596573, + "grad_norm": 1.6242922929093044, + "learning_rate": 1.6292360538656162e-06, + "loss": 0.9939, + "step": 7619 + }, + { + "epoch": 0.5726739816624079, + "grad_norm": 2.5505132607416483, + "learning_rate": 1.6287576613255105e-06, + "loss": 0.8728, + "step": 7620 + }, + { + "epoch": 0.5727491357282429, + "grad_norm": 1.291464861620342, + "learning_rate": 1.6282792907829341e-06, + "loss": 0.9991, + "step": 7621 + }, + { + "epoch": 0.5728242897940778, + "grad_norm": 1.791184478754023, + "learning_rate": 1.6278009422662345e-06, + "loss": 1.0648, + "step": 7622 + }, + { + "epoch": 0.5728994438599129, + "grad_norm": 1.449290065604413, + "learning_rate": 1.6273226158037528e-06, + "loss": 0.8936, + "step": 7623 + }, + { + "epoch": 0.5729745979257478, + "grad_norm": 1.9510241386683937, + "learning_rate": 1.6268443114238345e-06, + "loss": 0.9236, + "step": 7624 + }, + { + "epoch": 0.5730497519915827, + "grad_norm": 2.2650741433646195, + "learning_rate": 1.6263660291548191e-06, + "loss": 0.9878, + "step": 7625 + }, + { + "epoch": 0.5731249060574177, + "grad_norm": 1.8890844957508337, + "learning_rate": 1.6258877690250472e-06, + "loss": 1.0316, + "step": 7626 + }, + { + "epoch": 0.5732000601232526, + "grad_norm": 2.993939317678208, + "learning_rate": 1.6254095310628578e-06, + "loss": 1.0307, + "step": 7627 + }, + { + "epoch": 0.5732752141890877, + "grad_norm": 1.9231184371722205, + "learning_rate": 1.6249313152965876e-06, + "loss": 1.0485, + "step": 7628 + }, + { + "epoch": 0.5733503682549226, + "grad_norm": 2.7485894140057856, + "learning_rate": 1.6244531217545738e-06, + "loss": 1.0129, + "step": 7629 + }, + { + "epoch": 0.5734255223207576, + "grad_norm": 0.7077162990920941, + "learning_rate": 1.6239749504651505e-06, + "loss": 0.8268, + "step": 7630 + }, + { + "epoch": 0.5735006763865925, + "grad_norm": 1.2856349807077376, + "learning_rate": 1.6234968014566509e-06, + "loss": 1.0009, + "step": 7631 + }, + { + "epoch": 0.5735758304524274, + "grad_norm": 1.828471876312518, + "learning_rate": 1.6230186747574077e-06, + "loss": 0.9409, + "step": 7632 + }, + { + "epoch": 0.5736509845182625, + "grad_norm": 1.6164872132163444, + "learning_rate": 1.6225405703957515e-06, + "loss": 0.9286, + "step": 7633 + }, + { + "epoch": 0.5737261385840974, + "grad_norm": 1.62617420908357, + "learning_rate": 1.6220624884000123e-06, + "loss": 1.0008, + "step": 7634 + }, + { + "epoch": 0.5738012926499324, + "grad_norm": 1.546149960537454, + "learning_rate": 1.6215844287985178e-06, + "loss": 0.9217, + "step": 7635 + }, + { + "epoch": 0.5738764467157673, + "grad_norm": 1.5161816076340857, + "learning_rate": 1.6211063916195945e-06, + "loss": 0.9604, + "step": 7636 + }, + { + "epoch": 0.5739516007816022, + "grad_norm": 1.6124213561759515, + "learning_rate": 1.6206283768915687e-06, + "loss": 0.9238, + "step": 7637 + }, + { + "epoch": 0.5740267548474373, + "grad_norm": 2.470101288483169, + "learning_rate": 1.620150384642764e-06, + "loss": 1.0036, + "step": 7638 + }, + { + "epoch": 0.5741019089132722, + "grad_norm": 2.200296520710955, + "learning_rate": 1.619672414901504e-06, + "loss": 0.9929, + "step": 7639 + }, + { + "epoch": 0.5741770629791072, + "grad_norm": 2.0862283682458824, + "learning_rate": 1.6191944676961097e-06, + "loss": 0.9489, + "step": 7640 + }, + { + "epoch": 0.5742522170449421, + "grad_norm": 2.010200831500447, + "learning_rate": 1.6187165430549011e-06, + "loss": 1.054, + "step": 7641 + }, + { + "epoch": 0.5743273711107771, + "grad_norm": 2.044764238491965, + "learning_rate": 1.6182386410061976e-06, + "loss": 0.9104, + "step": 7642 + }, + { + "epoch": 0.574402525176612, + "grad_norm": 2.3798349255928004, + "learning_rate": 1.6177607615783158e-06, + "loss": 1.0326, + "step": 7643 + }, + { + "epoch": 0.574477679242447, + "grad_norm": 1.6427539636067474, + "learning_rate": 1.6172829047995733e-06, + "loss": 0.9841, + "step": 7644 + }, + { + "epoch": 0.574552833308282, + "grad_norm": 1.629050634923798, + "learning_rate": 1.616805070698284e-06, + "loss": 0.9201, + "step": 7645 + }, + { + "epoch": 0.5746279873741169, + "grad_norm": 3.7676115375656827, + "learning_rate": 1.6163272593027615e-06, + "loss": 0.9144, + "step": 7646 + }, + { + "epoch": 0.5747031414399519, + "grad_norm": 1.749127511102208, + "learning_rate": 1.6158494706413187e-06, + "loss": 0.9672, + "step": 7647 + }, + { + "epoch": 0.5747782955057869, + "grad_norm": 2.425395449613447, + "learning_rate": 1.6153717047422648e-06, + "loss": 1.0032, + "step": 7648 + }, + { + "epoch": 0.5748534495716219, + "grad_norm": 1.4931790717239426, + "learning_rate": 1.614893961633911e-06, + "loss": 1.079, + "step": 7649 + }, + { + "epoch": 0.5749286036374568, + "grad_norm": 3.789744144751493, + "learning_rate": 1.6144162413445642e-06, + "loss": 1.0336, + "step": 7650 + }, + { + "epoch": 0.5750037577032917, + "grad_norm": 1.6585708048774614, + "learning_rate": 1.6139385439025319e-06, + "loss": 1.0455, + "step": 7651 + }, + { + "epoch": 0.5750789117691267, + "grad_norm": 1.6526841699002393, + "learning_rate": 1.6134608693361193e-06, + "loss": 0.9446, + "step": 7652 + }, + { + "epoch": 0.5751540658349616, + "grad_norm": 1.3609696604523382, + "learning_rate": 1.61298321767363e-06, + "loss": 0.9755, + "step": 7653 + }, + { + "epoch": 0.5752292199007967, + "grad_norm": 0.7502176445832767, + "learning_rate": 1.6125055889433674e-06, + "loss": 0.8541, + "step": 7654 + }, + { + "epoch": 0.5753043739666316, + "grad_norm": 4.221665342872634, + "learning_rate": 1.612027983173632e-06, + "loss": 1.0241, + "step": 7655 + }, + { + "epoch": 0.5753795280324665, + "grad_norm": 1.3527850316449586, + "learning_rate": 1.6115504003927245e-06, + "loss": 0.9735, + "step": 7656 + }, + { + "epoch": 0.5754546820983015, + "grad_norm": 2.2454747151369974, + "learning_rate": 1.6110728406289436e-06, + "loss": 0.8267, + "step": 7657 + }, + { + "epoch": 0.5755298361641364, + "grad_norm": 1.3723309795346172, + "learning_rate": 1.6105953039105855e-06, + "loss": 1.0335, + "step": 7658 + }, + { + "epoch": 0.5756049902299715, + "grad_norm": 1.5495139067359056, + "learning_rate": 1.6101177902659474e-06, + "loss": 0.9154, + "step": 7659 + }, + { + "epoch": 0.5756801442958064, + "grad_norm": 2.358533797713587, + "learning_rate": 1.6096402997233225e-06, + "loss": 0.9645, + "step": 7660 + }, + { + "epoch": 0.5757552983616414, + "grad_norm": 1.8188618786146966, + "learning_rate": 1.6091628323110053e-06, + "loss": 0.8811, + "step": 7661 + }, + { + "epoch": 0.5758304524274763, + "grad_norm": 3.780694589729584, + "learning_rate": 1.6086853880572868e-06, + "loss": 0.9374, + "step": 7662 + }, + { + "epoch": 0.5759056064933112, + "grad_norm": 1.7641972533421402, + "learning_rate": 1.6082079669904572e-06, + "loss": 0.8977, + "step": 7663 + }, + { + "epoch": 0.5759807605591463, + "grad_norm": 2.267093051281249, + "learning_rate": 1.607730569138806e-06, + "loss": 0.9749, + "step": 7664 + }, + { + "epoch": 0.5760559146249812, + "grad_norm": 1.2663789081387042, + "learning_rate": 1.60725319453062e-06, + "loss": 0.9678, + "step": 7665 + }, + { + "epoch": 0.5761310686908162, + "grad_norm": 1.9308074913331617, + "learning_rate": 1.606775843194187e-06, + "loss": 1.0367, + "step": 7666 + }, + { + "epoch": 0.5762062227566511, + "grad_norm": 2.286679221731405, + "learning_rate": 1.6062985151577904e-06, + "loss": 1.0732, + "step": 7667 + }, + { + "epoch": 0.5762813768224861, + "grad_norm": 1.52614193045553, + "learning_rate": 1.605821210449715e-06, + "loss": 0.9818, + "step": 7668 + }, + { + "epoch": 0.5763565308883211, + "grad_norm": 1.712282253260701, + "learning_rate": 1.6053439290982422e-06, + "loss": 1.0463, + "step": 7669 + }, + { + "epoch": 0.576431684954156, + "grad_norm": 2.4593562783446012, + "learning_rate": 1.6048666711316526e-06, + "loss": 1.039, + "step": 7670 + }, + { + "epoch": 0.576506839019991, + "grad_norm": 2.1963875198515446, + "learning_rate": 1.6043894365782262e-06, + "loss": 1.0551, + "step": 7671 + }, + { + "epoch": 0.5765819930858259, + "grad_norm": 1.6709684962573708, + "learning_rate": 1.60391222546624e-06, + "loss": 0.9555, + "step": 7672 + }, + { + "epoch": 0.5766571471516609, + "grad_norm": 1.6368377569447177, + "learning_rate": 1.6034350378239715e-06, + "loss": 1.0145, + "step": 7673 + }, + { + "epoch": 0.5767323012174959, + "grad_norm": 1.5098008402163248, + "learning_rate": 1.6029578736796958e-06, + "loss": 0.9353, + "step": 7674 + }, + { + "epoch": 0.5768074552833309, + "grad_norm": 1.5200636845347975, + "learning_rate": 1.6024807330616858e-06, + "loss": 0.9587, + "step": 7675 + }, + { + "epoch": 0.5768826093491658, + "grad_norm": 1.5293425532470215, + "learning_rate": 1.6020036159982154e-06, + "loss": 0.9515, + "step": 7676 + }, + { + "epoch": 0.5769577634150007, + "grad_norm": 1.5948801708351383, + "learning_rate": 1.601526522517554e-06, + "loss": 1.0333, + "step": 7677 + }, + { + "epoch": 0.5770329174808357, + "grad_norm": 1.5387770489717674, + "learning_rate": 1.6010494526479726e-06, + "loss": 1.0109, + "step": 7678 + }, + { + "epoch": 0.5771080715466707, + "grad_norm": 2.0557450158495896, + "learning_rate": 1.6005724064177387e-06, + "loss": 0.979, + "step": 7679 + }, + { + "epoch": 0.5771832256125057, + "grad_norm": 2.519053048581185, + "learning_rate": 1.6000953838551187e-06, + "loss": 0.8674, + "step": 7680 + }, + { + "epoch": 0.5772583796783406, + "grad_norm": 3.1747883162931485, + "learning_rate": 1.5996183849883793e-06, + "loss": 0.9622, + "step": 7681 + }, + { + "epoch": 0.5773335337441755, + "grad_norm": 1.6305481756557072, + "learning_rate": 1.599141409845783e-06, + "loss": 0.9911, + "step": 7682 + }, + { + "epoch": 0.5774086878100105, + "grad_norm": 1.577829692140217, + "learning_rate": 1.5986644584555937e-06, + "loss": 0.9839, + "step": 7683 + }, + { + "epoch": 0.5774838418758454, + "grad_norm": 2.3537183239021413, + "learning_rate": 1.598187530846072e-06, + "loss": 0.8959, + "step": 7684 + }, + { + "epoch": 0.5775589959416805, + "grad_norm": 1.6604727409719022, + "learning_rate": 1.5977106270454775e-06, + "loss": 0.8929, + "step": 7685 + }, + { + "epoch": 0.5776341500075154, + "grad_norm": 1.5820817350644563, + "learning_rate": 1.597233747082069e-06, + "loss": 1.005, + "step": 7686 + }, + { + "epoch": 0.5777093040733504, + "grad_norm": 2.1844797264829534, + "learning_rate": 1.5967568909841026e-06, + "loss": 0.9654, + "step": 7687 + }, + { + "epoch": 0.5777844581391853, + "grad_norm": 3.966236189449394, + "learning_rate": 1.5962800587798352e-06, + "loss": 0.9822, + "step": 7688 + }, + { + "epoch": 0.5778596122050202, + "grad_norm": 1.7740215398519383, + "learning_rate": 1.59580325049752e-06, + "loss": 0.9987, + "step": 7689 + }, + { + "epoch": 0.5779347662708553, + "grad_norm": 3.358261136650704, + "learning_rate": 1.5953264661654104e-06, + "loss": 1.0148, + "step": 7690 + }, + { + "epoch": 0.5780099203366902, + "grad_norm": 1.7593903644222897, + "learning_rate": 1.5948497058117574e-06, + "loss": 0.9232, + "step": 7691 + }, + { + "epoch": 0.5780850744025252, + "grad_norm": 2.3398221481421455, + "learning_rate": 1.59437296946481e-06, + "loss": 0.8989, + "step": 7692 + }, + { + "epoch": 0.5781602284683601, + "grad_norm": 4.574291303598382, + "learning_rate": 1.593896257152818e-06, + "loss": 0.8829, + "step": 7693 + }, + { + "epoch": 0.5782353825341952, + "grad_norm": 1.7042266362589147, + "learning_rate": 1.5934195689040276e-06, + "loss": 0.9571, + "step": 7694 + }, + { + "epoch": 0.5783105366000301, + "grad_norm": 1.8499182631504356, + "learning_rate": 1.592942904746685e-06, + "loss": 0.9467, + "step": 7695 + }, + { + "epoch": 0.578385690665865, + "grad_norm": 2.415757951125086, + "learning_rate": 1.592466264709034e-06, + "loss": 1.0274, + "step": 7696 + }, + { + "epoch": 0.5784608447317, + "grad_norm": 1.8439015471682358, + "learning_rate": 1.5919896488193166e-06, + "loss": 1.0479, + "step": 7697 + }, + { + "epoch": 0.5785359987975349, + "grad_norm": 0.8968025922622103, + "learning_rate": 1.5915130571057755e-06, + "loss": 0.911, + "step": 7698 + }, + { + "epoch": 0.57861115286337, + "grad_norm": 0.709319904893432, + "learning_rate": 1.5910364895966498e-06, + "loss": 0.831, + "step": 7699 + }, + { + "epoch": 0.5786863069292049, + "grad_norm": 3.086209442131743, + "learning_rate": 1.5905599463201785e-06, + "loss": 1.0082, + "step": 7700 + }, + { + "epoch": 0.5787614609950398, + "grad_norm": 2.6495827868259534, + "learning_rate": 1.590083427304598e-06, + "loss": 0.7643, + "step": 7701 + }, + { + "epoch": 0.5788366150608748, + "grad_norm": 1.7294797116071405, + "learning_rate": 1.5896069325781435e-06, + "loss": 0.8988, + "step": 7702 + }, + { + "epoch": 0.5789117691267097, + "grad_norm": 1.9168500881609882, + "learning_rate": 1.5891304621690508e-06, + "loss": 0.9707, + "step": 7703 + }, + { + "epoch": 0.5789869231925447, + "grad_norm": 4.002675592119846, + "learning_rate": 1.5886540161055507e-06, + "loss": 0.9653, + "step": 7704 + }, + { + "epoch": 0.5790620772583797, + "grad_norm": 2.1621344064367976, + "learning_rate": 1.5881775944158755e-06, + "loss": 0.8688, + "step": 7705 + }, + { + "epoch": 0.5791372313242147, + "grad_norm": 2.2016758041519497, + "learning_rate": 1.5877011971282553e-06, + "loss": 1.0217, + "step": 7706 + }, + { + "epoch": 0.5792123853900496, + "grad_norm": 5.396975991563759, + "learning_rate": 1.5872248242709168e-06, + "loss": 0.9388, + "step": 7707 + }, + { + "epoch": 0.5792875394558845, + "grad_norm": 1.8112401859078113, + "learning_rate": 1.5867484758720894e-06, + "loss": 0.9675, + "step": 7708 + }, + { + "epoch": 0.5793626935217195, + "grad_norm": 1.6498604368660594, + "learning_rate": 1.5862721519599963e-06, + "loss": 1.0091, + "step": 7709 + }, + { + "epoch": 0.5794378475875545, + "grad_norm": 2.3208562040543272, + "learning_rate": 1.585795852562863e-06, + "loss": 0.9467, + "step": 7710 + }, + { + "epoch": 0.5795130016533895, + "grad_norm": 1.8722372682056758, + "learning_rate": 1.585319577708911e-06, + "loss": 0.8959, + "step": 7711 + }, + { + "epoch": 0.5795881557192244, + "grad_norm": 0.6903142981033081, + "learning_rate": 1.5848433274263627e-06, + "loss": 0.8407, + "step": 7712 + }, + { + "epoch": 0.5796633097850594, + "grad_norm": 1.4899454899232067, + "learning_rate": 1.5843671017434366e-06, + "loss": 0.9398, + "step": 7713 + }, + { + "epoch": 0.5797384638508943, + "grad_norm": 1.6999972259274456, + "learning_rate": 1.583890900688351e-06, + "loss": 0.9428, + "step": 7714 + }, + { + "epoch": 0.5798136179167293, + "grad_norm": 1.9300283524797588, + "learning_rate": 1.5834147242893234e-06, + "loss": 0.9606, + "step": 7715 + }, + { + "epoch": 0.5798887719825643, + "grad_norm": 1.5848083234586705, + "learning_rate": 1.5829385725745684e-06, + "loss": 0.9222, + "step": 7716 + }, + { + "epoch": 0.5799639260483992, + "grad_norm": 1.5467264123317, + "learning_rate": 1.5824624455723e-06, + "loss": 0.9963, + "step": 7717 + }, + { + "epoch": 0.5800390801142342, + "grad_norm": 4.8799806241135455, + "learning_rate": 1.581986343310731e-06, + "loss": 1.0159, + "step": 7718 + }, + { + "epoch": 0.5801142341800691, + "grad_norm": 1.3291647935971702, + "learning_rate": 1.581510265818071e-06, + "loss": 0.8405, + "step": 7719 + }, + { + "epoch": 0.5801893882459042, + "grad_norm": 1.7606239046385217, + "learning_rate": 1.5810342131225308e-06, + "loss": 0.9445, + "step": 7720 + }, + { + "epoch": 0.5802645423117391, + "grad_norm": 2.2336560139578756, + "learning_rate": 1.5805581852523176e-06, + "loss": 0.8816, + "step": 7721 + }, + { + "epoch": 0.580339696377574, + "grad_norm": 1.7299922519576485, + "learning_rate": 1.5800821822356383e-06, + "loss": 0.8776, + "step": 7722 + }, + { + "epoch": 0.580414850443409, + "grad_norm": 2.136926728052334, + "learning_rate": 1.5796062041006978e-06, + "loss": 1.0562, + "step": 7723 + }, + { + "epoch": 0.5804900045092439, + "grad_norm": 0.8306360142697308, + "learning_rate": 1.579130250875699e-06, + "loss": 0.9026, + "step": 7724 + }, + { + "epoch": 0.580565158575079, + "grad_norm": 1.996440401975663, + "learning_rate": 1.578654322588845e-06, + "loss": 0.9469, + "step": 7725 + }, + { + "epoch": 0.5806403126409139, + "grad_norm": 1.9230704582545848, + "learning_rate": 1.5781784192683351e-06, + "loss": 0.9926, + "step": 7726 + }, + { + "epoch": 0.5807154667067488, + "grad_norm": 1.5664439969734945, + "learning_rate": 1.57770254094237e-06, + "loss": 0.9816, + "step": 7727 + }, + { + "epoch": 0.5807906207725838, + "grad_norm": 1.7009651417600145, + "learning_rate": 1.577226687639146e-06, + "loss": 0.885, + "step": 7728 + }, + { + "epoch": 0.5808657748384187, + "grad_norm": 1.4896217376808987, + "learning_rate": 1.5767508593868588e-06, + "loss": 0.9996, + "step": 7729 + }, + { + "epoch": 0.5809409289042538, + "grad_norm": 2.1836972561838874, + "learning_rate": 1.5762750562137056e-06, + "loss": 1.0484, + "step": 7730 + }, + { + "epoch": 0.5810160829700887, + "grad_norm": 1.87702051015957, + "learning_rate": 1.5757992781478762e-06, + "loss": 0.9578, + "step": 7731 + }, + { + "epoch": 0.5810912370359237, + "grad_norm": 1.6934553248975026, + "learning_rate": 1.575323525217565e-06, + "loss": 0.9033, + "step": 7732 + }, + { + "epoch": 0.5811663911017586, + "grad_norm": 1.365654832876373, + "learning_rate": 1.5748477974509606e-06, + "loss": 0.9995, + "step": 7733 + }, + { + "epoch": 0.5812415451675935, + "grad_norm": 1.9891959074460397, + "learning_rate": 1.5743720948762527e-06, + "loss": 0.9648, + "step": 7734 + }, + { + "epoch": 0.5813166992334285, + "grad_norm": 1.5967482245898108, + "learning_rate": 1.573896417521628e-06, + "loss": 1.0417, + "step": 7735 + }, + { + "epoch": 0.5813918532992635, + "grad_norm": 1.803037210902461, + "learning_rate": 1.5734207654152718e-06, + "loss": 0.9603, + "step": 7736 + }, + { + "epoch": 0.5814670073650985, + "grad_norm": 4.451641491516825, + "learning_rate": 1.572945138585369e-06, + "loss": 1.0146, + "step": 7737 + }, + { + "epoch": 0.5815421614309334, + "grad_norm": 1.5881752050061304, + "learning_rate": 1.572469537060102e-06, + "loss": 0.9838, + "step": 7738 + }, + { + "epoch": 0.5816173154967684, + "grad_norm": 0.7610057526043931, + "learning_rate": 1.5719939608676523e-06, + "loss": 0.8015, + "step": 7739 + }, + { + "epoch": 0.5816924695626033, + "grad_norm": 1.4512620695073888, + "learning_rate": 1.5715184100361992e-06, + "loss": 0.9684, + "step": 7740 + }, + { + "epoch": 0.5817676236284383, + "grad_norm": 1.7110105176699377, + "learning_rate": 1.5710428845939207e-06, + "loss": 0.9726, + "step": 7741 + }, + { + "epoch": 0.5818427776942733, + "grad_norm": 10.031111196347895, + "learning_rate": 1.5705673845689945e-06, + "loss": 1.0633, + "step": 7742 + }, + { + "epoch": 0.5819179317601082, + "grad_norm": 2.063269847186933, + "learning_rate": 1.5700919099895943e-06, + "loss": 0.9677, + "step": 7743 + }, + { + "epoch": 0.5819930858259432, + "grad_norm": 1.4433015597563177, + "learning_rate": 1.5696164608838956e-06, + "loss": 1.0702, + "step": 7744 + }, + { + "epoch": 0.5820682398917781, + "grad_norm": 3.8626877624897773, + "learning_rate": 1.5691410372800696e-06, + "loss": 0.9777, + "step": 7745 + }, + { + "epoch": 0.582143393957613, + "grad_norm": 2.070699288566704, + "learning_rate": 1.5686656392062863e-06, + "loss": 1.0417, + "step": 7746 + }, + { + "epoch": 0.5822185480234481, + "grad_norm": 3.367242303743035, + "learning_rate": 1.5681902666907161e-06, + "loss": 0.87, + "step": 7747 + }, + { + "epoch": 0.582293702089283, + "grad_norm": 8.838751900367495, + "learning_rate": 1.5677149197615257e-06, + "loss": 1.0104, + "step": 7748 + }, + { + "epoch": 0.582368856155118, + "grad_norm": 2.2347237908285593, + "learning_rate": 1.567239598446882e-06, + "loss": 0.9029, + "step": 7749 + }, + { + "epoch": 0.5824440102209529, + "grad_norm": 2.225567297461981, + "learning_rate": 1.5667643027749492e-06, + "loss": 0.8439, + "step": 7750 + }, + { + "epoch": 0.582519164286788, + "grad_norm": 1.5025472470438321, + "learning_rate": 1.5662890327738897e-06, + "loss": 0.8946, + "step": 7751 + }, + { + "epoch": 0.5825943183526229, + "grad_norm": 1.8631289757779252, + "learning_rate": 1.5658137884718672e-06, + "loss": 0.9681, + "step": 7752 + }, + { + "epoch": 0.5826694724184578, + "grad_norm": 2.188337329280756, + "learning_rate": 1.565338569897039e-06, + "loss": 0.8614, + "step": 7753 + }, + { + "epoch": 0.5827446264842928, + "grad_norm": 2.444671546857651, + "learning_rate": 1.5648633770775656e-06, + "loss": 0.9627, + "step": 7754 + }, + { + "epoch": 0.5828197805501277, + "grad_norm": 1.9695182391307775, + "learning_rate": 1.564388210041603e-06, + "loss": 1.0227, + "step": 7755 + }, + { + "epoch": 0.5828949346159628, + "grad_norm": 1.9051200144051579, + "learning_rate": 1.5639130688173077e-06, + "loss": 1.0988, + "step": 7756 + }, + { + "epoch": 0.5829700886817977, + "grad_norm": 1.6721764408630573, + "learning_rate": 1.5634379534328326e-06, + "loss": 1.0011, + "step": 7757 + }, + { + "epoch": 0.5830452427476327, + "grad_norm": 1.8018783570363202, + "learning_rate": 1.5629628639163304e-06, + "loss": 0.9465, + "step": 7758 + }, + { + "epoch": 0.5831203968134676, + "grad_norm": 1.9344692521559037, + "learning_rate": 1.5624878002959521e-06, + "loss": 0.8736, + "step": 7759 + }, + { + "epoch": 0.5831955508793025, + "grad_norm": 2.1131528579152827, + "learning_rate": 1.5620127625998469e-06, + "loss": 0.9457, + "step": 7760 + }, + { + "epoch": 0.5832707049451376, + "grad_norm": 2.2427314337783777, + "learning_rate": 1.5615377508561628e-06, + "loss": 0.966, + "step": 7761 + }, + { + "epoch": 0.5833458590109725, + "grad_norm": 0.8524693201646919, + "learning_rate": 1.561062765093046e-06, + "loss": 0.8729, + "step": 7762 + }, + { + "epoch": 0.5834210130768075, + "grad_norm": 4.131027903624877, + "learning_rate": 1.560587805338641e-06, + "loss": 0.9907, + "step": 7763 + }, + { + "epoch": 0.5834961671426424, + "grad_norm": 1.8444763304674812, + "learning_rate": 1.5601128716210915e-06, + "loss": 0.9918, + "step": 7764 + }, + { + "epoch": 0.5835713212084774, + "grad_norm": 2.441175770370848, + "learning_rate": 1.5596379639685382e-06, + "loss": 1.0013, + "step": 7765 + }, + { + "epoch": 0.5836464752743123, + "grad_norm": 1.3995879571688197, + "learning_rate": 1.5591630824091224e-06, + "loss": 0.9531, + "step": 7766 + }, + { + "epoch": 0.5837216293401473, + "grad_norm": 1.4621489698596533, + "learning_rate": 1.5586882269709819e-06, + "loss": 0.9609, + "step": 7767 + }, + { + "epoch": 0.5837967834059823, + "grad_norm": 1.8992201836100808, + "learning_rate": 1.5582133976822534e-06, + "loss": 1.0097, + "step": 7768 + }, + { + "epoch": 0.5838719374718172, + "grad_norm": 3.305093366080344, + "learning_rate": 1.5577385945710732e-06, + "loss": 0.9912, + "step": 7769 + }, + { + "epoch": 0.5839470915376522, + "grad_norm": 0.845471074995215, + "learning_rate": 1.5572638176655742e-06, + "loss": 0.9325, + "step": 7770 + }, + { + "epoch": 0.5840222456034871, + "grad_norm": 1.7402347860111047, + "learning_rate": 1.5567890669938905e-06, + "loss": 1.0041, + "step": 7771 + }, + { + "epoch": 0.5840973996693221, + "grad_norm": 1.7491605652728022, + "learning_rate": 1.5563143425841512e-06, + "loss": 0.8864, + "step": 7772 + }, + { + "epoch": 0.5841725537351571, + "grad_norm": 3.37443919290868, + "learning_rate": 1.5558396444644854e-06, + "loss": 1.0169, + "step": 7773 + }, + { + "epoch": 0.584247707800992, + "grad_norm": 1.6227639292517462, + "learning_rate": 1.5553649726630222e-06, + "loss": 0.9713, + "step": 7774 + }, + { + "epoch": 0.584322861866827, + "grad_norm": 1.5052291006307785, + "learning_rate": 1.5548903272078865e-06, + "loss": 0.9405, + "step": 7775 + }, + { + "epoch": 0.5843980159326619, + "grad_norm": 2.1714026946124965, + "learning_rate": 1.554415708127204e-06, + "loss": 0.8768, + "step": 7776 + }, + { + "epoch": 0.584473169998497, + "grad_norm": 1.5476560836245021, + "learning_rate": 1.5539411154490967e-06, + "loss": 0.9889, + "step": 7777 + }, + { + "epoch": 0.5845483240643319, + "grad_norm": 1.738489442851959, + "learning_rate": 1.5534665492016865e-06, + "loss": 0.9416, + "step": 7778 + }, + { + "epoch": 0.5846234781301668, + "grad_norm": 1.7819118943102863, + "learning_rate": 1.5529920094130932e-06, + "loss": 1.0516, + "step": 7779 + }, + { + "epoch": 0.5846986321960018, + "grad_norm": 1.6880616560247483, + "learning_rate": 1.552517496111435e-06, + "loss": 0.8257, + "step": 7780 + }, + { + "epoch": 0.5847737862618367, + "grad_norm": 1.828528465677382, + "learning_rate": 1.5520430093248286e-06, + "loss": 0.8998, + "step": 7781 + }, + { + "epoch": 0.5848489403276718, + "grad_norm": 2.4118472308210848, + "learning_rate": 1.5515685490813891e-06, + "loss": 0.9211, + "step": 7782 + }, + { + "epoch": 0.5849240943935067, + "grad_norm": 2.174356045159257, + "learning_rate": 1.5510941154092304e-06, + "loss": 0.9787, + "step": 7783 + }, + { + "epoch": 0.5849992484593417, + "grad_norm": 2.317994163690153, + "learning_rate": 1.5506197083364647e-06, + "loss": 1.0128, + "step": 7784 + }, + { + "epoch": 0.5850744025251766, + "grad_norm": 1.8511653425910803, + "learning_rate": 1.5501453278912013e-06, + "loss": 0.9288, + "step": 7785 + }, + { + "epoch": 0.5851495565910115, + "grad_norm": 1.5711655458561176, + "learning_rate": 1.5496709741015505e-06, + "loss": 0.954, + "step": 7786 + }, + { + "epoch": 0.5852247106568466, + "grad_norm": 1.719245769784529, + "learning_rate": 1.5491966469956187e-06, + "loss": 0.9467, + "step": 7787 + }, + { + "epoch": 0.5852998647226815, + "grad_norm": 0.7049503580498998, + "learning_rate": 1.5487223466015118e-06, + "loss": 0.8677, + "step": 7788 + }, + { + "epoch": 0.5853750187885165, + "grad_norm": 2.213861479749755, + "learning_rate": 1.5482480729473339e-06, + "loss": 1.0952, + "step": 7789 + }, + { + "epoch": 0.5854501728543514, + "grad_norm": 1.6818171994396218, + "learning_rate": 1.5477738260611875e-06, + "loss": 0.9824, + "step": 7790 + }, + { + "epoch": 0.5855253269201863, + "grad_norm": 1.8173370929524264, + "learning_rate": 1.5472996059711738e-06, + "loss": 0.9751, + "step": 7791 + }, + { + "epoch": 0.5856004809860214, + "grad_norm": 1.6406087965392178, + "learning_rate": 1.546825412705391e-06, + "loss": 0.9966, + "step": 7792 + }, + { + "epoch": 0.5856756350518563, + "grad_norm": 1.6975199308053992, + "learning_rate": 1.5463512462919393e-06, + "loss": 0.9628, + "step": 7793 + }, + { + "epoch": 0.5857507891176913, + "grad_norm": 1.4973178566794942, + "learning_rate": 1.5458771067589128e-06, + "loss": 0.9416, + "step": 7794 + }, + { + "epoch": 0.5858259431835262, + "grad_norm": 2.8211728630287083, + "learning_rate": 1.545402994134406e-06, + "loss": 1.0385, + "step": 7795 + }, + { + "epoch": 0.5859010972493612, + "grad_norm": 2.9166979288514785, + "learning_rate": 1.544928908446513e-06, + "loss": 0.9917, + "step": 7796 + }, + { + "epoch": 0.5859762513151962, + "grad_norm": 1.6565725624226466, + "learning_rate": 1.544454849723325e-06, + "loss": 0.8666, + "step": 7797 + }, + { + "epoch": 0.5860514053810311, + "grad_norm": 1.6694410914709885, + "learning_rate": 1.5439808179929316e-06, + "loss": 0.9121, + "step": 7798 + }, + { + "epoch": 0.5861265594468661, + "grad_norm": 1.786187473000929, + "learning_rate": 1.5435068132834204e-06, + "loss": 0.9363, + "step": 7799 + }, + { + "epoch": 0.586201713512701, + "grad_norm": 1.6152083222987732, + "learning_rate": 1.543032835622879e-06, + "loss": 0.9077, + "step": 7800 + }, + { + "epoch": 0.586276867578536, + "grad_norm": 1.5945849755827528, + "learning_rate": 1.5425588850393922e-06, + "loss": 0.9821, + "step": 7801 + }, + { + "epoch": 0.586352021644371, + "grad_norm": 1.6080600588718206, + "learning_rate": 1.5420849615610424e-06, + "loss": 0.9756, + "step": 7802 + }, + { + "epoch": 0.586427175710206, + "grad_norm": 1.6117110843448414, + "learning_rate": 1.541611065215913e-06, + "loss": 0.9064, + "step": 7803 + }, + { + "epoch": 0.5865023297760409, + "grad_norm": 1.5905168687680793, + "learning_rate": 1.5411371960320822e-06, + "loss": 1.055, + "step": 7804 + }, + { + "epoch": 0.5865774838418758, + "grad_norm": 1.5587678204996078, + "learning_rate": 1.5406633540376307e-06, + "loss": 0.9371, + "step": 7805 + }, + { + "epoch": 0.5866526379077108, + "grad_norm": 1.460568386510188, + "learning_rate": 1.5401895392606339e-06, + "loss": 0.9906, + "step": 7806 + }, + { + "epoch": 0.5867277919735457, + "grad_norm": 3.3176748772116516, + "learning_rate": 1.5397157517291674e-06, + "loss": 1.0139, + "step": 7807 + }, + { + "epoch": 0.5868029460393808, + "grad_norm": 2.5472393126372572, + "learning_rate": 1.5392419914713054e-06, + "loss": 1.0688, + "step": 7808 + }, + { + "epoch": 0.5868781001052157, + "grad_norm": 2.483292445757202, + "learning_rate": 1.5387682585151195e-06, + "loss": 0.9714, + "step": 7809 + }, + { + "epoch": 0.5869532541710507, + "grad_norm": 1.5986703950988543, + "learning_rate": 1.538294552888681e-06, + "loss": 1.0128, + "step": 7810 + }, + { + "epoch": 0.5870284082368856, + "grad_norm": 2.044239239944629, + "learning_rate": 1.537820874620058e-06, + "loss": 0.8584, + "step": 7811 + }, + { + "epoch": 0.5871035623027205, + "grad_norm": 1.7191608978164832, + "learning_rate": 1.537347223737318e-06, + "loss": 1.0215, + "step": 7812 + }, + { + "epoch": 0.5871787163685556, + "grad_norm": 0.803783835237284, + "learning_rate": 1.5368736002685266e-06, + "loss": 0.8287, + "step": 7813 + }, + { + "epoch": 0.5872538704343905, + "grad_norm": 1.5950199378271233, + "learning_rate": 1.5364000042417468e-06, + "loss": 0.8765, + "step": 7814 + }, + { + "epoch": 0.5873290245002255, + "grad_norm": 1.7611901278908533, + "learning_rate": 1.5359264356850435e-06, + "loss": 0.9787, + "step": 7815 + }, + { + "epoch": 0.5874041785660604, + "grad_norm": 1.814071368738751, + "learning_rate": 1.5354528946264757e-06, + "loss": 0.8907, + "step": 7816 + }, + { + "epoch": 0.5874793326318953, + "grad_norm": 1.3851647064243462, + "learning_rate": 1.534979381094102e-06, + "loss": 0.8993, + "step": 7817 + }, + { + "epoch": 0.5875544866977304, + "grad_norm": 2.88220655074116, + "learning_rate": 1.534505895115981e-06, + "loss": 0.9758, + "step": 7818 + }, + { + "epoch": 0.5876296407635653, + "grad_norm": 1.877923847413566, + "learning_rate": 1.5340324367201681e-06, + "loss": 0.9403, + "step": 7819 + }, + { + "epoch": 0.5877047948294003, + "grad_norm": 2.330696580257682, + "learning_rate": 1.533559005934718e-06, + "loss": 0.9448, + "step": 7820 + }, + { + "epoch": 0.5877799488952352, + "grad_norm": 1.6101884730417215, + "learning_rate": 1.5330856027876827e-06, + "loss": 1.0467, + "step": 7821 + }, + { + "epoch": 0.5878551029610702, + "grad_norm": 2.178311149272117, + "learning_rate": 1.5326122273071133e-06, + "loss": 1.0247, + "step": 7822 + }, + { + "epoch": 0.5879302570269052, + "grad_norm": 2.4618491769885615, + "learning_rate": 1.5321388795210597e-06, + "loss": 1.0542, + "step": 7823 + }, + { + "epoch": 0.5880054110927401, + "grad_norm": 1.9195079337828287, + "learning_rate": 1.5316655594575685e-06, + "loss": 0.9626, + "step": 7824 + }, + { + "epoch": 0.5880805651585751, + "grad_norm": 1.6352230694455494, + "learning_rate": 1.5311922671446864e-06, + "loss": 0.8712, + "step": 7825 + }, + { + "epoch": 0.58815571922441, + "grad_norm": 1.8548321404815593, + "learning_rate": 1.5307190026104574e-06, + "loss": 1.0199, + "step": 7826 + }, + { + "epoch": 0.588230873290245, + "grad_norm": 1.7167715997321753, + "learning_rate": 1.530245765882925e-06, + "loss": 1.0121, + "step": 7827 + }, + { + "epoch": 0.58830602735608, + "grad_norm": 1.6458940454544164, + "learning_rate": 1.5297725569901298e-06, + "loss": 1.0129, + "step": 7828 + }, + { + "epoch": 0.588381181421915, + "grad_norm": 1.6767241130768327, + "learning_rate": 1.5292993759601107e-06, + "loss": 0.9783, + "step": 7829 + }, + { + "epoch": 0.5884563354877499, + "grad_norm": 1.6030892886818788, + "learning_rate": 1.5288262228209066e-06, + "loss": 0.9244, + "step": 7830 + }, + { + "epoch": 0.5885314895535848, + "grad_norm": 1.3723480052202306, + "learning_rate": 1.5283530976005524e-06, + "loss": 0.9709, + "step": 7831 + }, + { + "epoch": 0.5886066436194198, + "grad_norm": 1.6070221673971201, + "learning_rate": 1.5278800003270838e-06, + "loss": 0.9944, + "step": 7832 + }, + { + "epoch": 0.5886817976852547, + "grad_norm": 1.9637680597767873, + "learning_rate": 1.527406931028533e-06, + "loss": 0.9348, + "step": 7833 + }, + { + "epoch": 0.5887569517510898, + "grad_norm": 1.8357804900864325, + "learning_rate": 1.5269338897329308e-06, + "loss": 0.9514, + "step": 7834 + }, + { + "epoch": 0.5888321058169247, + "grad_norm": 1.3991376093985786, + "learning_rate": 1.5264608764683074e-06, + "loss": 1.0177, + "step": 7835 + }, + { + "epoch": 0.5889072598827596, + "grad_norm": 1.8850502234125217, + "learning_rate": 1.5259878912626896e-06, + "loss": 1.0701, + "step": 7836 + }, + { + "epoch": 0.5889824139485946, + "grad_norm": 1.4937487518433021, + "learning_rate": 1.5255149341441053e-06, + "loss": 0.9353, + "step": 7837 + }, + { + "epoch": 0.5890575680144295, + "grad_norm": 2.2028217512789623, + "learning_rate": 1.5250420051405783e-06, + "loss": 0.9459, + "step": 7838 + }, + { + "epoch": 0.5891327220802646, + "grad_norm": 1.4776626605814205, + "learning_rate": 1.5245691042801302e-06, + "loss": 1.1141, + "step": 7839 + }, + { + "epoch": 0.5892078761460995, + "grad_norm": 1.6509161659781555, + "learning_rate": 1.524096231590784e-06, + "loss": 0.9774, + "step": 7840 + }, + { + "epoch": 0.5892830302119345, + "grad_norm": 1.6049913128056235, + "learning_rate": 1.523623387100558e-06, + "loss": 1.0214, + "step": 7841 + }, + { + "epoch": 0.5893581842777694, + "grad_norm": 5.061429322604409, + "learning_rate": 1.5231505708374707e-06, + "loss": 1.0421, + "step": 7842 + }, + { + "epoch": 0.5894333383436043, + "grad_norm": 1.6329301855577787, + "learning_rate": 1.5226777828295378e-06, + "loss": 1.067, + "step": 7843 + }, + { + "epoch": 0.5895084924094394, + "grad_norm": 1.742055092471443, + "learning_rate": 1.5222050231047747e-06, + "loss": 0.9948, + "step": 7844 + }, + { + "epoch": 0.5895836464752743, + "grad_norm": 2.2929507617147373, + "learning_rate": 1.5217322916911934e-06, + "loss": 0.9712, + "step": 7845 + }, + { + "epoch": 0.5896588005411093, + "grad_norm": 1.928943691454218, + "learning_rate": 1.5212595886168046e-06, + "loss": 1.0259, + "step": 7846 + }, + { + "epoch": 0.5897339546069442, + "grad_norm": 2.3021047829941983, + "learning_rate": 1.5207869139096191e-06, + "loss": 1.0547, + "step": 7847 + }, + { + "epoch": 0.5898091086727792, + "grad_norm": 1.4499613460781806, + "learning_rate": 1.5203142675976434e-06, + "loss": 0.9324, + "step": 7848 + }, + { + "epoch": 0.5898842627386142, + "grad_norm": 19.281725613309437, + "learning_rate": 1.5198416497088849e-06, + "loss": 0.9806, + "step": 7849 + }, + { + "epoch": 0.5899594168044491, + "grad_norm": 2.266042736610028, + "learning_rate": 1.519369060271347e-06, + "loss": 1.0675, + "step": 7850 + }, + { + "epoch": 0.5900345708702841, + "grad_norm": 2.2023433083124115, + "learning_rate": 1.5188964993130321e-06, + "loss": 0.8718, + "step": 7851 + }, + { + "epoch": 0.590109724936119, + "grad_norm": 1.5729983310678484, + "learning_rate": 1.5184239668619427e-06, + "loss": 1.0209, + "step": 7852 + }, + { + "epoch": 0.590184879001954, + "grad_norm": 5.070937295985275, + "learning_rate": 1.517951462946077e-06, + "loss": 0.8775, + "step": 7853 + }, + { + "epoch": 0.590260033067789, + "grad_norm": 1.5604139517271143, + "learning_rate": 1.5174789875934332e-06, + "loss": 1.0638, + "step": 7854 + }, + { + "epoch": 0.590335187133624, + "grad_norm": 2.0419686022818957, + "learning_rate": 1.517006540832007e-06, + "loss": 0.9249, + "step": 7855 + }, + { + "epoch": 0.5904103411994589, + "grad_norm": 1.8255743507743194, + "learning_rate": 1.5165341226897926e-06, + "loss": 0.9986, + "step": 7856 + }, + { + "epoch": 0.5904854952652938, + "grad_norm": 1.8079421515367196, + "learning_rate": 1.5160617331947828e-06, + "loss": 0.9678, + "step": 7857 + }, + { + "epoch": 0.5905606493311288, + "grad_norm": 1.565591043496182, + "learning_rate": 1.515589372374968e-06, + "loss": 1.097, + "step": 7858 + }, + { + "epoch": 0.5906358033969638, + "grad_norm": 1.4265420634382096, + "learning_rate": 1.5151170402583384e-06, + "loss": 0.9303, + "step": 7859 + }, + { + "epoch": 0.5907109574627988, + "grad_norm": 1.80409997401261, + "learning_rate": 1.5146447368728814e-06, + "loss": 1.0492, + "step": 7860 + }, + { + "epoch": 0.5907861115286337, + "grad_norm": 1.6013823617252765, + "learning_rate": 1.514172462246581e-06, + "loss": 1.0289, + "step": 7861 + }, + { + "epoch": 0.5908612655944686, + "grad_norm": 2.1789886557505636, + "learning_rate": 1.5137002164074234e-06, + "loss": 0.9459, + "step": 7862 + }, + { + "epoch": 0.5909364196603036, + "grad_norm": 1.9146547607890267, + "learning_rate": 1.5132279993833898e-06, + "loss": 0.9568, + "step": 7863 + }, + { + "epoch": 0.5910115737261386, + "grad_norm": 1.9988662367201575, + "learning_rate": 1.5127558112024617e-06, + "loss": 1.038, + "step": 7864 + }, + { + "epoch": 0.5910867277919736, + "grad_norm": 2.428655767178465, + "learning_rate": 1.512283651892617e-06, + "loss": 0.9088, + "step": 7865 + }, + { + "epoch": 0.5911618818578085, + "grad_norm": 1.7428543798795504, + "learning_rate": 1.5118115214818339e-06, + "loss": 1.0074, + "step": 7866 + }, + { + "epoch": 0.5912370359236435, + "grad_norm": 1.366052996086336, + "learning_rate": 1.5113394199980877e-06, + "loss": 1.0422, + "step": 7867 + }, + { + "epoch": 0.5913121899894784, + "grad_norm": 1.990347637214344, + "learning_rate": 1.5108673474693516e-06, + "loss": 1.034, + "step": 7868 + }, + { + "epoch": 0.5913873440553133, + "grad_norm": 1.5750210896851131, + "learning_rate": 1.5103953039235986e-06, + "loss": 0.9301, + "step": 7869 + }, + { + "epoch": 0.5914624981211484, + "grad_norm": 1.6678484620935843, + "learning_rate": 1.5099232893887983e-06, + "loss": 0.9854, + "step": 7870 + }, + { + "epoch": 0.5915376521869833, + "grad_norm": 1.7450956696569655, + "learning_rate": 1.5094513038929199e-06, + "loss": 1.073, + "step": 7871 + }, + { + "epoch": 0.5916128062528183, + "grad_norm": 1.6938232349736704, + "learning_rate": 1.5089793474639305e-06, + "loss": 0.899, + "step": 7872 + }, + { + "epoch": 0.5916879603186532, + "grad_norm": 1.786288180093377, + "learning_rate": 1.5085074201297943e-06, + "loss": 1.0657, + "step": 7873 + }, + { + "epoch": 0.5917631143844883, + "grad_norm": 2.6244251893439237, + "learning_rate": 1.5080355219184762e-06, + "loss": 0.967, + "step": 7874 + }, + { + "epoch": 0.5918382684503232, + "grad_norm": 1.7554067467476304, + "learning_rate": 1.5075636528579366e-06, + "loss": 1.0325, + "step": 7875 + }, + { + "epoch": 0.5919134225161581, + "grad_norm": 1.522862174141598, + "learning_rate": 1.507091812976137e-06, + "loss": 1.0495, + "step": 7876 + }, + { + "epoch": 0.5919885765819931, + "grad_norm": 2.009491333626964, + "learning_rate": 1.5066200023010347e-06, + "loss": 0.9204, + "step": 7877 + }, + { + "epoch": 0.592063730647828, + "grad_norm": 2.929018966691539, + "learning_rate": 1.5061482208605856e-06, + "loss": 0.9655, + "step": 7878 + }, + { + "epoch": 0.592138884713663, + "grad_norm": 1.7765736011705162, + "learning_rate": 1.505676468682747e-06, + "loss": 0.9689, + "step": 7879 + }, + { + "epoch": 0.592214038779498, + "grad_norm": 2.754254691082658, + "learning_rate": 1.5052047457954691e-06, + "loss": 1.0261, + "step": 7880 + }, + { + "epoch": 0.5922891928453329, + "grad_norm": 1.809139951944751, + "learning_rate": 1.5047330522267056e-06, + "loss": 0.9868, + "step": 7881 + }, + { + "epoch": 0.5923643469111679, + "grad_norm": 1.6857475081788424, + "learning_rate": 1.5042613880044053e-06, + "loss": 0.9926, + "step": 7882 + }, + { + "epoch": 0.5924395009770028, + "grad_norm": 1.6859102885089314, + "learning_rate": 1.5037897531565155e-06, + "loss": 0.9951, + "step": 7883 + }, + { + "epoch": 0.5925146550428378, + "grad_norm": 1.8351831712692839, + "learning_rate": 1.5033181477109835e-06, + "loss": 0.9942, + "step": 7884 + }, + { + "epoch": 0.5925898091086728, + "grad_norm": 1.7523829040370857, + "learning_rate": 1.5028465716957527e-06, + "loss": 0.9396, + "step": 7885 + }, + { + "epoch": 0.5926649631745078, + "grad_norm": 1.3986367877487669, + "learning_rate": 1.5023750251387668e-06, + "loss": 0.9556, + "step": 7886 + }, + { + "epoch": 0.5927401172403427, + "grad_norm": 1.772775198659864, + "learning_rate": 1.501903508067966e-06, + "loss": 0.9323, + "step": 7887 + }, + { + "epoch": 0.5928152713061776, + "grad_norm": 1.6426381430167243, + "learning_rate": 1.5014320205112897e-06, + "loss": 0.813, + "step": 7888 + }, + { + "epoch": 0.5928904253720126, + "grad_norm": 2.078537634704519, + "learning_rate": 1.5009605624966753e-06, + "loss": 0.9282, + "step": 7889 + }, + { + "epoch": 0.5929655794378476, + "grad_norm": 1.7834909630865465, + "learning_rate": 1.5004891340520583e-06, + "loss": 0.9777, + "step": 7890 + }, + { + "epoch": 0.5930407335036826, + "grad_norm": 2.442325510529574, + "learning_rate": 1.5000177352053732e-06, + "loss": 0.9425, + "step": 7891 + }, + { + "epoch": 0.5931158875695175, + "grad_norm": 1.5152312878358196, + "learning_rate": 1.4995463659845512e-06, + "loss": 0.9915, + "step": 7892 + }, + { + "epoch": 0.5931910416353525, + "grad_norm": 1.2770311632229854, + "learning_rate": 1.499075026417524e-06, + "loss": 0.9616, + "step": 7893 + }, + { + "epoch": 0.5932661957011874, + "grad_norm": 1.7020758100675666, + "learning_rate": 1.4986037165322199e-06, + "loss": 1.0315, + "step": 7894 + }, + { + "epoch": 0.5933413497670224, + "grad_norm": 0.7578181237080919, + "learning_rate": 1.498132436356565e-06, + "loss": 0.8854, + "step": 7895 + }, + { + "epoch": 0.5934165038328574, + "grad_norm": 1.377757601735521, + "learning_rate": 1.4976611859184852e-06, + "loss": 1.0122, + "step": 7896 + }, + { + "epoch": 0.5934916578986923, + "grad_norm": 2.4665118345003383, + "learning_rate": 1.4971899652459034e-06, + "loss": 0.8854, + "step": 7897 + }, + { + "epoch": 0.5935668119645273, + "grad_norm": 1.5315693113638293, + "learning_rate": 1.4967187743667423e-06, + "loss": 0.9667, + "step": 7898 + }, + { + "epoch": 0.5936419660303622, + "grad_norm": 2.9067477548586798, + "learning_rate": 1.4962476133089207e-06, + "loss": 1.0121, + "step": 7899 + }, + { + "epoch": 0.5937171200961973, + "grad_norm": 11.590662624846253, + "learning_rate": 1.4957764821003562e-06, + "loss": 0.9336, + "step": 7900 + }, + { + "epoch": 0.5937922741620322, + "grad_norm": 4.502869221122703, + "learning_rate": 1.4953053807689671e-06, + "loss": 0.9062, + "step": 7901 + }, + { + "epoch": 0.5938674282278671, + "grad_norm": 4.514538274498199, + "learning_rate": 1.4948343093426656e-06, + "loss": 0.9281, + "step": 7902 + }, + { + "epoch": 0.5939425822937021, + "grad_norm": 1.9904005226511745, + "learning_rate": 1.4943632678493668e-06, + "loss": 0.9591, + "step": 7903 + }, + { + "epoch": 0.594017736359537, + "grad_norm": 1.9305376041349218, + "learning_rate": 1.4938922563169801e-06, + "loss": 0.9083, + "step": 7904 + }, + { + "epoch": 0.594092890425372, + "grad_norm": 1.7089430798582241, + "learning_rate": 1.4934212747734153e-06, + "loss": 0.9189, + "step": 7905 + }, + { + "epoch": 0.594168044491207, + "grad_norm": 1.8779609824363617, + "learning_rate": 1.49295032324658e-06, + "loss": 0.98, + "step": 7906 + }, + { + "epoch": 0.5942431985570419, + "grad_norm": 1.4354088541749346, + "learning_rate": 1.492479401764379e-06, + "loss": 0.9137, + "step": 7907 + }, + { + "epoch": 0.5943183526228769, + "grad_norm": 1.4329096437758915, + "learning_rate": 1.4920085103547177e-06, + "loss": 0.962, + "step": 7908 + }, + { + "epoch": 0.5943935066887118, + "grad_norm": 1.9013232993618652, + "learning_rate": 1.491537649045497e-06, + "loss": 1.0581, + "step": 7909 + }, + { + "epoch": 0.5944686607545469, + "grad_norm": 1.3906564173203388, + "learning_rate": 1.4910668178646178e-06, + "loss": 0.8988, + "step": 7910 + }, + { + "epoch": 0.5945438148203818, + "grad_norm": 1.9058710565852097, + "learning_rate": 1.4905960168399783e-06, + "loss": 1.0424, + "step": 7911 + }, + { + "epoch": 0.5946189688862168, + "grad_norm": 1.7209221725377022, + "learning_rate": 1.4901252459994753e-06, + "loss": 0.9334, + "step": 7912 + }, + { + "epoch": 0.5946941229520517, + "grad_norm": 1.3417341947838288, + "learning_rate": 1.4896545053710044e-06, + "loss": 0.8694, + "step": 7913 + }, + { + "epoch": 0.5947692770178866, + "grad_norm": 1.7052002608524333, + "learning_rate": 1.4891837949824578e-06, + "loss": 1.034, + "step": 7914 + }, + { + "epoch": 0.5948444310837216, + "grad_norm": 1.562214024122664, + "learning_rate": 1.4887131148617279e-06, + "loss": 1.045, + "step": 7915 + }, + { + "epoch": 0.5949195851495566, + "grad_norm": 1.3850964580094838, + "learning_rate": 1.4882424650367034e-06, + "loss": 0.9606, + "step": 7916 + }, + { + "epoch": 0.5949947392153916, + "grad_norm": 1.9139654018141057, + "learning_rate": 1.4877718455352723e-06, + "loss": 1.0378, + "step": 7917 + }, + { + "epoch": 0.5950698932812265, + "grad_norm": 1.9103326696513314, + "learning_rate": 1.4873012563853213e-06, + "loss": 1.0347, + "step": 7918 + }, + { + "epoch": 0.5951450473470615, + "grad_norm": 1.5985178967316145, + "learning_rate": 1.4868306976147337e-06, + "loss": 0.9502, + "step": 7919 + }, + { + "epoch": 0.5952202014128964, + "grad_norm": 2.0066055648262684, + "learning_rate": 1.4863601692513927e-06, + "loss": 0.9846, + "step": 7920 + }, + { + "epoch": 0.5952953554787314, + "grad_norm": 1.6572731744328768, + "learning_rate": 1.4858896713231786e-06, + "loss": 1.0389, + "step": 7921 + }, + { + "epoch": 0.5953705095445664, + "grad_norm": 1.6655047260539175, + "learning_rate": 1.485419203857969e-06, + "loss": 1.0064, + "step": 7922 + }, + { + "epoch": 0.5954456636104013, + "grad_norm": 1.532841147681014, + "learning_rate": 1.4849487668836439e-06, + "loss": 1.0321, + "step": 7923 + }, + { + "epoch": 0.5955208176762363, + "grad_norm": 1.9008839232611034, + "learning_rate": 1.4844783604280746e-06, + "loss": 0.9132, + "step": 7924 + }, + { + "epoch": 0.5955959717420712, + "grad_norm": 1.2995309863973747, + "learning_rate": 1.4840079845191379e-06, + "loss": 0.9089, + "step": 7925 + }, + { + "epoch": 0.5956711258079062, + "grad_norm": 3.8542906595166264, + "learning_rate": 1.483537639184704e-06, + "loss": 1.0253, + "step": 7926 + }, + { + "epoch": 0.5957462798737412, + "grad_norm": 1.7154952703737076, + "learning_rate": 1.4830673244526418e-06, + "loss": 1.0583, + "step": 7927 + }, + { + "epoch": 0.5958214339395761, + "grad_norm": 1.770594532142641, + "learning_rate": 1.4825970403508208e-06, + "loss": 0.9706, + "step": 7928 + }, + { + "epoch": 0.5958965880054111, + "grad_norm": 1.7079685178307593, + "learning_rate": 1.482126786907106e-06, + "loss": 0.9673, + "step": 7929 + }, + { + "epoch": 0.595971742071246, + "grad_norm": 1.6757973979180831, + "learning_rate": 1.4816565641493623e-06, + "loss": 0.9178, + "step": 7930 + }, + { + "epoch": 0.5960468961370811, + "grad_norm": 1.757930293904781, + "learning_rate": 1.481186372105452e-06, + "loss": 0.9297, + "step": 7931 + }, + { + "epoch": 0.596122050202916, + "grad_norm": 1.3089051744343478, + "learning_rate": 1.4807162108032363e-06, + "loss": 1.0385, + "step": 7932 + }, + { + "epoch": 0.5961972042687509, + "grad_norm": 2.173905116771676, + "learning_rate": 1.4802460802705731e-06, + "loss": 0.9658, + "step": 7933 + }, + { + "epoch": 0.5962723583345859, + "grad_norm": 0.751454572006329, + "learning_rate": 1.4797759805353199e-06, + "loss": 0.8468, + "step": 7934 + }, + { + "epoch": 0.5963475124004208, + "grad_norm": 2.4606950912976653, + "learning_rate": 1.4793059116253322e-06, + "loss": 1.0321, + "step": 7935 + }, + { + "epoch": 0.5964226664662559, + "grad_norm": 16.835631160206024, + "learning_rate": 1.4788358735684626e-06, + "loss": 0.9794, + "step": 7936 + }, + { + "epoch": 0.5964978205320908, + "grad_norm": 1.7320812970038508, + "learning_rate": 1.4783658663925637e-06, + "loss": 1.0047, + "step": 7937 + }, + { + "epoch": 0.5965729745979258, + "grad_norm": 0.8285225327926352, + "learning_rate": 1.4778958901254847e-06, + "loss": 0.9165, + "step": 7938 + }, + { + "epoch": 0.5966481286637607, + "grad_norm": 1.5826259672582117, + "learning_rate": 1.477425944795073e-06, + "loss": 0.9097, + "step": 7939 + }, + { + "epoch": 0.5967232827295956, + "grad_norm": 1.6002057539251069, + "learning_rate": 1.4769560304291755e-06, + "loss": 0.9288, + "step": 7940 + }, + { + "epoch": 0.5967984367954307, + "grad_norm": 1.305685883413162, + "learning_rate": 1.4764861470556357e-06, + "loss": 0.9651, + "step": 7941 + }, + { + "epoch": 0.5968735908612656, + "grad_norm": 1.5337730174680866, + "learning_rate": 1.4760162947022972e-06, + "loss": 1.0951, + "step": 7942 + }, + { + "epoch": 0.5969487449271006, + "grad_norm": 1.4302550476550584, + "learning_rate": 1.475546473396999e-06, + "loss": 1.0428, + "step": 7943 + }, + { + "epoch": 0.5970238989929355, + "grad_norm": 2.297652028795585, + "learning_rate": 1.47507668316758e-06, + "loss": 0.8927, + "step": 7944 + }, + { + "epoch": 0.5970990530587705, + "grad_norm": 1.9722650423311614, + "learning_rate": 1.4746069240418785e-06, + "loss": 0.9798, + "step": 7945 + }, + { + "epoch": 0.5971742071246054, + "grad_norm": 1.9293897144187797, + "learning_rate": 1.474137196047728e-06, + "loss": 0.9339, + "step": 7946 + }, + { + "epoch": 0.5972493611904404, + "grad_norm": 2.8262746247843125, + "learning_rate": 1.473667499212963e-06, + "loss": 0.885, + "step": 7947 + }, + { + "epoch": 0.5973245152562754, + "grad_norm": 1.6823881218090475, + "learning_rate": 1.4731978335654138e-06, + "loss": 0.9378, + "step": 7948 + }, + { + "epoch": 0.5973996693221103, + "grad_norm": 1.7789962966906132, + "learning_rate": 1.47272819913291e-06, + "loss": 1.0642, + "step": 7949 + }, + { + "epoch": 0.5974748233879453, + "grad_norm": 2.324343586622489, + "learning_rate": 1.4722585959432802e-06, + "loss": 0.9646, + "step": 7950 + }, + { + "epoch": 0.5975499774537802, + "grad_norm": 1.8033193711582347, + "learning_rate": 1.4717890240243484e-06, + "loss": 1.0112, + "step": 7951 + }, + { + "epoch": 0.5976251315196152, + "grad_norm": 1.2963569709540086, + "learning_rate": 1.4713194834039401e-06, + "loss": 1.0149, + "step": 7952 + }, + { + "epoch": 0.5977002855854502, + "grad_norm": 1.7787435196673367, + "learning_rate": 1.470849974109877e-06, + "loss": 0.9787, + "step": 7953 + }, + { + "epoch": 0.5977754396512851, + "grad_norm": 1.4170242400589599, + "learning_rate": 1.470380496169979e-06, + "loss": 0.9331, + "step": 7954 + }, + { + "epoch": 0.5978505937171201, + "grad_norm": 2.0434295543067558, + "learning_rate": 1.4699110496120648e-06, + "loss": 0.9596, + "step": 7955 + }, + { + "epoch": 0.597925747782955, + "grad_norm": 12.684180816795543, + "learning_rate": 1.4694416344639503e-06, + "loss": 0.9149, + "step": 7956 + }, + { + "epoch": 0.5980009018487901, + "grad_norm": 1.7526713146328494, + "learning_rate": 1.4689722507534514e-06, + "loss": 0.9263, + "step": 7957 + }, + { + "epoch": 0.598076055914625, + "grad_norm": 1.618772605237388, + "learning_rate": 1.4685028985083794e-06, + "loss": 1.0324, + "step": 7958 + }, + { + "epoch": 0.5981512099804599, + "grad_norm": 2.031948263020323, + "learning_rate": 1.4680335777565462e-06, + "loss": 1.0248, + "step": 7959 + }, + { + "epoch": 0.5982263640462949, + "grad_norm": 1.3023689821358528, + "learning_rate": 1.467564288525761e-06, + "loss": 1.0574, + "step": 7960 + }, + { + "epoch": 0.5983015181121298, + "grad_norm": 1.2805972413628657, + "learning_rate": 1.4670950308438298e-06, + "loss": 1.0533, + "step": 7961 + }, + { + "epoch": 0.5983766721779649, + "grad_norm": 0.7581355792484411, + "learning_rate": 1.4666258047385588e-06, + "loss": 0.8536, + "step": 7962 + }, + { + "epoch": 0.5984518262437998, + "grad_norm": 1.8036766358786642, + "learning_rate": 1.4661566102377507e-06, + "loss": 0.9503, + "step": 7963 + }, + { + "epoch": 0.5985269803096348, + "grad_norm": 1.5654734738057146, + "learning_rate": 1.465687447369209e-06, + "loss": 1.021, + "step": 7964 + }, + { + "epoch": 0.5986021343754697, + "grad_norm": 1.7890817566214268, + "learning_rate": 1.4652183161607314e-06, + "loss": 0.8485, + "step": 7965 + }, + { + "epoch": 0.5986772884413046, + "grad_norm": 3.5147150515099437, + "learning_rate": 1.4647492166401159e-06, + "loss": 1.0061, + "step": 7966 + }, + { + "epoch": 0.5987524425071397, + "grad_norm": 1.8451669065452165, + "learning_rate": 1.4642801488351598e-06, + "loss": 0.9118, + "step": 7967 + }, + { + "epoch": 0.5988275965729746, + "grad_norm": 2.889766256258126, + "learning_rate": 1.4638111127736555e-06, + "loss": 0.9837, + "step": 7968 + }, + { + "epoch": 0.5989027506388096, + "grad_norm": 2.0326507433631162, + "learning_rate": 1.4633421084833965e-06, + "loss": 1.0579, + "step": 7969 + }, + { + "epoch": 0.5989779047046445, + "grad_norm": 1.8580434824040777, + "learning_rate": 1.4628731359921727e-06, + "loss": 0.9054, + "step": 7970 + }, + { + "epoch": 0.5990530587704794, + "grad_norm": 0.747683244973224, + "learning_rate": 1.462404195327772e-06, + "loss": 0.8675, + "step": 7971 + }, + { + "epoch": 0.5991282128363145, + "grad_norm": 1.599992978557439, + "learning_rate": 1.4619352865179814e-06, + "loss": 0.844, + "step": 7972 + }, + { + "epoch": 0.5992033669021494, + "grad_norm": 1.6126151938285522, + "learning_rate": 1.4614664095905856e-06, + "loss": 0.9765, + "step": 7973 + }, + { + "epoch": 0.5992785209679844, + "grad_norm": 1.4876971362617235, + "learning_rate": 1.460997564573367e-06, + "loss": 1.0094, + "step": 7974 + }, + { + "epoch": 0.5993536750338193, + "grad_norm": 2.0928154752908683, + "learning_rate": 1.4605287514941068e-06, + "loss": 1.0495, + "step": 7975 + }, + { + "epoch": 0.5994288290996543, + "grad_norm": 1.9375682941828922, + "learning_rate": 1.460059970380584e-06, + "loss": 1.0308, + "step": 7976 + }, + { + "epoch": 0.5995039831654893, + "grad_norm": 1.8041877880431199, + "learning_rate": 1.4595912212605755e-06, + "loss": 0.9464, + "step": 7977 + }, + { + "epoch": 0.5995791372313242, + "grad_norm": 1.8474958483245412, + "learning_rate": 1.459122504161856e-06, + "loss": 1.1106, + "step": 7978 + }, + { + "epoch": 0.5996542912971592, + "grad_norm": 4.87628352556827, + "learning_rate": 1.4586538191121999e-06, + "loss": 0.978, + "step": 7979 + }, + { + "epoch": 0.5997294453629941, + "grad_norm": 2.0495144664498604, + "learning_rate": 1.4581851661393776e-06, + "loss": 0.9508, + "step": 7980 + }, + { + "epoch": 0.5998045994288291, + "grad_norm": 1.7238691908117707, + "learning_rate": 1.4577165452711592e-06, + "loss": 1.0214, + "step": 7981 + }, + { + "epoch": 0.599879753494664, + "grad_norm": 1.6000396187168688, + "learning_rate": 1.4572479565353122e-06, + "loss": 0.9099, + "step": 7982 + }, + { + "epoch": 0.5999549075604991, + "grad_norm": 1.6935175521522154, + "learning_rate": 1.4567793999596014e-06, + "loss": 1.0206, + "step": 7983 + }, + { + "epoch": 0.600030061626334, + "grad_norm": 2.1805558201276076, + "learning_rate": 1.456310875571792e-06, + "loss": 0.9764, + "step": 7984 + }, + { + "epoch": 0.6001052156921689, + "grad_norm": 1.5467728089765191, + "learning_rate": 1.4558423833996443e-06, + "loss": 0.8794, + "step": 7985 + }, + { + "epoch": 0.6001803697580039, + "grad_norm": 1.6465642328229682, + "learning_rate": 1.4553739234709199e-06, + "loss": 0.9498, + "step": 7986 + }, + { + "epoch": 0.6002555238238388, + "grad_norm": 2.149475401818137, + "learning_rate": 1.4549054958133765e-06, + "loss": 0.9436, + "step": 7987 + }, + { + "epoch": 0.6003306778896739, + "grad_norm": 1.944564920810552, + "learning_rate": 1.4544371004547685e-06, + "loss": 0.8699, + "step": 7988 + }, + { + "epoch": 0.6004058319555088, + "grad_norm": 1.8337097937421873, + "learning_rate": 1.453968737422852e-06, + "loss": 0.925, + "step": 7989 + }, + { + "epoch": 0.6004809860213438, + "grad_norm": 1.8295636507332427, + "learning_rate": 1.4535004067453785e-06, + "loss": 1.049, + "step": 7990 + }, + { + "epoch": 0.6005561400871787, + "grad_norm": 1.5799259447295204, + "learning_rate": 1.453032108450099e-06, + "loss": 0.9111, + "step": 7991 + }, + { + "epoch": 0.6006312941530136, + "grad_norm": 1.4199130597162313, + "learning_rate": 1.4525638425647615e-06, + "loss": 0.9863, + "step": 7992 + }, + { + "epoch": 0.6007064482188487, + "grad_norm": 1.8123677961870779, + "learning_rate": 1.4520956091171121e-06, + "loss": 1.0391, + "step": 7993 + }, + { + "epoch": 0.6007816022846836, + "grad_norm": 1.5053596904469977, + "learning_rate": 1.4516274081348965e-06, + "loss": 0.9259, + "step": 7994 + }, + { + "epoch": 0.6008567563505186, + "grad_norm": 1.7264445683175422, + "learning_rate": 1.4511592396458565e-06, + "loss": 1.0611, + "step": 7995 + }, + { + "epoch": 0.6009319104163535, + "grad_norm": 1.942092804813703, + "learning_rate": 1.4506911036777335e-06, + "loss": 0.9851, + "step": 7996 + }, + { + "epoch": 0.6010070644821884, + "grad_norm": 1.6838011293075115, + "learning_rate": 1.4502230002582655e-06, + "loss": 1.0781, + "step": 7997 + }, + { + "epoch": 0.6010822185480235, + "grad_norm": 1.6823764303325885, + "learning_rate": 1.4497549294151905e-06, + "loss": 1.0095, + "step": 7998 + }, + { + "epoch": 0.6011573726138584, + "grad_norm": 1.9777751112978292, + "learning_rate": 1.4492868911762428e-06, + "loss": 0.9776, + "step": 7999 + }, + { + "epoch": 0.6012325266796934, + "grad_norm": 1.7545748585309329, + "learning_rate": 1.4488188855691555e-06, + "loss": 0.9408, + "step": 8000 + }, + { + "epoch": 0.6013076807455283, + "grad_norm": 8.92634024507796, + "learning_rate": 1.44835091262166e-06, + "loss": 0.9762, + "step": 8001 + }, + { + "epoch": 0.6013828348113633, + "grad_norm": 2.0582002846449727, + "learning_rate": 1.447882972361485e-06, + "loss": 1.0129, + "step": 8002 + }, + { + "epoch": 0.6014579888771983, + "grad_norm": 1.3898169495098398, + "learning_rate": 1.4474150648163588e-06, + "loss": 0.997, + "step": 8003 + }, + { + "epoch": 0.6015331429430332, + "grad_norm": 2.3178721737904318, + "learning_rate": 1.4469471900140056e-06, + "loss": 0.9743, + "step": 8004 + }, + { + "epoch": 0.6016082970088682, + "grad_norm": 1.5910979251756219, + "learning_rate": 1.4464793479821489e-06, + "loss": 0.9979, + "step": 8005 + }, + { + "epoch": 0.6016834510747031, + "grad_norm": 1.4997838977277436, + "learning_rate": 1.446011538748511e-06, + "loss": 0.9327, + "step": 8006 + }, + { + "epoch": 0.6017586051405381, + "grad_norm": 1.4791231104209177, + "learning_rate": 1.4455437623408097e-06, + "loss": 0.9472, + "step": 8007 + }, + { + "epoch": 0.601833759206373, + "grad_norm": 2.121428011976263, + "learning_rate": 1.4450760187867644e-06, + "loss": 0.9514, + "step": 8008 + }, + { + "epoch": 0.6019089132722081, + "grad_norm": 1.7664960214984144, + "learning_rate": 1.4446083081140904e-06, + "loss": 0.9417, + "step": 8009 + }, + { + "epoch": 0.601984067338043, + "grad_norm": 3.078213453232194, + "learning_rate": 1.4441406303504998e-06, + "loss": 0.9042, + "step": 8010 + }, + { + "epoch": 0.6020592214038779, + "grad_norm": 2.425312777139387, + "learning_rate": 1.4436729855237063e-06, + "loss": 0.8875, + "step": 8011 + }, + { + "epoch": 0.6021343754697129, + "grad_norm": 1.649942587807316, + "learning_rate": 1.443205373661418e-06, + "loss": 0.8725, + "step": 8012 + }, + { + "epoch": 0.6022095295355478, + "grad_norm": 0.8285541142395203, + "learning_rate": 1.442737794791344e-06, + "loss": 0.9253, + "step": 8013 + }, + { + "epoch": 0.6022846836013829, + "grad_norm": 2.0130903290419138, + "learning_rate": 1.4422702489411896e-06, + "loss": 0.894, + "step": 8014 + }, + { + "epoch": 0.6023598376672178, + "grad_norm": 1.4367556925551823, + "learning_rate": 1.441802736138658e-06, + "loss": 0.9523, + "step": 8015 + }, + { + "epoch": 0.6024349917330527, + "grad_norm": 2.324225315116778, + "learning_rate": 1.4413352564114525e-06, + "loss": 0.9126, + "step": 8016 + }, + { + "epoch": 0.6025101457988877, + "grad_norm": 0.7133136487039254, + "learning_rate": 1.4408678097872717e-06, + "loss": 0.8563, + "step": 8017 + }, + { + "epoch": 0.6025852998647226, + "grad_norm": 0.7299367834179615, + "learning_rate": 1.440400396293815e-06, + "loss": 0.8207, + "step": 8018 + }, + { + "epoch": 0.6026604539305577, + "grad_norm": 1.6754823075706224, + "learning_rate": 1.439933015958777e-06, + "loss": 0.9607, + "step": 8019 + }, + { + "epoch": 0.6027356079963926, + "grad_norm": 1.6288654862021306, + "learning_rate": 1.4394656688098526e-06, + "loss": 1.0029, + "step": 8020 + }, + { + "epoch": 0.6028107620622276, + "grad_norm": 1.6312807920626327, + "learning_rate": 1.4389983548747337e-06, + "loss": 1.0721, + "step": 8021 + }, + { + "epoch": 0.6028859161280625, + "grad_norm": 1.635232182036946, + "learning_rate": 1.4385310741811106e-06, + "loss": 0.9221, + "step": 8022 + }, + { + "epoch": 0.6029610701938974, + "grad_norm": 2.207553065905289, + "learning_rate": 1.4380638267566716e-06, + "loss": 0.9714, + "step": 8023 + }, + { + "epoch": 0.6030362242597325, + "grad_norm": 1.7454081646580215, + "learning_rate": 1.4375966126291022e-06, + "loss": 1.0221, + "step": 8024 + }, + { + "epoch": 0.6031113783255674, + "grad_norm": 1.6279087442882738, + "learning_rate": 1.4371294318260874e-06, + "loss": 0.9635, + "step": 8025 + }, + { + "epoch": 0.6031865323914024, + "grad_norm": 2.297502808179997, + "learning_rate": 1.4366622843753092e-06, + "loss": 1.002, + "step": 8026 + }, + { + "epoch": 0.6032616864572373, + "grad_norm": 1.4697287524641076, + "learning_rate": 1.4361951703044475e-06, + "loss": 0.9893, + "step": 8027 + }, + { + "epoch": 0.6033368405230723, + "grad_norm": 2.342360020773695, + "learning_rate": 1.4357280896411813e-06, + "loss": 0.9965, + "step": 8028 + }, + { + "epoch": 0.6034119945889073, + "grad_norm": 3.183064235344704, + "learning_rate": 1.4352610424131854e-06, + "loss": 1.0141, + "step": 8029 + }, + { + "epoch": 0.6034871486547422, + "grad_norm": 2.072226506905761, + "learning_rate": 1.4347940286481364e-06, + "loss": 0.8635, + "step": 8030 + }, + { + "epoch": 0.6035623027205772, + "grad_norm": 1.5723092967537282, + "learning_rate": 1.434327048373706e-06, + "loss": 1.0204, + "step": 8031 + }, + { + "epoch": 0.6036374567864121, + "grad_norm": 1.5685107014430542, + "learning_rate": 1.4338601016175624e-06, + "loss": 0.8828, + "step": 8032 + }, + { + "epoch": 0.6037126108522471, + "grad_norm": 2.0255272873448447, + "learning_rate": 1.4333931884073769e-06, + "loss": 0.9421, + "step": 8033 + }, + { + "epoch": 0.6037877649180821, + "grad_norm": 1.2634114818172641, + "learning_rate": 1.4329263087708144e-06, + "loss": 1.0278, + "step": 8034 + }, + { + "epoch": 0.6038629189839171, + "grad_norm": 1.7675690202847347, + "learning_rate": 1.4324594627355397e-06, + "loss": 0.9249, + "step": 8035 + }, + { + "epoch": 0.603938073049752, + "grad_norm": 1.6300581922850472, + "learning_rate": 1.431992650329215e-06, + "loss": 1.0092, + "step": 8036 + }, + { + "epoch": 0.6040132271155869, + "grad_norm": 2.054189643447644, + "learning_rate": 1.4315258715795007e-06, + "loss": 1.1011, + "step": 8037 + }, + { + "epoch": 0.6040883811814219, + "grad_norm": 1.4319600039176459, + "learning_rate": 1.4310591265140555e-06, + "loss": 0.9841, + "step": 8038 + }, + { + "epoch": 0.6041635352472569, + "grad_norm": 1.4974620329003427, + "learning_rate": 1.4305924151605354e-06, + "loss": 0.941, + "step": 8039 + }, + { + "epoch": 0.6042386893130919, + "grad_norm": 2.412982785634222, + "learning_rate": 1.4301257375465956e-06, + "loss": 1.1101, + "step": 8040 + }, + { + "epoch": 0.6043138433789268, + "grad_norm": 1.795723255734216, + "learning_rate": 1.4296590936998874e-06, + "loss": 0.9326, + "step": 8041 + }, + { + "epoch": 0.6043889974447617, + "grad_norm": 1.7077547892516045, + "learning_rate": 1.4291924836480625e-06, + "loss": 0.9094, + "step": 8042 + }, + { + "epoch": 0.6044641515105967, + "grad_norm": 1.7039790680589169, + "learning_rate": 1.4287259074187685e-06, + "loss": 0.9284, + "step": 8043 + }, + { + "epoch": 0.6045393055764317, + "grad_norm": 1.8658720354521892, + "learning_rate": 1.428259365039652e-06, + "loss": 0.9629, + "step": 8044 + }, + { + "epoch": 0.6046144596422667, + "grad_norm": 3.2055909131893556, + "learning_rate": 1.4277928565383577e-06, + "loss": 0.8721, + "step": 8045 + }, + { + "epoch": 0.6046896137081016, + "grad_norm": 1.4390316142890043, + "learning_rate": 1.4273263819425272e-06, + "loss": 1.0087, + "step": 8046 + }, + { + "epoch": 0.6047647677739366, + "grad_norm": 1.6043284280970014, + "learning_rate": 1.426859941279802e-06, + "loss": 0.9237, + "step": 8047 + }, + { + "epoch": 0.6048399218397715, + "grad_norm": 1.6804274237821522, + "learning_rate": 1.42639353457782e-06, + "loss": 0.9187, + "step": 8048 + }, + { + "epoch": 0.6049150759056064, + "grad_norm": 1.8436574104466172, + "learning_rate": 1.4259271618642166e-06, + "loss": 0.9894, + "step": 8049 + }, + { + "epoch": 0.6049902299714415, + "grad_norm": 1.341806470012091, + "learning_rate": 1.4254608231666286e-06, + "loss": 0.9311, + "step": 8050 + }, + { + "epoch": 0.6050653840372764, + "grad_norm": 1.3156598972213736, + "learning_rate": 1.4249945185126855e-06, + "loss": 0.9744, + "step": 8051 + }, + { + "epoch": 0.6051405381031114, + "grad_norm": 2.0155605335424136, + "learning_rate": 1.4245282479300199e-06, + "loss": 0.9535, + "step": 8052 + }, + { + "epoch": 0.6052156921689463, + "grad_norm": 1.3537262371510208, + "learning_rate": 1.424062011446259e-06, + "loss": 1.024, + "step": 8053 + }, + { + "epoch": 0.6052908462347814, + "grad_norm": 3.7738306065859115, + "learning_rate": 1.4235958090890293e-06, + "loss": 0.8931, + "step": 8054 + }, + { + "epoch": 0.6053660003006163, + "grad_norm": 2.0295109273386562, + "learning_rate": 1.4231296408859553e-06, + "loss": 1.0694, + "step": 8055 + }, + { + "epoch": 0.6054411543664512, + "grad_norm": 1.5856223709391388, + "learning_rate": 1.4226635068646586e-06, + "loss": 0.9912, + "step": 8056 + }, + { + "epoch": 0.6055163084322862, + "grad_norm": 1.9113846088158168, + "learning_rate": 1.4221974070527606e-06, + "loss": 0.9167, + "step": 8057 + }, + { + "epoch": 0.6055914624981211, + "grad_norm": 1.4766627686379494, + "learning_rate": 1.4217313414778786e-06, + "loss": 0.9488, + "step": 8058 + }, + { + "epoch": 0.6056666165639562, + "grad_norm": 1.623313101590851, + "learning_rate": 1.4212653101676285e-06, + "loss": 0.9227, + "step": 8059 + }, + { + "epoch": 0.6057417706297911, + "grad_norm": 2.500854822880227, + "learning_rate": 1.4207993131496254e-06, + "loss": 0.9248, + "step": 8060 + }, + { + "epoch": 0.605816924695626, + "grad_norm": 1.4363234307464254, + "learning_rate": 1.4203333504514805e-06, + "loss": 0.952, + "step": 8061 + }, + { + "epoch": 0.605892078761461, + "grad_norm": 1.358982204820392, + "learning_rate": 1.4198674221008045e-06, + "loss": 0.9485, + "step": 8062 + }, + { + "epoch": 0.6059672328272959, + "grad_norm": 1.4007549583405854, + "learning_rate": 1.419401528125205e-06, + "loss": 0.9039, + "step": 8063 + }, + { + "epoch": 0.606042386893131, + "grad_norm": 1.5120520409150688, + "learning_rate": 1.4189356685522884e-06, + "loss": 1.0702, + "step": 8064 + }, + { + "epoch": 0.6061175409589659, + "grad_norm": 1.3765040321360325, + "learning_rate": 1.4184698434096586e-06, + "loss": 0.8575, + "step": 8065 + }, + { + "epoch": 0.6061926950248009, + "grad_norm": 1.9131058774165026, + "learning_rate": 1.4180040527249172e-06, + "loss": 1.0395, + "step": 8066 + }, + { + "epoch": 0.6062678490906358, + "grad_norm": 1.5018124756357225, + "learning_rate": 1.4175382965256644e-06, + "loss": 1.0409, + "step": 8067 + }, + { + "epoch": 0.6063430031564707, + "grad_norm": 2.9488636737822818, + "learning_rate": 1.4170725748394977e-06, + "loss": 0.9061, + "step": 8068 + }, + { + "epoch": 0.6064181572223057, + "grad_norm": 1.9812879203753535, + "learning_rate": 1.4166068876940135e-06, + "loss": 1.0309, + "step": 8069 + }, + { + "epoch": 0.6064933112881407, + "grad_norm": 1.763264115723131, + "learning_rate": 1.4161412351168053e-06, + "loss": 0.8239, + "step": 8070 + }, + { + "epoch": 0.6065684653539757, + "grad_norm": 1.9670536665351606, + "learning_rate": 1.4156756171354637e-06, + "loss": 0.9948, + "step": 8071 + }, + { + "epoch": 0.6066436194198106, + "grad_norm": 2.0045098479503127, + "learning_rate": 1.4152100337775804e-06, + "loss": 1.094, + "step": 8072 + }, + { + "epoch": 0.6067187734856456, + "grad_norm": 1.5228652270492813, + "learning_rate": 1.414744485070741e-06, + "loss": 1.0603, + "step": 8073 + }, + { + "epoch": 0.6067939275514805, + "grad_norm": 1.4633546304709852, + "learning_rate": 1.4142789710425325e-06, + "loss": 1.0722, + "step": 8074 + }, + { + "epoch": 0.6068690816173155, + "grad_norm": 1.5269119159547064, + "learning_rate": 1.4138134917205377e-06, + "loss": 0.9404, + "step": 8075 + }, + { + "epoch": 0.6069442356831505, + "grad_norm": 1.3419451458119613, + "learning_rate": 1.413348047132338e-06, + "loss": 0.967, + "step": 8076 + }, + { + "epoch": 0.6070193897489854, + "grad_norm": 2.0314112672611055, + "learning_rate": 1.4128826373055134e-06, + "loss": 0.9744, + "step": 8077 + }, + { + "epoch": 0.6070945438148204, + "grad_norm": 2.1552859376407447, + "learning_rate": 1.4124172622676406e-06, + "loss": 0.9973, + "step": 8078 + }, + { + "epoch": 0.6071696978806553, + "grad_norm": 1.5520388022910827, + "learning_rate": 1.411951922046295e-06, + "loss": 1.1076, + "step": 8079 + }, + { + "epoch": 0.6072448519464904, + "grad_norm": 1.6348873261757708, + "learning_rate": 1.4114866166690498e-06, + "loss": 0.9822, + "step": 8080 + }, + { + "epoch": 0.6073200060123253, + "grad_norm": 1.55486116445743, + "learning_rate": 1.411021346163476e-06, + "loss": 0.8942, + "step": 8081 + }, + { + "epoch": 0.6073951600781602, + "grad_norm": 2.2060633334547264, + "learning_rate": 1.4105561105571428e-06, + "loss": 1.0335, + "step": 8082 + }, + { + "epoch": 0.6074703141439952, + "grad_norm": 1.48551046209975, + "learning_rate": 1.410090909877617e-06, + "loss": 0.9333, + "step": 8083 + }, + { + "epoch": 0.6075454682098301, + "grad_norm": 1.8765392343316032, + "learning_rate": 1.4096257441524643e-06, + "loss": 0.9459, + "step": 8084 + }, + { + "epoch": 0.6076206222756652, + "grad_norm": 1.8608843023093673, + "learning_rate": 1.4091606134092465e-06, + "loss": 0.9485, + "step": 8085 + }, + { + "epoch": 0.6076957763415001, + "grad_norm": 1.755311195083739, + "learning_rate": 1.4086955176755248e-06, + "loss": 0.8908, + "step": 8086 + }, + { + "epoch": 0.607770930407335, + "grad_norm": 1.7358493209129473, + "learning_rate": 1.4082304569788582e-06, + "loss": 0.9438, + "step": 8087 + }, + { + "epoch": 0.60784608447317, + "grad_norm": 8.748855079949658, + "learning_rate": 1.407765431346803e-06, + "loss": 0.9665, + "step": 8088 + }, + { + "epoch": 0.6079212385390049, + "grad_norm": 0.7939385933492009, + "learning_rate": 1.4073004408069138e-06, + "loss": 0.9489, + "step": 8089 + }, + { + "epoch": 0.60799639260484, + "grad_norm": 1.7457040100027126, + "learning_rate": 1.4068354853867429e-06, + "loss": 1.0265, + "step": 8090 + }, + { + "epoch": 0.6080715466706749, + "grad_norm": 1.4943654408581921, + "learning_rate": 1.406370565113841e-06, + "loss": 0.9696, + "step": 8091 + }, + { + "epoch": 0.6081467007365099, + "grad_norm": 1.981262048149179, + "learning_rate": 1.4059056800157567e-06, + "loss": 1.0715, + "step": 8092 + }, + { + "epoch": 0.6082218548023448, + "grad_norm": 1.4624938834298844, + "learning_rate": 1.4054408301200345e-06, + "loss": 1.0222, + "step": 8093 + }, + { + "epoch": 0.6082970088681797, + "grad_norm": 2.1179496266153564, + "learning_rate": 1.4049760154542214e-06, + "loss": 0.9724, + "step": 8094 + }, + { + "epoch": 0.6083721629340147, + "grad_norm": 1.5002956099096798, + "learning_rate": 1.4045112360458564e-06, + "loss": 0.9751, + "step": 8095 + }, + { + "epoch": 0.6084473169998497, + "grad_norm": 1.5137164496965034, + "learning_rate": 1.404046491922482e-06, + "loss": 0.9368, + "step": 8096 + }, + { + "epoch": 0.6085224710656847, + "grad_norm": 1.5365477799926026, + "learning_rate": 1.403581783111635e-06, + "loss": 1.0352, + "step": 8097 + }, + { + "epoch": 0.6085976251315196, + "grad_norm": 1.619987091672356, + "learning_rate": 1.4031171096408506e-06, + "loss": 0.9683, + "step": 8098 + }, + { + "epoch": 0.6086727791973546, + "grad_norm": 1.6551705207450773, + "learning_rate": 1.4026524715376637e-06, + "loss": 0.9499, + "step": 8099 + }, + { + "epoch": 0.6087479332631895, + "grad_norm": 1.7784506903443935, + "learning_rate": 1.4021878688296047e-06, + "loss": 0.9989, + "step": 8100 + }, + { + "epoch": 0.6088230873290245, + "grad_norm": 1.6707417577950923, + "learning_rate": 1.401723301544204e-06, + "loss": 1.0013, + "step": 8101 + }, + { + "epoch": 0.6088982413948595, + "grad_norm": 5.015476497935443, + "learning_rate": 1.4012587697089885e-06, + "loss": 1.0177, + "step": 8102 + }, + { + "epoch": 0.6089733954606944, + "grad_norm": 1.4644700044039236, + "learning_rate": 1.4007942733514836e-06, + "loss": 0.9112, + "step": 8103 + }, + { + "epoch": 0.6090485495265294, + "grad_norm": 2.656401775736216, + "learning_rate": 1.400329812499213e-06, + "loss": 1.0817, + "step": 8104 + }, + { + "epoch": 0.6091237035923643, + "grad_norm": 1.7788238734160524, + "learning_rate": 1.3998653871796964e-06, + "loss": 1.0822, + "step": 8105 + }, + { + "epoch": 0.6091988576581993, + "grad_norm": 2.4375106615664768, + "learning_rate": 1.3994009974204547e-06, + "loss": 0.9459, + "step": 8106 + }, + { + "epoch": 0.6092740117240343, + "grad_norm": 2.09055277837931, + "learning_rate": 1.3989366432490028e-06, + "loss": 1.0247, + "step": 8107 + }, + { + "epoch": 0.6093491657898692, + "grad_norm": 1.7168313848609533, + "learning_rate": 1.3984723246928569e-06, + "loss": 1.0148, + "step": 8108 + }, + { + "epoch": 0.6094243198557042, + "grad_norm": 1.6996831149791465, + "learning_rate": 1.3980080417795296e-06, + "loss": 0.9535, + "step": 8109 + }, + { + "epoch": 0.6094994739215391, + "grad_norm": 1.6218816474087416, + "learning_rate": 1.39754379453653e-06, + "loss": 1.0642, + "step": 8110 + }, + { + "epoch": 0.6095746279873742, + "grad_norm": 1.8790643510503262, + "learning_rate": 1.3970795829913682e-06, + "loss": 0.9886, + "step": 8111 + }, + { + "epoch": 0.6096497820532091, + "grad_norm": 2.383190787883214, + "learning_rate": 1.396615407171549e-06, + "loss": 1.0124, + "step": 8112 + }, + { + "epoch": 0.609724936119044, + "grad_norm": 0.6979672831120052, + "learning_rate": 1.3961512671045787e-06, + "loss": 0.8517, + "step": 8113 + }, + { + "epoch": 0.609800090184879, + "grad_norm": 1.5109139599391452, + "learning_rate": 1.3956871628179577e-06, + "loss": 1.0572, + "step": 8114 + }, + { + "epoch": 0.6098752442507139, + "grad_norm": 1.5224219434049768, + "learning_rate": 1.3952230943391856e-06, + "loss": 1.0053, + "step": 8115 + }, + { + "epoch": 0.609950398316549, + "grad_norm": 1.6766642487181767, + "learning_rate": 1.3947590616957618e-06, + "loss": 0.8482, + "step": 8116 + }, + { + "epoch": 0.6100255523823839, + "grad_norm": 1.563263210819755, + "learning_rate": 1.3942950649151808e-06, + "loss": 1.0063, + "step": 8117 + }, + { + "epoch": 0.6101007064482189, + "grad_norm": 2.0050024414047636, + "learning_rate": 1.3938311040249371e-06, + "loss": 0.9912, + "step": 8118 + }, + { + "epoch": 0.6101758605140538, + "grad_norm": 1.7170805018709505, + "learning_rate": 1.3933671790525215e-06, + "loss": 1.0157, + "step": 8119 + }, + { + "epoch": 0.6102510145798887, + "grad_norm": 1.2918803540743098, + "learning_rate": 1.3929032900254232e-06, + "loss": 1.0025, + "step": 8120 + }, + { + "epoch": 0.6103261686457238, + "grad_norm": 1.8079334479042035, + "learning_rate": 1.39243943697113e-06, + "loss": 0.9852, + "step": 8121 + }, + { + "epoch": 0.6104013227115587, + "grad_norm": 1.6514782814632767, + "learning_rate": 1.3919756199171263e-06, + "loss": 0.9957, + "step": 8122 + }, + { + "epoch": 0.6104764767773937, + "grad_norm": 1.9072845653302426, + "learning_rate": 1.3915118388908958e-06, + "loss": 1.0128, + "step": 8123 + }, + { + "epoch": 0.6105516308432286, + "grad_norm": 0.6928953310361184, + "learning_rate": 1.3910480939199184e-06, + "loss": 0.8337, + "step": 8124 + }, + { + "epoch": 0.6106267849090636, + "grad_norm": 1.7153297378694106, + "learning_rate": 1.3905843850316738e-06, + "loss": 0.8413, + "step": 8125 + }, + { + "epoch": 0.6107019389748986, + "grad_norm": 2.2701588483511075, + "learning_rate": 1.3901207122536383e-06, + "loss": 1.0205, + "step": 8126 + }, + { + "epoch": 0.6107770930407335, + "grad_norm": 1.5471647357286884, + "learning_rate": 1.3896570756132851e-06, + "loss": 0.9716, + "step": 8127 + }, + { + "epoch": 0.6108522471065685, + "grad_norm": 1.9308009468142675, + "learning_rate": 1.3891934751380879e-06, + "loss": 0.9634, + "step": 8128 + }, + { + "epoch": 0.6109274011724034, + "grad_norm": 1.4893369759953643, + "learning_rate": 1.3887299108555158e-06, + "loss": 0.9848, + "step": 8129 + }, + { + "epoch": 0.6110025552382384, + "grad_norm": 1.4542246933575127, + "learning_rate": 1.3882663827930375e-06, + "loss": 1.0447, + "step": 8130 + }, + { + "epoch": 0.6110777093040733, + "grad_norm": 1.354528359258139, + "learning_rate": 1.3878028909781187e-06, + "loss": 0.9489, + "step": 8131 + }, + { + "epoch": 0.6111528633699083, + "grad_norm": 2.668128348138656, + "learning_rate": 1.3873394354382225e-06, + "loss": 0.9894, + "step": 8132 + }, + { + "epoch": 0.6112280174357433, + "grad_norm": 2.0700888458940643, + "learning_rate": 1.3868760162008108e-06, + "loss": 0.9708, + "step": 8133 + }, + { + "epoch": 0.6113031715015782, + "grad_norm": 1.4642979126265965, + "learning_rate": 1.3864126332933425e-06, + "loss": 0.9677, + "step": 8134 + }, + { + "epoch": 0.6113783255674132, + "grad_norm": 3.203965090862695, + "learning_rate": 1.3859492867432765e-06, + "loss": 0.9992, + "step": 8135 + }, + { + "epoch": 0.6114534796332481, + "grad_norm": 1.983324929494711, + "learning_rate": 1.385485976578066e-06, + "loss": 0.9426, + "step": 8136 + }, + { + "epoch": 0.6115286336990832, + "grad_norm": 1.4949727054005295, + "learning_rate": 1.3850227028251639e-06, + "loss": 1.0097, + "step": 8137 + }, + { + "epoch": 0.6116037877649181, + "grad_norm": 1.6812109137887656, + "learning_rate": 1.3845594655120224e-06, + "loss": 0.9749, + "step": 8138 + }, + { + "epoch": 0.611678941830753, + "grad_norm": 1.8429754537644012, + "learning_rate": 1.3840962646660885e-06, + "loss": 0.889, + "step": 8139 + }, + { + "epoch": 0.611754095896588, + "grad_norm": 1.6701713509509961, + "learning_rate": 1.3836331003148101e-06, + "loss": 1.0246, + "step": 8140 + }, + { + "epoch": 0.6118292499624229, + "grad_norm": 2.195773147636465, + "learning_rate": 1.3831699724856307e-06, + "loss": 0.9453, + "step": 8141 + }, + { + "epoch": 0.611904404028258, + "grad_norm": 1.6874992935744326, + "learning_rate": 1.3827068812059918e-06, + "loss": 0.9034, + "step": 8142 + }, + { + "epoch": 0.6119795580940929, + "grad_norm": 1.3911216791822683, + "learning_rate": 1.3822438265033345e-06, + "loss": 0.9723, + "step": 8143 + }, + { + "epoch": 0.6120547121599279, + "grad_norm": 2.7256862686153394, + "learning_rate": 1.3817808084050957e-06, + "loss": 0.8987, + "step": 8144 + }, + { + "epoch": 0.6121298662257628, + "grad_norm": 1.4834543685966535, + "learning_rate": 1.3813178269387119e-06, + "loss": 1.0116, + "step": 8145 + }, + { + "epoch": 0.6122050202915977, + "grad_norm": 2.154392382112018, + "learning_rate": 1.380854882131615e-06, + "loss": 0.936, + "step": 8146 + }, + { + "epoch": 0.6122801743574328, + "grad_norm": 2.226886069896334, + "learning_rate": 1.3803919740112383e-06, + "loss": 0.8675, + "step": 8147 + }, + { + "epoch": 0.6123553284232677, + "grad_norm": 2.040084290539512, + "learning_rate": 1.379929102605009e-06, + "loss": 1.0466, + "step": 8148 + }, + { + "epoch": 0.6124304824891027, + "grad_norm": 1.5400783845843895, + "learning_rate": 1.379466267940355e-06, + "loss": 0.9108, + "step": 8149 + }, + { + "epoch": 0.6125056365549376, + "grad_norm": 2.477682251602248, + "learning_rate": 1.3790034700447008e-06, + "loss": 0.9659, + "step": 8150 + }, + { + "epoch": 0.6125807906207725, + "grad_norm": 1.4023409447602078, + "learning_rate": 1.378540708945469e-06, + "loss": 1.0548, + "step": 8151 + }, + { + "epoch": 0.6126559446866076, + "grad_norm": 1.669491218226192, + "learning_rate": 1.3780779846700799e-06, + "loss": 1.0157, + "step": 8152 + }, + { + "epoch": 0.6127310987524425, + "grad_norm": 1.6475903833861107, + "learning_rate": 1.3776152972459517e-06, + "loss": 0.9028, + "step": 8153 + }, + { + "epoch": 0.6128062528182775, + "grad_norm": 1.6369477978948617, + "learning_rate": 1.3771526467005004e-06, + "loss": 0.9633, + "step": 8154 + }, + { + "epoch": 0.6128814068841124, + "grad_norm": 1.407060897997714, + "learning_rate": 1.37669003306114e-06, + "loss": 1.0573, + "step": 8155 + }, + { + "epoch": 0.6129565609499474, + "grad_norm": 1.9314401057726045, + "learning_rate": 1.3762274563552811e-06, + "loss": 0.9776, + "step": 8156 + }, + { + "epoch": 0.6130317150157824, + "grad_norm": 1.6494676887282294, + "learning_rate": 1.375764916610335e-06, + "loss": 0.9215, + "step": 8157 + }, + { + "epoch": 0.6131068690816173, + "grad_norm": 0.7276401526349989, + "learning_rate": 1.3753024138537082e-06, + "loss": 0.8227, + "step": 8158 + }, + { + "epoch": 0.6131820231474523, + "grad_norm": 1.823942829315271, + "learning_rate": 1.3748399481128043e-06, + "loss": 0.9233, + "step": 8159 + }, + { + "epoch": 0.6132571772132872, + "grad_norm": 1.6749147251114798, + "learning_rate": 1.3743775194150281e-06, + "loss": 1.0432, + "step": 8160 + }, + { + "epoch": 0.6133323312791222, + "grad_norm": 1.6875415373034561, + "learning_rate": 1.3739151277877792e-06, + "loss": 0.9448, + "step": 8161 + }, + { + "epoch": 0.6134074853449571, + "grad_norm": 1.8607164560738896, + "learning_rate": 1.3734527732584568e-06, + "loss": 0.9892, + "step": 8162 + }, + { + "epoch": 0.6134826394107922, + "grad_norm": 1.4439830542386378, + "learning_rate": 1.372990455854457e-06, + "loss": 0.9559, + "step": 8163 + }, + { + "epoch": 0.6135577934766271, + "grad_norm": 1.5056597110964005, + "learning_rate": 1.372528175603173e-06, + "loss": 0.9452, + "step": 8164 + }, + { + "epoch": 0.613632947542462, + "grad_norm": 1.9143879516442415, + "learning_rate": 1.372065932531998e-06, + "loss": 0.9584, + "step": 8165 + }, + { + "epoch": 0.613708101608297, + "grad_norm": 2.343901464018606, + "learning_rate": 1.3716037266683203e-06, + "loss": 0.9628, + "step": 8166 + }, + { + "epoch": 0.613783255674132, + "grad_norm": 1.7706978091571017, + "learning_rate": 1.3711415580395288e-06, + "loss": 0.8713, + "step": 8167 + }, + { + "epoch": 0.613858409739967, + "grad_norm": 1.614994792398377, + "learning_rate": 1.3706794266730072e-06, + "loss": 1.0177, + "step": 8168 + }, + { + "epoch": 0.6139335638058019, + "grad_norm": 2.3041257011906287, + "learning_rate": 1.37021733259614e-06, + "loss": 0.9091, + "step": 8169 + }, + { + "epoch": 0.6140087178716369, + "grad_norm": 1.9667196552228765, + "learning_rate": 1.3697552758363079e-06, + "loss": 0.8089, + "step": 8170 + }, + { + "epoch": 0.6140838719374718, + "grad_norm": 1.890814464312985, + "learning_rate": 1.3692932564208884e-06, + "loss": 0.9784, + "step": 8171 + }, + { + "epoch": 0.6141590260033067, + "grad_norm": 3.907218629904782, + "learning_rate": 1.3688312743772588e-06, + "loss": 0.9849, + "step": 8172 + }, + { + "epoch": 0.6142341800691418, + "grad_norm": 1.4425618984488406, + "learning_rate": 1.3683693297327927e-06, + "loss": 0.9987, + "step": 8173 + }, + { + "epoch": 0.6143093341349767, + "grad_norm": 2.5933973693132977, + "learning_rate": 1.367907422514863e-06, + "loss": 0.9497, + "step": 8174 + }, + { + "epoch": 0.6143844882008117, + "grad_norm": 1.5677635799909282, + "learning_rate": 1.367445552750839e-06, + "loss": 0.9968, + "step": 8175 + }, + { + "epoch": 0.6144596422666466, + "grad_norm": 1.714664736638879, + "learning_rate": 1.3669837204680876e-06, + "loss": 0.856, + "step": 8176 + }, + { + "epoch": 0.6145347963324815, + "grad_norm": 1.736339933786475, + "learning_rate": 1.3665219256939753e-06, + "loss": 0.9949, + "step": 8177 + }, + { + "epoch": 0.6146099503983166, + "grad_norm": 1.396655945808822, + "learning_rate": 1.3660601684558639e-06, + "loss": 0.8573, + "step": 8178 + }, + { + "epoch": 0.6146851044641515, + "grad_norm": 1.5782482741147204, + "learning_rate": 1.3655984487811158e-06, + "loss": 1.0102, + "step": 8179 + }, + { + "epoch": 0.6147602585299865, + "grad_norm": 1.7338280030011883, + "learning_rate": 1.3651367666970895e-06, + "loss": 1.0478, + "step": 8180 + }, + { + "epoch": 0.6148354125958214, + "grad_norm": 1.4955116514047306, + "learning_rate": 1.3646751222311392e-06, + "loss": 0.926, + "step": 8181 + }, + { + "epoch": 0.6149105666616564, + "grad_norm": 1.4833191178029994, + "learning_rate": 1.3642135154106219e-06, + "loss": 0.9852, + "step": 8182 + }, + { + "epoch": 0.6149857207274914, + "grad_norm": 0.6461036921224003, + "learning_rate": 1.3637519462628876e-06, + "loss": 0.7817, + "step": 8183 + }, + { + "epoch": 0.6150608747933263, + "grad_norm": 1.6495132190629092, + "learning_rate": 1.3632904148152877e-06, + "loss": 0.9383, + "step": 8184 + }, + { + "epoch": 0.6151360288591613, + "grad_norm": 1.6976471045327177, + "learning_rate": 1.3628289210951687e-06, + "loss": 0.9948, + "step": 8185 + }, + { + "epoch": 0.6152111829249962, + "grad_norm": 1.52859055929856, + "learning_rate": 1.3623674651298752e-06, + "loss": 0.8594, + "step": 8186 + }, + { + "epoch": 0.6152863369908312, + "grad_norm": 2.6837882203293106, + "learning_rate": 1.361906046946752e-06, + "loss": 0.9987, + "step": 8187 + }, + { + "epoch": 0.6153614910566662, + "grad_norm": 1.5846626156113364, + "learning_rate": 1.3614446665731385e-06, + "loss": 0.858, + "step": 8188 + }, + { + "epoch": 0.6154366451225012, + "grad_norm": 1.5160396048171114, + "learning_rate": 1.3609833240363738e-06, + "loss": 0.9847, + "step": 8189 + }, + { + "epoch": 0.6155117991883361, + "grad_norm": 2.676885268507631, + "learning_rate": 1.3605220193637942e-06, + "loss": 0.9181, + "step": 8190 + }, + { + "epoch": 0.615586953254171, + "grad_norm": 2.5073882602732964, + "learning_rate": 1.3600607525827335e-06, + "loss": 0.9598, + "step": 8191 + }, + { + "epoch": 0.615662107320006, + "grad_norm": 0.8135098637133524, + "learning_rate": 1.359599523720524e-06, + "loss": 0.9127, + "step": 8192 + }, + { + "epoch": 0.615737261385841, + "grad_norm": 1.4786173736352373, + "learning_rate": 1.3591383328044943e-06, + "loss": 0.9057, + "step": 8193 + }, + { + "epoch": 0.615812415451676, + "grad_norm": 1.4667400648795865, + "learning_rate": 1.358677179861973e-06, + "loss": 1.0969, + "step": 8194 + }, + { + "epoch": 0.6158875695175109, + "grad_norm": 1.8787362067350142, + "learning_rate": 1.3582160649202844e-06, + "loss": 1.078, + "step": 8195 + }, + { + "epoch": 0.6159627235833458, + "grad_norm": 2.1356428321225636, + "learning_rate": 1.3577549880067516e-06, + "loss": 1.0294, + "step": 8196 + }, + { + "epoch": 0.6160378776491808, + "grad_norm": 1.9798940693811018, + "learning_rate": 1.3572939491486952e-06, + "loss": 1.0411, + "step": 8197 + }, + { + "epoch": 0.6161130317150157, + "grad_norm": 2.325390432246601, + "learning_rate": 1.3568329483734329e-06, + "loss": 1.029, + "step": 8198 + }, + { + "epoch": 0.6161881857808508, + "grad_norm": 2.2511098032013757, + "learning_rate": 1.3563719857082817e-06, + "loss": 0.9785, + "step": 8199 + }, + { + "epoch": 0.6162633398466857, + "grad_norm": 1.820338040283205, + "learning_rate": 1.3559110611805542e-06, + "loss": 1.007, + "step": 8200 + }, + { + "epoch": 0.6163384939125207, + "grad_norm": 1.9262389592789413, + "learning_rate": 1.3554501748175637e-06, + "loss": 0.9337, + "step": 8201 + }, + { + "epoch": 0.6164136479783556, + "grad_norm": 2.3933285653466183, + "learning_rate": 1.3549893266466188e-06, + "loss": 0.9725, + "step": 8202 + }, + { + "epoch": 0.6164888020441905, + "grad_norm": 1.9685056625606454, + "learning_rate": 1.3545285166950246e-06, + "loss": 0.9484, + "step": 8203 + }, + { + "epoch": 0.6165639561100256, + "grad_norm": 5.410366012968726, + "learning_rate": 1.3540677449900887e-06, + "loss": 0.9971, + "step": 8204 + }, + { + "epoch": 0.6166391101758605, + "grad_norm": 2.0595631352856216, + "learning_rate": 1.3536070115591118e-06, + "loss": 0.9547, + "step": 8205 + }, + { + "epoch": 0.6167142642416955, + "grad_norm": 2.029567197504382, + "learning_rate": 1.3531463164293952e-06, + "loss": 1.0753, + "step": 8206 + }, + { + "epoch": 0.6167894183075304, + "grad_norm": 1.4772833623703363, + "learning_rate": 1.352685659628236e-06, + "loss": 0.8824, + "step": 8207 + }, + { + "epoch": 0.6168645723733654, + "grad_norm": 1.438383742974462, + "learning_rate": 1.3522250411829301e-06, + "loss": 0.984, + "step": 8208 + }, + { + "epoch": 0.6169397264392004, + "grad_norm": 1.8802439791046992, + "learning_rate": 1.3517644611207715e-06, + "loss": 1.0003, + "step": 8209 + }, + { + "epoch": 0.6170148805050353, + "grad_norm": 2.5598475987549176, + "learning_rate": 1.35130391946905e-06, + "loss": 0.9449, + "step": 8210 + }, + { + "epoch": 0.6170900345708703, + "grad_norm": 1.4626350536100343, + "learning_rate": 1.350843416255056e-06, + "loss": 1.0313, + "step": 8211 + }, + { + "epoch": 0.6171651886367052, + "grad_norm": 1.8360938472324138, + "learning_rate": 1.350382951506075e-06, + "loss": 0.928, + "step": 8212 + }, + { + "epoch": 0.6172403427025402, + "grad_norm": 0.7751745504269397, + "learning_rate": 1.3499225252493918e-06, + "loss": 0.8702, + "step": 8213 + }, + { + "epoch": 0.6173154967683752, + "grad_norm": 1.4928925131950992, + "learning_rate": 1.3494621375122886e-06, + "loss": 1.0841, + "step": 8214 + }, + { + "epoch": 0.6173906508342102, + "grad_norm": 2.1478336109737333, + "learning_rate": 1.3490017883220443e-06, + "loss": 1.0158, + "step": 8215 + }, + { + "epoch": 0.6174658049000451, + "grad_norm": 0.6833206503164047, + "learning_rate": 1.3485414777059375e-06, + "loss": 0.8273, + "step": 8216 + }, + { + "epoch": 0.61754095896588, + "grad_norm": 2.1856452708385605, + "learning_rate": 1.3480812056912417e-06, + "loss": 0.9174, + "step": 8217 + }, + { + "epoch": 0.617616113031715, + "grad_norm": 1.8181333800929533, + "learning_rate": 1.3476209723052318e-06, + "loss": 0.9017, + "step": 8218 + }, + { + "epoch": 0.61769126709755, + "grad_norm": 1.6410343204514823, + "learning_rate": 1.3471607775751774e-06, + "loss": 0.9257, + "step": 8219 + }, + { + "epoch": 0.617766421163385, + "grad_norm": 1.7109411230876308, + "learning_rate": 1.3467006215283459e-06, + "loss": 0.9062, + "step": 8220 + }, + { + "epoch": 0.6178415752292199, + "grad_norm": 0.7393938591971143, + "learning_rate": 1.3462405041920053e-06, + "loss": 0.8465, + "step": 8221 + }, + { + "epoch": 0.6179167292950548, + "grad_norm": 1.5317202254773397, + "learning_rate": 1.3457804255934172e-06, + "loss": 0.9551, + "step": 8222 + }, + { + "epoch": 0.6179918833608898, + "grad_norm": 2.172336008008304, + "learning_rate": 1.3453203857598449e-06, + "loss": 0.8802, + "step": 8223 + }, + { + "epoch": 0.6180670374267248, + "grad_norm": 1.9570809546457473, + "learning_rate": 1.3448603847185464e-06, + "loss": 0.9052, + "step": 8224 + }, + { + "epoch": 0.6181421914925598, + "grad_norm": 1.4281363488879535, + "learning_rate": 1.3444004224967787e-06, + "loss": 0.8874, + "step": 8225 + }, + { + "epoch": 0.6182173455583947, + "grad_norm": 1.535385619444421, + "learning_rate": 1.3439404991217968e-06, + "loss": 0.9151, + "step": 8226 + }, + { + "epoch": 0.6182924996242297, + "grad_norm": 1.5891349393374288, + "learning_rate": 1.343480614620852e-06, + "loss": 0.9618, + "step": 8227 + }, + { + "epoch": 0.6183676536900646, + "grad_norm": 0.6423925810199042, + "learning_rate": 1.3430207690211953e-06, + "loss": 0.7782, + "step": 8228 + }, + { + "epoch": 0.6184428077558995, + "grad_norm": 4.682609346480661, + "learning_rate": 1.3425609623500738e-06, + "loss": 0.856, + "step": 8229 + }, + { + "epoch": 0.6185179618217346, + "grad_norm": 2.7350101278281422, + "learning_rate": 1.3421011946347323e-06, + "loss": 0.9149, + "step": 8230 + }, + { + "epoch": 0.6185931158875695, + "grad_norm": 2.3232526550441994, + "learning_rate": 1.3416414659024147e-06, + "loss": 1.0315, + "step": 8231 + }, + { + "epoch": 0.6186682699534045, + "grad_norm": 1.6754341388440568, + "learning_rate": 1.3411817761803608e-06, + "loss": 0.9872, + "step": 8232 + }, + { + "epoch": 0.6187434240192394, + "grad_norm": 1.9352916314422663, + "learning_rate": 1.34072212549581e-06, + "loss": 1.0665, + "step": 8233 + }, + { + "epoch": 0.6188185780850745, + "grad_norm": 4.064603759672587, + "learning_rate": 1.3402625138759972e-06, + "loss": 0.9718, + "step": 8234 + }, + { + "epoch": 0.6188937321509094, + "grad_norm": 0.6598126798374588, + "learning_rate": 1.3398029413481573e-06, + "loss": 0.78, + "step": 8235 + }, + { + "epoch": 0.6189688862167443, + "grad_norm": 1.6684670340551047, + "learning_rate": 1.3393434079395212e-06, + "loss": 0.9053, + "step": 8236 + }, + { + "epoch": 0.6190440402825793, + "grad_norm": 1.3954734290807713, + "learning_rate": 1.3388839136773174e-06, + "loss": 1.014, + "step": 8237 + }, + { + "epoch": 0.6191191943484142, + "grad_norm": 1.755240563644778, + "learning_rate": 1.3384244585887738e-06, + "loss": 1.0235, + "step": 8238 + }, + { + "epoch": 0.6191943484142493, + "grad_norm": 0.7553003890834651, + "learning_rate": 1.3379650427011141e-06, + "loss": 0.8322, + "step": 8239 + }, + { + "epoch": 0.6192695024800842, + "grad_norm": 1.6384982902604077, + "learning_rate": 1.337505666041561e-06, + "loss": 1.0783, + "step": 8240 + }, + { + "epoch": 0.6193446565459191, + "grad_norm": 2.3464919263992234, + "learning_rate": 1.337046328637334e-06, + "loss": 0.9388, + "step": 8241 + }, + { + "epoch": 0.6194198106117541, + "grad_norm": 1.432407314371131, + "learning_rate": 1.3365870305156502e-06, + "loss": 0.9723, + "step": 8242 + }, + { + "epoch": 0.619494964677589, + "grad_norm": 1.8432520177724854, + "learning_rate": 1.336127771703726e-06, + "loss": 0.9845, + "step": 8243 + }, + { + "epoch": 0.619570118743424, + "grad_norm": 2.0774265348846916, + "learning_rate": 1.3356685522287724e-06, + "loss": 1.0041, + "step": 8244 + }, + { + "epoch": 0.619645272809259, + "grad_norm": 0.7157295034140639, + "learning_rate": 1.3352093721180017e-06, + "loss": 0.7841, + "step": 8245 + }, + { + "epoch": 0.619720426875094, + "grad_norm": 2.284445314590661, + "learning_rate": 1.3347502313986216e-06, + "loss": 0.8033, + "step": 8246 + }, + { + "epoch": 0.6197955809409289, + "grad_norm": 1.528893038654117, + "learning_rate": 1.3342911300978373e-06, + "loss": 0.9478, + "step": 8247 + }, + { + "epoch": 0.6198707350067638, + "grad_norm": 1.956531940654751, + "learning_rate": 1.3338320682428527e-06, + "loss": 0.9974, + "step": 8248 + }, + { + "epoch": 0.6199458890725988, + "grad_norm": 1.6753782130017545, + "learning_rate": 1.3333730458608688e-06, + "loss": 0.9222, + "step": 8249 + }, + { + "epoch": 0.6200210431384338, + "grad_norm": 1.7490397952543277, + "learning_rate": 1.3329140629790851e-06, + "loss": 1.0299, + "step": 8250 + }, + { + "epoch": 0.6200961972042688, + "grad_norm": 1.6081989027907897, + "learning_rate": 1.3324551196246977e-06, + "loss": 0.9736, + "step": 8251 + }, + { + "epoch": 0.6201713512701037, + "grad_norm": 1.3141044619071192, + "learning_rate": 1.3319962158249e-06, + "loss": 0.9977, + "step": 8252 + }, + { + "epoch": 0.6202465053359387, + "grad_norm": 1.6404880284852412, + "learning_rate": 1.331537351606885e-06, + "loss": 1.0287, + "step": 8253 + }, + { + "epoch": 0.6203216594017736, + "grad_norm": 1.816464134032332, + "learning_rate": 1.3310785269978413e-06, + "loss": 0.8479, + "step": 8254 + }, + { + "epoch": 0.6203968134676086, + "grad_norm": 1.8078695540463177, + "learning_rate": 1.3306197420249566e-06, + "loss": 0.9717, + "step": 8255 + }, + { + "epoch": 0.6204719675334436, + "grad_norm": 1.9261632699025744, + "learning_rate": 1.3301609967154152e-06, + "loss": 1.039, + "step": 8256 + }, + { + "epoch": 0.6205471215992785, + "grad_norm": 1.7976941439179537, + "learning_rate": 1.3297022910964e-06, + "loss": 0.9541, + "step": 8257 + }, + { + "epoch": 0.6206222756651135, + "grad_norm": 2.333204424794025, + "learning_rate": 1.3292436251950906e-06, + "loss": 1.0044, + "step": 8258 + }, + { + "epoch": 0.6206974297309484, + "grad_norm": 1.683217159202995, + "learning_rate": 1.3287849990386647e-06, + "loss": 1.003, + "step": 8259 + }, + { + "epoch": 0.6207725837967835, + "grad_norm": 2.827651094806207, + "learning_rate": 1.3283264126542986e-06, + "loss": 0.912, + "step": 8260 + }, + { + "epoch": 0.6208477378626184, + "grad_norm": 1.548346234046455, + "learning_rate": 1.3278678660691638e-06, + "loss": 0.9649, + "step": 8261 + }, + { + "epoch": 0.6209228919284533, + "grad_norm": 1.4838079875358274, + "learning_rate": 1.327409359310432e-06, + "loss": 0.9192, + "step": 8262 + }, + { + "epoch": 0.6209980459942883, + "grad_norm": 1.9152779593070743, + "learning_rate": 1.3269508924052715e-06, + "loss": 0.9398, + "step": 8263 + }, + { + "epoch": 0.6210732000601232, + "grad_norm": 2.0137822206975264, + "learning_rate": 1.326492465380847e-06, + "loss": 0.9389, + "step": 8264 + }, + { + "epoch": 0.6211483541259583, + "grad_norm": 1.8355812538760845, + "learning_rate": 1.326034078264324e-06, + "loss": 0.9929, + "step": 8265 + }, + { + "epoch": 0.6212235081917932, + "grad_norm": 1.7319037211971282, + "learning_rate": 1.3255757310828614e-06, + "loss": 0.9695, + "step": 8266 + }, + { + "epoch": 0.6212986622576281, + "grad_norm": 1.0015878587865936, + "learning_rate": 1.3251174238636202e-06, + "loss": 0.878, + "step": 8267 + }, + { + "epoch": 0.6213738163234631, + "grad_norm": 1.9630617589033716, + "learning_rate": 1.3246591566337563e-06, + "loss": 0.9927, + "step": 8268 + }, + { + "epoch": 0.621448970389298, + "grad_norm": 2.213497984212057, + "learning_rate": 1.3242009294204223e-06, + "loss": 0.8782, + "step": 8269 + }, + { + "epoch": 0.621524124455133, + "grad_norm": 1.5369603085974137, + "learning_rate": 1.3237427422507721e-06, + "loss": 1.0084, + "step": 8270 + }, + { + "epoch": 0.621599278520968, + "grad_norm": 3.0058606440868014, + "learning_rate": 1.323284595151953e-06, + "loss": 1.0868, + "step": 8271 + }, + { + "epoch": 0.621674432586803, + "grad_norm": 1.4027668094311116, + "learning_rate": 1.3228264881511137e-06, + "loss": 0.9637, + "step": 8272 + }, + { + "epoch": 0.6217495866526379, + "grad_norm": 3.2021774751370966, + "learning_rate": 1.322368421275398e-06, + "loss": 0.8724, + "step": 8273 + }, + { + "epoch": 0.6218247407184728, + "grad_norm": 1.675983550125277, + "learning_rate": 1.3219103945519479e-06, + "loss": 0.9312, + "step": 8274 + }, + { + "epoch": 0.6218998947843078, + "grad_norm": 1.9322413174906348, + "learning_rate": 1.3214524080079038e-06, + "loss": 0.9365, + "step": 8275 + }, + { + "epoch": 0.6219750488501428, + "grad_norm": 1.820285387614948, + "learning_rate": 1.3209944616704023e-06, + "loss": 0.8949, + "step": 8276 + }, + { + "epoch": 0.6220502029159778, + "grad_norm": 1.8047139219521444, + "learning_rate": 1.3205365555665795e-06, + "loss": 1.0066, + "step": 8277 + }, + { + "epoch": 0.6221253569818127, + "grad_norm": 1.332916234262692, + "learning_rate": 1.3200786897235675e-06, + "loss": 0.9531, + "step": 8278 + }, + { + "epoch": 0.6222005110476477, + "grad_norm": 2.087510219708856, + "learning_rate": 1.3196208641684968e-06, + "loss": 0.9994, + "step": 8279 + }, + { + "epoch": 0.6222756651134826, + "grad_norm": 4.122411898567415, + "learning_rate": 1.3191630789284954e-06, + "loss": 1.0179, + "step": 8280 + }, + { + "epoch": 0.6223508191793176, + "grad_norm": 2.1145618105640236, + "learning_rate": 1.318705334030688e-06, + "loss": 0.9966, + "step": 8281 + }, + { + "epoch": 0.6224259732451526, + "grad_norm": 1.6958641174421614, + "learning_rate": 1.318247629502199e-06, + "loss": 1.0347, + "step": 8282 + }, + { + "epoch": 0.6225011273109875, + "grad_norm": 2.181825962919752, + "learning_rate": 1.317789965370148e-06, + "loss": 0.9439, + "step": 8283 + }, + { + "epoch": 0.6225762813768225, + "grad_norm": 3.139034342888743, + "learning_rate": 1.3173323416616549e-06, + "loss": 0.9634, + "step": 8284 + }, + { + "epoch": 0.6226514354426574, + "grad_norm": 1.5089766203710433, + "learning_rate": 1.3168747584038341e-06, + "loss": 0.9338, + "step": 8285 + }, + { + "epoch": 0.6227265895084924, + "grad_norm": 0.6528450872560861, + "learning_rate": 1.3164172156237992e-06, + "loss": 0.8084, + "step": 8286 + }, + { + "epoch": 0.6228017435743274, + "grad_norm": 1.6161402753151874, + "learning_rate": 1.3159597133486625e-06, + "loss": 0.9503, + "step": 8287 + }, + { + "epoch": 0.6228768976401623, + "grad_norm": 4.340718480772658, + "learning_rate": 1.315502251605532e-06, + "loss": 0.9899, + "step": 8288 + }, + { + "epoch": 0.6229520517059973, + "grad_norm": 1.3942886697195678, + "learning_rate": 1.3150448304215142e-06, + "loss": 0.961, + "step": 8289 + }, + { + "epoch": 0.6230272057718322, + "grad_norm": 0.7492596230281917, + "learning_rate": 1.3145874498237133e-06, + "loss": 0.8222, + "step": 8290 + }, + { + "epoch": 0.6231023598376673, + "grad_norm": 1.8084164361565838, + "learning_rate": 1.3141301098392302e-06, + "loss": 0.9293, + "step": 8291 + }, + { + "epoch": 0.6231775139035022, + "grad_norm": 1.7806392760422525, + "learning_rate": 1.3136728104951652e-06, + "loss": 1.0064, + "step": 8292 + }, + { + "epoch": 0.6232526679693371, + "grad_norm": 2.171529069973391, + "learning_rate": 1.3132155518186135e-06, + "loss": 0.9852, + "step": 8293 + }, + { + "epoch": 0.6233278220351721, + "grad_norm": 12.416531623942307, + "learning_rate": 1.3127583338366707e-06, + "loss": 0.9631, + "step": 8294 + }, + { + "epoch": 0.623402976101007, + "grad_norm": 1.6561061148852678, + "learning_rate": 1.312301156576428e-06, + "loss": 0.9847, + "step": 8295 + }, + { + "epoch": 0.6234781301668421, + "grad_norm": 2.4274018260145898, + "learning_rate": 1.3118440200649752e-06, + "loss": 0.9846, + "step": 8296 + }, + { + "epoch": 0.623553284232677, + "grad_norm": 4.805982263298183, + "learning_rate": 1.3113869243293993e-06, + "loss": 1.0673, + "step": 8297 + }, + { + "epoch": 0.623628438298512, + "grad_norm": 2.1019805152127904, + "learning_rate": 1.310929869396785e-06, + "loss": 0.8222, + "step": 8298 + }, + { + "epoch": 0.6237035923643469, + "grad_norm": 3.180950092201182, + "learning_rate": 1.3104728552942149e-06, + "loss": 0.8407, + "step": 8299 + }, + { + "epoch": 0.6237787464301818, + "grad_norm": 1.612594421342816, + "learning_rate": 1.3100158820487679e-06, + "loss": 1.0207, + "step": 8300 + }, + { + "epoch": 0.6238539004960169, + "grad_norm": 2.027155814160632, + "learning_rate": 1.3095589496875224e-06, + "loss": 1.0367, + "step": 8301 + }, + { + "epoch": 0.6239290545618518, + "grad_norm": 1.833639776743601, + "learning_rate": 1.309102058237553e-06, + "loss": 0.8854, + "step": 8302 + }, + { + "epoch": 0.6240042086276868, + "grad_norm": 2.86121363492149, + "learning_rate": 1.3086452077259323e-06, + "loss": 0.9799, + "step": 8303 + }, + { + "epoch": 0.6240793626935217, + "grad_norm": 1.6995031359992367, + "learning_rate": 1.3081883981797303e-06, + "loss": 0.9478, + "step": 8304 + }, + { + "epoch": 0.6241545167593567, + "grad_norm": 1.718042609555683, + "learning_rate": 1.3077316296260144e-06, + "loss": 0.9778, + "step": 8305 + }, + { + "epoch": 0.6242296708251917, + "grad_norm": 1.5276075364639394, + "learning_rate": 1.3072749020918514e-06, + "loss": 0.8868, + "step": 8306 + }, + { + "epoch": 0.6243048248910266, + "grad_norm": 1.5737096375007817, + "learning_rate": 1.3068182156043026e-06, + "loss": 0.9359, + "step": 8307 + }, + { + "epoch": 0.6243799789568616, + "grad_norm": 2.53374414835679, + "learning_rate": 1.306361570190428e-06, + "loss": 1.0252, + "step": 8308 + }, + { + "epoch": 0.6244551330226965, + "grad_norm": 1.8560838824094237, + "learning_rate": 1.3059049658772875e-06, + "loss": 0.9511, + "step": 8309 + }, + { + "epoch": 0.6245302870885315, + "grad_norm": 1.8605143157974113, + "learning_rate": 1.305448402691935e-06, + "loss": 0.9785, + "step": 8310 + }, + { + "epoch": 0.6246054411543664, + "grad_norm": 1.9847002138389203, + "learning_rate": 1.304991880661425e-06, + "loss": 0.9246, + "step": 8311 + }, + { + "epoch": 0.6246805952202014, + "grad_norm": 0.6874577552647627, + "learning_rate": 1.3045353998128073e-06, + "loss": 0.8172, + "step": 8312 + }, + { + "epoch": 0.6247557492860364, + "grad_norm": 4.564524580233377, + "learning_rate": 1.30407896017313e-06, + "loss": 0.9823, + "step": 8313 + }, + { + "epoch": 0.6248309033518713, + "grad_norm": 1.5759339938936976, + "learning_rate": 1.3036225617694391e-06, + "loss": 1.0177, + "step": 8314 + }, + { + "epoch": 0.6249060574177063, + "grad_norm": 3.960691425572499, + "learning_rate": 1.3031662046287778e-06, + "loss": 0.9687, + "step": 8315 + }, + { + "epoch": 0.6249812114835412, + "grad_norm": 1.6248358130038771, + "learning_rate": 1.302709888778188e-06, + "loss": 0.9023, + "step": 8316 + }, + { + "epoch": 0.6250563655493763, + "grad_norm": 1.6353168902565116, + "learning_rate": 1.3022536142447069e-06, + "loss": 0.9237, + "step": 8317 + }, + { + "epoch": 0.6251315196152112, + "grad_norm": 1.6461856883354158, + "learning_rate": 1.3017973810553709e-06, + "loss": 0.8864, + "step": 8318 + }, + { + "epoch": 0.6252066736810461, + "grad_norm": 1.7898510523580098, + "learning_rate": 1.301341189237214e-06, + "loss": 1.0278, + "step": 8319 + }, + { + "epoch": 0.6252818277468811, + "grad_norm": 1.5213666877723233, + "learning_rate": 1.3008850388172668e-06, + "loss": 0.9707, + "step": 8320 + }, + { + "epoch": 0.625356981812716, + "grad_norm": 1.6048729897985727, + "learning_rate": 1.3004289298225582e-06, + "loss": 0.8999, + "step": 8321 + }, + { + "epoch": 0.6254321358785511, + "grad_norm": 1.3118473882799873, + "learning_rate": 1.299972862280114e-06, + "loss": 0.9477, + "step": 8322 + }, + { + "epoch": 0.625507289944386, + "grad_norm": 2.082071558005692, + "learning_rate": 1.299516836216959e-06, + "loss": 0.9834, + "step": 8323 + }, + { + "epoch": 0.625582444010221, + "grad_norm": 1.6434886100845545, + "learning_rate": 1.2990608516601133e-06, + "loss": 0.9767, + "step": 8324 + }, + { + "epoch": 0.6256575980760559, + "grad_norm": 1.4186194742128584, + "learning_rate": 1.2986049086365963e-06, + "loss": 0.9083, + "step": 8325 + }, + { + "epoch": 0.6257327521418908, + "grad_norm": 1.638359904004254, + "learning_rate": 1.2981490071734244e-06, + "loss": 0.955, + "step": 8326 + }, + { + "epoch": 0.6258079062077259, + "grad_norm": 2.3078516043189112, + "learning_rate": 1.2976931472976106e-06, + "loss": 0.8668, + "step": 8327 + }, + { + "epoch": 0.6258830602735608, + "grad_norm": 3.0580148356085926, + "learning_rate": 1.2972373290361683e-06, + "loss": 0.8826, + "step": 8328 + }, + { + "epoch": 0.6259582143393958, + "grad_norm": 1.2976965542573724, + "learning_rate": 1.296781552416105e-06, + "loss": 0.9857, + "step": 8329 + }, + { + "epoch": 0.6260333684052307, + "grad_norm": 1.7294582059893897, + "learning_rate": 1.2963258174644266e-06, + "loss": 0.9693, + "step": 8330 + }, + { + "epoch": 0.6261085224710656, + "grad_norm": 1.930463997600762, + "learning_rate": 1.295870124208139e-06, + "loss": 0.9911, + "step": 8331 + }, + { + "epoch": 0.6261836765369007, + "grad_norm": 1.99970439872153, + "learning_rate": 1.2954144726742424e-06, + "loss": 0.8551, + "step": 8332 + }, + { + "epoch": 0.6262588306027356, + "grad_norm": 1.9018418043475747, + "learning_rate": 1.2949588628897367e-06, + "loss": 0.9892, + "step": 8333 + }, + { + "epoch": 0.6263339846685706, + "grad_norm": 5.022879704524392, + "learning_rate": 1.2945032948816183e-06, + "loss": 0.8673, + "step": 8334 + }, + { + "epoch": 0.6264091387344055, + "grad_norm": 2.2393121126348086, + "learning_rate": 1.2940477686768806e-06, + "loss": 1.0293, + "step": 8335 + }, + { + "epoch": 0.6264842928002405, + "grad_norm": 2.0696332861003, + "learning_rate": 1.2935922843025165e-06, + "loss": 0.953, + "step": 8336 + }, + { + "epoch": 0.6265594468660755, + "grad_norm": 1.537826118928309, + "learning_rate": 1.293136841785514e-06, + "loss": 0.9976, + "step": 8337 + }, + { + "epoch": 0.6266346009319104, + "grad_norm": 2.1126354873818545, + "learning_rate": 1.292681441152861e-06, + "loss": 0.907, + "step": 8338 + }, + { + "epoch": 0.6267097549977454, + "grad_norm": 1.5766668098051773, + "learning_rate": 1.2922260824315409e-06, + "loss": 0.9725, + "step": 8339 + }, + { + "epoch": 0.6267849090635803, + "grad_norm": 13.890296545962062, + "learning_rate": 1.2917707656485352e-06, + "loss": 1.0216, + "step": 8340 + }, + { + "epoch": 0.6268600631294153, + "grad_norm": 4.793621537975315, + "learning_rate": 1.2913154908308244e-06, + "loss": 0.9637, + "step": 8341 + }, + { + "epoch": 0.6269352171952502, + "grad_norm": 3.289794985288149, + "learning_rate": 1.2908602580053836e-06, + "loss": 0.9038, + "step": 8342 + }, + { + "epoch": 0.6270103712610853, + "grad_norm": 1.878495708827219, + "learning_rate": 1.2904050671991887e-06, + "loss": 0.8736, + "step": 8343 + }, + { + "epoch": 0.6270855253269202, + "grad_norm": 1.794235214374454, + "learning_rate": 1.2899499184392105e-06, + "loss": 0.8734, + "step": 8344 + }, + { + "epoch": 0.6271606793927551, + "grad_norm": 1.9657294095545106, + "learning_rate": 1.2894948117524188e-06, + "loss": 0.8157, + "step": 8345 + }, + { + "epoch": 0.6272358334585901, + "grad_norm": 1.3064424966127945, + "learning_rate": 1.2890397471657802e-06, + "loss": 1.0341, + "step": 8346 + }, + { + "epoch": 0.627310987524425, + "grad_norm": 1.5619988973078822, + "learning_rate": 1.2885847247062587e-06, + "loss": 1.0492, + "step": 8347 + }, + { + "epoch": 0.6273861415902601, + "grad_norm": 1.3026620367232096, + "learning_rate": 1.2881297444008165e-06, + "loss": 1.0062, + "step": 8348 + }, + { + "epoch": 0.627461295656095, + "grad_norm": 2.2944574413696035, + "learning_rate": 1.2876748062764127e-06, + "loss": 0.9427, + "step": 8349 + }, + { + "epoch": 0.6275364497219299, + "grad_norm": 3.173218541219523, + "learning_rate": 1.2872199103600046e-06, + "loss": 0.9689, + "step": 8350 + }, + { + "epoch": 0.6276116037877649, + "grad_norm": 1.7336517062992418, + "learning_rate": 1.286765056678547e-06, + "loss": 1.0307, + "step": 8351 + }, + { + "epoch": 0.6276867578535998, + "grad_norm": 2.0393394779030865, + "learning_rate": 1.2863102452589893e-06, + "loss": 0.984, + "step": 8352 + }, + { + "epoch": 0.6277619119194349, + "grad_norm": 4.565409947316949, + "learning_rate": 1.2858554761282837e-06, + "loss": 0.9962, + "step": 8353 + }, + { + "epoch": 0.6278370659852698, + "grad_norm": 1.456137129703064, + "learning_rate": 1.2854007493133754e-06, + "loss": 1.0296, + "step": 8354 + }, + { + "epoch": 0.6279122200511048, + "grad_norm": 1.32884475054386, + "learning_rate": 1.2849460648412092e-06, + "loss": 0.9476, + "step": 8355 + }, + { + "epoch": 0.6279873741169397, + "grad_norm": 2.0083666085089105, + "learning_rate": 1.2844914227387266e-06, + "loss": 0.9632, + "step": 8356 + }, + { + "epoch": 0.6280625281827746, + "grad_norm": 2.829187857352343, + "learning_rate": 1.2840368230328672e-06, + "loss": 0.8717, + "step": 8357 + }, + { + "epoch": 0.6281376822486097, + "grad_norm": 1.5677421371854061, + "learning_rate": 1.2835822657505678e-06, + "loss": 1.0667, + "step": 8358 + }, + { + "epoch": 0.6282128363144446, + "grad_norm": 2.4105842494709884, + "learning_rate": 1.2831277509187622e-06, + "loss": 0.9733, + "step": 8359 + }, + { + "epoch": 0.6282879903802796, + "grad_norm": 0.8277070052197626, + "learning_rate": 1.2826732785643826e-06, + "loss": 0.8365, + "step": 8360 + }, + { + "epoch": 0.6283631444461145, + "grad_norm": 1.5170790785796442, + "learning_rate": 1.2822188487143581e-06, + "loss": 0.9181, + "step": 8361 + }, + { + "epoch": 0.6284382985119495, + "grad_norm": 2.0150487504840653, + "learning_rate": 1.2817644613956153e-06, + "loss": 1.0006, + "step": 8362 + }, + { + "epoch": 0.6285134525777845, + "grad_norm": 1.4289477993538455, + "learning_rate": 1.2813101166350786e-06, + "loss": 1.0125, + "step": 8363 + }, + { + "epoch": 0.6285886066436194, + "grad_norm": 1.5826739477463063, + "learning_rate": 1.2808558144596692e-06, + "loss": 0.9377, + "step": 8364 + }, + { + "epoch": 0.6286637607094544, + "grad_norm": 2.301722698243243, + "learning_rate": 1.280401554896307e-06, + "loss": 1.0306, + "step": 8365 + }, + { + "epoch": 0.6287389147752893, + "grad_norm": 5.373354105345251, + "learning_rate": 1.2799473379719077e-06, + "loss": 1.054, + "step": 8366 + }, + { + "epoch": 0.6288140688411243, + "grad_norm": 1.621323608280387, + "learning_rate": 1.2794931637133863e-06, + "loss": 0.9628, + "step": 8367 + }, + { + "epoch": 0.6288892229069593, + "grad_norm": 3.8262623517729057, + "learning_rate": 1.2790390321476542e-06, + "loss": 1.0066, + "step": 8368 + }, + { + "epoch": 0.6289643769727943, + "grad_norm": 2.729115774442125, + "learning_rate": 1.2785849433016198e-06, + "loss": 0.9847, + "step": 8369 + }, + { + "epoch": 0.6290395310386292, + "grad_norm": 2.1049208343914447, + "learning_rate": 1.27813089720219e-06, + "loss": 1.0381, + "step": 8370 + }, + { + "epoch": 0.6291146851044641, + "grad_norm": 1.6439959817034395, + "learning_rate": 1.277676893876268e-06, + "loss": 1.0358, + "step": 8371 + }, + { + "epoch": 0.6291898391702991, + "grad_norm": 2.681958629092791, + "learning_rate": 1.277222933350757e-06, + "loss": 0.9294, + "step": 8372 + }, + { + "epoch": 0.629264993236134, + "grad_norm": 1.5456460251479127, + "learning_rate": 1.2767690156525554e-06, + "loss": 0.8868, + "step": 8373 + }, + { + "epoch": 0.6293401473019691, + "grad_norm": 1.2618707146082944, + "learning_rate": 1.276315140808558e-06, + "loss": 1.0082, + "step": 8374 + }, + { + "epoch": 0.629415301367804, + "grad_norm": 1.6938006432200863, + "learning_rate": 1.27586130884566e-06, + "loss": 0.9559, + "step": 8375 + }, + { + "epoch": 0.6294904554336389, + "grad_norm": 1.9201267012871759, + "learning_rate": 1.275407519790752e-06, + "loss": 1.0677, + "step": 8376 + }, + { + "epoch": 0.6295656094994739, + "grad_norm": 1.6987564334930014, + "learning_rate": 1.2749537736707239e-06, + "loss": 0.9378, + "step": 8377 + }, + { + "epoch": 0.6296407635653088, + "grad_norm": 1.5971281862450302, + "learning_rate": 1.274500070512461e-06, + "loss": 1.0942, + "step": 8378 + }, + { + "epoch": 0.6297159176311439, + "grad_norm": 1.632171139855132, + "learning_rate": 1.2740464103428463e-06, + "loss": 0.9674, + "step": 8379 + }, + { + "epoch": 0.6297910716969788, + "grad_norm": 1.5280301242075824, + "learning_rate": 1.2735927931887625e-06, + "loss": 0.9761, + "step": 8380 + }, + { + "epoch": 0.6298662257628138, + "grad_norm": 1.4844705651289052, + "learning_rate": 1.2731392190770866e-06, + "loss": 0.9138, + "step": 8381 + }, + { + "epoch": 0.6299413798286487, + "grad_norm": 1.6882440551445266, + "learning_rate": 1.2726856880346956e-06, + "loss": 0.999, + "step": 8382 + }, + { + "epoch": 0.6300165338944836, + "grad_norm": 1.8379728033359504, + "learning_rate": 1.2722322000884628e-06, + "loss": 1.0626, + "step": 8383 + }, + { + "epoch": 0.6300916879603187, + "grad_norm": 1.5742156452309255, + "learning_rate": 1.2717787552652585e-06, + "loss": 0.9438, + "step": 8384 + }, + { + "epoch": 0.6301668420261536, + "grad_norm": 3.3680165414200105, + "learning_rate": 1.2713253535919521e-06, + "loss": 0.9266, + "step": 8385 + }, + { + "epoch": 0.6302419960919886, + "grad_norm": 1.8337350824187462, + "learning_rate": 1.2708719950954082e-06, + "loss": 1.0141, + "step": 8386 + }, + { + "epoch": 0.6303171501578235, + "grad_norm": 1.4774669319492948, + "learning_rate": 1.2704186798024913e-06, + "loss": 1.0049, + "step": 8387 + }, + { + "epoch": 0.6303923042236586, + "grad_norm": 0.6975689396577633, + "learning_rate": 1.2699654077400608e-06, + "loss": 0.838, + "step": 8388 + }, + { + "epoch": 0.6304674582894935, + "grad_norm": 2.1043734559024303, + "learning_rate": 1.2695121789349757e-06, + "loss": 0.9725, + "step": 8389 + }, + { + "epoch": 0.6305426123553284, + "grad_norm": 4.265512541801125, + "learning_rate": 1.2690589934140912e-06, + "loss": 0.8369, + "step": 8390 + }, + { + "epoch": 0.6306177664211634, + "grad_norm": 3.163077193952102, + "learning_rate": 1.2686058512042594e-06, + "loss": 0.9408, + "step": 8391 + }, + { + "epoch": 0.6306929204869983, + "grad_norm": 1.5385654634440273, + "learning_rate": 1.268152752332333e-06, + "loss": 0.9917, + "step": 8392 + }, + { + "epoch": 0.6307680745528333, + "grad_norm": 1.5487100524922908, + "learning_rate": 1.2676996968251574e-06, + "loss": 0.9967, + "step": 8393 + }, + { + "epoch": 0.6308432286186683, + "grad_norm": 2.002872430884142, + "learning_rate": 1.2672466847095793e-06, + "loss": 1.0567, + "step": 8394 + }, + { + "epoch": 0.6309183826845032, + "grad_norm": 2.0524563006675445, + "learning_rate": 1.2667937160124416e-06, + "loss": 0.9841, + "step": 8395 + }, + { + "epoch": 0.6309935367503382, + "grad_norm": 1.988216736352063, + "learning_rate": 1.266340790760583e-06, + "loss": 0.9354, + "step": 8396 + }, + { + "epoch": 0.6310686908161731, + "grad_norm": 1.7790854838280539, + "learning_rate": 1.2658879089808423e-06, + "loss": 1.0736, + "step": 8397 + }, + { + "epoch": 0.6311438448820081, + "grad_norm": 1.9264611210566884, + "learning_rate": 1.2654350707000538e-06, + "loss": 1.0222, + "step": 8398 + }, + { + "epoch": 0.6312189989478431, + "grad_norm": 1.9080255088020652, + "learning_rate": 1.264982275945051e-06, + "loss": 0.9779, + "step": 8399 + }, + { + "epoch": 0.6312941530136781, + "grad_norm": 1.7684403650584508, + "learning_rate": 1.2645295247426625e-06, + "loss": 1.0175, + "step": 8400 + }, + { + "epoch": 0.631369307079513, + "grad_norm": 2.201300440421026, + "learning_rate": 1.2640768171197156e-06, + "loss": 0.9676, + "step": 8401 + }, + { + "epoch": 0.6314444611453479, + "grad_norm": 1.5034029824531165, + "learning_rate": 1.2636241531030355e-06, + "loss": 1.0087, + "step": 8402 + }, + { + "epoch": 0.6315196152111829, + "grad_norm": 1.748174669080439, + "learning_rate": 1.263171532719444e-06, + "loss": 0.9603, + "step": 8403 + }, + { + "epoch": 0.6315947692770179, + "grad_norm": 0.6439194058224703, + "learning_rate": 1.2627189559957612e-06, + "loss": 0.8652, + "step": 8404 + }, + { + "epoch": 0.6316699233428529, + "grad_norm": 2.615065804960962, + "learning_rate": 1.2622664229588033e-06, + "loss": 0.8685, + "step": 8405 + }, + { + "epoch": 0.6317450774086878, + "grad_norm": 1.9924402413207718, + "learning_rate": 1.2618139336353846e-06, + "loss": 1.0079, + "step": 8406 + }, + { + "epoch": 0.6318202314745228, + "grad_norm": 1.7268909815056697, + "learning_rate": 1.2613614880523172e-06, + "loss": 0.9074, + "step": 8407 + }, + { + "epoch": 0.6318953855403577, + "grad_norm": 1.6594710787384737, + "learning_rate": 1.2609090862364099e-06, + "loss": 0.9224, + "step": 8408 + }, + { + "epoch": 0.6319705396061926, + "grad_norm": 1.7855222230894015, + "learning_rate": 1.2604567282144696e-06, + "loss": 0.9673, + "step": 8409 + }, + { + "epoch": 0.6320456936720277, + "grad_norm": 1.4845349677372273, + "learning_rate": 1.2600044140132994e-06, + "loss": 0.9776, + "step": 8410 + }, + { + "epoch": 0.6321208477378626, + "grad_norm": 1.46349840564074, + "learning_rate": 1.259552143659702e-06, + "loss": 1.0574, + "step": 8411 + }, + { + "epoch": 0.6321960018036976, + "grad_norm": 0.6469092917452123, + "learning_rate": 1.2590999171804758e-06, + "loss": 0.8292, + "step": 8412 + }, + { + "epoch": 0.6322711558695325, + "grad_norm": 1.5233061122479417, + "learning_rate": 1.2586477346024154e-06, + "loss": 0.8266, + "step": 8413 + }, + { + "epoch": 0.6323463099353676, + "grad_norm": 1.5850381039151826, + "learning_rate": 1.258195595952317e-06, + "loss": 0.9478, + "step": 8414 + }, + { + "epoch": 0.6324214640012025, + "grad_norm": 2.0293808773743978, + "learning_rate": 1.2577435012569684e-06, + "loss": 1.0404, + "step": 8415 + }, + { + "epoch": 0.6324966180670374, + "grad_norm": 1.8000620566373153, + "learning_rate": 1.2572914505431609e-06, + "loss": 1.0692, + "step": 8416 + }, + { + "epoch": 0.6325717721328724, + "grad_norm": 1.572851529151597, + "learning_rate": 1.2568394438376788e-06, + "loss": 0.9236, + "step": 8417 + }, + { + "epoch": 0.6326469261987073, + "grad_norm": 1.7902823874361038, + "learning_rate": 1.2563874811673053e-06, + "loss": 0.9857, + "step": 8418 + }, + { + "epoch": 0.6327220802645424, + "grad_norm": 1.5820421006572338, + "learning_rate": 1.2559355625588208e-06, + "loss": 0.9606, + "step": 8419 + }, + { + "epoch": 0.6327972343303773, + "grad_norm": 1.8333248297176241, + "learning_rate": 1.2554836880390033e-06, + "loss": 0.9642, + "step": 8420 + }, + { + "epoch": 0.6328723883962122, + "grad_norm": 1.5674730886920527, + "learning_rate": 1.2550318576346287e-06, + "loss": 0.949, + "step": 8421 + }, + { + "epoch": 0.6329475424620472, + "grad_norm": 1.6007931233641075, + "learning_rate": 1.2545800713724694e-06, + "loss": 1.0431, + "step": 8422 + }, + { + "epoch": 0.6330226965278821, + "grad_norm": 2.3658690353159546, + "learning_rate": 1.2541283292792949e-06, + "loss": 0.9043, + "step": 8423 + }, + { + "epoch": 0.6330978505937171, + "grad_norm": 1.3710713180924086, + "learning_rate": 1.2536766313818732e-06, + "loss": 0.9967, + "step": 8424 + }, + { + "epoch": 0.6331730046595521, + "grad_norm": 2.009182355079278, + "learning_rate": 1.2532249777069686e-06, + "loss": 0.9978, + "step": 8425 + }, + { + "epoch": 0.6332481587253871, + "grad_norm": 1.9928195922974208, + "learning_rate": 1.252773368281344e-06, + "loss": 1.1121, + "step": 8426 + }, + { + "epoch": 0.633323312791222, + "grad_norm": 1.7928956791899884, + "learning_rate": 1.2523218031317586e-06, + "loss": 1.1068, + "step": 8427 + }, + { + "epoch": 0.6333984668570569, + "grad_norm": 2.013470229848357, + "learning_rate": 1.2518702822849694e-06, + "loss": 0.9466, + "step": 8428 + }, + { + "epoch": 0.633473620922892, + "grad_norm": 1.5881670233861245, + "learning_rate": 1.2514188057677309e-06, + "loss": 0.8907, + "step": 8429 + }, + { + "epoch": 0.6335487749887269, + "grad_norm": 1.5185404786061534, + "learning_rate": 1.250967373606794e-06, + "loss": 0.9249, + "step": 8430 + }, + { + "epoch": 0.6336239290545619, + "grad_norm": 1.9913573685966341, + "learning_rate": 1.2505159858289092e-06, + "loss": 1.0011, + "step": 8431 + }, + { + "epoch": 0.6336990831203968, + "grad_norm": 1.7619029285048247, + "learning_rate": 1.2500646424608217e-06, + "loss": 0.9344, + "step": 8432 + }, + { + "epoch": 0.6337742371862318, + "grad_norm": 1.7122782152891962, + "learning_rate": 1.2496133435292762e-06, + "loss": 0.9265, + "step": 8433 + }, + { + "epoch": 0.6338493912520667, + "grad_norm": 0.722789978850751, + "learning_rate": 1.2491620890610135e-06, + "loss": 0.8122, + "step": 8434 + }, + { + "epoch": 0.6339245453179017, + "grad_norm": 1.595356972548084, + "learning_rate": 1.2487108790827714e-06, + "loss": 0.9442, + "step": 8435 + }, + { + "epoch": 0.6339996993837367, + "grad_norm": 1.9546186915300294, + "learning_rate": 1.2482597136212877e-06, + "loss": 0.9861, + "step": 8436 + }, + { + "epoch": 0.6340748534495716, + "grad_norm": 1.7512083650114731, + "learning_rate": 1.2478085927032935e-06, + "loss": 1.0152, + "step": 8437 + }, + { + "epoch": 0.6341500075154066, + "grad_norm": 8.276735392344182, + "learning_rate": 1.2473575163555215e-06, + "loss": 0.9929, + "step": 8438 + }, + { + "epoch": 0.6342251615812415, + "grad_norm": 0.7955094773658884, + "learning_rate": 1.2469064846046986e-06, + "loss": 0.9172, + "step": 8439 + }, + { + "epoch": 0.6343003156470765, + "grad_norm": 1.6884889177495834, + "learning_rate": 1.2464554974775496e-06, + "loss": 0.9772, + "step": 8440 + }, + { + "epoch": 0.6343754697129115, + "grad_norm": 2.2547394321564935, + "learning_rate": 1.2460045550007985e-06, + "loss": 0.9667, + "step": 8441 + }, + { + "epoch": 0.6344506237787464, + "grad_norm": 4.115033681182406, + "learning_rate": 1.2455536572011643e-06, + "loss": 0.9876, + "step": 8442 + }, + { + "epoch": 0.6345257778445814, + "grad_norm": 1.993667531475336, + "learning_rate": 1.2451028041053656e-06, + "loss": 0.9883, + "step": 8443 + }, + { + "epoch": 0.6346009319104163, + "grad_norm": 1.9463713801269893, + "learning_rate": 1.2446519957401157e-06, + "loss": 0.9152, + "step": 8444 + }, + { + "epoch": 0.6346760859762514, + "grad_norm": 2.2799977039860995, + "learning_rate": 1.2442012321321277e-06, + "loss": 0.9431, + "step": 8445 + }, + { + "epoch": 0.6347512400420863, + "grad_norm": 1.5842940360605866, + "learning_rate": 1.2437505133081112e-06, + "loss": 0.9513, + "step": 8446 + }, + { + "epoch": 0.6348263941079212, + "grad_norm": 0.7955927911368538, + "learning_rate": 1.2432998392947723e-06, + "loss": 0.8329, + "step": 8447 + }, + { + "epoch": 0.6349015481737562, + "grad_norm": 1.9700542549949993, + "learning_rate": 1.2428492101188156e-06, + "loss": 1.0912, + "step": 8448 + }, + { + "epoch": 0.6349767022395911, + "grad_norm": 2.9403461709224645, + "learning_rate": 1.2423986258069428e-06, + "loss": 0.9412, + "step": 8449 + }, + { + "epoch": 0.6350518563054262, + "grad_norm": 2.359805844612556, + "learning_rate": 1.241948086385852e-06, + "loss": 1.0022, + "step": 8450 + }, + { + "epoch": 0.6351270103712611, + "grad_norm": 0.6837402622708074, + "learning_rate": 1.24149759188224e-06, + "loss": 0.8469, + "step": 8451 + }, + { + "epoch": 0.6352021644370961, + "grad_norm": 0.7346144346756465, + "learning_rate": 1.2410471423228002e-06, + "loss": 0.8468, + "step": 8452 + }, + { + "epoch": 0.635277318502931, + "grad_norm": 1.5974567913725004, + "learning_rate": 1.2405967377342236e-06, + "loss": 0.8416, + "step": 8453 + }, + { + "epoch": 0.6353524725687659, + "grad_norm": 1.5768652694226533, + "learning_rate": 1.2401463781431974e-06, + "loss": 0.9903, + "step": 8454 + }, + { + "epoch": 0.635427626634601, + "grad_norm": 1.516868160123929, + "learning_rate": 1.2396960635764093e-06, + "loss": 1.0347, + "step": 8455 + }, + { + "epoch": 0.6355027807004359, + "grad_norm": 1.5145522578364634, + "learning_rate": 1.2392457940605402e-06, + "loss": 0.8997, + "step": 8456 + }, + { + "epoch": 0.6355779347662709, + "grad_norm": 1.4353233941958528, + "learning_rate": 1.2387955696222702e-06, + "loss": 1.0037, + "step": 8457 + }, + { + "epoch": 0.6356530888321058, + "grad_norm": 1.6003993579048026, + "learning_rate": 1.2383453902882787e-06, + "loss": 1.0839, + "step": 8458 + }, + { + "epoch": 0.6357282428979408, + "grad_norm": 2.1970362293449326, + "learning_rate": 1.2378952560852386e-06, + "loss": 1.0276, + "step": 8459 + }, + { + "epoch": 0.6358033969637757, + "grad_norm": 1.5296792518151765, + "learning_rate": 1.2374451670398233e-06, + "loss": 0.9371, + "step": 8460 + }, + { + "epoch": 0.6358785510296107, + "grad_norm": 1.6141943760457838, + "learning_rate": 1.236995123178702e-06, + "loss": 1.0026, + "step": 8461 + }, + { + "epoch": 0.6359537050954457, + "grad_norm": 1.7360287591687913, + "learning_rate": 1.2365451245285413e-06, + "loss": 0.9516, + "step": 8462 + }, + { + "epoch": 0.6360288591612806, + "grad_norm": 1.5828902561561422, + "learning_rate": 1.2360951711160055e-06, + "loss": 0.9772, + "step": 8463 + }, + { + "epoch": 0.6361040132271156, + "grad_norm": 1.711936902657104, + "learning_rate": 1.2356452629677554e-06, + "loss": 0.9955, + "step": 8464 + }, + { + "epoch": 0.6361791672929505, + "grad_norm": 2.1318831322793623, + "learning_rate": 1.235195400110451e-06, + "loss": 1.036, + "step": 8465 + }, + { + "epoch": 0.6362543213587855, + "grad_norm": 1.6336986564618097, + "learning_rate": 1.2347455825707477e-06, + "loss": 0.816, + "step": 8466 + }, + { + "epoch": 0.6363294754246205, + "grad_norm": 1.5238289354640484, + "learning_rate": 1.2342958103752987e-06, + "loss": 0.9111, + "step": 8467 + }, + { + "epoch": 0.6364046294904554, + "grad_norm": 1.6312126827721662, + "learning_rate": 1.2338460835507554e-06, + "loss": 0.9427, + "step": 8468 + }, + { + "epoch": 0.6364797835562904, + "grad_norm": 2.0590488574996955, + "learning_rate": 1.233396402123765e-06, + "loss": 1.0225, + "step": 8469 + }, + { + "epoch": 0.6365549376221253, + "grad_norm": 1.6722418837097273, + "learning_rate": 1.2329467661209738e-06, + "loss": 1.018, + "step": 8470 + }, + { + "epoch": 0.6366300916879604, + "grad_norm": 2.1421991950341206, + "learning_rate": 1.232497175569024e-06, + "loss": 0.9136, + "step": 8471 + }, + { + "epoch": 0.6367052457537953, + "grad_norm": 1.7660384242715157, + "learning_rate": 1.2320476304945548e-06, + "loss": 0.9513, + "step": 8472 + }, + { + "epoch": 0.6367803998196302, + "grad_norm": 1.645360017472519, + "learning_rate": 1.2315981309242046e-06, + "loss": 0.9202, + "step": 8473 + }, + { + "epoch": 0.6368555538854652, + "grad_norm": 1.757272120172783, + "learning_rate": 1.2311486768846075e-06, + "loss": 0.9522, + "step": 8474 + }, + { + "epoch": 0.6369307079513001, + "grad_norm": 1.7636498400218872, + "learning_rate": 1.2306992684023955e-06, + "loss": 0.9251, + "step": 8475 + }, + { + "epoch": 0.6370058620171352, + "grad_norm": 0.7506602083535909, + "learning_rate": 1.230249905504197e-06, + "loss": 0.8005, + "step": 8476 + }, + { + "epoch": 0.6370810160829701, + "grad_norm": 1.818485964456871, + "learning_rate": 1.2298005882166406e-06, + "loss": 0.9816, + "step": 8477 + }, + { + "epoch": 0.6371561701488051, + "grad_norm": 3.5500439171022227, + "learning_rate": 1.229351316566348e-06, + "loss": 0.9917, + "step": 8478 + }, + { + "epoch": 0.63723132421464, + "grad_norm": 1.5820399154614144, + "learning_rate": 1.2289020905799401e-06, + "loss": 0.9412, + "step": 8479 + }, + { + "epoch": 0.6373064782804749, + "grad_norm": 2.8730933459871886, + "learning_rate": 1.2284529102840369e-06, + "loss": 1.0383, + "step": 8480 + }, + { + "epoch": 0.63738163234631, + "grad_norm": 0.784443242117465, + "learning_rate": 1.2280037757052527e-06, + "loss": 0.8722, + "step": 8481 + }, + { + "epoch": 0.6374567864121449, + "grad_norm": 1.7880466692293473, + "learning_rate": 1.2275546868702017e-06, + "loss": 0.9655, + "step": 8482 + }, + { + "epoch": 0.6375319404779799, + "grad_norm": 2.1191590761005497, + "learning_rate": 1.2271056438054933e-06, + "loss": 0.9061, + "step": 8483 + }, + { + "epoch": 0.6376070945438148, + "grad_norm": 1.849151844769225, + "learning_rate": 1.2266566465377343e-06, + "loss": 0.9423, + "step": 8484 + }, + { + "epoch": 0.6376822486096497, + "grad_norm": 1.7582312869189825, + "learning_rate": 1.2262076950935311e-06, + "loss": 0.9242, + "step": 8485 + }, + { + "epoch": 0.6377574026754848, + "grad_norm": 2.5752831719997524, + "learning_rate": 1.2257587894994842e-06, + "loss": 0.9384, + "step": 8486 + }, + { + "epoch": 0.6378325567413197, + "grad_norm": 1.535327697964418, + "learning_rate": 1.2253099297821948e-06, + "loss": 1.0271, + "step": 8487 + }, + { + "epoch": 0.6379077108071547, + "grad_norm": 4.685344963604976, + "learning_rate": 1.2248611159682582e-06, + "loss": 0.9341, + "step": 8488 + }, + { + "epoch": 0.6379828648729896, + "grad_norm": 2.055490779142453, + "learning_rate": 1.2244123480842685e-06, + "loss": 1.0211, + "step": 8489 + }, + { + "epoch": 0.6380580189388246, + "grad_norm": 3.289348381466792, + "learning_rate": 1.2239636261568174e-06, + "loss": 0.9291, + "step": 8490 + }, + { + "epoch": 0.6381331730046595, + "grad_norm": 2.459273487211193, + "learning_rate": 1.2235149502124924e-06, + "loss": 0.9881, + "step": 8491 + }, + { + "epoch": 0.6382083270704945, + "grad_norm": 3.4991687059922683, + "learning_rate": 1.2230663202778806e-06, + "loss": 0.9686, + "step": 8492 + }, + { + "epoch": 0.6382834811363295, + "grad_norm": 1.683707744518453, + "learning_rate": 1.2226177363795645e-06, + "loss": 1.0329, + "step": 8493 + }, + { + "epoch": 0.6383586352021644, + "grad_norm": 2.848663444995137, + "learning_rate": 1.2221691985441238e-06, + "loss": 0.9186, + "step": 8494 + }, + { + "epoch": 0.6384337892679994, + "grad_norm": 1.6564025808624228, + "learning_rate": 1.221720706798137e-06, + "loss": 1.0024, + "step": 8495 + }, + { + "epoch": 0.6385089433338343, + "grad_norm": 0.8326088776286733, + "learning_rate": 1.221272261168178e-06, + "loss": 0.8662, + "step": 8496 + }, + { + "epoch": 0.6385840973996694, + "grad_norm": 1.5307606284884474, + "learning_rate": 1.2208238616808202e-06, + "loss": 0.9291, + "step": 8497 + }, + { + "epoch": 0.6386592514655043, + "grad_norm": 1.7394937808196187, + "learning_rate": 1.2203755083626312e-06, + "loss": 0.9029, + "step": 8498 + }, + { + "epoch": 0.6387344055313392, + "grad_norm": 1.7927418816674832, + "learning_rate": 1.21992720124018e-06, + "loss": 1.0219, + "step": 8499 + }, + { + "epoch": 0.6388095595971742, + "grad_norm": 1.4923114026142252, + "learning_rate": 1.2194789403400289e-06, + "loss": 0.8966, + "step": 8500 + }, + { + "epoch": 0.6388847136630091, + "grad_norm": 2.048041324436976, + "learning_rate": 1.2190307256887384e-06, + "loss": 0.9674, + "step": 8501 + }, + { + "epoch": 0.6389598677288442, + "grad_norm": 1.934384640868386, + "learning_rate": 1.218582557312869e-06, + "loss": 0.897, + "step": 8502 + }, + { + "epoch": 0.6390350217946791, + "grad_norm": 3.0858719830861574, + "learning_rate": 1.2181344352389746e-06, + "loss": 1.022, + "step": 8503 + }, + { + "epoch": 0.6391101758605141, + "grad_norm": 1.7050393221395552, + "learning_rate": 1.2176863594936095e-06, + "loss": 0.9126, + "step": 8504 + }, + { + "epoch": 0.639185329926349, + "grad_norm": 1.751977280330098, + "learning_rate": 1.2172383301033233e-06, + "loss": 1.0232, + "step": 8505 + }, + { + "epoch": 0.6392604839921839, + "grad_norm": 1.5647642133499657, + "learning_rate": 1.216790347094663e-06, + "loss": 1.0475, + "step": 8506 + }, + { + "epoch": 0.639335638058019, + "grad_norm": 2.850719254066056, + "learning_rate": 1.2163424104941743e-06, + "loss": 0.9426, + "step": 8507 + }, + { + "epoch": 0.6394107921238539, + "grad_norm": 1.5431732428523521, + "learning_rate": 1.215894520328398e-06, + "loss": 0.9481, + "step": 8508 + }, + { + "epoch": 0.6394859461896889, + "grad_norm": 1.8045933025927063, + "learning_rate": 1.2154466766238742e-06, + "loss": 1.0026, + "step": 8509 + }, + { + "epoch": 0.6395611002555238, + "grad_norm": 1.5175454468811171, + "learning_rate": 1.2149988794071392e-06, + "loss": 0.9336, + "step": 8510 + }, + { + "epoch": 0.6396362543213587, + "grad_norm": 1.7675848017613884, + "learning_rate": 1.214551128704726e-06, + "loss": 1.0251, + "step": 8511 + }, + { + "epoch": 0.6397114083871938, + "grad_norm": 1.7032875193477095, + "learning_rate": 1.214103424543167e-06, + "loss": 0.9837, + "step": 8512 + }, + { + "epoch": 0.6397865624530287, + "grad_norm": 1.6794759927699257, + "learning_rate": 1.2136557669489886e-06, + "loss": 0.9728, + "step": 8513 + }, + { + "epoch": 0.6398617165188637, + "grad_norm": 1.703308865482799, + "learning_rate": 1.2132081559487177e-06, + "loss": 0.9122, + "step": 8514 + }, + { + "epoch": 0.6399368705846986, + "grad_norm": 2.3149603050569265, + "learning_rate": 1.2127605915688764e-06, + "loss": 1.0163, + "step": 8515 + }, + { + "epoch": 0.6400120246505336, + "grad_norm": 0.7541076312133389, + "learning_rate": 1.2123130738359842e-06, + "loss": 0.8689, + "step": 8516 + }, + { + "epoch": 0.6400871787163686, + "grad_norm": 1.2558121027140159, + "learning_rate": 1.2118656027765591e-06, + "loss": 0.9538, + "step": 8517 + }, + { + "epoch": 0.6401623327822035, + "grad_norm": 1.7026967901469232, + "learning_rate": 1.2114181784171144e-06, + "loss": 0.9424, + "step": 8518 + }, + { + "epoch": 0.6402374868480385, + "grad_norm": 1.5238398876386487, + "learning_rate": 1.2109708007841629e-06, + "loss": 0.9919, + "step": 8519 + }, + { + "epoch": 0.6403126409138734, + "grad_norm": 1.7181144926624798, + "learning_rate": 1.2105234699042117e-06, + "loss": 0.9752, + "step": 8520 + }, + { + "epoch": 0.6403877949797084, + "grad_norm": 3.2888121158500097, + "learning_rate": 1.2100761858037692e-06, + "loss": 1.026, + "step": 8521 + }, + { + "epoch": 0.6404629490455434, + "grad_norm": 1.7196059436511697, + "learning_rate": 1.2096289485093379e-06, + "loss": 0.9836, + "step": 8522 + }, + { + "epoch": 0.6405381031113784, + "grad_norm": 2.0561230652191447, + "learning_rate": 1.2091817580474164e-06, + "loss": 0.9123, + "step": 8523 + }, + { + "epoch": 0.6406132571772133, + "grad_norm": 1.5975439503195255, + "learning_rate": 1.2087346144445053e-06, + "loss": 0.9607, + "step": 8524 + }, + { + "epoch": 0.6406884112430482, + "grad_norm": 1.8963665753756325, + "learning_rate": 1.2082875177270974e-06, + "loss": 0.8994, + "step": 8525 + }, + { + "epoch": 0.6407635653088832, + "grad_norm": 1.669894605268937, + "learning_rate": 1.2078404679216862e-06, + "loss": 0.9671, + "step": 8526 + }, + { + "epoch": 0.6408387193747181, + "grad_norm": 1.964529569772742, + "learning_rate": 1.207393465054761e-06, + "loss": 1.0405, + "step": 8527 + }, + { + "epoch": 0.6409138734405532, + "grad_norm": 1.9729362666745267, + "learning_rate": 1.2069465091528074e-06, + "loss": 0.9344, + "step": 8528 + }, + { + "epoch": 0.6409890275063881, + "grad_norm": 2.321043676390253, + "learning_rate": 1.2064996002423105e-06, + "loss": 0.9778, + "step": 8529 + }, + { + "epoch": 0.641064181572223, + "grad_norm": 1.5572192410199217, + "learning_rate": 1.2060527383497501e-06, + "loss": 0.9736, + "step": 8530 + }, + { + "epoch": 0.641139335638058, + "grad_norm": 2.082490623425535, + "learning_rate": 1.2056059235016056e-06, + "loss": 1.0094, + "step": 8531 + }, + { + "epoch": 0.6412144897038929, + "grad_norm": 2.1235611476083838, + "learning_rate": 1.2051591557243526e-06, + "loss": 0.9944, + "step": 8532 + }, + { + "epoch": 0.641289643769728, + "grad_norm": 1.9659725154923053, + "learning_rate": 1.2047124350444624e-06, + "loss": 0.9206, + "step": 8533 + }, + { + "epoch": 0.6413647978355629, + "grad_norm": 0.7181194690992677, + "learning_rate": 1.2042657614884062e-06, + "loss": 0.8776, + "step": 8534 + }, + { + "epoch": 0.6414399519013979, + "grad_norm": 1.4997479703889147, + "learning_rate": 1.2038191350826506e-06, + "loss": 0.9534, + "step": 8535 + }, + { + "epoch": 0.6415151059672328, + "grad_norm": 2.3852388629419434, + "learning_rate": 1.20337255585366e-06, + "loss": 0.8297, + "step": 8536 + }, + { + "epoch": 0.6415902600330677, + "grad_norm": 1.5009702087942947, + "learning_rate": 1.2029260238278962e-06, + "loss": 0.9584, + "step": 8537 + }, + { + "epoch": 0.6416654140989028, + "grad_norm": 1.2737165742031629, + "learning_rate": 1.2024795390318172e-06, + "loss": 0.9228, + "step": 8538 + }, + { + "epoch": 0.6417405681647377, + "grad_norm": 1.8096858572882455, + "learning_rate": 1.2020331014918799e-06, + "loss": 0.9881, + "step": 8539 + }, + { + "epoch": 0.6418157222305727, + "grad_norm": 1.1914637817500144, + "learning_rate": 1.2015867112345367e-06, + "loss": 0.9933, + "step": 8540 + }, + { + "epoch": 0.6418908762964076, + "grad_norm": 3.528101910362679, + "learning_rate": 1.2011403682862384e-06, + "loss": 0.9937, + "step": 8541 + }, + { + "epoch": 0.6419660303622426, + "grad_norm": 1.760579419559971, + "learning_rate": 1.2006940726734315e-06, + "loss": 1.0228, + "step": 8542 + }, + { + "epoch": 0.6420411844280776, + "grad_norm": 1.6903172041802332, + "learning_rate": 1.2002478244225623e-06, + "loss": 0.9006, + "step": 8543 + }, + { + "epoch": 0.6421163384939125, + "grad_norm": 8.430591156528417, + "learning_rate": 1.1998016235600726e-06, + "loss": 1.0097, + "step": 8544 + }, + { + "epoch": 0.6421914925597475, + "grad_norm": 1.7347311393129208, + "learning_rate": 1.1993554701123993e-06, + "loss": 0.8901, + "step": 8545 + }, + { + "epoch": 0.6422666466255824, + "grad_norm": 1.8671137583707609, + "learning_rate": 1.1989093641059813e-06, + "loss": 1.0645, + "step": 8546 + }, + { + "epoch": 0.6423418006914174, + "grad_norm": 1.6866908075969729, + "learning_rate": 1.1984633055672508e-06, + "loss": 0.9175, + "step": 8547 + }, + { + "epoch": 0.6424169547572524, + "grad_norm": 1.4678520541986295, + "learning_rate": 1.1980172945226389e-06, + "loss": 1.0237, + "step": 8548 + }, + { + "epoch": 0.6424921088230874, + "grad_norm": 1.5259692160774863, + "learning_rate": 1.1975713309985732e-06, + "loss": 1.0685, + "step": 8549 + }, + { + "epoch": 0.6425672628889223, + "grad_norm": 1.927417238192864, + "learning_rate": 1.1971254150214788e-06, + "loss": 0.8357, + "step": 8550 + }, + { + "epoch": 0.6426424169547572, + "grad_norm": 2.0706851264014516, + "learning_rate": 1.1966795466177782e-06, + "loss": 1.067, + "step": 8551 + }, + { + "epoch": 0.6427175710205922, + "grad_norm": 1.9209754513388781, + "learning_rate": 1.1962337258138902e-06, + "loss": 0.9491, + "step": 8552 + }, + { + "epoch": 0.6427927250864272, + "grad_norm": 1.5105479830078608, + "learning_rate": 1.1957879526362323e-06, + "loss": 0.9074, + "step": 8553 + }, + { + "epoch": 0.6428678791522622, + "grad_norm": 2.6897664826015126, + "learning_rate": 1.1953422271112175e-06, + "loss": 0.9688, + "step": 8554 + }, + { + "epoch": 0.6429430332180971, + "grad_norm": 1.7485903102993294, + "learning_rate": 1.1948965492652565e-06, + "loss": 0.8632, + "step": 8555 + }, + { + "epoch": 0.643018187283932, + "grad_norm": 1.788680323130744, + "learning_rate": 1.1944509191247585e-06, + "loss": 1.0263, + "step": 8556 + }, + { + "epoch": 0.643093341349767, + "grad_norm": 1.9043892079979188, + "learning_rate": 1.1940053367161278e-06, + "loss": 0.9537, + "step": 8557 + }, + { + "epoch": 0.643168495415602, + "grad_norm": 1.9457748660248828, + "learning_rate": 1.1935598020657676e-06, + "loss": 0.8892, + "step": 8558 + }, + { + "epoch": 0.643243649481437, + "grad_norm": 2.379343077292905, + "learning_rate": 1.193114315200077e-06, + "loss": 0.9825, + "step": 8559 + }, + { + "epoch": 0.6433188035472719, + "grad_norm": 1.7430835694607134, + "learning_rate": 1.1926688761454531e-06, + "loss": 0.9074, + "step": 8560 + }, + { + "epoch": 0.6433939576131069, + "grad_norm": 2.103945377409558, + "learning_rate": 1.1922234849282897e-06, + "loss": 0.9515, + "step": 8561 + }, + { + "epoch": 0.6434691116789418, + "grad_norm": 1.7055568307104876, + "learning_rate": 1.1917781415749774e-06, + "loss": 1.1178, + "step": 8562 + }, + { + "epoch": 0.6435442657447767, + "grad_norm": 1.8843319408754182, + "learning_rate": 1.1913328461119062e-06, + "loss": 1.0233, + "step": 8563 + }, + { + "epoch": 0.6436194198106118, + "grad_norm": 1.8628146629020166, + "learning_rate": 1.1908875985654593e-06, + "loss": 0.9257, + "step": 8564 + }, + { + "epoch": 0.6436945738764467, + "grad_norm": 1.6930843072664281, + "learning_rate": 1.1904423989620216e-06, + "loss": 1.0008, + "step": 8565 + }, + { + "epoch": 0.6437697279422817, + "grad_norm": 1.992296402890447, + "learning_rate": 1.1899972473279713e-06, + "loss": 0.9484, + "step": 8566 + }, + { + "epoch": 0.6438448820081166, + "grad_norm": 1.6879666707468262, + "learning_rate": 1.1895521436896857e-06, + "loss": 0.9646, + "step": 8567 + }, + { + "epoch": 0.6439200360739517, + "grad_norm": 1.922374071280361, + "learning_rate": 1.1891070880735395e-06, + "loss": 1.0684, + "step": 8568 + }, + { + "epoch": 0.6439951901397866, + "grad_norm": 1.8861385740784877, + "learning_rate": 1.1886620805059027e-06, + "loss": 0.9261, + "step": 8569 + }, + { + "epoch": 0.6440703442056215, + "grad_norm": 1.2951055198009247, + "learning_rate": 1.1882171210131452e-06, + "loss": 0.9574, + "step": 8570 + }, + { + "epoch": 0.6441454982714565, + "grad_norm": 1.8303455006830616, + "learning_rate": 1.1877722096216313e-06, + "loss": 0.9023, + "step": 8571 + }, + { + "epoch": 0.6442206523372914, + "grad_norm": 12.64625256182328, + "learning_rate": 1.187327346357724e-06, + "loss": 0.916, + "step": 8572 + }, + { + "epoch": 0.6442958064031264, + "grad_norm": 1.521157696069959, + "learning_rate": 1.186882531247784e-06, + "loss": 0.9395, + "step": 8573 + }, + { + "epoch": 0.6443709604689614, + "grad_norm": 1.2440197465224794, + "learning_rate": 1.1864377643181671e-06, + "loss": 0.915, + "step": 8574 + }, + { + "epoch": 0.6444461145347963, + "grad_norm": 1.9002960902946637, + "learning_rate": 1.1859930455952283e-06, + "loss": 0.8915, + "step": 8575 + }, + { + "epoch": 0.6445212686006313, + "grad_norm": 2.3267847081857553, + "learning_rate": 1.185548375105319e-06, + "loss": 1.0674, + "step": 8576 + }, + { + "epoch": 0.6445964226664662, + "grad_norm": 1.7399887099941145, + "learning_rate": 1.1851037528747863e-06, + "loss": 0.9433, + "step": 8577 + }, + { + "epoch": 0.6446715767323012, + "grad_norm": 2.0585708182575937, + "learning_rate": 1.1846591789299774e-06, + "loss": 1.0025, + "step": 8578 + }, + { + "epoch": 0.6447467307981362, + "grad_norm": 1.4971484259067123, + "learning_rate": 1.184214653297234e-06, + "loss": 1.0038, + "step": 8579 + }, + { + "epoch": 0.6448218848639712, + "grad_norm": 2.294929147262414, + "learning_rate": 1.1837701760028962e-06, + "loss": 0.995, + "step": 8580 + }, + { + "epoch": 0.6448970389298061, + "grad_norm": 1.4839509860321947, + "learning_rate": 1.1833257470733013e-06, + "loss": 0.9596, + "step": 8581 + }, + { + "epoch": 0.644972192995641, + "grad_norm": 1.9038712715501627, + "learning_rate": 1.1828813665347828e-06, + "loss": 0.9403, + "step": 8582 + }, + { + "epoch": 0.645047347061476, + "grad_norm": 1.9442928966622994, + "learning_rate": 1.1824370344136724e-06, + "loss": 0.9443, + "step": 8583 + }, + { + "epoch": 0.645122501127311, + "grad_norm": 2.4597685444502386, + "learning_rate": 1.181992750736298e-06, + "loss": 1.0103, + "step": 8584 + }, + { + "epoch": 0.645197655193146, + "grad_norm": 0.7794671506022242, + "learning_rate": 1.1815485155289864e-06, + "loss": 0.9427, + "step": 8585 + }, + { + "epoch": 0.6452728092589809, + "grad_norm": 0.7577255828407118, + "learning_rate": 1.1811043288180583e-06, + "loss": 0.8117, + "step": 8586 + }, + { + "epoch": 0.6453479633248159, + "grad_norm": 1.9780563921126737, + "learning_rate": 1.180660190629835e-06, + "loss": 0.9511, + "step": 8587 + }, + { + "epoch": 0.6454231173906508, + "grad_norm": 2.009447078094336, + "learning_rate": 1.180216100990633e-06, + "loss": 1.039, + "step": 8588 + }, + { + "epoch": 0.6454982714564858, + "grad_norm": 1.7713592720649842, + "learning_rate": 1.179772059926766e-06, + "loss": 0.9511, + "step": 8589 + }, + { + "epoch": 0.6455734255223208, + "grad_norm": 2.095824167928, + "learning_rate": 1.1793280674645454e-06, + "loss": 0.982, + "step": 8590 + }, + { + "epoch": 0.6456485795881557, + "grad_norm": 7.25962296192553, + "learning_rate": 1.1788841236302789e-06, + "loss": 0.9255, + "step": 8591 + }, + { + "epoch": 0.6457237336539907, + "grad_norm": 2.3725902981552442, + "learning_rate": 1.178440228450273e-06, + "loss": 0.8823, + "step": 8592 + }, + { + "epoch": 0.6457988877198256, + "grad_norm": 1.8175422643603658, + "learning_rate": 1.1779963819508293e-06, + "loss": 1.0852, + "step": 8593 + }, + { + "epoch": 0.6458740417856607, + "grad_norm": 1.8147698363131657, + "learning_rate": 1.1775525841582475e-06, + "loss": 0.9887, + "step": 8594 + }, + { + "epoch": 0.6459491958514956, + "grad_norm": 1.74822969812546, + "learning_rate": 1.1771088350988247e-06, + "loss": 1.0146, + "step": 8595 + }, + { + "epoch": 0.6460243499173305, + "grad_norm": 1.830435377037356, + "learning_rate": 1.1766651347988542e-06, + "loss": 1.014, + "step": 8596 + }, + { + "epoch": 0.6460995039831655, + "grad_norm": 1.5454442507100061, + "learning_rate": 1.1762214832846274e-06, + "loss": 0.9923, + "step": 8597 + }, + { + "epoch": 0.6461746580490004, + "grad_norm": 1.4668952919036373, + "learning_rate": 1.1757778805824324e-06, + "loss": 1.0159, + "step": 8598 + }, + { + "epoch": 0.6462498121148355, + "grad_norm": 4.884828294842386, + "learning_rate": 1.1753343267185535e-06, + "loss": 0.9335, + "step": 8599 + }, + { + "epoch": 0.6463249661806704, + "grad_norm": 1.8086098985383616, + "learning_rate": 1.1748908217192744e-06, + "loss": 1.0385, + "step": 8600 + }, + { + "epoch": 0.6464001202465053, + "grad_norm": 1.9253549471502311, + "learning_rate": 1.1744473656108729e-06, + "loss": 0.9077, + "step": 8601 + }, + { + "epoch": 0.6464752743123403, + "grad_norm": 2.079647796962654, + "learning_rate": 1.1740039584196265e-06, + "loss": 0.9675, + "step": 8602 + }, + { + "epoch": 0.6465504283781752, + "grad_norm": 1.671181508203647, + "learning_rate": 1.1735606001718087e-06, + "loss": 1.0043, + "step": 8603 + }, + { + "epoch": 0.6466255824440102, + "grad_norm": 1.419337933088679, + "learning_rate": 1.17311729089369e-06, + "loss": 0.9884, + "step": 8604 + }, + { + "epoch": 0.6467007365098452, + "grad_norm": 2.8302075216277895, + "learning_rate": 1.172674030611538e-06, + "loss": 1.0134, + "step": 8605 + }, + { + "epoch": 0.6467758905756802, + "grad_norm": 1.6085101275873732, + "learning_rate": 1.172230819351617e-06, + "loss": 1.0305, + "step": 8606 + }, + { + "epoch": 0.6468510446415151, + "grad_norm": 0.7564778323660292, + "learning_rate": 1.1717876571401913e-06, + "loss": 0.8328, + "step": 8607 + }, + { + "epoch": 0.64692619870735, + "grad_norm": 1.4318181405335788, + "learning_rate": 1.1713445440035168e-06, + "loss": 0.9108, + "step": 8608 + }, + { + "epoch": 0.647001352773185, + "grad_norm": 1.7438011052562712, + "learning_rate": 1.170901479967852e-06, + "loss": 1.0339, + "step": 8609 + }, + { + "epoch": 0.64707650683902, + "grad_norm": 1.915239991769757, + "learning_rate": 1.1704584650594495e-06, + "loss": 0.9805, + "step": 8610 + }, + { + "epoch": 0.647151660904855, + "grad_norm": 1.5079696652424757, + "learning_rate": 1.1700154993045588e-06, + "loss": 0.9007, + "step": 8611 + }, + { + "epoch": 0.6472268149706899, + "grad_norm": 2.2060993218772045, + "learning_rate": 1.1695725827294286e-06, + "loss": 0.992, + "step": 8612 + }, + { + "epoch": 0.6473019690365249, + "grad_norm": 1.5120820784316116, + "learning_rate": 1.1691297153603023e-06, + "loss": 0.9891, + "step": 8613 + }, + { + "epoch": 0.6473771231023598, + "grad_norm": 1.524901249299159, + "learning_rate": 1.1686868972234227e-06, + "loss": 0.9918, + "step": 8614 + }, + { + "epoch": 0.6474522771681948, + "grad_norm": 1.5730099260922796, + "learning_rate": 1.1682441283450275e-06, + "loss": 0.9757, + "step": 8615 + }, + { + "epoch": 0.6475274312340298, + "grad_norm": 1.4117347568603023, + "learning_rate": 1.1678014087513522e-06, + "loss": 0.9009, + "step": 8616 + }, + { + "epoch": 0.6476025852998647, + "grad_norm": 1.7598727211662781, + "learning_rate": 1.1673587384686308e-06, + "loss": 0.9902, + "step": 8617 + }, + { + "epoch": 0.6476777393656997, + "grad_norm": 1.442906371845588, + "learning_rate": 1.1669161175230913e-06, + "loss": 0.9243, + "step": 8618 + }, + { + "epoch": 0.6477528934315346, + "grad_norm": 1.5925688012799644, + "learning_rate": 1.1664735459409632e-06, + "loss": 0.8857, + "step": 8619 + }, + { + "epoch": 0.6478280474973696, + "grad_norm": 1.3773381553878152, + "learning_rate": 1.1660310237484691e-06, + "loss": 0.9411, + "step": 8620 + }, + { + "epoch": 0.6479032015632046, + "grad_norm": 2.2282562773607655, + "learning_rate": 1.1655885509718304e-06, + "loss": 0.9316, + "step": 8621 + }, + { + "epoch": 0.6479783556290395, + "grad_norm": 0.7214691309203969, + "learning_rate": 1.165146127637265e-06, + "loss": 0.8635, + "step": 8622 + }, + { + "epoch": 0.6480535096948745, + "grad_norm": 1.7580194648125111, + "learning_rate": 1.1647037537709876e-06, + "loss": 0.9622, + "step": 8623 + }, + { + "epoch": 0.6481286637607094, + "grad_norm": 1.9143720104108253, + "learning_rate": 1.1642614293992123e-06, + "loss": 0.892, + "step": 8624 + }, + { + "epoch": 0.6482038178265445, + "grad_norm": 1.7027325659276493, + "learning_rate": 1.1638191545481476e-06, + "loss": 0.9449, + "step": 8625 + }, + { + "epoch": 0.6482789718923794, + "grad_norm": 1.7907234045021818, + "learning_rate": 1.163376929244e-06, + "loss": 0.9929, + "step": 8626 + }, + { + "epoch": 0.6483541259582143, + "grad_norm": 2.070733369477945, + "learning_rate": 1.1629347535129728e-06, + "loss": 0.8764, + "step": 8627 + }, + { + "epoch": 0.6484292800240493, + "grad_norm": 2.1911241210364696, + "learning_rate": 1.1624926273812664e-06, + "loss": 0.9951, + "step": 8628 + }, + { + "epoch": 0.6485044340898842, + "grad_norm": 0.8160817354629981, + "learning_rate": 1.162050550875079e-06, + "loss": 0.8464, + "step": 8629 + }, + { + "epoch": 0.6485795881557193, + "grad_norm": 1.8770501055159805, + "learning_rate": 1.1616085240206058e-06, + "loss": 0.9479, + "step": 8630 + }, + { + "epoch": 0.6486547422215542, + "grad_norm": 5.648337208977378, + "learning_rate": 1.1611665468440376e-06, + "loss": 1.0309, + "step": 8631 + }, + { + "epoch": 0.6487298962873892, + "grad_norm": 2.0582716398478844, + "learning_rate": 1.1607246193715637e-06, + "loss": 0.9659, + "step": 8632 + }, + { + "epoch": 0.6488050503532241, + "grad_norm": 1.8564129484948535, + "learning_rate": 1.160282741629369e-06, + "loss": 1.0267, + "step": 8633 + }, + { + "epoch": 0.648880204419059, + "grad_norm": 1.5557552485732242, + "learning_rate": 1.1598409136436385e-06, + "loss": 0.9467, + "step": 8634 + }, + { + "epoch": 0.648955358484894, + "grad_norm": 1.8513338615797692, + "learning_rate": 1.1593991354405505e-06, + "loss": 0.9805, + "step": 8635 + }, + { + "epoch": 0.649030512550729, + "grad_norm": 0.7149722291314959, + "learning_rate": 1.158957407046283e-06, + "loss": 0.8134, + "step": 8636 + }, + { + "epoch": 0.649105666616564, + "grad_norm": 1.7859373157731409, + "learning_rate": 1.1585157284870097e-06, + "loss": 1.0227, + "step": 8637 + }, + { + "epoch": 0.6491808206823989, + "grad_norm": 1.6798283007846926, + "learning_rate": 1.1580740997889008e-06, + "loss": 0.9906, + "step": 8638 + }, + { + "epoch": 0.6492559747482339, + "grad_norm": 1.4436326445736247, + "learning_rate": 1.1576325209781263e-06, + "loss": 1.0433, + "step": 8639 + }, + { + "epoch": 0.6493311288140688, + "grad_norm": 2.2011257325881277, + "learning_rate": 1.1571909920808498e-06, + "loss": 0.9302, + "step": 8640 + }, + { + "epoch": 0.6494062828799038, + "grad_norm": 0.7009181251518264, + "learning_rate": 1.156749513123235e-06, + "loss": 0.8552, + "step": 8641 + }, + { + "epoch": 0.6494814369457388, + "grad_norm": 0.6765762900341272, + "learning_rate": 1.1563080841314408e-06, + "loss": 0.7866, + "step": 8642 + }, + { + "epoch": 0.6495565910115737, + "grad_norm": 2.039426924488438, + "learning_rate": 1.155866705131623e-06, + "loss": 0.899, + "step": 8643 + }, + { + "epoch": 0.6496317450774087, + "grad_norm": 1.63779779885792, + "learning_rate": 1.1554253761499358e-06, + "loss": 0.9081, + "step": 8644 + }, + { + "epoch": 0.6497068991432436, + "grad_norm": 3.11315029352279, + "learning_rate": 1.154984097212528e-06, + "loss": 0.9781, + "step": 8645 + }, + { + "epoch": 0.6497820532090786, + "grad_norm": 2.8516589840804625, + "learning_rate": 1.154542868345549e-06, + "loss": 0.997, + "step": 8646 + }, + { + "epoch": 0.6498572072749136, + "grad_norm": 1.451384127100546, + "learning_rate": 1.1541016895751425e-06, + "loss": 0.9059, + "step": 8647 + }, + { + "epoch": 0.6499323613407485, + "grad_norm": 2.158304093014319, + "learning_rate": 1.1536605609274504e-06, + "loss": 0.8446, + "step": 8648 + }, + { + "epoch": 0.6500075154065835, + "grad_norm": 2.2069301345752845, + "learning_rate": 1.1532194824286107e-06, + "loss": 0.8719, + "step": 8649 + }, + { + "epoch": 0.6500826694724184, + "grad_norm": 1.6401929740650156, + "learning_rate": 1.1527784541047583e-06, + "loss": 1.0104, + "step": 8650 + }, + { + "epoch": 0.6501578235382535, + "grad_norm": 1.807126531467975, + "learning_rate": 1.1523374759820276e-06, + "loss": 0.9731, + "step": 8651 + }, + { + "epoch": 0.6502329776040884, + "grad_norm": 1.8420402228554043, + "learning_rate": 1.1518965480865474e-06, + "loss": 0.9806, + "step": 8652 + }, + { + "epoch": 0.6503081316699233, + "grad_norm": 1.5899691145674246, + "learning_rate": 1.1514556704444446e-06, + "loss": 0.9965, + "step": 8653 + }, + { + "epoch": 0.6503832857357583, + "grad_norm": 1.8366756821957473, + "learning_rate": 1.151014843081842e-06, + "loss": 0.9918, + "step": 8654 + }, + { + "epoch": 0.6504584398015932, + "grad_norm": 1.6245474552022485, + "learning_rate": 1.1505740660248606e-06, + "loss": 0.9886, + "step": 8655 + }, + { + "epoch": 0.6505335938674283, + "grad_norm": 1.5979121593161436, + "learning_rate": 1.1501333392996194e-06, + "loss": 0.9175, + "step": 8656 + }, + { + "epoch": 0.6506087479332632, + "grad_norm": 2.3064621122247324, + "learning_rate": 1.1496926629322316e-06, + "loss": 1.0524, + "step": 8657 + }, + { + "epoch": 0.6506839019990982, + "grad_norm": 0.7580981551887217, + "learning_rate": 1.14925203694881e-06, + "loss": 0.8487, + "step": 8658 + }, + { + "epoch": 0.6507590560649331, + "grad_norm": 2.1893998751260364, + "learning_rate": 1.148811461375463e-06, + "loss": 1.0106, + "step": 8659 + }, + { + "epoch": 0.650834210130768, + "grad_norm": 1.487610513124092, + "learning_rate": 1.1483709362382953e-06, + "loss": 1.0575, + "step": 8660 + }, + { + "epoch": 0.6509093641966031, + "grad_norm": 1.746186665701929, + "learning_rate": 1.1479304615634115e-06, + "loss": 0.7952, + "step": 8661 + }, + { + "epoch": 0.650984518262438, + "grad_norm": 2.102442787471532, + "learning_rate": 1.14749003737691e-06, + "loss": 0.9848, + "step": 8662 + }, + { + "epoch": 0.651059672328273, + "grad_norm": 1.6762716647233729, + "learning_rate": 1.147049663704889e-06, + "loss": 0.9975, + "step": 8663 + }, + { + "epoch": 0.6511348263941079, + "grad_norm": 2.3123391585217776, + "learning_rate": 1.1466093405734417e-06, + "loss": 1.0604, + "step": 8664 + }, + { + "epoch": 0.6512099804599428, + "grad_norm": 2.117203715920372, + "learning_rate": 1.1461690680086587e-06, + "loss": 1.0256, + "step": 8665 + }, + { + "epoch": 0.6512851345257779, + "grad_norm": 1.8278224727353989, + "learning_rate": 1.145728846036628e-06, + "loss": 1.0411, + "step": 8666 + }, + { + "epoch": 0.6513602885916128, + "grad_norm": 1.4292425912128224, + "learning_rate": 1.1452886746834335e-06, + "loss": 0.9289, + "step": 8667 + }, + { + "epoch": 0.6514354426574478, + "grad_norm": 2.085294245716187, + "learning_rate": 1.1448485539751586e-06, + "loss": 0.942, + "step": 8668 + }, + { + "epoch": 0.6515105967232827, + "grad_norm": 1.5238175920573713, + "learning_rate": 1.144408483937882e-06, + "loss": 0.881, + "step": 8669 + }, + { + "epoch": 0.6515857507891177, + "grad_norm": 1.6655150328902244, + "learning_rate": 1.1439684645976787e-06, + "loss": 0.9282, + "step": 8670 + }, + { + "epoch": 0.6516609048549526, + "grad_norm": 1.8901013997039207, + "learning_rate": 1.1435284959806218e-06, + "loss": 0.8497, + "step": 8671 + }, + { + "epoch": 0.6517360589207876, + "grad_norm": 2.0126749377304476, + "learning_rate": 1.1430885781127803e-06, + "loss": 0.9353, + "step": 8672 + }, + { + "epoch": 0.6518112129866226, + "grad_norm": 0.8958344607383081, + "learning_rate": 1.1426487110202228e-06, + "loss": 0.9179, + "step": 8673 + }, + { + "epoch": 0.6518863670524575, + "grad_norm": 2.0913286014430796, + "learning_rate": 1.142208894729012e-06, + "loss": 0.9194, + "step": 8674 + }, + { + "epoch": 0.6519615211182925, + "grad_norm": 2.490184876193101, + "learning_rate": 1.1417691292652091e-06, + "loss": 0.914, + "step": 8675 + }, + { + "epoch": 0.6520366751841274, + "grad_norm": 2.1470857563988472, + "learning_rate": 1.1413294146548716e-06, + "loss": 0.9993, + "step": 8676 + }, + { + "epoch": 0.6521118292499625, + "grad_norm": 2.260770028851243, + "learning_rate": 1.1408897509240537e-06, + "loss": 0.9842, + "step": 8677 + }, + { + "epoch": 0.6521869833157974, + "grad_norm": 1.605504464821917, + "learning_rate": 1.1404501380988084e-06, + "loss": 0.8982, + "step": 8678 + }, + { + "epoch": 0.6522621373816323, + "grad_norm": 1.899471864580812, + "learning_rate": 1.1400105762051833e-06, + "loss": 0.9049, + "step": 8679 + }, + { + "epoch": 0.6523372914474673, + "grad_norm": 1.6315028582918205, + "learning_rate": 1.139571065269226e-06, + "loss": 0.906, + "step": 8680 + }, + { + "epoch": 0.6524124455133022, + "grad_norm": 4.164935222731069, + "learning_rate": 1.1391316053169773e-06, + "loss": 0.9792, + "step": 8681 + }, + { + "epoch": 0.6524875995791373, + "grad_norm": 2.218640176983261, + "learning_rate": 1.1386921963744765e-06, + "loss": 0.9, + "step": 8682 + }, + { + "epoch": 0.6525627536449722, + "grad_norm": 1.3758147166837382, + "learning_rate": 1.1382528384677619e-06, + "loss": 1.0717, + "step": 8683 + }, + { + "epoch": 0.6526379077108072, + "grad_norm": 2.613163556241262, + "learning_rate": 1.137813531622866e-06, + "loss": 1.0017, + "step": 8684 + }, + { + "epoch": 0.6527130617766421, + "grad_norm": 1.9166125068752473, + "learning_rate": 1.1373742758658206e-06, + "loss": 0.9358, + "step": 8685 + }, + { + "epoch": 0.652788215842477, + "grad_norm": 1.6617918327555976, + "learning_rate": 1.1369350712226525e-06, + "loss": 1.0561, + "step": 8686 + }, + { + "epoch": 0.6528633699083121, + "grad_norm": 0.8046194621830854, + "learning_rate": 1.1364959177193863e-06, + "loss": 0.8489, + "step": 8687 + }, + { + "epoch": 0.652938523974147, + "grad_norm": 2.493751346138421, + "learning_rate": 1.1360568153820436e-06, + "loss": 1.033, + "step": 8688 + }, + { + "epoch": 0.653013678039982, + "grad_norm": 0.7227125816040743, + "learning_rate": 1.1356177642366422e-06, + "loss": 0.8946, + "step": 8689 + }, + { + "epoch": 0.6530888321058169, + "grad_norm": 1.4552119865544564, + "learning_rate": 1.1351787643091988e-06, + "loss": 0.9417, + "step": 8690 + }, + { + "epoch": 0.6531639861716518, + "grad_norm": 1.9611906822867091, + "learning_rate": 1.1347398156257253e-06, + "loss": 1.0124, + "step": 8691 + }, + { + "epoch": 0.6532391402374869, + "grad_norm": 2.4147722009668438, + "learning_rate": 1.134300918212231e-06, + "loss": 1.0308, + "step": 8692 + }, + { + "epoch": 0.6533142943033218, + "grad_norm": 2.0845323037151364, + "learning_rate": 1.1338620720947223e-06, + "loss": 0.8499, + "step": 8693 + }, + { + "epoch": 0.6533894483691568, + "grad_norm": 2.6508228284125397, + "learning_rate": 1.1334232772992018e-06, + "loss": 0.9618, + "step": 8694 + }, + { + "epoch": 0.6534646024349917, + "grad_norm": 2.3508350349680645, + "learning_rate": 1.132984533851671e-06, + "loss": 1.0842, + "step": 8695 + }, + { + "epoch": 0.6535397565008267, + "grad_norm": 1.4624670660761219, + "learning_rate": 1.132545841778127e-06, + "loss": 0.9969, + "step": 8696 + }, + { + "epoch": 0.6536149105666617, + "grad_norm": 1.2579261301970768, + "learning_rate": 1.1321072011045631e-06, + "loss": 0.9788, + "step": 8697 + }, + { + "epoch": 0.6536900646324966, + "grad_norm": 1.516609029757655, + "learning_rate": 1.1316686118569712e-06, + "loss": 1.0256, + "step": 8698 + }, + { + "epoch": 0.6537652186983316, + "grad_norm": 1.6007496329406479, + "learning_rate": 1.1312300740613382e-06, + "loss": 0.9887, + "step": 8699 + }, + { + "epoch": 0.6538403727641665, + "grad_norm": 2.206914469912358, + "learning_rate": 1.130791587743651e-06, + "loss": 0.9816, + "step": 8700 + }, + { + "epoch": 0.6539155268300015, + "grad_norm": 1.6090320851438416, + "learning_rate": 1.1303531529298898e-06, + "loss": 0.8948, + "step": 8701 + }, + { + "epoch": 0.6539906808958365, + "grad_norm": 0.7724534913293727, + "learning_rate": 1.1299147696460361e-06, + "loss": 0.8326, + "step": 8702 + }, + { + "epoch": 0.6540658349616715, + "grad_norm": 0.8437184928380203, + "learning_rate": 1.129476437918063e-06, + "loss": 0.8573, + "step": 8703 + }, + { + "epoch": 0.6541409890275064, + "grad_norm": 1.9073054409172099, + "learning_rate": 1.1290381577719436e-06, + "loss": 1.0406, + "step": 8704 + }, + { + "epoch": 0.6542161430933413, + "grad_norm": 9.097935503174435, + "learning_rate": 1.1285999292336495e-06, + "loss": 0.9104, + "step": 8705 + }, + { + "epoch": 0.6542912971591763, + "grad_norm": 1.6924491644793718, + "learning_rate": 1.1281617523291456e-06, + "loss": 1.0594, + "step": 8706 + }, + { + "epoch": 0.6543664512250112, + "grad_norm": 1.9041428097733628, + "learning_rate": 1.127723627084397e-06, + "loss": 0.8511, + "step": 8707 + }, + { + "epoch": 0.6544416052908463, + "grad_norm": 1.553841288701334, + "learning_rate": 1.1272855535253637e-06, + "loss": 0.966, + "step": 8708 + }, + { + "epoch": 0.6545167593566812, + "grad_norm": 2.5124221220330405, + "learning_rate": 1.1268475316780036e-06, + "loss": 0.999, + "step": 8709 + }, + { + "epoch": 0.6545919134225161, + "grad_norm": 2.468880082578354, + "learning_rate": 1.1264095615682704e-06, + "loss": 0.9449, + "step": 8710 + }, + { + "epoch": 0.6546670674883511, + "grad_norm": 2.2492655509062622, + "learning_rate": 1.125971643222115e-06, + "loss": 0.8929, + "step": 8711 + }, + { + "epoch": 0.654742221554186, + "grad_norm": 1.4803536000423843, + "learning_rate": 1.1255337766654873e-06, + "loss": 1.0032, + "step": 8712 + }, + { + "epoch": 0.6548173756200211, + "grad_norm": 1.4726772964552108, + "learning_rate": 1.1250959619243322e-06, + "loss": 0.9629, + "step": 8713 + }, + { + "epoch": 0.654892529685856, + "grad_norm": 2.3283179286409865, + "learning_rate": 1.1246581990245916e-06, + "loss": 0.9551, + "step": 8714 + }, + { + "epoch": 0.654967683751691, + "grad_norm": 2.1330437726958564, + "learning_rate": 1.1242204879922045e-06, + "loss": 1.0366, + "step": 8715 + }, + { + "epoch": 0.6550428378175259, + "grad_norm": 1.4353284604783771, + "learning_rate": 1.1237828288531063e-06, + "loss": 0.9647, + "step": 8716 + }, + { + "epoch": 0.6551179918833608, + "grad_norm": 1.7372476655244753, + "learning_rate": 1.1233452216332316e-06, + "loss": 0.8758, + "step": 8717 + }, + { + "epoch": 0.6551931459491959, + "grad_norm": 1.5751474069309834, + "learning_rate": 1.1229076663585094e-06, + "loss": 0.973, + "step": 8718 + }, + { + "epoch": 0.6552683000150308, + "grad_norm": 1.4907057830225472, + "learning_rate": 1.1224701630548665e-06, + "loss": 0.9485, + "step": 8719 + }, + { + "epoch": 0.6553434540808658, + "grad_norm": 1.4441656562961596, + "learning_rate": 1.122032711748227e-06, + "loss": 0.9555, + "step": 8720 + }, + { + "epoch": 0.6554186081467007, + "grad_norm": 1.4176621959540143, + "learning_rate": 1.12159531246451e-06, + "loss": 0.9886, + "step": 8721 + }, + { + "epoch": 0.6554937622125357, + "grad_norm": 4.171431957250964, + "learning_rate": 1.1211579652296355e-06, + "loss": 1.0371, + "step": 8722 + }, + { + "epoch": 0.6555689162783707, + "grad_norm": 2.1549065869258612, + "learning_rate": 1.1207206700695161e-06, + "loss": 0.9897, + "step": 8723 + }, + { + "epoch": 0.6556440703442056, + "grad_norm": 2.1708475880023665, + "learning_rate": 1.1202834270100655e-06, + "loss": 0.9305, + "step": 8724 + }, + { + "epoch": 0.6557192244100406, + "grad_norm": 1.5622772821005888, + "learning_rate": 1.1198462360771895e-06, + "loss": 0.9804, + "step": 8725 + }, + { + "epoch": 0.6557943784758755, + "grad_norm": 1.988496959803854, + "learning_rate": 1.1194090972967943e-06, + "loss": 0.9536, + "step": 8726 + }, + { + "epoch": 0.6558695325417105, + "grad_norm": 2.66638717577483, + "learning_rate": 1.1189720106947823e-06, + "loss": 1.0182, + "step": 8727 + }, + { + "epoch": 0.6559446866075455, + "grad_norm": 1.9494245951790026, + "learning_rate": 1.1185349762970515e-06, + "loss": 0.9019, + "step": 8728 + }, + { + "epoch": 0.6560198406733805, + "grad_norm": 1.7220371370774787, + "learning_rate": 1.1180979941294998e-06, + "loss": 0.9366, + "step": 8729 + }, + { + "epoch": 0.6560949947392154, + "grad_norm": 2.1683131588762854, + "learning_rate": 1.1176610642180184e-06, + "loss": 0.8662, + "step": 8730 + }, + { + "epoch": 0.6561701488050503, + "grad_norm": 1.7460513525814298, + "learning_rate": 1.117224186588498e-06, + "loss": 1.0656, + "step": 8731 + }, + { + "epoch": 0.6562453028708853, + "grad_norm": 1.8399823373485846, + "learning_rate": 1.1167873612668252e-06, + "loss": 1.1113, + "step": 8732 + }, + { + "epoch": 0.6563204569367203, + "grad_norm": 1.5759817242700356, + "learning_rate": 1.1163505882788821e-06, + "loss": 0.966, + "step": 8733 + }, + { + "epoch": 0.6563956110025553, + "grad_norm": 1.6922589766886598, + "learning_rate": 1.1159138676505516e-06, + "loss": 0.9218, + "step": 8734 + }, + { + "epoch": 0.6564707650683902, + "grad_norm": 1.3804036512067377, + "learning_rate": 1.1154771994077095e-06, + "loss": 1.0171, + "step": 8735 + }, + { + "epoch": 0.6565459191342251, + "grad_norm": 0.662846293003858, + "learning_rate": 1.1150405835762304e-06, + "loss": 0.8242, + "step": 8736 + }, + { + "epoch": 0.6566210732000601, + "grad_norm": 1.6591255129748186, + "learning_rate": 1.1146040201819855e-06, + "loss": 1.0068, + "step": 8737 + }, + { + "epoch": 0.656696227265895, + "grad_norm": 1.7001991688237068, + "learning_rate": 1.114167509250842e-06, + "loss": 1.0045, + "step": 8738 + }, + { + "epoch": 0.6567713813317301, + "grad_norm": 1.64173655320852, + "learning_rate": 1.1137310508086666e-06, + "loss": 0.9949, + "step": 8739 + }, + { + "epoch": 0.656846535397565, + "grad_norm": 1.8515955282235181, + "learning_rate": 1.11329464488132e-06, + "loss": 0.9361, + "step": 8740 + }, + { + "epoch": 0.6569216894634, + "grad_norm": 1.655550467115657, + "learning_rate": 1.112858291494661e-06, + "loss": 0.9907, + "step": 8741 + }, + { + "epoch": 0.6569968435292349, + "grad_norm": 2.042890441444036, + "learning_rate": 1.1124219906745458e-06, + "loss": 0.9042, + "step": 8742 + }, + { + "epoch": 0.6570719975950698, + "grad_norm": 0.8013430601760495, + "learning_rate": 1.1119857424468252e-06, + "loss": 0.8703, + "step": 8743 + }, + { + "epoch": 0.6571471516609049, + "grad_norm": 3.2889469516056464, + "learning_rate": 1.1115495468373505e-06, + "loss": 0.9837, + "step": 8744 + }, + { + "epoch": 0.6572223057267398, + "grad_norm": 2.0266801339672424, + "learning_rate": 1.111113403871967e-06, + "loss": 1.0564, + "step": 8745 + }, + { + "epoch": 0.6572974597925748, + "grad_norm": 1.8955135355199217, + "learning_rate": 1.1106773135765183e-06, + "loss": 0.9564, + "step": 8746 + }, + { + "epoch": 0.6573726138584097, + "grad_norm": 1.4602608159047885, + "learning_rate": 1.1102412759768455e-06, + "loss": 0.989, + "step": 8747 + }, + { + "epoch": 0.6574477679242448, + "grad_norm": 1.639374028975286, + "learning_rate": 1.1098052910987824e-06, + "loss": 0.8424, + "step": 8748 + }, + { + "epoch": 0.6575229219900797, + "grad_norm": 2.0954877559638203, + "learning_rate": 1.1093693589681654e-06, + "loss": 0.9368, + "step": 8749 + }, + { + "epoch": 0.6575980760559146, + "grad_norm": 1.4246949956972441, + "learning_rate": 1.1089334796108235e-06, + "loss": 0.929, + "step": 8750 + }, + { + "epoch": 0.6576732301217496, + "grad_norm": 1.3176614906318098, + "learning_rate": 1.1084976530525858e-06, + "loss": 0.996, + "step": 8751 + }, + { + "epoch": 0.6577483841875845, + "grad_norm": 1.9274711081272802, + "learning_rate": 1.108061879319276e-06, + "loss": 0.9837, + "step": 8752 + }, + { + "epoch": 0.6578235382534195, + "grad_norm": 1.80799410871373, + "learning_rate": 1.1076261584367154e-06, + "loss": 1.0152, + "step": 8753 + }, + { + "epoch": 0.6578986923192545, + "grad_norm": 1.6226898057651846, + "learning_rate": 1.107190490430722e-06, + "loss": 1.0153, + "step": 8754 + }, + { + "epoch": 0.6579738463850894, + "grad_norm": 2.625467440448133, + "learning_rate": 1.10675487532711e-06, + "loss": 1.0362, + "step": 8755 + }, + { + "epoch": 0.6580490004509244, + "grad_norm": 1.8291574154955148, + "learning_rate": 1.1063193131516928e-06, + "loss": 1.0228, + "step": 8756 + }, + { + "epoch": 0.6581241545167593, + "grad_norm": 2.8187702220521245, + "learning_rate": 1.1058838039302788e-06, + "loss": 0.8976, + "step": 8757 + }, + { + "epoch": 0.6581993085825943, + "grad_norm": 1.5346147586871979, + "learning_rate": 1.1054483476886727e-06, + "loss": 0.8971, + "step": 8758 + }, + { + "epoch": 0.6582744626484293, + "grad_norm": 1.9796388230214237, + "learning_rate": 1.1050129444526777e-06, + "loss": 0.9655, + "step": 8759 + }, + { + "epoch": 0.6583496167142643, + "grad_norm": 3.964975319168609, + "learning_rate": 1.104577594248092e-06, + "loss": 0.9441, + "step": 8760 + }, + { + "epoch": 0.6584247707800992, + "grad_norm": 1.7847246111479858, + "learning_rate": 1.1041422971007137e-06, + "loss": 0.9286, + "step": 8761 + }, + { + "epoch": 0.6584999248459341, + "grad_norm": 1.9396078272774278, + "learning_rate": 1.1037070530363343e-06, + "loss": 0.9279, + "step": 8762 + }, + { + "epoch": 0.6585750789117691, + "grad_norm": 2.6112595565569894, + "learning_rate": 1.1032718620807446e-06, + "loss": 0.886, + "step": 8763 + }, + { + "epoch": 0.658650232977604, + "grad_norm": 1.7197226806458177, + "learning_rate": 1.1028367242597307e-06, + "loss": 0.9315, + "step": 8764 + }, + { + "epoch": 0.6587253870434391, + "grad_norm": 2.03357668916021, + "learning_rate": 1.1024016395990756e-06, + "loss": 1.0074, + "step": 8765 + }, + { + "epoch": 0.658800541109274, + "grad_norm": 1.4462769460846705, + "learning_rate": 1.1019666081245613e-06, + "loss": 1.0059, + "step": 8766 + }, + { + "epoch": 0.658875695175109, + "grad_norm": 1.575876579370416, + "learning_rate": 1.1015316298619634e-06, + "loss": 0.9628, + "step": 8767 + }, + { + "epoch": 0.6589508492409439, + "grad_norm": 1.7046850915991452, + "learning_rate": 1.1010967048370577e-06, + "loss": 0.992, + "step": 8768 + }, + { + "epoch": 0.6590260033067789, + "grad_norm": 1.5468961926174145, + "learning_rate": 1.1006618330756153e-06, + "loss": 0.8755, + "step": 8769 + }, + { + "epoch": 0.6591011573726139, + "grad_norm": 2.316238062704816, + "learning_rate": 1.1002270146034013e-06, + "loss": 1.0139, + "step": 8770 + }, + { + "epoch": 0.6591763114384488, + "grad_norm": 1.727901026196045, + "learning_rate": 1.099792249446183e-06, + "loss": 0.9342, + "step": 8771 + }, + { + "epoch": 0.6592514655042838, + "grad_norm": 0.6950465561689992, + "learning_rate": 1.0993575376297201e-06, + "loss": 0.8444, + "step": 8772 + }, + { + "epoch": 0.6593266195701187, + "grad_norm": 1.859514567803034, + "learning_rate": 1.0989228791797729e-06, + "loss": 1.0615, + "step": 8773 + }, + { + "epoch": 0.6594017736359538, + "grad_norm": 1.7374946100165105, + "learning_rate": 1.0984882741220957e-06, + "loss": 1.037, + "step": 8774 + }, + { + "epoch": 0.6594769277017887, + "grad_norm": 8.326878552801492, + "learning_rate": 1.0980537224824403e-06, + "loss": 0.9807, + "step": 8775 + }, + { + "epoch": 0.6595520817676236, + "grad_norm": 2.0220099756389067, + "learning_rate": 1.0976192242865554e-06, + "loss": 0.9337, + "step": 8776 + }, + { + "epoch": 0.6596272358334586, + "grad_norm": 1.3538287792724846, + "learning_rate": 1.097184779560186e-06, + "loss": 0.9819, + "step": 8777 + }, + { + "epoch": 0.6597023898992935, + "grad_norm": 2.2366169172497083, + "learning_rate": 1.0967503883290768e-06, + "loss": 1.0384, + "step": 8778 + }, + { + "epoch": 0.6597775439651286, + "grad_norm": 1.8843185289819104, + "learning_rate": 1.0963160506189655e-06, + "loss": 0.9765, + "step": 8779 + }, + { + "epoch": 0.6598526980309635, + "grad_norm": 1.6187989687785076, + "learning_rate": 1.0958817664555886e-06, + "loss": 0.99, + "step": 8780 + }, + { + "epoch": 0.6599278520967984, + "grad_norm": 1.628326532407606, + "learning_rate": 1.0954475358646793e-06, + "loss": 0.9232, + "step": 8781 + }, + { + "epoch": 0.6600030061626334, + "grad_norm": 1.739914578938976, + "learning_rate": 1.0950133588719665e-06, + "loss": 0.9354, + "step": 8782 + }, + { + "epoch": 0.6600781602284683, + "grad_norm": 1.3464876719922194, + "learning_rate": 1.0945792355031785e-06, + "loss": 0.9879, + "step": 8783 + }, + { + "epoch": 0.6601533142943034, + "grad_norm": 2.0239847393214463, + "learning_rate": 1.0941451657840377e-06, + "loss": 0.9517, + "step": 8784 + }, + { + "epoch": 0.6602284683601383, + "grad_norm": 1.8338305131520216, + "learning_rate": 1.0937111497402648e-06, + "loss": 0.9825, + "step": 8785 + }, + { + "epoch": 0.6603036224259733, + "grad_norm": 1.578485277805059, + "learning_rate": 1.0932771873975764e-06, + "loss": 1.0329, + "step": 8786 + }, + { + "epoch": 0.6603787764918082, + "grad_norm": 0.7565950662567511, + "learning_rate": 1.0928432787816859e-06, + "loss": 0.8486, + "step": 8787 + }, + { + "epoch": 0.6604539305576431, + "grad_norm": 4.274000280727436, + "learning_rate": 1.092409423918306e-06, + "loss": 0.9031, + "step": 8788 + }, + { + "epoch": 0.6605290846234781, + "grad_norm": 0.7708602891968576, + "learning_rate": 1.091975622833142e-06, + "loss": 0.8159, + "step": 8789 + }, + { + "epoch": 0.6606042386893131, + "grad_norm": 1.5661554682792265, + "learning_rate": 1.0915418755519004e-06, + "loss": 0.9005, + "step": 8790 + }, + { + "epoch": 0.6606793927551481, + "grad_norm": 0.6815035007132209, + "learning_rate": 1.0911081821002811e-06, + "loss": 0.8623, + "step": 8791 + }, + { + "epoch": 0.660754546820983, + "grad_norm": 1.4945111781366232, + "learning_rate": 1.0906745425039829e-06, + "loss": 0.9793, + "step": 8792 + }, + { + "epoch": 0.660829700886818, + "grad_norm": 1.6492355367510043, + "learning_rate": 1.0902409567886996e-06, + "loss": 1.0058, + "step": 8793 + }, + { + "epoch": 0.6609048549526529, + "grad_norm": 2.20266047781813, + "learning_rate": 1.0898074249801227e-06, + "loss": 0.9416, + "step": 8794 + }, + { + "epoch": 0.6609800090184879, + "grad_norm": 1.8740993879948746, + "learning_rate": 1.089373947103942e-06, + "loss": 1.0057, + "step": 8795 + }, + { + "epoch": 0.6610551630843229, + "grad_norm": 2.2354456330550128, + "learning_rate": 1.0889405231858422e-06, + "loss": 0.8687, + "step": 8796 + }, + { + "epoch": 0.6611303171501578, + "grad_norm": 2.03589615212624, + "learning_rate": 1.0885071532515049e-06, + "loss": 1.007, + "step": 8797 + }, + { + "epoch": 0.6612054712159928, + "grad_norm": 2.748535112878064, + "learning_rate": 1.088073837326609e-06, + "loss": 0.9708, + "step": 8798 + }, + { + "epoch": 0.6612806252818277, + "grad_norm": 1.7713857873571066, + "learning_rate": 1.0876405754368296e-06, + "loss": 0.9197, + "step": 8799 + }, + { + "epoch": 0.6613557793476627, + "grad_norm": 1.5728260628918693, + "learning_rate": 1.0872073676078405e-06, + "loss": 0.9623, + "step": 8800 + }, + { + "epoch": 0.6614309334134977, + "grad_norm": 1.82656467721514, + "learning_rate": 1.0867742138653103e-06, + "loss": 1.0424, + "step": 8801 + }, + { + "epoch": 0.6615060874793326, + "grad_norm": 1.7039396804825282, + "learning_rate": 1.0863411142349046e-06, + "loss": 1.0062, + "step": 8802 + }, + { + "epoch": 0.6615812415451676, + "grad_norm": 1.7242528734862135, + "learning_rate": 1.0859080687422868e-06, + "loss": 0.9783, + "step": 8803 + }, + { + "epoch": 0.6616563956110025, + "grad_norm": 1.4092588237237853, + "learning_rate": 1.0854750774131153e-06, + "loss": 0.9088, + "step": 8804 + }, + { + "epoch": 0.6617315496768376, + "grad_norm": 2.0643251608578956, + "learning_rate": 1.0850421402730482e-06, + "loss": 1.0474, + "step": 8805 + }, + { + "epoch": 0.6618067037426725, + "grad_norm": 1.2888220360470974, + "learning_rate": 1.084609257347738e-06, + "loss": 0.8983, + "step": 8806 + }, + { + "epoch": 0.6618818578085074, + "grad_norm": 1.5935071124327174, + "learning_rate": 1.0841764286628344e-06, + "loss": 0.9979, + "step": 8807 + }, + { + "epoch": 0.6619570118743424, + "grad_norm": 1.6642388621973705, + "learning_rate": 1.0837436542439843e-06, + "loss": 1.0504, + "step": 8808 + }, + { + "epoch": 0.6620321659401773, + "grad_norm": 1.789520471584216, + "learning_rate": 1.0833109341168308e-06, + "loss": 0.9529, + "step": 8809 + }, + { + "epoch": 0.6621073200060124, + "grad_norm": 2.304041160858233, + "learning_rate": 1.0828782683070153e-06, + "loss": 0.9986, + "step": 8810 + }, + { + "epoch": 0.6621824740718473, + "grad_norm": 1.4807478910808238, + "learning_rate": 1.0824456568401735e-06, + "loss": 1.0017, + "step": 8811 + }, + { + "epoch": 0.6622576281376823, + "grad_norm": 2.8173603870153863, + "learning_rate": 1.0820130997419407e-06, + "loss": 1.0197, + "step": 8812 + }, + { + "epoch": 0.6623327822035172, + "grad_norm": 1.7342326475267615, + "learning_rate": 1.0815805970379473e-06, + "loss": 0.9832, + "step": 8813 + }, + { + "epoch": 0.6624079362693521, + "grad_norm": 1.6422453371964052, + "learning_rate": 1.08114814875382e-06, + "loss": 1.0757, + "step": 8814 + }, + { + "epoch": 0.6624830903351872, + "grad_norm": 2.570918513225954, + "learning_rate": 1.0807157549151838e-06, + "loss": 0.9092, + "step": 8815 + }, + { + "epoch": 0.6625582444010221, + "grad_norm": 1.8406133965481095, + "learning_rate": 1.0802834155476582e-06, + "loss": 0.8445, + "step": 8816 + }, + { + "epoch": 0.6626333984668571, + "grad_norm": 0.6767923585735963, + "learning_rate": 1.0798511306768628e-06, + "loss": 0.837, + "step": 8817 + }, + { + "epoch": 0.662708552532692, + "grad_norm": 1.3717496775603641, + "learning_rate": 1.0794189003284118e-06, + "loss": 0.9971, + "step": 8818 + }, + { + "epoch": 0.662783706598527, + "grad_norm": 1.7492938660631585, + "learning_rate": 1.0789867245279157e-06, + "loss": 0.9377, + "step": 8819 + }, + { + "epoch": 0.662858860664362, + "grad_norm": 1.4791990285565915, + "learning_rate": 1.0785546033009829e-06, + "loss": 1.0081, + "step": 8820 + }, + { + "epoch": 0.6629340147301969, + "grad_norm": 4.245855217684774, + "learning_rate": 1.0781225366732179e-06, + "loss": 0.9415, + "step": 8821 + }, + { + "epoch": 0.6630091687960319, + "grad_norm": 1.9566636035230187, + "learning_rate": 1.0776905246702233e-06, + "loss": 1.0135, + "step": 8822 + }, + { + "epoch": 0.6630843228618668, + "grad_norm": 1.6934660951629084, + "learning_rate": 1.077258567317597e-06, + "loss": 1.0548, + "step": 8823 + }, + { + "epoch": 0.6631594769277018, + "grad_norm": 1.7224538255988724, + "learning_rate": 1.076826664640934e-06, + "loss": 0.961, + "step": 8824 + }, + { + "epoch": 0.6632346309935367, + "grad_norm": 1.349235471832787, + "learning_rate": 1.076394816665826e-06, + "loss": 0.9817, + "step": 8825 + }, + { + "epoch": 0.6633097850593717, + "grad_norm": 1.7733773376746287, + "learning_rate": 1.075963023417861e-06, + "loss": 0.9536, + "step": 8826 + }, + { + "epoch": 0.6633849391252067, + "grad_norm": 3.954658904286863, + "learning_rate": 1.075531284922626e-06, + "loss": 1.0431, + "step": 8827 + }, + { + "epoch": 0.6634600931910416, + "grad_norm": 1.4393012330287893, + "learning_rate": 1.0750996012057028e-06, + "loss": 0.9529, + "step": 8828 + }, + { + "epoch": 0.6635352472568766, + "grad_norm": 0.7191328604640074, + "learning_rate": 1.0746679722926695e-06, + "loss": 0.841, + "step": 8829 + }, + { + "epoch": 0.6636104013227115, + "grad_norm": 1.5462444060462661, + "learning_rate": 1.0742363982091023e-06, + "loss": 0.87, + "step": 8830 + }, + { + "epoch": 0.6636855553885466, + "grad_norm": 1.6158435784353333, + "learning_rate": 1.0738048789805727e-06, + "loss": 0.9907, + "step": 8831 + }, + { + "epoch": 0.6637607094543815, + "grad_norm": 2.8036552378490667, + "learning_rate": 1.0733734146326513e-06, + "loss": 0.9447, + "step": 8832 + }, + { + "epoch": 0.6638358635202164, + "grad_norm": 1.5851045873654404, + "learning_rate": 1.072942005190903e-06, + "loss": 1.058, + "step": 8833 + }, + { + "epoch": 0.6639110175860514, + "grad_norm": 2.500845575384243, + "learning_rate": 1.0725106506808912e-06, + "loss": 0.9362, + "step": 8834 + }, + { + "epoch": 0.6639861716518863, + "grad_norm": 1.6786091026322052, + "learning_rate": 1.0720793511281754e-06, + "loss": 0.9822, + "step": 8835 + }, + { + "epoch": 0.6640613257177214, + "grad_norm": 4.341944916363897, + "learning_rate": 1.0716481065583108e-06, + "loss": 0.9565, + "step": 8836 + }, + { + "epoch": 0.6641364797835563, + "grad_norm": 2.1487630701471994, + "learning_rate": 1.071216916996851e-06, + "loss": 1.037, + "step": 8837 + }, + { + "epoch": 0.6642116338493913, + "grad_norm": 1.7618783679725634, + "learning_rate": 1.0707857824693446e-06, + "loss": 0.8877, + "step": 8838 + }, + { + "epoch": 0.6642867879152262, + "grad_norm": 2.187266528068111, + "learning_rate": 1.0703547030013399e-06, + "loss": 0.952, + "step": 8839 + }, + { + "epoch": 0.6643619419810611, + "grad_norm": 1.780008854040349, + "learning_rate": 1.0699236786183786e-06, + "loss": 0.9648, + "step": 8840 + }, + { + "epoch": 0.6644370960468962, + "grad_norm": 1.8268681958098263, + "learning_rate": 1.0694927093460007e-06, + "loss": 0.9615, + "step": 8841 + }, + { + "epoch": 0.6645122501127311, + "grad_norm": 1.9055056134764108, + "learning_rate": 1.069061795209743e-06, + "loss": 0.8208, + "step": 8842 + }, + { + "epoch": 0.6645874041785661, + "grad_norm": 1.5893685195316711, + "learning_rate": 1.068630936235138e-06, + "loss": 1.016, + "step": 8843 + }, + { + "epoch": 0.664662558244401, + "grad_norm": 1.831453537289406, + "learning_rate": 1.0682001324477173e-06, + "loss": 0.9519, + "step": 8844 + }, + { + "epoch": 0.6647377123102359, + "grad_norm": 5.038188438607663, + "learning_rate": 1.0677693838730068e-06, + "loss": 0.9451, + "step": 8845 + }, + { + "epoch": 0.664812866376071, + "grad_norm": 0.6620076869420347, + "learning_rate": 1.06733869053653e-06, + "loss": 0.862, + "step": 8846 + }, + { + "epoch": 0.6648880204419059, + "grad_norm": 1.6937502505594328, + "learning_rate": 1.0669080524638072e-06, + "loss": 0.9422, + "step": 8847 + }, + { + "epoch": 0.6649631745077409, + "grad_norm": 3.537581986734189, + "learning_rate": 1.0664774696803548e-06, + "loss": 0.9555, + "step": 8848 + }, + { + "epoch": 0.6650383285735758, + "grad_norm": 1.5476748007558478, + "learning_rate": 1.0660469422116876e-06, + "loss": 1.0352, + "step": 8849 + }, + { + "epoch": 0.6651134826394108, + "grad_norm": 1.5692900560148932, + "learning_rate": 1.0656164700833148e-06, + "loss": 0.9497, + "step": 8850 + }, + { + "epoch": 0.6651886367052458, + "grad_norm": 1.4606921147999983, + "learning_rate": 1.0651860533207452e-06, + "loss": 0.9904, + "step": 8851 + }, + { + "epoch": 0.6652637907710807, + "grad_norm": 1.8136531844481354, + "learning_rate": 1.0647556919494814e-06, + "loss": 0.933, + "step": 8852 + }, + { + "epoch": 0.6653389448369157, + "grad_norm": 2.4746495509329924, + "learning_rate": 1.0643253859950231e-06, + "loss": 0.9519, + "step": 8853 + }, + { + "epoch": 0.6654140989027506, + "grad_norm": 1.4859288966494417, + "learning_rate": 1.0638951354828693e-06, + "loss": 0.9574, + "step": 8854 + }, + { + "epoch": 0.6654892529685856, + "grad_norm": 1.630519150748697, + "learning_rate": 1.0634649404385127e-06, + "loss": 1.0544, + "step": 8855 + }, + { + "epoch": 0.6655644070344205, + "grad_norm": 1.5757103756052115, + "learning_rate": 1.0630348008874452e-06, + "loss": 0.9262, + "step": 8856 + }, + { + "epoch": 0.6656395611002556, + "grad_norm": 2.5380268508356534, + "learning_rate": 1.062604716855154e-06, + "loss": 1.0099, + "step": 8857 + }, + { + "epoch": 0.6657147151660905, + "grad_norm": 1.7914607868635448, + "learning_rate": 1.0621746883671226e-06, + "loss": 0.9795, + "step": 8858 + }, + { + "epoch": 0.6657898692319254, + "grad_norm": 3.04874835988306, + "learning_rate": 1.0617447154488322e-06, + "loss": 0.9525, + "step": 8859 + }, + { + "epoch": 0.6658650232977604, + "grad_norm": 1.845819960029786, + "learning_rate": 1.061314798125759e-06, + "loss": 1.0291, + "step": 8860 + }, + { + "epoch": 0.6659401773635953, + "grad_norm": 1.600324943447918, + "learning_rate": 1.0608849364233798e-06, + "loss": 0.8723, + "step": 8861 + }, + { + "epoch": 0.6660153314294304, + "grad_norm": 1.8252533749835373, + "learning_rate": 1.0604551303671641e-06, + "loss": 0.9576, + "step": 8862 + }, + { + "epoch": 0.6660904854952653, + "grad_norm": 1.4744307049766958, + "learning_rate": 1.0600253799825797e-06, + "loss": 1.0017, + "step": 8863 + }, + { + "epoch": 0.6661656395611003, + "grad_norm": 1.9624197172035132, + "learning_rate": 1.0595956852950907e-06, + "loss": 0.9221, + "step": 8864 + }, + { + "epoch": 0.6662407936269352, + "grad_norm": 1.8615809586118752, + "learning_rate": 1.0591660463301578e-06, + "loss": 0.9467, + "step": 8865 + }, + { + "epoch": 0.6663159476927701, + "grad_norm": 0.8020268127211643, + "learning_rate": 1.0587364631132402e-06, + "loss": 0.8476, + "step": 8866 + }, + { + "epoch": 0.6663911017586052, + "grad_norm": 2.1577050648433365, + "learning_rate": 1.0583069356697913e-06, + "loss": 0.9472, + "step": 8867 + }, + { + "epoch": 0.6664662558244401, + "grad_norm": 1.6963925063809868, + "learning_rate": 1.0578774640252626e-06, + "loss": 1.0546, + "step": 8868 + }, + { + "epoch": 0.6665414098902751, + "grad_norm": 1.6479989141812497, + "learning_rate": 1.0574480482051017e-06, + "loss": 1.0297, + "step": 8869 + }, + { + "epoch": 0.66661656395611, + "grad_norm": 2.7872786600820834, + "learning_rate": 1.0570186882347525e-06, + "loss": 1.043, + "step": 8870 + }, + { + "epoch": 0.6666917180219449, + "grad_norm": 2.157779648604642, + "learning_rate": 1.0565893841396575e-06, + "loss": 0.9648, + "step": 8871 + }, + { + "epoch": 0.66676687208778, + "grad_norm": 1.5076375276705931, + "learning_rate": 1.0561601359452537e-06, + "loss": 0.9906, + "step": 8872 + }, + { + "epoch": 0.6668420261536149, + "grad_norm": 1.6711349275356742, + "learning_rate": 1.0557309436769776e-06, + "loss": 0.9318, + "step": 8873 + }, + { + "epoch": 0.6669171802194499, + "grad_norm": 2.140290394269375, + "learning_rate": 1.0553018073602577e-06, + "loss": 0.9797, + "step": 8874 + }, + { + "epoch": 0.6669923342852848, + "grad_norm": 1.4560374943967314, + "learning_rate": 1.0548727270205227e-06, + "loss": 1.0109, + "step": 8875 + }, + { + "epoch": 0.6670674883511198, + "grad_norm": 1.474085915731097, + "learning_rate": 1.0544437026831983e-06, + "loss": 1.0625, + "step": 8876 + }, + { + "epoch": 0.6671426424169548, + "grad_norm": 1.6575938836575859, + "learning_rate": 1.0540147343737049e-06, + "loss": 1.0011, + "step": 8877 + }, + { + "epoch": 0.6672177964827897, + "grad_norm": 1.4326482248097105, + "learning_rate": 1.0535858221174614e-06, + "loss": 0.9972, + "step": 8878 + }, + { + "epoch": 0.6672929505486247, + "grad_norm": 1.6705767145111212, + "learning_rate": 1.0531569659398821e-06, + "loss": 1.0076, + "step": 8879 + }, + { + "epoch": 0.6673681046144596, + "grad_norm": 1.9637431101250176, + "learning_rate": 1.0527281658663782e-06, + "loss": 1.037, + "step": 8880 + }, + { + "epoch": 0.6674432586802946, + "grad_norm": 1.4249080126335043, + "learning_rate": 1.052299421922358e-06, + "loss": 0.9704, + "step": 8881 + }, + { + "epoch": 0.6675184127461296, + "grad_norm": 1.7430889038637907, + "learning_rate": 1.051870734133225e-06, + "loss": 1.0156, + "step": 8882 + }, + { + "epoch": 0.6675935668119646, + "grad_norm": 2.070892598195106, + "learning_rate": 1.0514421025243825e-06, + "loss": 0.9608, + "step": 8883 + }, + { + "epoch": 0.6676687208777995, + "grad_norm": 1.892300769493246, + "learning_rate": 1.0510135271212278e-06, + "loss": 0.9315, + "step": 8884 + }, + { + "epoch": 0.6677438749436344, + "grad_norm": 3.18011441750687, + "learning_rate": 1.0505850079491553e-06, + "loss": 0.9931, + "step": 8885 + }, + { + "epoch": 0.6678190290094694, + "grad_norm": 2.4193824985932904, + "learning_rate": 1.0501565450335573e-06, + "loss": 0.9062, + "step": 8886 + }, + { + "epoch": 0.6678941830753043, + "grad_norm": 1.715336756994273, + "learning_rate": 1.0497281383998195e-06, + "loss": 0.9976, + "step": 8887 + }, + { + "epoch": 0.6679693371411394, + "grad_norm": 1.8476559274040085, + "learning_rate": 1.0492997880733297e-06, + "loss": 0.9808, + "step": 8888 + }, + { + "epoch": 0.6680444912069743, + "grad_norm": 1.456520952550043, + "learning_rate": 1.0488714940794677e-06, + "loss": 0.9854, + "step": 8889 + }, + { + "epoch": 0.6681196452728092, + "grad_norm": 0.6777293400206411, + "learning_rate": 1.048443256443612e-06, + "loss": 0.8476, + "step": 8890 + }, + { + "epoch": 0.6681947993386442, + "grad_norm": 1.5279033445525276, + "learning_rate": 1.048015075191137e-06, + "loss": 0.9142, + "step": 8891 + }, + { + "epoch": 0.6682699534044791, + "grad_norm": 7.8946037308028485, + "learning_rate": 1.0475869503474133e-06, + "loss": 1.0196, + "step": 8892 + }, + { + "epoch": 0.6683451074703142, + "grad_norm": 1.6103691067731765, + "learning_rate": 1.0471588819378107e-06, + "loss": 0.8883, + "step": 8893 + }, + { + "epoch": 0.6684202615361491, + "grad_norm": 2.0118588296952047, + "learning_rate": 1.0467308699876922e-06, + "loss": 0.9507, + "step": 8894 + }, + { + "epoch": 0.6684954156019841, + "grad_norm": 1.7106127147630634, + "learning_rate": 1.0463029145224216e-06, + "loss": 1.0184, + "step": 8895 + }, + { + "epoch": 0.668570569667819, + "grad_norm": 1.4215905041136525, + "learning_rate": 1.0458750155673546e-06, + "loss": 1.0331, + "step": 8896 + }, + { + "epoch": 0.6686457237336539, + "grad_norm": 2.077794328352799, + "learning_rate": 1.0454471731478455e-06, + "loss": 0.93, + "step": 8897 + }, + { + "epoch": 0.668720877799489, + "grad_norm": 0.671589480213581, + "learning_rate": 1.0450193872892472e-06, + "loss": 0.8734, + "step": 8898 + }, + { + "epoch": 0.6687960318653239, + "grad_norm": 1.480742738679223, + "learning_rate": 1.0445916580169067e-06, + "loss": 0.9373, + "step": 8899 + }, + { + "epoch": 0.6688711859311589, + "grad_norm": 1.873446329460304, + "learning_rate": 1.0441639853561697e-06, + "loss": 0.9905, + "step": 8900 + }, + { + "epoch": 0.6689463399969938, + "grad_norm": 2.6066985880888276, + "learning_rate": 1.0437363693323765e-06, + "loss": 1.0388, + "step": 8901 + }, + { + "epoch": 0.6690214940628288, + "grad_norm": 2.0068752609818254, + "learning_rate": 1.0433088099708653e-06, + "loss": 0.8815, + "step": 8902 + }, + { + "epoch": 0.6690966481286638, + "grad_norm": 1.6607646231339857, + "learning_rate": 1.0428813072969704e-06, + "loss": 1.0287, + "step": 8903 + }, + { + "epoch": 0.6691718021944987, + "grad_norm": 1.5563610787945932, + "learning_rate": 1.0424538613360226e-06, + "loss": 0.97, + "step": 8904 + }, + { + "epoch": 0.6692469562603337, + "grad_norm": 1.5294614973656016, + "learning_rate": 1.0420264721133508e-06, + "loss": 1.004, + "step": 8905 + }, + { + "epoch": 0.6693221103261686, + "grad_norm": 1.5746621390161069, + "learning_rate": 1.041599139654279e-06, + "loss": 1.0383, + "step": 8906 + }, + { + "epoch": 0.6693972643920036, + "grad_norm": 1.8690867321571918, + "learning_rate": 1.041171863984128e-06, + "loss": 1.0478, + "step": 8907 + }, + { + "epoch": 0.6694724184578386, + "grad_norm": 1.8412369881330095, + "learning_rate": 1.040744645128216e-06, + "loss": 0.8364, + "step": 8908 + }, + { + "epoch": 0.6695475725236736, + "grad_norm": 0.7107504713933765, + "learning_rate": 1.0403174831118556e-06, + "loss": 0.8115, + "step": 8909 + }, + { + "epoch": 0.6696227265895085, + "grad_norm": 1.6799875390635168, + "learning_rate": 1.0398903779603604e-06, + "loss": 0.9422, + "step": 8910 + }, + { + "epoch": 0.6696978806553434, + "grad_norm": 1.93859573881924, + "learning_rate": 1.0394633296990364e-06, + "loss": 1.0343, + "step": 8911 + }, + { + "epoch": 0.6697730347211784, + "grad_norm": 2.395225337185116, + "learning_rate": 1.0390363383531888e-06, + "loss": 0.9513, + "step": 8912 + }, + { + "epoch": 0.6698481887870134, + "grad_norm": 3.8061273815028938, + "learning_rate": 1.0386094039481177e-06, + "loss": 0.9254, + "step": 8913 + }, + { + "epoch": 0.6699233428528484, + "grad_norm": 1.5275491639761576, + "learning_rate": 1.0381825265091197e-06, + "loss": 0.985, + "step": 8914 + }, + { + "epoch": 0.6699984969186833, + "grad_norm": 1.8821228157087027, + "learning_rate": 1.0377557060614913e-06, + "loss": 0.9292, + "step": 8915 + }, + { + "epoch": 0.6700736509845182, + "grad_norm": 2.669859097795481, + "learning_rate": 1.0373289426305211e-06, + "loss": 1.0151, + "step": 8916 + }, + { + "epoch": 0.6701488050503532, + "grad_norm": 1.737037196374371, + "learning_rate": 1.036902236241498e-06, + "loss": 1.0293, + "step": 8917 + }, + { + "epoch": 0.6702239591161882, + "grad_norm": 2.6542363331832637, + "learning_rate": 1.0364755869197064e-06, + "loss": 1.0191, + "step": 8918 + }, + { + "epoch": 0.6702991131820232, + "grad_norm": 1.7158168386685808, + "learning_rate": 1.0360489946904241e-06, + "loss": 1.0062, + "step": 8919 + }, + { + "epoch": 0.6703742672478581, + "grad_norm": 1.917822744868319, + "learning_rate": 1.0356224595789309e-06, + "loss": 0.9622, + "step": 8920 + }, + { + "epoch": 0.6704494213136931, + "grad_norm": 2.0670364659178913, + "learning_rate": 1.035195981610499e-06, + "loss": 1.0546, + "step": 8921 + }, + { + "epoch": 0.670524575379528, + "grad_norm": 1.5185443252258992, + "learning_rate": 1.0347695608104006e-06, + "loss": 0.964, + "step": 8922 + }, + { + "epoch": 0.670599729445363, + "grad_norm": 1.5578315419953062, + "learning_rate": 1.0343431972039017e-06, + "loss": 0.95, + "step": 8923 + }, + { + "epoch": 0.670674883511198, + "grad_norm": 1.6048372609235062, + "learning_rate": 1.0339168908162662e-06, + "loss": 0.9779, + "step": 8924 + }, + { + "epoch": 0.6707500375770329, + "grad_norm": 1.8384626189647062, + "learning_rate": 1.033490641672754e-06, + "loss": 0.9426, + "step": 8925 + }, + { + "epoch": 0.6708251916428679, + "grad_norm": 2.4292079047426367, + "learning_rate": 1.0330644497986218e-06, + "loss": 0.9141, + "step": 8926 + }, + { + "epoch": 0.6709003457087028, + "grad_norm": 0.8997144166524662, + "learning_rate": 1.0326383152191245e-06, + "loss": 0.8937, + "step": 8927 + }, + { + "epoch": 0.6709754997745379, + "grad_norm": 1.574983018450546, + "learning_rate": 1.0322122379595112e-06, + "loss": 1.0051, + "step": 8928 + }, + { + "epoch": 0.6710506538403728, + "grad_norm": 1.4492591379301016, + "learning_rate": 1.0317862180450285e-06, + "loss": 1.0121, + "step": 8929 + }, + { + "epoch": 0.6711258079062077, + "grad_norm": 1.9484748720502167, + "learning_rate": 1.03136025550092e-06, + "loss": 1.0205, + "step": 8930 + }, + { + "epoch": 0.6712009619720427, + "grad_norm": 1.9555793295063, + "learning_rate": 1.0309343503524248e-06, + "loss": 1.0128, + "step": 8931 + }, + { + "epoch": 0.6712761160378776, + "grad_norm": 1.7712830888649753, + "learning_rate": 1.030508502624781e-06, + "loss": 1.0176, + "step": 8932 + }, + { + "epoch": 0.6713512701037126, + "grad_norm": 1.914362732053698, + "learning_rate": 1.0300827123432206e-06, + "loss": 0.9271, + "step": 8933 + }, + { + "epoch": 0.6714264241695476, + "grad_norm": 1.4141345348240713, + "learning_rate": 1.0296569795329739e-06, + "loss": 0.9773, + "step": 8934 + }, + { + "epoch": 0.6715015782353825, + "grad_norm": 0.6228949143259532, + "learning_rate": 1.0292313042192664e-06, + "loss": 0.8154, + "step": 8935 + }, + { + "epoch": 0.6715767323012175, + "grad_norm": 1.4037120381196115, + "learning_rate": 1.0288056864273212e-06, + "loss": 0.9443, + "step": 8936 + }, + { + "epoch": 0.6716518863670524, + "grad_norm": 1.6536388326982312, + "learning_rate": 1.0283801261823587e-06, + "loss": 0.9926, + "step": 8937 + }, + { + "epoch": 0.6717270404328874, + "grad_norm": 2.2721964276856763, + "learning_rate": 1.027954623509593e-06, + "loss": 0.9389, + "step": 8938 + }, + { + "epoch": 0.6718021944987224, + "grad_norm": 1.680868682729175, + "learning_rate": 1.0275291784342394e-06, + "loss": 0.9782, + "step": 8939 + }, + { + "epoch": 0.6718773485645574, + "grad_norm": 1.4683592256189237, + "learning_rate": 1.0271037909815063e-06, + "loss": 0.9685, + "step": 8940 + }, + { + "epoch": 0.6719525026303923, + "grad_norm": 3.3678667485297056, + "learning_rate": 1.0266784611765976e-06, + "loss": 1.062, + "step": 8941 + }, + { + "epoch": 0.6720276566962272, + "grad_norm": 1.6515108358482158, + "learning_rate": 1.0262531890447182e-06, + "loss": 0.9236, + "step": 8942 + }, + { + "epoch": 0.6721028107620622, + "grad_norm": 1.4823479715654895, + "learning_rate": 1.025827974611065e-06, + "loss": 0.9422, + "step": 8943 + }, + { + "epoch": 0.6721779648278972, + "grad_norm": 1.5911236234406179, + "learning_rate": 1.0254028179008355e-06, + "loss": 0.9199, + "step": 8944 + }, + { + "epoch": 0.6722531188937322, + "grad_norm": 1.8177236066320483, + "learning_rate": 1.024977718939221e-06, + "loss": 0.8861, + "step": 8945 + }, + { + "epoch": 0.6723282729595671, + "grad_norm": 1.538252331649325, + "learning_rate": 1.0245526777514104e-06, + "loss": 1.0036, + "step": 8946 + }, + { + "epoch": 0.6724034270254021, + "grad_norm": 0.8912712563311243, + "learning_rate": 1.0241276943625887e-06, + "loss": 0.8382, + "step": 8947 + }, + { + "epoch": 0.672478581091237, + "grad_norm": 3.1799127372428777, + "learning_rate": 1.0237027687979371e-06, + "loss": 0.9149, + "step": 8948 + }, + { + "epoch": 0.672553735157072, + "grad_norm": 2.2333891501023007, + "learning_rate": 1.0232779010826361e-06, + "loss": 0.9903, + "step": 8949 + }, + { + "epoch": 0.672628889222907, + "grad_norm": 0.8043473645185408, + "learning_rate": 1.0228530912418594e-06, + "loss": 0.8426, + "step": 8950 + }, + { + "epoch": 0.6727040432887419, + "grad_norm": 1.8586526958391123, + "learning_rate": 1.0224283393007786e-06, + "loss": 1.0441, + "step": 8951 + }, + { + "epoch": 0.6727791973545769, + "grad_norm": 1.7311782243633633, + "learning_rate": 1.022003645284562e-06, + "loss": 0.9702, + "step": 8952 + }, + { + "epoch": 0.6728543514204118, + "grad_norm": 2.5364058931061764, + "learning_rate": 1.021579009218374e-06, + "loss": 1.0353, + "step": 8953 + }, + { + "epoch": 0.6729295054862469, + "grad_norm": 3.0542867646450675, + "learning_rate": 1.021154431127377e-06, + "loss": 0.9708, + "step": 8954 + }, + { + "epoch": 0.6730046595520818, + "grad_norm": 1.7595779708406016, + "learning_rate": 1.0207299110367282e-06, + "loss": 0.9523, + "step": 8955 + }, + { + "epoch": 0.6730798136179167, + "grad_norm": 1.8321378306800857, + "learning_rate": 1.020305448971582e-06, + "loss": 1.0337, + "step": 8956 + }, + { + "epoch": 0.6731549676837517, + "grad_norm": 1.5454471818728126, + "learning_rate": 1.0198810449570894e-06, + "loss": 0.9707, + "step": 8957 + }, + { + "epoch": 0.6732301217495866, + "grad_norm": 1.788076270518482, + "learning_rate": 1.0194566990183972e-06, + "loss": 0.9518, + "step": 8958 + }, + { + "epoch": 0.6733052758154217, + "grad_norm": 1.6448028412677478, + "learning_rate": 1.0190324111806514e-06, + "loss": 0.8904, + "step": 8959 + }, + { + "epoch": 0.6733804298812566, + "grad_norm": 1.5255603573742007, + "learning_rate": 1.0186081814689907e-06, + "loss": 0.987, + "step": 8960 + }, + { + "epoch": 0.6734555839470915, + "grad_norm": 1.46184695082784, + "learning_rate": 1.018184009908554e-06, + "loss": 0.9413, + "step": 8961 + }, + { + "epoch": 0.6735307380129265, + "grad_norm": 1.7902626110283157, + "learning_rate": 1.0177598965244744e-06, + "loss": 0.9147, + "step": 8962 + }, + { + "epoch": 0.6736058920787614, + "grad_norm": 1.980891135654531, + "learning_rate": 1.017335841341882e-06, + "loss": 0.9041, + "step": 8963 + }, + { + "epoch": 0.6736810461445965, + "grad_norm": 1.7498545586229783, + "learning_rate": 1.0169118443859037e-06, + "loss": 0.9434, + "step": 8964 + }, + { + "epoch": 0.6737562002104314, + "grad_norm": 1.6737778432409034, + "learning_rate": 1.0164879056816627e-06, + "loss": 1.0267, + "step": 8965 + }, + { + "epoch": 0.6738313542762664, + "grad_norm": 1.441315356027483, + "learning_rate": 1.01606402525428e-06, + "loss": 1.0355, + "step": 8966 + }, + { + "epoch": 0.6739065083421013, + "grad_norm": 4.81688322132637, + "learning_rate": 1.0156402031288717e-06, + "loss": 0.9604, + "step": 8967 + }, + { + "epoch": 0.6739816624079362, + "grad_norm": 1.7353847416768702, + "learning_rate": 1.0152164393305506e-06, + "loss": 0.9917, + "step": 8968 + }, + { + "epoch": 0.6740568164737712, + "grad_norm": 1.6082898527734537, + "learning_rate": 1.0147927338844267e-06, + "loss": 1.0065, + "step": 8969 + }, + { + "epoch": 0.6741319705396062, + "grad_norm": 1.9477258136566158, + "learning_rate": 1.014369086815605e-06, + "loss": 0.9608, + "step": 8970 + }, + { + "epoch": 0.6742071246054412, + "grad_norm": 1.7403329463832395, + "learning_rate": 1.0139454981491898e-06, + "loss": 1.0512, + "step": 8971 + }, + { + "epoch": 0.6742822786712761, + "grad_norm": 1.6693685406810201, + "learning_rate": 1.0135219679102797e-06, + "loss": 1.0182, + "step": 8972 + }, + { + "epoch": 0.6743574327371111, + "grad_norm": 1.680206715377467, + "learning_rate": 1.0130984961239706e-06, + "loss": 0.8589, + "step": 8973 + }, + { + "epoch": 0.674432586802946, + "grad_norm": 1.2793064262674223, + "learning_rate": 1.0126750828153546e-06, + "loss": 0.9489, + "step": 8974 + }, + { + "epoch": 0.674507740868781, + "grad_norm": 1.6703642677442507, + "learning_rate": 1.01225172800952e-06, + "loss": 1.0107, + "step": 8975 + }, + { + "epoch": 0.674582894934616, + "grad_norm": 2.3939884435554815, + "learning_rate": 1.0118284317315535e-06, + "loss": 1.0243, + "step": 8976 + }, + { + "epoch": 0.6746580490004509, + "grad_norm": 2.7427954366674094, + "learning_rate": 1.0114051940065365e-06, + "loss": 0.9256, + "step": 8977 + }, + { + "epoch": 0.6747332030662859, + "grad_norm": 1.8408822208588518, + "learning_rate": 1.0109820148595473e-06, + "loss": 1.1711, + "step": 8978 + }, + { + "epoch": 0.6748083571321208, + "grad_norm": 1.7151566135308456, + "learning_rate": 1.010558894315661e-06, + "loss": 0.9564, + "step": 8979 + }, + { + "epoch": 0.6748835111979558, + "grad_norm": 1.5014832633524635, + "learning_rate": 1.010135832399948e-06, + "loss": 0.9594, + "step": 8980 + }, + { + "epoch": 0.6749586652637908, + "grad_norm": 1.8077802043591884, + "learning_rate": 1.0097128291374786e-06, + "loss": 0.9741, + "step": 8981 + }, + { + "epoch": 0.6750338193296257, + "grad_norm": 1.7032796806998005, + "learning_rate": 1.009289884553315e-06, + "loss": 0.8631, + "step": 8982 + }, + { + "epoch": 0.6751089733954607, + "grad_norm": 2.0952812402997716, + "learning_rate": 1.0088669986725201e-06, + "loss": 1.0268, + "step": 8983 + }, + { + "epoch": 0.6751841274612956, + "grad_norm": 2.0628866353472235, + "learning_rate": 1.0084441715201513e-06, + "loss": 0.8886, + "step": 8984 + }, + { + "epoch": 0.6752592815271307, + "grad_norm": 1.6785776419365162, + "learning_rate": 1.008021403121262e-06, + "loss": 1.0088, + "step": 8985 + }, + { + "epoch": 0.6753344355929656, + "grad_norm": 1.7271776441248377, + "learning_rate": 1.0075986935009028e-06, + "loss": 0.8611, + "step": 8986 + }, + { + "epoch": 0.6754095896588005, + "grad_norm": 1.7014066794713087, + "learning_rate": 1.0071760426841208e-06, + "loss": 0.9005, + "step": 8987 + }, + { + "epoch": 0.6754847437246355, + "grad_norm": 1.928861433507405, + "learning_rate": 1.0067534506959608e-06, + "loss": 0.9743, + "step": 8988 + }, + { + "epoch": 0.6755598977904704, + "grad_norm": 2.675192574672798, + "learning_rate": 1.006330917561462e-06, + "loss": 0.8583, + "step": 8989 + }, + { + "epoch": 0.6756350518563055, + "grad_norm": 1.5999575132451027, + "learning_rate": 1.0059084433056616e-06, + "loss": 0.9848, + "step": 8990 + }, + { + "epoch": 0.6757102059221404, + "grad_norm": 1.9307894012410003, + "learning_rate": 1.0054860279535922e-06, + "loss": 0.9135, + "step": 8991 + }, + { + "epoch": 0.6757853599879754, + "grad_norm": 1.835938603826962, + "learning_rate": 1.0050636715302837e-06, + "loss": 0.9977, + "step": 8992 + }, + { + "epoch": 0.6758605140538103, + "grad_norm": 1.5266522259147237, + "learning_rate": 1.0046413740607626e-06, + "loss": 0.9641, + "step": 8993 + }, + { + "epoch": 0.6759356681196452, + "grad_norm": 1.8080003065514094, + "learning_rate": 1.004219135570052e-06, + "loss": 1.0046, + "step": 8994 + }, + { + "epoch": 0.6760108221854803, + "grad_norm": 1.8620045764674276, + "learning_rate": 1.0037969560831708e-06, + "loss": 0.9939, + "step": 8995 + }, + { + "epoch": 0.6760859762513152, + "grad_norm": 3.2829226635583093, + "learning_rate": 1.0033748356251343e-06, + "loss": 0.9331, + "step": 8996 + }, + { + "epoch": 0.6761611303171502, + "grad_norm": 1.5712553012366988, + "learning_rate": 1.0029527742209547e-06, + "loss": 0.9324, + "step": 8997 + }, + { + "epoch": 0.6762362843829851, + "grad_norm": 1.6850225134684065, + "learning_rate": 1.0025307718956417e-06, + "loss": 1.0645, + "step": 8998 + }, + { + "epoch": 0.6763114384488201, + "grad_norm": 1.7840624908057043, + "learning_rate": 1.0021088286742003e-06, + "loss": 0.9647, + "step": 8999 + }, + { + "epoch": 0.676386592514655, + "grad_norm": 2.5946380749460447, + "learning_rate": 1.0016869445816318e-06, + "loss": 1.0445, + "step": 9000 + }, + { + "epoch": 0.67646174658049, + "grad_norm": 3.8779268593779035, + "learning_rate": 1.0012651196429347e-06, + "loss": 0.9037, + "step": 9001 + }, + { + "epoch": 0.676536900646325, + "grad_norm": 2.583139801979805, + "learning_rate": 1.0008433538831028e-06, + "loss": 0.9834, + "step": 9002 + }, + { + "epoch": 0.6766120547121599, + "grad_norm": 2.022150285661655, + "learning_rate": 1.000421647327129e-06, + "loss": 1.0109, + "step": 9003 + }, + { + "epoch": 0.6766872087779949, + "grad_norm": 1.699832919268884, + "learning_rate": 9.999999999999995e-07, + "loss": 0.9825, + "step": 9004 + }, + { + "epoch": 0.6767623628438298, + "grad_norm": 1.392198936338365, + "learning_rate": 9.995784119267e-07, + "loss": 0.9446, + "step": 9005 + }, + { + "epoch": 0.6768375169096648, + "grad_norm": 1.914036840150069, + "learning_rate": 9.991568831322105e-07, + "loss": 0.9031, + "step": 9006 + }, + { + "epoch": 0.6769126709754998, + "grad_norm": 1.6200247675568293, + "learning_rate": 9.987354136415083e-07, + "loss": 0.9394, + "step": 9007 + }, + { + "epoch": 0.6769878250413347, + "grad_norm": 2.348612103598929, + "learning_rate": 9.983140034795667e-07, + "loss": 1.0444, + "step": 9008 + }, + { + "epoch": 0.6770629791071697, + "grad_norm": 1.8739441759961144, + "learning_rate": 9.978926526713556e-07, + "loss": 0.8305, + "step": 9009 + }, + { + "epoch": 0.6771381331730046, + "grad_norm": 1.5581790687920798, + "learning_rate": 9.974713612418427e-07, + "loss": 0.9123, + "step": 9010 + }, + { + "epoch": 0.6772132872388397, + "grad_norm": 1.4578628416865176, + "learning_rate": 9.970501292159904e-07, + "loss": 0.8944, + "step": 9011 + }, + { + "epoch": 0.6772884413046746, + "grad_norm": 1.6018606513158786, + "learning_rate": 9.96628956618759e-07, + "loss": 0.9921, + "step": 9012 + }, + { + "epoch": 0.6773635953705095, + "grad_norm": 1.7319647047089453, + "learning_rate": 9.962078434751038e-07, + "loss": 0.8761, + "step": 9013 + }, + { + "epoch": 0.6774387494363445, + "grad_norm": 0.8606898219810157, + "learning_rate": 9.957867898099768e-07, + "loss": 0.9374, + "step": 9014 + }, + { + "epoch": 0.6775139035021794, + "grad_norm": 3.5064944911311366, + "learning_rate": 9.953657956483287e-07, + "loss": 1.0573, + "step": 9015 + }, + { + "epoch": 0.6775890575680145, + "grad_norm": 1.9292848607129651, + "learning_rate": 9.949448610151043e-07, + "loss": 0.9596, + "step": 9016 + }, + { + "epoch": 0.6776642116338494, + "grad_norm": 1.4684654122456102, + "learning_rate": 9.945239859352455e-07, + "loss": 0.9457, + "step": 9017 + }, + { + "epoch": 0.6777393656996844, + "grad_norm": 1.9858531702896831, + "learning_rate": 9.941031704336908e-07, + "loss": 1.0558, + "step": 9018 + }, + { + "epoch": 0.6778145197655193, + "grad_norm": 2.0240449325037093, + "learning_rate": 9.936824145353742e-07, + "loss": 1.0283, + "step": 9019 + }, + { + "epoch": 0.6778896738313542, + "grad_norm": 1.6814954387093624, + "learning_rate": 9.932617182652288e-07, + "loss": 0.9822, + "step": 9020 + }, + { + "epoch": 0.6779648278971893, + "grad_norm": 1.8485279567380206, + "learning_rate": 9.928410816481808e-07, + "loss": 1.0171, + "step": 9021 + }, + { + "epoch": 0.6780399819630242, + "grad_norm": 1.758189928013333, + "learning_rate": 9.924205047091572e-07, + "loss": 0.9697, + "step": 9022 + }, + { + "epoch": 0.6781151360288592, + "grad_norm": 1.8646389475223641, + "learning_rate": 9.91999987473076e-07, + "loss": 1.0067, + "step": 9023 + }, + { + "epoch": 0.6781902900946941, + "grad_norm": 3.201281654472503, + "learning_rate": 9.915795299648545e-07, + "loss": 1.0795, + "step": 9024 + }, + { + "epoch": 0.678265444160529, + "grad_norm": 1.6320498935187895, + "learning_rate": 9.911591322094085e-07, + "loss": 0.9644, + "step": 9025 + }, + { + "epoch": 0.678340598226364, + "grad_norm": 1.3808210440445459, + "learning_rate": 9.90738794231646e-07, + "loss": 0.9785, + "step": 9026 + }, + { + "epoch": 0.678415752292199, + "grad_norm": 1.7360830745552478, + "learning_rate": 9.903185160564756e-07, + "loss": 0.959, + "step": 9027 + }, + { + "epoch": 0.678490906358034, + "grad_norm": 2.118573400288715, + "learning_rate": 9.89898297708799e-07, + "loss": 1.1084, + "step": 9028 + }, + { + "epoch": 0.6785660604238689, + "grad_norm": 1.8993747108913652, + "learning_rate": 9.894781392135167e-07, + "loss": 0.9656, + "step": 9029 + }, + { + "epoch": 0.6786412144897039, + "grad_norm": 2.473589246556742, + "learning_rate": 9.89058040595524e-07, + "loss": 0.9862, + "step": 9030 + }, + { + "epoch": 0.6787163685555389, + "grad_norm": 1.5966619181446244, + "learning_rate": 9.88638001879713e-07, + "loss": 0.9145, + "step": 9031 + }, + { + "epoch": 0.6787915226213738, + "grad_norm": 2.3612823860528307, + "learning_rate": 9.882180230909736e-07, + "loss": 1.0433, + "step": 9032 + }, + { + "epoch": 0.6788666766872088, + "grad_norm": 1.7071619049281945, + "learning_rate": 9.877981042541908e-07, + "loss": 1.0876, + "step": 9033 + }, + { + "epoch": 0.6789418307530437, + "grad_norm": 11.55298317700177, + "learning_rate": 9.873782453942462e-07, + "loss": 0.9652, + "step": 9034 + }, + { + "epoch": 0.6790169848188787, + "grad_norm": 1.4792246560921427, + "learning_rate": 9.869584465360184e-07, + "loss": 0.9748, + "step": 9035 + }, + { + "epoch": 0.6790921388847136, + "grad_norm": 1.71850424656858, + "learning_rate": 9.86538707704381e-07, + "loss": 0.9686, + "step": 9036 + }, + { + "epoch": 0.6791672929505487, + "grad_norm": 4.6945320788348, + "learning_rate": 9.861190289242067e-07, + "loss": 0.9886, + "step": 9037 + }, + { + "epoch": 0.6792424470163836, + "grad_norm": 1.9508366721655774, + "learning_rate": 9.856994102203623e-07, + "loss": 0.9203, + "step": 9038 + }, + { + "epoch": 0.6793176010822185, + "grad_norm": 1.8316375374437481, + "learning_rate": 9.852798516177119e-07, + "loss": 0.9793, + "step": 9039 + }, + { + "epoch": 0.6793927551480535, + "grad_norm": 1.3006122926149009, + "learning_rate": 9.848603531411159e-07, + "loss": 1.0349, + "step": 9040 + }, + { + "epoch": 0.6794679092138884, + "grad_norm": 1.719245769784529, + "learning_rate": 9.844409148154304e-07, + "loss": 0.9841, + "step": 9041 + }, + { + "epoch": 0.6795430632797235, + "grad_norm": 1.6088653053630007, + "learning_rate": 9.840215366655104e-07, + "loss": 0.9363, + "step": 9042 + }, + { + "epoch": 0.6796182173455584, + "grad_norm": 1.6534464157691795, + "learning_rate": 9.83602218716204e-07, + "loss": 0.9508, + "step": 9043 + }, + { + "epoch": 0.6796933714113934, + "grad_norm": 2.0850239442689977, + "learning_rate": 9.831829609923596e-07, + "loss": 0.9489, + "step": 9044 + }, + { + "epoch": 0.6797685254772283, + "grad_norm": 1.5525653099291699, + "learning_rate": 9.82763763518818e-07, + "loss": 0.8756, + "step": 9045 + }, + { + "epoch": 0.6798436795430632, + "grad_norm": 0.7871157738210165, + "learning_rate": 9.823446263204175e-07, + "loss": 0.9306, + "step": 9046 + }, + { + "epoch": 0.6799188336088983, + "grad_norm": 11.121113644827656, + "learning_rate": 9.819255494219957e-07, + "loss": 1.0359, + "step": 9047 + }, + { + "epoch": 0.6799939876747332, + "grad_norm": 5.64291850879268, + "learning_rate": 9.815065328483827e-07, + "loss": 0.9978, + "step": 9048 + }, + { + "epoch": 0.6800691417405682, + "grad_norm": 1.3205162658556244, + "learning_rate": 9.810875766244086e-07, + "loss": 1.0415, + "step": 9049 + }, + { + "epoch": 0.6801442958064031, + "grad_norm": 1.8880222153932622, + "learning_rate": 9.806686807748972e-07, + "loss": 0.9279, + "step": 9050 + }, + { + "epoch": 0.680219449872238, + "grad_norm": 1.7347405538140974, + "learning_rate": 9.802498453246697e-07, + "loss": 1.02, + "step": 9051 + }, + { + "epoch": 0.6802946039380731, + "grad_norm": 1.5576596627976251, + "learning_rate": 9.79831070298544e-07, + "loss": 0.9118, + "step": 9052 + }, + { + "epoch": 0.680369758003908, + "grad_norm": 1.8197934962064923, + "learning_rate": 9.794123557213328e-07, + "loss": 1.0032, + "step": 9053 + }, + { + "epoch": 0.680444912069743, + "grad_norm": 0.766025049442439, + "learning_rate": 9.789937016178485e-07, + "loss": 0.8903, + "step": 9054 + }, + { + "epoch": 0.6805200661355779, + "grad_norm": 1.40626593686716, + "learning_rate": 9.78575108012897e-07, + "loss": 1.0694, + "step": 9055 + }, + { + "epoch": 0.6805952202014129, + "grad_norm": 4.169772986583749, + "learning_rate": 9.781565749312816e-07, + "loss": 1.0082, + "step": 9056 + }, + { + "epoch": 0.6806703742672479, + "grad_norm": 2.0391411894014895, + "learning_rate": 9.777381023978022e-07, + "loss": 0.9678, + "step": 9057 + }, + { + "epoch": 0.6807455283330828, + "grad_norm": 1.4520980323401458, + "learning_rate": 9.773196904372539e-07, + "loss": 0.9623, + "step": 9058 + }, + { + "epoch": 0.6808206823989178, + "grad_norm": 1.8836489517053059, + "learning_rate": 9.769013390744307e-07, + "loss": 0.9767, + "step": 9059 + }, + { + "epoch": 0.6808958364647527, + "grad_norm": 2.475274458030488, + "learning_rate": 9.76483048334121e-07, + "loss": 0.9041, + "step": 9060 + }, + { + "epoch": 0.6809709905305877, + "grad_norm": 1.6625378568540996, + "learning_rate": 9.760648182411102e-07, + "loss": 0.9206, + "step": 9061 + }, + { + "epoch": 0.6810461445964227, + "grad_norm": 2.0240715535750455, + "learning_rate": 9.756466488201795e-07, + "loss": 0.8927, + "step": 9062 + }, + { + "epoch": 0.6811212986622577, + "grad_norm": 1.6642835586646156, + "learning_rate": 9.752285400961067e-07, + "loss": 0.9787, + "step": 9063 + }, + { + "epoch": 0.6811964527280926, + "grad_norm": 1.7029119935829902, + "learning_rate": 9.748104920936678e-07, + "loss": 0.9374, + "step": 9064 + }, + { + "epoch": 0.6812716067939275, + "grad_norm": 2.1599225819806938, + "learning_rate": 9.743925048376322e-07, + "loss": 0.9517, + "step": 9065 + }, + { + "epoch": 0.6813467608597625, + "grad_norm": 1.6644652689607369, + "learning_rate": 9.739745783527695e-07, + "loss": 0.9978, + "step": 9066 + }, + { + "epoch": 0.6814219149255974, + "grad_norm": 1.7945624023709563, + "learning_rate": 9.73556712663841e-07, + "loss": 0.9174, + "step": 9067 + }, + { + "epoch": 0.6814970689914325, + "grad_norm": 1.7195353707693553, + "learning_rate": 9.731389077956073e-07, + "loss": 1.0229, + "step": 9068 + }, + { + "epoch": 0.6815722230572674, + "grad_norm": 3.5994106658053457, + "learning_rate": 9.727211637728261e-07, + "loss": 1.0683, + "step": 9069 + }, + { + "epoch": 0.6816473771231023, + "grad_norm": 1.8296309569081308, + "learning_rate": 9.72303480620249e-07, + "loss": 0.9665, + "step": 9070 + }, + { + "epoch": 0.6817225311889373, + "grad_norm": 1.5679710734093086, + "learning_rate": 9.718858583626266e-07, + "loss": 0.9702, + "step": 9071 + }, + { + "epoch": 0.6817976852547722, + "grad_norm": 3.5535230804462437, + "learning_rate": 9.714682970247042e-07, + "loss": 0.9799, + "step": 9072 + }, + { + "epoch": 0.6818728393206073, + "grad_norm": 1.4605560620761975, + "learning_rate": 9.710507966312233e-07, + "loss": 0.9741, + "step": 9073 + }, + { + "epoch": 0.6819479933864422, + "grad_norm": 4.522334094359985, + "learning_rate": 9.706333572069232e-07, + "loss": 1.0197, + "step": 9074 + }, + { + "epoch": 0.6820231474522772, + "grad_norm": 1.7696857174448921, + "learning_rate": 9.702159787765376e-07, + "loss": 0.914, + "step": 9075 + }, + { + "epoch": 0.6820983015181121, + "grad_norm": 1.9004972610273247, + "learning_rate": 9.697986613647992e-07, + "loss": 1.0374, + "step": 9076 + }, + { + "epoch": 0.682173455583947, + "grad_norm": 1.8840313523755408, + "learning_rate": 9.69381404996435e-07, + "loss": 0.9783, + "step": 9077 + }, + { + "epoch": 0.6822486096497821, + "grad_norm": 1.5530027529464265, + "learning_rate": 9.689642096961692e-07, + "loss": 0.9286, + "step": 9078 + }, + { + "epoch": 0.682323763715617, + "grad_norm": 1.6979799161336573, + "learning_rate": 9.68547075488722e-07, + "loss": 0.949, + "step": 9079 + }, + { + "epoch": 0.682398917781452, + "grad_norm": 3.551879041657035, + "learning_rate": 9.681300023988095e-07, + "loss": 0.9217, + "step": 9080 + }, + { + "epoch": 0.6824740718472869, + "grad_norm": 2.008668116079762, + "learning_rate": 9.677129904511462e-07, + "loss": 0.951, + "step": 9081 + }, + { + "epoch": 0.682549225913122, + "grad_norm": 2.209201630045027, + "learning_rate": 9.672960396704416e-07, + "loss": 0.9925, + "step": 9082 + }, + { + "epoch": 0.6826243799789569, + "grad_norm": 1.7815761602172782, + "learning_rate": 9.668791500814007e-07, + "loss": 0.9073, + "step": 9083 + }, + { + "epoch": 0.6826995340447918, + "grad_norm": 2.537687428134158, + "learning_rate": 9.664623217087264e-07, + "loss": 0.9361, + "step": 9084 + }, + { + "epoch": 0.6827746881106268, + "grad_norm": 0.8290882985701941, + "learning_rate": 9.660455545771164e-07, + "loss": 0.9116, + "step": 9085 + }, + { + "epoch": 0.6828498421764617, + "grad_norm": 1.637323308832212, + "learning_rate": 9.656288487112673e-07, + "loss": 0.9755, + "step": 9086 + }, + { + "epoch": 0.6829249962422967, + "grad_norm": 2.0475643249393665, + "learning_rate": 9.652122041358693e-07, + "loss": 0.8378, + "step": 9087 + }, + { + "epoch": 0.6830001503081317, + "grad_norm": 2.988661000176268, + "learning_rate": 9.647956208756113e-07, + "loss": 1.0368, + "step": 9088 + }, + { + "epoch": 0.6830753043739667, + "grad_norm": 2.810797197064584, + "learning_rate": 9.643790989551775e-07, + "loss": 1.0482, + "step": 9089 + }, + { + "epoch": 0.6831504584398016, + "grad_norm": 1.5750563596299474, + "learning_rate": 9.63962638399246e-07, + "loss": 0.9222, + "step": 9090 + }, + { + "epoch": 0.6832256125056365, + "grad_norm": 2.040894251960833, + "learning_rate": 9.635462392324967e-07, + "loss": 1.0203, + "step": 9091 + }, + { + "epoch": 0.6833007665714715, + "grad_norm": 4.079105887708684, + "learning_rate": 9.631299014796003e-07, + "loss": 0.9909, + "step": 9092 + }, + { + "epoch": 0.6833759206373065, + "grad_norm": 1.5765461341454583, + "learning_rate": 9.62713625165229e-07, + "loss": 0.9251, + "step": 9093 + }, + { + "epoch": 0.6834510747031415, + "grad_norm": 2.1257243324407136, + "learning_rate": 9.622974103140468e-07, + "loss": 0.9438, + "step": 9094 + }, + { + "epoch": 0.6835262287689764, + "grad_norm": 4.846283477754158, + "learning_rate": 9.61881256950717e-07, + "loss": 0.8867, + "step": 9095 + }, + { + "epoch": 0.6836013828348113, + "grad_norm": 2.106398226901886, + "learning_rate": 9.614651650998982e-07, + "loss": 1.0481, + "step": 9096 + }, + { + "epoch": 0.6836765369006463, + "grad_norm": 2.8487846326208377, + "learning_rate": 9.610491347862439e-07, + "loss": 0.9379, + "step": 9097 + }, + { + "epoch": 0.6837516909664813, + "grad_norm": 2.305220099509771, + "learning_rate": 9.60633166034408e-07, + "loss": 0.9813, + "step": 9098 + }, + { + "epoch": 0.6838268450323163, + "grad_norm": 1.5759561573068643, + "learning_rate": 9.602172588690368e-07, + "loss": 0.9378, + "step": 9099 + }, + { + "epoch": 0.6839019990981512, + "grad_norm": 2.5048280825222204, + "learning_rate": 9.598014133147744e-07, + "loss": 0.8762, + "step": 9100 + }, + { + "epoch": 0.6839771531639862, + "grad_norm": 2.3135730212848036, + "learning_rate": 9.593856293962619e-07, + "loss": 0.9082, + "step": 9101 + }, + { + "epoch": 0.6840523072298211, + "grad_norm": 1.5187502229164493, + "learning_rate": 9.589699071381346e-07, + "loss": 1.013, + "step": 9102 + }, + { + "epoch": 0.684127461295656, + "grad_norm": 4.521189287414522, + "learning_rate": 9.585542465650274e-07, + "loss": 1.0317, + "step": 9103 + }, + { + "epoch": 0.6842026153614911, + "grad_norm": 1.7368118075872139, + "learning_rate": 9.581386477015691e-07, + "loss": 0.8697, + "step": 9104 + }, + { + "epoch": 0.684277769427326, + "grad_norm": 1.8361687697172637, + "learning_rate": 9.577231105723856e-07, + "loss": 0.9712, + "step": 9105 + }, + { + "epoch": 0.684352923493161, + "grad_norm": 2.8560945631341506, + "learning_rate": 9.573076352020989e-07, + "loss": 1.0599, + "step": 9106 + }, + { + "epoch": 0.6844280775589959, + "grad_norm": 1.4941901703642053, + "learning_rate": 9.568922216153266e-07, + "loss": 0.9725, + "step": 9107 + }, + { + "epoch": 0.684503231624831, + "grad_norm": 3.8311850014877904, + "learning_rate": 9.564768698366855e-07, + "loss": 1.0082, + "step": 9108 + }, + { + "epoch": 0.6845783856906659, + "grad_norm": 1.379129969789569, + "learning_rate": 9.560615798907849e-07, + "loss": 0.9176, + "step": 9109 + }, + { + "epoch": 0.6846535397565008, + "grad_norm": 1.619483828551913, + "learning_rate": 9.55646351802234e-07, + "loss": 0.9119, + "step": 9110 + }, + { + "epoch": 0.6847286938223358, + "grad_norm": 1.884950176157306, + "learning_rate": 9.552311855956364e-07, + "loss": 1.0344, + "step": 9111 + }, + { + "epoch": 0.6848038478881707, + "grad_norm": 2.1987827878484105, + "learning_rate": 9.548160812955905e-07, + "loss": 0.9178, + "step": 9112 + }, + { + "epoch": 0.6848790019540058, + "grad_norm": 2.273417679218526, + "learning_rate": 9.544010389266948e-07, + "loss": 1.0035, + "step": 9113 + }, + { + "epoch": 0.6849541560198407, + "grad_norm": 2.0422769399961145, + "learning_rate": 9.539860585135405e-07, + "loss": 1.0369, + "step": 9114 + }, + { + "epoch": 0.6850293100856756, + "grad_norm": 2.344115165237824, + "learning_rate": 9.535711400807185e-07, + "loss": 0.9685, + "step": 9115 + }, + { + "epoch": 0.6851044641515106, + "grad_norm": 1.8629532691307873, + "learning_rate": 9.531562836528135e-07, + "loss": 0.9746, + "step": 9116 + }, + { + "epoch": 0.6851796182173455, + "grad_norm": 1.6661859136850252, + "learning_rate": 9.527414892544075e-07, + "loss": 0.818, + "step": 9117 + }, + { + "epoch": 0.6852547722831805, + "grad_norm": 1.3275724720255084, + "learning_rate": 9.523267569100784e-07, + "loss": 0.9939, + "step": 9118 + }, + { + "epoch": 0.6853299263490155, + "grad_norm": 1.8317867677147572, + "learning_rate": 9.519120866443997e-07, + "loss": 1.0296, + "step": 9119 + }, + { + "epoch": 0.6854050804148505, + "grad_norm": 2.033564144333754, + "learning_rate": 9.514974784819443e-07, + "loss": 0.9822, + "step": 9120 + }, + { + "epoch": 0.6854802344806854, + "grad_norm": 1.962453491728937, + "learning_rate": 9.510829324472782e-07, + "loss": 1.0299, + "step": 9121 + }, + { + "epoch": 0.6855553885465203, + "grad_norm": 1.6587843308852361, + "learning_rate": 9.50668448564965e-07, + "loss": 0.9222, + "step": 9122 + }, + { + "epoch": 0.6856305426123553, + "grad_norm": 1.8996652785148487, + "learning_rate": 9.502540268595645e-07, + "loss": 1.0088, + "step": 9123 + }, + { + "epoch": 0.6857056966781903, + "grad_norm": 2.254581978193036, + "learning_rate": 9.498396673556317e-07, + "loss": 0.896, + "step": 9124 + }, + { + "epoch": 0.6857808507440253, + "grad_norm": 1.3766244915481802, + "learning_rate": 9.494253700777207e-07, + "loss": 0.9703, + "step": 9125 + }, + { + "epoch": 0.6858560048098602, + "grad_norm": 1.6021854584697457, + "learning_rate": 9.490111350503793e-07, + "loss": 0.9393, + "step": 9126 + }, + { + "epoch": 0.6859311588756952, + "grad_norm": 1.7057911713568372, + "learning_rate": 9.485969622981528e-07, + "loss": 0.9893, + "step": 9127 + }, + { + "epoch": 0.6860063129415301, + "grad_norm": 2.1582017993491402, + "learning_rate": 9.481828518455825e-07, + "loss": 0.9374, + "step": 9128 + }, + { + "epoch": 0.686081467007365, + "grad_norm": 2.5661487188852723, + "learning_rate": 9.477688037172051e-07, + "loss": 0.9798, + "step": 9129 + }, + { + "epoch": 0.6861566210732001, + "grad_norm": 1.9733960265862962, + "learning_rate": 9.473548179375561e-07, + "loss": 0.9572, + "step": 9130 + }, + { + "epoch": 0.686231775139035, + "grad_norm": 1.6673361943441678, + "learning_rate": 9.469408945311641e-07, + "loss": 0.914, + "step": 9131 + }, + { + "epoch": 0.68630692920487, + "grad_norm": 2.6571182627059744, + "learning_rate": 9.465270335225575e-07, + "loss": 1.0916, + "step": 9132 + }, + { + "epoch": 0.6863820832707049, + "grad_norm": 2.8165435545912776, + "learning_rate": 9.46113234936258e-07, + "loss": 0.9881, + "step": 9133 + }, + { + "epoch": 0.68645723733654, + "grad_norm": 1.5577164477826204, + "learning_rate": 9.45699498796785e-07, + "loss": 0.9399, + "step": 9134 + }, + { + "epoch": 0.6865323914023749, + "grad_norm": 2.211388100580805, + "learning_rate": 9.452858251286537e-07, + "loss": 0.9409, + "step": 9135 + }, + { + "epoch": 0.6866075454682098, + "grad_norm": 1.7691293558986074, + "learning_rate": 9.448722139563756e-07, + "loss": 0.8662, + "step": 9136 + }, + { + "epoch": 0.6866826995340448, + "grad_norm": 2.4205165868110856, + "learning_rate": 9.444586653044597e-07, + "loss": 0.9309, + "step": 9137 + }, + { + "epoch": 0.6867578535998797, + "grad_norm": 2.7756897283959447, + "learning_rate": 9.4404517919741e-07, + "loss": 1.0014, + "step": 9138 + }, + { + "epoch": 0.6868330076657148, + "grad_norm": 0.7477596518385601, + "learning_rate": 9.436317556597269e-07, + "loss": 0.8417, + "step": 9139 + }, + { + "epoch": 0.6869081617315497, + "grad_norm": 3.9541676469378904, + "learning_rate": 9.432183947159071e-07, + "loss": 1.0291, + "step": 9140 + }, + { + "epoch": 0.6869833157973846, + "grad_norm": 1.3043528601646028, + "learning_rate": 9.428050963904437e-07, + "loss": 1.0356, + "step": 9141 + }, + { + "epoch": 0.6870584698632196, + "grad_norm": 2.279668390352137, + "learning_rate": 9.423918607078272e-07, + "loss": 1.0024, + "step": 9142 + }, + { + "epoch": 0.6871336239290545, + "grad_norm": 1.838164452029213, + "learning_rate": 9.419786876925428e-07, + "loss": 1.0037, + "step": 9143 + }, + { + "epoch": 0.6872087779948896, + "grad_norm": 0.7248875991244718, + "learning_rate": 9.415655773690727e-07, + "loss": 0.8263, + "step": 9144 + }, + { + "epoch": 0.6872839320607245, + "grad_norm": 1.6342322642796743, + "learning_rate": 9.41152529761895e-07, + "loss": 0.9224, + "step": 9145 + }, + { + "epoch": 0.6873590861265595, + "grad_norm": 1.3009673664184347, + "learning_rate": 9.40739544895484e-07, + "loss": 0.8924, + "step": 9146 + }, + { + "epoch": 0.6874342401923944, + "grad_norm": 2.1903182267952177, + "learning_rate": 9.403266227943116e-07, + "loss": 0.8992, + "step": 9147 + }, + { + "epoch": 0.6875093942582293, + "grad_norm": 2.426303973061917, + "learning_rate": 9.399137634828447e-07, + "loss": 0.8228, + "step": 9148 + }, + { + "epoch": 0.6875845483240643, + "grad_norm": 4.990575300242706, + "learning_rate": 9.395009669855467e-07, + "loss": 1.0066, + "step": 9149 + }, + { + "epoch": 0.6876597023898993, + "grad_norm": 0.7771221592657782, + "learning_rate": 9.390882333268772e-07, + "loss": 0.8464, + "step": 9150 + }, + { + "epoch": 0.6877348564557343, + "grad_norm": 1.6804577857687995, + "learning_rate": 9.386755625312919e-07, + "loss": 0.9317, + "step": 9151 + }, + { + "epoch": 0.6878100105215692, + "grad_norm": 1.931645253545232, + "learning_rate": 9.382629546232442e-07, + "loss": 0.9621, + "step": 9152 + }, + { + "epoch": 0.6878851645874042, + "grad_norm": 1.79507595060962, + "learning_rate": 9.37850409627181e-07, + "loss": 0.9477, + "step": 9153 + }, + { + "epoch": 0.6879603186532391, + "grad_norm": 0.6957880601221985, + "learning_rate": 9.374379275675495e-07, + "loss": 0.8347, + "step": 9154 + }, + { + "epoch": 0.6880354727190741, + "grad_norm": 2.186534559554674, + "learning_rate": 9.370255084687895e-07, + "loss": 0.9074, + "step": 9155 + }, + { + "epoch": 0.6881106267849091, + "grad_norm": 2.102093937837325, + "learning_rate": 9.366131523553385e-07, + "loss": 0.897, + "step": 9156 + }, + { + "epoch": 0.688185780850744, + "grad_norm": 1.679878969147083, + "learning_rate": 9.362008592516302e-07, + "loss": 0.9946, + "step": 9157 + }, + { + "epoch": 0.688260934916579, + "grad_norm": 1.7391629501081285, + "learning_rate": 9.357886291820938e-07, + "loss": 1.0192, + "step": 9158 + }, + { + "epoch": 0.6883360889824139, + "grad_norm": 4.500988110034818, + "learning_rate": 9.353764621711568e-07, + "loss": 0.9837, + "step": 9159 + }, + { + "epoch": 0.6884112430482489, + "grad_norm": 2.1988300637246625, + "learning_rate": 9.349643582432414e-07, + "loss": 0.9786, + "step": 9160 + }, + { + "epoch": 0.6884863971140839, + "grad_norm": 1.6360510873209604, + "learning_rate": 9.345523174227658e-07, + "loss": 0.9544, + "step": 9161 + }, + { + "epoch": 0.6885615511799188, + "grad_norm": 1.7293434359113964, + "learning_rate": 9.341403397341457e-07, + "loss": 0.9515, + "step": 9162 + }, + { + "epoch": 0.6886367052457538, + "grad_norm": 2.8452603866941284, + "learning_rate": 9.337284252017907e-07, + "loss": 0.9174, + "step": 9163 + }, + { + "epoch": 0.6887118593115887, + "grad_norm": 1.5229912519998967, + "learning_rate": 9.333165738501105e-07, + "loss": 1.0001, + "step": 9164 + }, + { + "epoch": 0.6887870133774238, + "grad_norm": 1.897159978584959, + "learning_rate": 9.32904785703508e-07, + "loss": 0.8997, + "step": 9165 + }, + { + "epoch": 0.6888621674432587, + "grad_norm": 1.5152913149003564, + "learning_rate": 9.32493060786383e-07, + "loss": 0.9644, + "step": 9166 + }, + { + "epoch": 0.6889373215090936, + "grad_norm": 1.3353743230540713, + "learning_rate": 9.32081399123132e-07, + "loss": 0.8966, + "step": 9167 + }, + { + "epoch": 0.6890124755749286, + "grad_norm": 1.2653448948725567, + "learning_rate": 9.316698007381467e-07, + "loss": 0.9946, + "step": 9168 + }, + { + "epoch": 0.6890876296407635, + "grad_norm": 2.0781889023671836, + "learning_rate": 9.312582656558173e-07, + "loss": 1.0091, + "step": 9169 + }, + { + "epoch": 0.6891627837065986, + "grad_norm": 0.7154421776866647, + "learning_rate": 9.30846793900528e-07, + "loss": 0.8101, + "step": 9170 + }, + { + "epoch": 0.6892379377724335, + "grad_norm": 1.7827785692655358, + "learning_rate": 9.304353854966605e-07, + "loss": 0.889, + "step": 9171 + }, + { + "epoch": 0.6893130918382685, + "grad_norm": 3.0217578569368313, + "learning_rate": 9.300240404685917e-07, + "loss": 0.9195, + "step": 9172 + }, + { + "epoch": 0.6893882459041034, + "grad_norm": 3.290082397356711, + "learning_rate": 9.296127588406952e-07, + "loss": 1.0998, + "step": 9173 + }, + { + "epoch": 0.6894633999699383, + "grad_norm": 1.8609689243843954, + "learning_rate": 9.292015406373423e-07, + "loss": 0.8965, + "step": 9174 + }, + { + "epoch": 0.6895385540357734, + "grad_norm": 0.6650512290883579, + "learning_rate": 9.287903858828976e-07, + "loss": 0.8693, + "step": 9175 + }, + { + "epoch": 0.6896137081016083, + "grad_norm": 0.7252727669578054, + "learning_rate": 9.283792946017253e-07, + "loss": 0.7992, + "step": 9176 + }, + { + "epoch": 0.6896888621674433, + "grad_norm": 1.6412210744036064, + "learning_rate": 9.279682668181835e-07, + "loss": 0.9219, + "step": 9177 + }, + { + "epoch": 0.6897640162332782, + "grad_norm": 3.2320211940108248, + "learning_rate": 9.275573025566266e-07, + "loss": 0.9596, + "step": 9178 + }, + { + "epoch": 0.6898391702991132, + "grad_norm": 1.6429833310396726, + "learning_rate": 9.271464018414064e-07, + "loss": 0.9543, + "step": 9179 + }, + { + "epoch": 0.6899143243649482, + "grad_norm": 4.868491890447546, + "learning_rate": 9.267355646968694e-07, + "loss": 0.9112, + "step": 9180 + }, + { + "epoch": 0.6899894784307831, + "grad_norm": 2.5232357735799256, + "learning_rate": 9.263247911473606e-07, + "loss": 1.0959, + "step": 9181 + }, + { + "epoch": 0.6900646324966181, + "grad_norm": 1.8946795533328333, + "learning_rate": 9.259140812172192e-07, + "loss": 0.9731, + "step": 9182 + }, + { + "epoch": 0.690139786562453, + "grad_norm": 1.7679721456335804, + "learning_rate": 9.255034349307818e-07, + "loss": 1.0144, + "step": 9183 + }, + { + "epoch": 0.690214940628288, + "grad_norm": 1.844666657250095, + "learning_rate": 9.250928523123802e-07, + "loss": 0.9425, + "step": 9184 + }, + { + "epoch": 0.690290094694123, + "grad_norm": 1.5269664874093551, + "learning_rate": 9.246823333863425e-07, + "loss": 0.9914, + "step": 9185 + }, + { + "epoch": 0.6903652487599579, + "grad_norm": 0.6815083328893512, + "learning_rate": 9.242718781769949e-07, + "loss": 0.8442, + "step": 9186 + }, + { + "epoch": 0.6904404028257929, + "grad_norm": 0.743385833253336, + "learning_rate": 9.238614867086578e-07, + "loss": 0.8818, + "step": 9187 + }, + { + "epoch": 0.6905155568916278, + "grad_norm": 3.2611204775044462, + "learning_rate": 9.234511590056484e-07, + "loss": 1.012, + "step": 9188 + }, + { + "epoch": 0.6905907109574628, + "grad_norm": 2.280210702429305, + "learning_rate": 9.230408950922801e-07, + "loss": 0.9869, + "step": 9189 + }, + { + "epoch": 0.6906658650232977, + "grad_norm": 1.610689829778746, + "learning_rate": 9.226306949928622e-07, + "loss": 0.8685, + "step": 9190 + }, + { + "epoch": 0.6907410190891328, + "grad_norm": 1.4385796514528553, + "learning_rate": 9.222205587317015e-07, + "loss": 0.9472, + "step": 9191 + }, + { + "epoch": 0.6908161731549677, + "grad_norm": 1.724182075971231, + "learning_rate": 9.218104863330996e-07, + "loss": 0.9784, + "step": 9192 + }, + { + "epoch": 0.6908913272208026, + "grad_norm": 2.0242843917633144, + "learning_rate": 9.214004778213562e-07, + "loss": 1.0111, + "step": 9193 + }, + { + "epoch": 0.6909664812866376, + "grad_norm": 3.07324017044937, + "learning_rate": 9.209905332207639e-07, + "loss": 0.8336, + "step": 9194 + }, + { + "epoch": 0.6910416353524725, + "grad_norm": 2.370899274162675, + "learning_rate": 9.205806525556136e-07, + "loss": 0.9354, + "step": 9195 + }, + { + "epoch": 0.6911167894183076, + "grad_norm": 2.0408060504693544, + "learning_rate": 9.20170835850194e-07, + "loss": 0.9489, + "step": 9196 + }, + { + "epoch": 0.6911919434841425, + "grad_norm": 4.205700166993996, + "learning_rate": 9.197610831287863e-07, + "loss": 0.8475, + "step": 9197 + }, + { + "epoch": 0.6912670975499775, + "grad_norm": 1.7885089001091532, + "learning_rate": 9.193513944156719e-07, + "loss": 0.9476, + "step": 9198 + }, + { + "epoch": 0.6913422516158124, + "grad_norm": 1.9466298862906193, + "learning_rate": 9.189417697351254e-07, + "loss": 0.8758, + "step": 9199 + }, + { + "epoch": 0.6914174056816473, + "grad_norm": 1.7987992970893407, + "learning_rate": 9.185322091114187e-07, + "loss": 0.9743, + "step": 9200 + }, + { + "epoch": 0.6914925597474824, + "grad_norm": 0.8527454943419064, + "learning_rate": 9.181227125688197e-07, + "loss": 0.887, + "step": 9201 + }, + { + "epoch": 0.6915677138133173, + "grad_norm": 1.4086045368985416, + "learning_rate": 9.177132801315921e-07, + "loss": 1.0231, + "step": 9202 + }, + { + "epoch": 0.6916428678791523, + "grad_norm": 0.7503369289808013, + "learning_rate": 9.173039118239978e-07, + "loss": 0.838, + "step": 9203 + }, + { + "epoch": 0.6917180219449872, + "grad_norm": 2.3002927096323615, + "learning_rate": 9.168946076702926e-07, + "loss": 0.9646, + "step": 9204 + }, + { + "epoch": 0.6917931760108221, + "grad_norm": 1.6210416854402785, + "learning_rate": 9.164853676947293e-07, + "loss": 1.0027, + "step": 9205 + }, + { + "epoch": 0.6918683300766572, + "grad_norm": 2.158725257647807, + "learning_rate": 9.160761919215572e-07, + "loss": 1.0052, + "step": 9206 + }, + { + "epoch": 0.6919434841424921, + "grad_norm": 2.614864400115932, + "learning_rate": 9.156670803750203e-07, + "loss": 1.0012, + "step": 9207 + }, + { + "epoch": 0.6920186382083271, + "grad_norm": 1.672284227716402, + "learning_rate": 9.15258033079362e-07, + "loss": 0.985, + "step": 9208 + }, + { + "epoch": 0.692093792274162, + "grad_norm": 0.7146119122216696, + "learning_rate": 9.148490500588191e-07, + "loss": 0.8441, + "step": 9209 + }, + { + "epoch": 0.692168946339997, + "grad_norm": 1.5258613280237487, + "learning_rate": 9.144401313376253e-07, + "loss": 0.9516, + "step": 9210 + }, + { + "epoch": 0.692244100405832, + "grad_norm": 1.5303062333205952, + "learning_rate": 9.140312769400105e-07, + "loss": 0.9179, + "step": 9211 + }, + { + "epoch": 0.6923192544716669, + "grad_norm": 1.8028684754260524, + "learning_rate": 9.136224868902003e-07, + "loss": 0.9601, + "step": 9212 + }, + { + "epoch": 0.6923944085375019, + "grad_norm": 1.537885651642168, + "learning_rate": 9.132137612124184e-07, + "loss": 1.0484, + "step": 9213 + }, + { + "epoch": 0.6924695626033368, + "grad_norm": 1.5829782924452966, + "learning_rate": 9.12805099930882e-07, + "loss": 0.9605, + "step": 9214 + }, + { + "epoch": 0.6925447166691718, + "grad_norm": 1.7126700630114715, + "learning_rate": 9.123965030698082e-07, + "loss": 0.9505, + "step": 9215 + }, + { + "epoch": 0.6926198707350067, + "grad_norm": 2.979421129007622, + "learning_rate": 9.119879706534054e-07, + "loss": 0.9263, + "step": 9216 + }, + { + "epoch": 0.6926950248008418, + "grad_norm": 1.661297645890211, + "learning_rate": 9.11579502705881e-07, + "loss": 0.9119, + "step": 9217 + }, + { + "epoch": 0.6927701788666767, + "grad_norm": 2.464149824783037, + "learning_rate": 9.111710992514397e-07, + "loss": 1.0495, + "step": 9218 + }, + { + "epoch": 0.6928453329325116, + "grad_norm": 1.88819388462495, + "learning_rate": 9.107627603142793e-07, + "loss": 0.8882, + "step": 9219 + }, + { + "epoch": 0.6929204869983466, + "grad_norm": 1.812751686774066, + "learning_rate": 9.103544859185972e-07, + "loss": 0.9818, + "step": 9220 + }, + { + "epoch": 0.6929956410641815, + "grad_norm": 2.10945084046966, + "learning_rate": 9.099462760885843e-07, + "loss": 0.9934, + "step": 9221 + }, + { + "epoch": 0.6930707951300166, + "grad_norm": 1.5829313755113477, + "learning_rate": 9.095381308484284e-07, + "loss": 0.8357, + "step": 9222 + }, + { + "epoch": 0.6931459491958515, + "grad_norm": 1.687207938457496, + "learning_rate": 9.091300502223142e-07, + "loss": 1.0251, + "step": 9223 + }, + { + "epoch": 0.6932211032616865, + "grad_norm": 1.8880597200444873, + "learning_rate": 9.087220342344209e-07, + "loss": 0.9602, + "step": 9224 + }, + { + "epoch": 0.6932962573275214, + "grad_norm": 1.8501565480366904, + "learning_rate": 9.083140829089266e-07, + "loss": 0.9935, + "step": 9225 + }, + { + "epoch": 0.6933714113933563, + "grad_norm": 1.640785572731798, + "learning_rate": 9.079061962700032e-07, + "loss": 0.9523, + "step": 9226 + }, + { + "epoch": 0.6934465654591914, + "grad_norm": 1.5198465133019097, + "learning_rate": 9.074983743418196e-07, + "loss": 1.0301, + "step": 9227 + }, + { + "epoch": 0.6935217195250263, + "grad_norm": 1.6567772080187784, + "learning_rate": 9.070906171485408e-07, + "loss": 1.0327, + "step": 9228 + }, + { + "epoch": 0.6935968735908613, + "grad_norm": 0.8672463336953956, + "learning_rate": 9.066829247143273e-07, + "loss": 0.7869, + "step": 9229 + }, + { + "epoch": 0.6936720276566962, + "grad_norm": 1.8190485941858847, + "learning_rate": 9.062752970633376e-07, + "loss": 1.0111, + "step": 9230 + }, + { + "epoch": 0.6937471817225311, + "grad_norm": 1.7365390919756443, + "learning_rate": 9.058677342197249e-07, + "loss": 0.895, + "step": 9231 + }, + { + "epoch": 0.6938223357883662, + "grad_norm": 1.5700589255190553, + "learning_rate": 9.054602362076387e-07, + "loss": 0.9744, + "step": 9232 + }, + { + "epoch": 0.6938974898542011, + "grad_norm": 1.602015510317313, + "learning_rate": 9.050528030512246e-07, + "loss": 0.9302, + "step": 9233 + }, + { + "epoch": 0.6939726439200361, + "grad_norm": 8.943699107368523, + "learning_rate": 9.046454347746242e-07, + "loss": 0.979, + "step": 9234 + }, + { + "epoch": 0.694047797985871, + "grad_norm": 1.6167451557625403, + "learning_rate": 9.042381314019766e-07, + "loss": 0.8143, + "step": 9235 + }, + { + "epoch": 0.694122952051706, + "grad_norm": 1.5178255446534972, + "learning_rate": 9.038308929574152e-07, + "loss": 1.0177, + "step": 9236 + }, + { + "epoch": 0.694198106117541, + "grad_norm": 1.567639937656716, + "learning_rate": 9.034237194650724e-07, + "loss": 0.9406, + "step": 9237 + }, + { + "epoch": 0.6942732601833759, + "grad_norm": 1.806560122784151, + "learning_rate": 9.030166109490724e-07, + "loss": 0.9325, + "step": 9238 + }, + { + "epoch": 0.6943484142492109, + "grad_norm": 1.706237608476853, + "learning_rate": 9.026095674335384e-07, + "loss": 1.018, + "step": 9239 + }, + { + "epoch": 0.6944235683150458, + "grad_norm": 2.068783061982079, + "learning_rate": 9.022025889425902e-07, + "loss": 1.0061, + "step": 9240 + }, + { + "epoch": 0.6944987223808808, + "grad_norm": 1.7631969129196114, + "learning_rate": 9.017956755003415e-07, + "loss": 0.8678, + "step": 9241 + }, + { + "epoch": 0.6945738764467158, + "grad_norm": 0.7624642660804246, + "learning_rate": 9.013888271309053e-07, + "loss": 0.832, + "step": 9242 + }, + { + "epoch": 0.6946490305125508, + "grad_norm": 0.9426321697905864, + "learning_rate": 9.00982043858388e-07, + "loss": 0.8442, + "step": 9243 + }, + { + "epoch": 0.6947241845783857, + "grad_norm": 3.535010437039476, + "learning_rate": 9.005753257068929e-07, + "loss": 0.9707, + "step": 9244 + }, + { + "epoch": 0.6947993386442206, + "grad_norm": 1.9904172324800742, + "learning_rate": 9.001686727005196e-07, + "loss": 1.0121, + "step": 9245 + }, + { + "epoch": 0.6948744927100556, + "grad_norm": 1.5006815633476454, + "learning_rate": 8.997620848633634e-07, + "loss": 0.9457, + "step": 9246 + }, + { + "epoch": 0.6949496467758906, + "grad_norm": 1.6100938543365535, + "learning_rate": 8.993555622195175e-07, + "loss": 0.7962, + "step": 9247 + }, + { + "epoch": 0.6950248008417256, + "grad_norm": 1.9566134008644898, + "learning_rate": 8.98949104793069e-07, + "loss": 1.005, + "step": 9248 + }, + { + "epoch": 0.6950999549075605, + "grad_norm": 0.7618579835063213, + "learning_rate": 8.985427126081024e-07, + "loss": 0.8948, + "step": 9249 + }, + { + "epoch": 0.6951751089733954, + "grad_norm": 0.7315117636469364, + "learning_rate": 8.981363856886979e-07, + "loss": 0.8771, + "step": 9250 + }, + { + "epoch": 0.6952502630392304, + "grad_norm": 1.655485372563395, + "learning_rate": 8.977301240589313e-07, + "loss": 0.9985, + "step": 9251 + }, + { + "epoch": 0.6953254171050653, + "grad_norm": 1.451406139086718, + "learning_rate": 8.973239277428761e-07, + "loss": 0.9612, + "step": 9252 + }, + { + "epoch": 0.6954005711709004, + "grad_norm": 1.5898404507989417, + "learning_rate": 8.969177967646007e-07, + "loss": 0.9003, + "step": 9253 + }, + { + "epoch": 0.6954757252367353, + "grad_norm": 3.1392787496190997, + "learning_rate": 8.965117311481698e-07, + "loss": 1.0053, + "step": 9254 + }, + { + "epoch": 0.6955508793025703, + "grad_norm": 1.4072862833495654, + "learning_rate": 8.961057309176445e-07, + "loss": 0.9779, + "step": 9255 + }, + { + "epoch": 0.6956260333684052, + "grad_norm": 1.438279148081622, + "learning_rate": 8.956997960970809e-07, + "loss": 0.9964, + "step": 9256 + }, + { + "epoch": 0.6957011874342401, + "grad_norm": 2.4154109213497876, + "learning_rate": 8.952939267105339e-07, + "loss": 1.0238, + "step": 9257 + }, + { + "epoch": 0.6957763415000752, + "grad_norm": 1.621962937045659, + "learning_rate": 8.94888122782051e-07, + "loss": 0.9762, + "step": 9258 + }, + { + "epoch": 0.6958514955659101, + "grad_norm": 1.5819736799903243, + "learning_rate": 8.944823843356795e-07, + "loss": 1.0118, + "step": 9259 + }, + { + "epoch": 0.6959266496317451, + "grad_norm": 2.029726131738181, + "learning_rate": 8.940767113954608e-07, + "loss": 1.0431, + "step": 9260 + }, + { + "epoch": 0.69600180369758, + "grad_norm": 0.7275456985068585, + "learning_rate": 8.936711039854301e-07, + "loss": 0.7623, + "step": 9261 + }, + { + "epoch": 0.696076957763415, + "grad_norm": 1.745138090293463, + "learning_rate": 8.932655621296239e-07, + "loss": 0.9419, + "step": 9262 + }, + { + "epoch": 0.69615211182925, + "grad_norm": 4.145694493919452, + "learning_rate": 8.928600858520703e-07, + "loss": 1.0511, + "step": 9263 + }, + { + "epoch": 0.6962272658950849, + "grad_norm": 5.430002414709858, + "learning_rate": 8.924546751767968e-07, + "loss": 0.9744, + "step": 9264 + }, + { + "epoch": 0.6963024199609199, + "grad_norm": 3.3623140475109334, + "learning_rate": 8.920493301278249e-07, + "loss": 0.9361, + "step": 9265 + }, + { + "epoch": 0.6963775740267548, + "grad_norm": 1.685249311188106, + "learning_rate": 8.916440507291727e-07, + "loss": 1.0193, + "step": 9266 + }, + { + "epoch": 0.6964527280925898, + "grad_norm": 2.032461421361126, + "learning_rate": 8.912388370048549e-07, + "loss": 0.9558, + "step": 9267 + }, + { + "epoch": 0.6965278821584248, + "grad_norm": 1.5312106069051947, + "learning_rate": 8.908336889788807e-07, + "loss": 0.8728, + "step": 9268 + }, + { + "epoch": 0.6966030362242598, + "grad_norm": 3.4754645167484783, + "learning_rate": 8.904286066752589e-07, + "loss": 0.9859, + "step": 9269 + }, + { + "epoch": 0.6966781902900947, + "grad_norm": 0.6448234127119943, + "learning_rate": 8.900235901179907e-07, + "loss": 0.8347, + "step": 9270 + }, + { + "epoch": 0.6967533443559296, + "grad_norm": 0.7468086931490925, + "learning_rate": 8.896186393310752e-07, + "loss": 0.8333, + "step": 9271 + }, + { + "epoch": 0.6968284984217646, + "grad_norm": 1.5812825256138663, + "learning_rate": 8.892137543385072e-07, + "loss": 0.9892, + "step": 9272 + }, + { + "epoch": 0.6969036524875996, + "grad_norm": 2.347992579381807, + "learning_rate": 8.888089351642769e-07, + "loss": 0.8949, + "step": 9273 + }, + { + "epoch": 0.6969788065534346, + "grad_norm": 1.9280106521407154, + "learning_rate": 8.884041818323733e-07, + "loss": 0.8983, + "step": 9274 + }, + { + "epoch": 0.6970539606192695, + "grad_norm": 1.9227780337246099, + "learning_rate": 8.879994943667784e-07, + "loss": 1.0103, + "step": 9275 + }, + { + "epoch": 0.6971291146851044, + "grad_norm": 2.031172765583938, + "learning_rate": 8.875948727914713e-07, + "loss": 0.9891, + "step": 9276 + }, + { + "epoch": 0.6972042687509394, + "grad_norm": 1.7569413802866127, + "learning_rate": 8.87190317130428e-07, + "loss": 0.994, + "step": 9277 + }, + { + "epoch": 0.6972794228167744, + "grad_norm": 1.7443734042503085, + "learning_rate": 8.867858274076188e-07, + "loss": 0.9347, + "step": 9278 + }, + { + "epoch": 0.6973545768826094, + "grad_norm": 1.5416173497465415, + "learning_rate": 8.863814036470128e-07, + "loss": 1.0144, + "step": 9279 + }, + { + "epoch": 0.6974297309484443, + "grad_norm": 1.7604251688313626, + "learning_rate": 8.859770458725722e-07, + "loss": 0.9114, + "step": 9280 + }, + { + "epoch": 0.6975048850142793, + "grad_norm": 2.1229283949877726, + "learning_rate": 8.855727541082583e-07, + "loss": 1.0198, + "step": 9281 + }, + { + "epoch": 0.6975800390801142, + "grad_norm": 2.060242428819733, + "learning_rate": 8.85168528378027e-07, + "loss": 1.0443, + "step": 9282 + }, + { + "epoch": 0.6976551931459491, + "grad_norm": 3.8607284087896607, + "learning_rate": 8.847643687058277e-07, + "loss": 0.9491, + "step": 9283 + }, + { + "epoch": 0.6977303472117842, + "grad_norm": 2.609697698866245, + "learning_rate": 8.84360275115611e-07, + "loss": 0.8704, + "step": 9284 + }, + { + "epoch": 0.6978055012776191, + "grad_norm": 2.1137408252158263, + "learning_rate": 8.839562476313192e-07, + "loss": 0.8832, + "step": 9285 + }, + { + "epoch": 0.6978806553434541, + "grad_norm": 1.422248875017236, + "learning_rate": 8.83552286276894e-07, + "loss": 0.9589, + "step": 9286 + }, + { + "epoch": 0.697955809409289, + "grad_norm": 0.7789624483731513, + "learning_rate": 8.831483910762711e-07, + "loss": 0.8644, + "step": 9287 + }, + { + "epoch": 0.698030963475124, + "grad_norm": 0.829560026157231, + "learning_rate": 8.827445620533829e-07, + "loss": 0.9082, + "step": 9288 + }, + { + "epoch": 0.698106117540959, + "grad_norm": 1.853492985899676, + "learning_rate": 8.823407992321574e-07, + "loss": 0.982, + "step": 9289 + }, + { + "epoch": 0.6981812716067939, + "grad_norm": 1.7482381534881468, + "learning_rate": 8.819371026365188e-07, + "loss": 0.98, + "step": 9290 + }, + { + "epoch": 0.6982564256726289, + "grad_norm": 2.326531396046912, + "learning_rate": 8.815334722903889e-07, + "loss": 1.0487, + "step": 9291 + }, + { + "epoch": 0.6983315797384638, + "grad_norm": 1.8369677311197306, + "learning_rate": 8.811299082176837e-07, + "loss": 0.8912, + "step": 9292 + }, + { + "epoch": 0.6984067338042989, + "grad_norm": 1.8696480344330324, + "learning_rate": 8.807264104423158e-07, + "loss": 0.9535, + "step": 9293 + }, + { + "epoch": 0.6984818878701338, + "grad_norm": 2.0107968486453864, + "learning_rate": 8.80322978988194e-07, + "loss": 1.0032, + "step": 9294 + }, + { + "epoch": 0.6985570419359687, + "grad_norm": 1.923453450388589, + "learning_rate": 8.799196138792227e-07, + "loss": 0.9829, + "step": 9295 + }, + { + "epoch": 0.6986321960018037, + "grad_norm": 2.1408631129685145, + "learning_rate": 8.795163151393039e-07, + "loss": 0.8954, + "step": 9296 + }, + { + "epoch": 0.6987073500676386, + "grad_norm": 1.9966019731823856, + "learning_rate": 8.791130827923341e-07, + "loss": 0.9754, + "step": 9297 + }, + { + "epoch": 0.6987825041334736, + "grad_norm": 1.8021211418051815, + "learning_rate": 8.787099168622063e-07, + "loss": 0.9821, + "step": 9298 + }, + { + "epoch": 0.6988576581993086, + "grad_norm": 0.8369611729078547, + "learning_rate": 8.783068173728097e-07, + "loss": 0.9064, + "step": 9299 + }, + { + "epoch": 0.6989328122651436, + "grad_norm": 1.7644069149578006, + "learning_rate": 8.779037843480285e-07, + "loss": 0.953, + "step": 9300 + }, + { + "epoch": 0.6990079663309785, + "grad_norm": 1.4076960440341963, + "learning_rate": 8.775008178117458e-07, + "loss": 1.0087, + "step": 9301 + }, + { + "epoch": 0.6990831203968134, + "grad_norm": 1.7826266406684133, + "learning_rate": 8.770979177878373e-07, + "loss": 0.8988, + "step": 9302 + }, + { + "epoch": 0.6991582744626484, + "grad_norm": 13.32687475818365, + "learning_rate": 8.766950843001776e-07, + "loss": 1.0059, + "step": 9303 + }, + { + "epoch": 0.6992334285284834, + "grad_norm": 1.9897529835962002, + "learning_rate": 8.762923173726358e-07, + "loss": 1.0185, + "step": 9304 + }, + { + "epoch": 0.6993085825943184, + "grad_norm": 2.0477620306712967, + "learning_rate": 8.758896170290768e-07, + "loss": 0.9475, + "step": 9305 + }, + { + "epoch": 0.6993837366601533, + "grad_norm": 1.99024449758604, + "learning_rate": 8.754869832933629e-07, + "loss": 1.0819, + "step": 9306 + }, + { + "epoch": 0.6994588907259883, + "grad_norm": 0.8269746816327478, + "learning_rate": 8.750844161893503e-07, + "loss": 0.9085, + "step": 9307 + }, + { + "epoch": 0.6995340447918232, + "grad_norm": 3.3046303602107985, + "learning_rate": 8.746819157408944e-07, + "loss": 0.8739, + "step": 9308 + }, + { + "epoch": 0.6996091988576582, + "grad_norm": 2.5359618989025736, + "learning_rate": 8.742794819718442e-07, + "loss": 0.8287, + "step": 9309 + }, + { + "epoch": 0.6996843529234932, + "grad_norm": 1.6484052465094774, + "learning_rate": 8.738771149060453e-07, + "loss": 1.0419, + "step": 9310 + }, + { + "epoch": 0.6997595069893281, + "grad_norm": 2.0905323640972733, + "learning_rate": 8.734748145673396e-07, + "loss": 1.0256, + "step": 9311 + }, + { + "epoch": 0.6998346610551631, + "grad_norm": 3.0681184884159807, + "learning_rate": 8.730725809795641e-07, + "loss": 1.0094, + "step": 9312 + }, + { + "epoch": 0.699909815120998, + "grad_norm": 1.5102451446308653, + "learning_rate": 8.726704141665542e-07, + "loss": 0.9228, + "step": 9313 + }, + { + "epoch": 0.6999849691868331, + "grad_norm": 2.7456859215243434, + "learning_rate": 8.722683141521392e-07, + "loss": 0.8407, + "step": 9314 + }, + { + "epoch": 0.700060123252668, + "grad_norm": 2.2887061561039324, + "learning_rate": 8.718662809601447e-07, + "loss": 0.8585, + "step": 9315 + }, + { + "epoch": 0.7001352773185029, + "grad_norm": 3.4742120576746456, + "learning_rate": 8.714643146143932e-07, + "loss": 1.0357, + "step": 9316 + }, + { + "epoch": 0.7002104313843379, + "grad_norm": 1.7349716397476818, + "learning_rate": 8.710624151387018e-07, + "loss": 0.9793, + "step": 9317 + }, + { + "epoch": 0.7002855854501728, + "grad_norm": 1.629861382400121, + "learning_rate": 8.70660582556886e-07, + "loss": 0.9906, + "step": 9318 + }, + { + "epoch": 0.7003607395160079, + "grad_norm": 1.9157021486762045, + "learning_rate": 8.702588168927551e-07, + "loss": 0.9955, + "step": 9319 + }, + { + "epoch": 0.7004358935818428, + "grad_norm": 1.9824728071312894, + "learning_rate": 8.698571181701154e-07, + "loss": 0.924, + "step": 9320 + }, + { + "epoch": 0.7005110476476777, + "grad_norm": 5.658300207356646, + "learning_rate": 8.69455486412769e-07, + "loss": 1.019, + "step": 9321 + }, + { + "epoch": 0.7005862017135127, + "grad_norm": 1.2962406399424795, + "learning_rate": 8.690539216445136e-07, + "loss": 1.0231, + "step": 9322 + }, + { + "epoch": 0.7006613557793476, + "grad_norm": 2.3251375834457417, + "learning_rate": 8.686524238891446e-07, + "loss": 0.9052, + "step": 9323 + }, + { + "epoch": 0.7007365098451827, + "grad_norm": 2.404785304965097, + "learning_rate": 8.682509931704511e-07, + "loss": 0.9404, + "step": 9324 + }, + { + "epoch": 0.7008116639110176, + "grad_norm": 1.969147596628795, + "learning_rate": 8.678496295122208e-07, + "loss": 0.9567, + "step": 9325 + }, + { + "epoch": 0.7008868179768526, + "grad_norm": 2.158788320337111, + "learning_rate": 8.674483329382351e-07, + "loss": 1.0284, + "step": 9326 + }, + { + "epoch": 0.7009619720426875, + "grad_norm": 1.585475008572389, + "learning_rate": 8.670471034722726e-07, + "loss": 0.9416, + "step": 9327 + }, + { + "epoch": 0.7010371261085224, + "grad_norm": 1.690063648492509, + "learning_rate": 8.666459411381075e-07, + "loss": 0.9146, + "step": 9328 + }, + { + "epoch": 0.7011122801743574, + "grad_norm": 1.4733140523578128, + "learning_rate": 8.662448459595095e-07, + "loss": 0.9179, + "step": 9329 + }, + { + "epoch": 0.7011874342401924, + "grad_norm": 2.806603587416544, + "learning_rate": 8.658438179602468e-07, + "loss": 0.9261, + "step": 9330 + }, + { + "epoch": 0.7012625883060274, + "grad_norm": 1.6876853558503027, + "learning_rate": 8.654428571640806e-07, + "loss": 0.9176, + "step": 9331 + }, + { + "epoch": 0.7013377423718623, + "grad_norm": 1.4788085163627565, + "learning_rate": 8.650419635947696e-07, + "loss": 0.9629, + "step": 9332 + }, + { + "epoch": 0.7014128964376973, + "grad_norm": 1.4728399915940031, + "learning_rate": 8.646411372760685e-07, + "loss": 1.0423, + "step": 9333 + }, + { + "epoch": 0.7014880505035322, + "grad_norm": 2.935827712624726, + "learning_rate": 8.642403782317269e-07, + "loss": 0.9455, + "step": 9334 + }, + { + "epoch": 0.7015632045693672, + "grad_norm": 2.1816713336091316, + "learning_rate": 8.638396864854927e-07, + "loss": 0.8973, + "step": 9335 + }, + { + "epoch": 0.7016383586352022, + "grad_norm": 1.8249000077746327, + "learning_rate": 8.634390620611076e-07, + "loss": 0.9842, + "step": 9336 + }, + { + "epoch": 0.7017135127010371, + "grad_norm": 1.6719607037119861, + "learning_rate": 8.630385049823101e-07, + "loss": 0.984, + "step": 9337 + }, + { + "epoch": 0.7017886667668721, + "grad_norm": 2.0123583915598986, + "learning_rate": 8.626380152728352e-07, + "loss": 0.8975, + "step": 9338 + }, + { + "epoch": 0.701863820832707, + "grad_norm": 1.57363093076463, + "learning_rate": 8.622375929564123e-07, + "loss": 1.0437, + "step": 9339 + }, + { + "epoch": 0.701938974898542, + "grad_norm": 1.855421912957373, + "learning_rate": 8.618372380567696e-07, + "loss": 0.9389, + "step": 9340 + }, + { + "epoch": 0.702014128964377, + "grad_norm": 2.0188641211787175, + "learning_rate": 8.614369505976287e-07, + "loss": 0.9835, + "step": 9341 + }, + { + "epoch": 0.7020892830302119, + "grad_norm": 1.7040230019722769, + "learning_rate": 8.610367306027084e-07, + "loss": 1.0019, + "step": 9342 + }, + { + "epoch": 0.7021644370960469, + "grad_norm": 2.1171522524669153, + "learning_rate": 8.606365780957232e-07, + "loss": 0.9512, + "step": 9343 + }, + { + "epoch": 0.7022395911618818, + "grad_norm": 1.8010239179040763, + "learning_rate": 8.602364931003831e-07, + "loss": 0.9817, + "step": 9344 + }, + { + "epoch": 0.7023147452277169, + "grad_norm": 2.465701478417484, + "learning_rate": 8.598364756403957e-07, + "loss": 0.9628, + "step": 9345 + }, + { + "epoch": 0.7023898992935518, + "grad_norm": 1.7530692297348567, + "learning_rate": 8.594365257394626e-07, + "loss": 0.9197, + "step": 9346 + }, + { + "epoch": 0.7024650533593867, + "grad_norm": 1.4239250746635566, + "learning_rate": 8.590366434212835e-07, + "loss": 1.0413, + "step": 9347 + }, + { + "epoch": 0.7025402074252217, + "grad_norm": 1.806186994357906, + "learning_rate": 8.586368287095522e-07, + "loss": 0.9004, + "step": 9348 + }, + { + "epoch": 0.7026153614910566, + "grad_norm": 3.491591298117639, + "learning_rate": 8.582370816279594e-07, + "loss": 0.9292, + "step": 9349 + }, + { + "epoch": 0.7026905155568917, + "grad_norm": 4.48645821807462, + "learning_rate": 8.578374022001917e-07, + "loss": 0.8825, + "step": 9350 + }, + { + "epoch": 0.7027656696227266, + "grad_norm": 1.724392316730596, + "learning_rate": 8.574377904499308e-07, + "loss": 0.9492, + "step": 9351 + }, + { + "epoch": 0.7028408236885616, + "grad_norm": 1.8129789443100537, + "learning_rate": 8.570382464008568e-07, + "loss": 0.9807, + "step": 9352 + }, + { + "epoch": 0.7029159777543965, + "grad_norm": 1.6184679941920004, + "learning_rate": 8.566387700766434e-07, + "loss": 1.0624, + "step": 9353 + }, + { + "epoch": 0.7029911318202314, + "grad_norm": 1.6620972539905787, + "learning_rate": 8.562393615009609e-07, + "loss": 0.9488, + "step": 9354 + }, + { + "epoch": 0.7030662858860665, + "grad_norm": 2.0108885005503185, + "learning_rate": 8.558400206974761e-07, + "loss": 0.9801, + "step": 9355 + }, + { + "epoch": 0.7031414399519014, + "grad_norm": 1.5002074098242066, + "learning_rate": 8.554407476898506e-07, + "loss": 0.9599, + "step": 9356 + }, + { + "epoch": 0.7032165940177364, + "grad_norm": 1.8217807464036166, + "learning_rate": 8.550415425017443e-07, + "loss": 0.9108, + "step": 9357 + }, + { + "epoch": 0.7032917480835713, + "grad_norm": 1.4147541188370358, + "learning_rate": 8.546424051568111e-07, + "loss": 1.0598, + "step": 9358 + }, + { + "epoch": 0.7033669021494063, + "grad_norm": 2.966363208409264, + "learning_rate": 8.542433356787011e-07, + "loss": 0.997, + "step": 9359 + }, + { + "epoch": 0.7034420562152413, + "grad_norm": 1.8371585762302363, + "learning_rate": 8.538443340910608e-07, + "loss": 0.898, + "step": 9360 + }, + { + "epoch": 0.7035172102810762, + "grad_norm": 2.2910736935176086, + "learning_rate": 8.53445400417532e-07, + "loss": 0.9316, + "step": 9361 + }, + { + "epoch": 0.7035923643469112, + "grad_norm": 1.6423937754197675, + "learning_rate": 8.530465346817543e-07, + "loss": 0.9072, + "step": 9362 + }, + { + "epoch": 0.7036675184127461, + "grad_norm": 3.438775398163374, + "learning_rate": 8.526477369073616e-07, + "loss": 0.9447, + "step": 9363 + }, + { + "epoch": 0.7037426724785811, + "grad_norm": 1.6015142666136297, + "learning_rate": 8.522490071179839e-07, + "loss": 0.9578, + "step": 9364 + }, + { + "epoch": 0.703817826544416, + "grad_norm": 1.3391132198240399, + "learning_rate": 8.518503453372477e-07, + "loss": 1.0275, + "step": 9365 + }, + { + "epoch": 0.703892980610251, + "grad_norm": 2.8613142931086326, + "learning_rate": 8.51451751588774e-07, + "loss": 0.9988, + "step": 9366 + }, + { + "epoch": 0.703968134676086, + "grad_norm": 2.1004537046870246, + "learning_rate": 8.510532258961831e-07, + "loss": 0.91, + "step": 9367 + }, + { + "epoch": 0.7040432887419209, + "grad_norm": 2.2460603131003025, + "learning_rate": 8.506547682830876e-07, + "loss": 1.0323, + "step": 9368 + }, + { + "epoch": 0.7041184428077559, + "grad_norm": 2.8359099247540214, + "learning_rate": 8.502563787730987e-07, + "loss": 0.9364, + "step": 9369 + }, + { + "epoch": 0.7041935968735908, + "grad_norm": 2.3045135949682116, + "learning_rate": 8.498580573898219e-07, + "loss": 0.939, + "step": 9370 + }, + { + "epoch": 0.7042687509394259, + "grad_norm": 1.6482685599153644, + "learning_rate": 8.494598041568597e-07, + "loss": 0.9695, + "step": 9371 + }, + { + "epoch": 0.7043439050052608, + "grad_norm": 1.7957500751192104, + "learning_rate": 8.490616190978097e-07, + "loss": 1.0264, + "step": 9372 + }, + { + "epoch": 0.7044190590710957, + "grad_norm": 1.5850618698126686, + "learning_rate": 8.486635022362651e-07, + "loss": 1.0136, + "step": 9373 + }, + { + "epoch": 0.7044942131369307, + "grad_norm": 1.6481835047580247, + "learning_rate": 8.482654535958178e-07, + "loss": 0.9219, + "step": 9374 + }, + { + "epoch": 0.7045693672027656, + "grad_norm": 1.996785083352212, + "learning_rate": 8.478674732000524e-07, + "loss": 1.0325, + "step": 9375 + }, + { + "epoch": 0.7046445212686007, + "grad_norm": 1.6658335113270855, + "learning_rate": 8.474695610725513e-07, + "loss": 0.9954, + "step": 9376 + }, + { + "epoch": 0.7047196753344356, + "grad_norm": 1.9436093883619745, + "learning_rate": 8.470717172368917e-07, + "loss": 1.0245, + "step": 9377 + }, + { + "epoch": 0.7047948294002706, + "grad_norm": 1.4402157382157612, + "learning_rate": 8.466739417166473e-07, + "loss": 0.9622, + "step": 9378 + }, + { + "epoch": 0.7048699834661055, + "grad_norm": 2.1858652818705595, + "learning_rate": 8.462762345353887e-07, + "loss": 0.912, + "step": 9379 + }, + { + "epoch": 0.7049451375319404, + "grad_norm": 1.3906773332328177, + "learning_rate": 8.458785957166812e-07, + "loss": 0.9675, + "step": 9380 + }, + { + "epoch": 0.7050202915977755, + "grad_norm": 1.533297512769389, + "learning_rate": 8.454810252840863e-07, + "loss": 0.9475, + "step": 9381 + }, + { + "epoch": 0.7050954456636104, + "grad_norm": 1.8785320075818568, + "learning_rate": 8.450835232611618e-07, + "loss": 0.9888, + "step": 9382 + }, + { + "epoch": 0.7051705997294454, + "grad_norm": 0.8640001160656886, + "learning_rate": 8.4468608967146e-07, + "loss": 0.8891, + "step": 9383 + }, + { + "epoch": 0.7052457537952803, + "grad_norm": 4.603532612267131, + "learning_rate": 8.442887245385324e-07, + "loss": 0.9804, + "step": 9384 + }, + { + "epoch": 0.7053209078611152, + "grad_norm": 1.7149144464401158, + "learning_rate": 8.438914278859231e-07, + "loss": 0.8789, + "step": 9385 + }, + { + "epoch": 0.7053960619269503, + "grad_norm": 1.5816092976307938, + "learning_rate": 8.434941997371738e-07, + "loss": 0.9407, + "step": 9386 + }, + { + "epoch": 0.7054712159927852, + "grad_norm": 2.16998560114008, + "learning_rate": 8.43097040115822e-07, + "loss": 0.9807, + "step": 9387 + }, + { + "epoch": 0.7055463700586202, + "grad_norm": 8.917253088755851, + "learning_rate": 8.426999490453996e-07, + "loss": 0.9917, + "step": 9388 + }, + { + "epoch": 0.7056215241244551, + "grad_norm": 2.7421570691976926, + "learning_rate": 8.423029265494377e-07, + "loss": 0.9297, + "step": 9389 + }, + { + "epoch": 0.7056966781902901, + "grad_norm": 0.7650621641993652, + "learning_rate": 8.419059726514597e-07, + "loss": 0.8194, + "step": 9390 + }, + { + "epoch": 0.705771832256125, + "grad_norm": 1.733048146274095, + "learning_rate": 8.415090873749882e-07, + "loss": 0.9483, + "step": 9391 + }, + { + "epoch": 0.70584698632196, + "grad_norm": 1.789982854665127, + "learning_rate": 8.411122707435394e-07, + "loss": 0.9086, + "step": 9392 + }, + { + "epoch": 0.705922140387795, + "grad_norm": 7.245945026691411, + "learning_rate": 8.407155227806264e-07, + "loss": 1.0006, + "step": 9393 + }, + { + "epoch": 0.7059972944536299, + "grad_norm": 2.6665821856786107, + "learning_rate": 8.403188435097576e-07, + "loss": 0.9884, + "step": 9394 + }, + { + "epoch": 0.7060724485194649, + "grad_norm": 1.5425679519252182, + "learning_rate": 8.399222329544375e-07, + "loss": 1.0737, + "step": 9395 + }, + { + "epoch": 0.7061476025852998, + "grad_norm": 1.852848169964928, + "learning_rate": 8.395256911381681e-07, + "loss": 0.971, + "step": 9396 + }, + { + "epoch": 0.7062227566511349, + "grad_norm": 1.965343191932596, + "learning_rate": 8.391292180844451e-07, + "loss": 1.054, + "step": 9397 + }, + { + "epoch": 0.7062979107169698, + "grad_norm": 1.622193333544206, + "learning_rate": 8.387328138167613e-07, + "loss": 0.9693, + "step": 9398 + }, + { + "epoch": 0.7063730647828047, + "grad_norm": 1.5604403844982346, + "learning_rate": 8.383364783586051e-07, + "loss": 1.1061, + "step": 9399 + }, + { + "epoch": 0.7064482188486397, + "grad_norm": 1.6962925060459075, + "learning_rate": 8.379402117334601e-07, + "loss": 1.057, + "step": 9400 + }, + { + "epoch": 0.7065233729144746, + "grad_norm": 1.9803912205916991, + "learning_rate": 8.375440139648082e-07, + "loss": 0.9187, + "step": 9401 + }, + { + "epoch": 0.7065985269803097, + "grad_norm": 1.7948710005796862, + "learning_rate": 8.371478850761247e-07, + "loss": 0.9863, + "step": 9402 + }, + { + "epoch": 0.7066736810461446, + "grad_norm": 2.4887970729574014, + "learning_rate": 8.367518250908818e-07, + "loss": 1.0132, + "step": 9403 + }, + { + "epoch": 0.7067488351119796, + "grad_norm": 1.5106500204189899, + "learning_rate": 8.363558340325478e-07, + "loss": 0.9574, + "step": 9404 + }, + { + "epoch": 0.7068239891778145, + "grad_norm": 1.4307013913645437, + "learning_rate": 8.359599119245857e-07, + "loss": 1.0386, + "step": 9405 + }, + { + "epoch": 0.7068991432436494, + "grad_norm": 2.0676573803931086, + "learning_rate": 8.355640587904569e-07, + "loss": 0.9983, + "step": 9406 + }, + { + "epoch": 0.7069742973094845, + "grad_norm": 0.8002067179631702, + "learning_rate": 8.351682746536166e-07, + "loss": 0.8325, + "step": 9407 + }, + { + "epoch": 0.7070494513753194, + "grad_norm": 5.334378418253701, + "learning_rate": 8.347725595375165e-07, + "loss": 1.1093, + "step": 9408 + }, + { + "epoch": 0.7071246054411544, + "grad_norm": 2.152588762852303, + "learning_rate": 8.343769134656043e-07, + "loss": 0.9451, + "step": 9409 + }, + { + "epoch": 0.7071997595069893, + "grad_norm": 1.5048733858856689, + "learning_rate": 8.339813364613224e-07, + "loss": 1.0386, + "step": 9410 + }, + { + "epoch": 0.7072749135728242, + "grad_norm": 2.4954520820078696, + "learning_rate": 8.335858285481124e-07, + "loss": 0.9535, + "step": 9411 + }, + { + "epoch": 0.7073500676386593, + "grad_norm": 1.80890806033441, + "learning_rate": 8.331903897494077e-07, + "loss": 0.9689, + "step": 9412 + }, + { + "epoch": 0.7074252217044942, + "grad_norm": 2.142153897011474, + "learning_rate": 8.327950200886409e-07, + "loss": 0.9909, + "step": 9413 + }, + { + "epoch": 0.7075003757703292, + "grad_norm": 1.7058668551795295, + "learning_rate": 8.323997195892389e-07, + "loss": 0.9383, + "step": 9414 + }, + { + "epoch": 0.7075755298361641, + "grad_norm": 7.789159225123478, + "learning_rate": 8.320044882746246e-07, + "loss": 1.0136, + "step": 9415 + }, + { + "epoch": 0.7076506839019991, + "grad_norm": 1.5594804961271247, + "learning_rate": 8.316093261682169e-07, + "loss": 0.9278, + "step": 9416 + }, + { + "epoch": 0.7077258379678341, + "grad_norm": 2.4272058698858547, + "learning_rate": 8.312142332934299e-07, + "loss": 0.9643, + "step": 9417 + }, + { + "epoch": 0.707800992033669, + "grad_norm": 1.5339597742309738, + "learning_rate": 8.308192096736759e-07, + "loss": 0.9652, + "step": 9418 + }, + { + "epoch": 0.707876146099504, + "grad_norm": 1.6746663828872836, + "learning_rate": 8.304242553323608e-07, + "loss": 0.9909, + "step": 9419 + }, + { + "epoch": 0.7079513001653389, + "grad_norm": 2.137684605529376, + "learning_rate": 8.300293702928873e-07, + "loss": 0.993, + "step": 9420 + }, + { + "epoch": 0.7080264542311739, + "grad_norm": 2.0447095525742593, + "learning_rate": 8.296345545786536e-07, + "loss": 0.9475, + "step": 9421 + }, + { + "epoch": 0.7081016082970089, + "grad_norm": 1.857203907015567, + "learning_rate": 8.292398082130534e-07, + "loss": 0.9346, + "step": 9422 + }, + { + "epoch": 0.7081767623628439, + "grad_norm": 1.52302874436683, + "learning_rate": 8.288451312194787e-07, + "loss": 0.898, + "step": 9423 + }, + { + "epoch": 0.7082519164286788, + "grad_norm": 1.498429032064145, + "learning_rate": 8.284505236213144e-07, + "loss": 0.8674, + "step": 9424 + }, + { + "epoch": 0.7083270704945137, + "grad_norm": 1.9194660130656993, + "learning_rate": 8.280559854419427e-07, + "loss": 1.04, + "step": 9425 + }, + { + "epoch": 0.7084022245603487, + "grad_norm": 0.6743393879592072, + "learning_rate": 8.276615167047416e-07, + "loss": 0.765, + "step": 9426 + }, + { + "epoch": 0.7084773786261837, + "grad_norm": 2.2283587789212076, + "learning_rate": 8.272671174330841e-07, + "loss": 0.943, + "step": 9427 + }, + { + "epoch": 0.7085525326920187, + "grad_norm": 1.7045434067453742, + "learning_rate": 8.268727876503411e-07, + "loss": 0.886, + "step": 9428 + }, + { + "epoch": 0.7086276867578536, + "grad_norm": 2.379691960411375, + "learning_rate": 8.26478527379878e-07, + "loss": 0.9629, + "step": 9429 + }, + { + "epoch": 0.7087028408236885, + "grad_norm": 1.7264734996106335, + "learning_rate": 8.260843366450549e-07, + "loss": 0.9644, + "step": 9430 + }, + { + "epoch": 0.7087779948895235, + "grad_norm": 2.455624612879559, + "learning_rate": 8.256902154692318e-07, + "loss": 0.8988, + "step": 9431 + }, + { + "epoch": 0.7088531489553584, + "grad_norm": 2.1274070570621855, + "learning_rate": 8.252961638757585e-07, + "loss": 1.1674, + "step": 9432 + }, + { + "epoch": 0.7089283030211935, + "grad_norm": 1.8272692641937551, + "learning_rate": 8.249021818879865e-07, + "loss": 0.9555, + "step": 9433 + }, + { + "epoch": 0.7090034570870284, + "grad_norm": 0.6811897513669749, + "learning_rate": 8.245082695292592e-07, + "loss": 0.8716, + "step": 9434 + }, + { + "epoch": 0.7090786111528634, + "grad_norm": 1.9560309802412261, + "learning_rate": 8.24114426822919e-07, + "loss": 0.9679, + "step": 9435 + }, + { + "epoch": 0.7091537652186983, + "grad_norm": 1.4906563617598056, + "learning_rate": 8.237206537923016e-07, + "loss": 0.9381, + "step": 9436 + }, + { + "epoch": 0.7092289192845332, + "grad_norm": 2.917166603475304, + "learning_rate": 8.233269504607398e-07, + "loss": 1.0038, + "step": 9437 + }, + { + "epoch": 0.7093040733503683, + "grad_norm": 2.934989180796761, + "learning_rate": 8.229333168515622e-07, + "loss": 1.0125, + "step": 9438 + }, + { + "epoch": 0.7093792274162032, + "grad_norm": 1.8934555932764088, + "learning_rate": 8.225397529880919e-07, + "loss": 1.0155, + "step": 9439 + }, + { + "epoch": 0.7094543814820382, + "grad_norm": 1.735191496689021, + "learning_rate": 8.22146258893651e-07, + "loss": 0.8729, + "step": 9440 + }, + { + "epoch": 0.7095295355478731, + "grad_norm": 3.235334023826217, + "learning_rate": 8.217528345915543e-07, + "loss": 0.9518, + "step": 9441 + }, + { + "epoch": 0.7096046896137082, + "grad_norm": 2.290691746341662, + "learning_rate": 8.21359480105114e-07, + "loss": 0.9511, + "step": 9442 + }, + { + "epoch": 0.7096798436795431, + "grad_norm": 1.8439833283281, + "learning_rate": 8.209661954576379e-07, + "loss": 0.9271, + "step": 9443 + }, + { + "epoch": 0.709754997745378, + "grad_norm": 0.7258736000395699, + "learning_rate": 8.205729806724288e-07, + "loss": 0.8587, + "step": 9444 + }, + { + "epoch": 0.709830151811213, + "grad_norm": 1.6567662712001576, + "learning_rate": 8.201798357727876e-07, + "loss": 1.0139, + "step": 9445 + }, + { + "epoch": 0.7099053058770479, + "grad_norm": 1.753631502309517, + "learning_rate": 8.19786760782009e-07, + "loss": 0.8755, + "step": 9446 + }, + { + "epoch": 0.709980459942883, + "grad_norm": 2.3491905644574658, + "learning_rate": 8.193937557233841e-07, + "loss": 0.9871, + "step": 9447 + }, + { + "epoch": 0.7100556140087179, + "grad_norm": 0.7952275168950168, + "learning_rate": 8.190008206202002e-07, + "loss": 0.8381, + "step": 9448 + }, + { + "epoch": 0.7101307680745529, + "grad_norm": 2.8222347059833166, + "learning_rate": 8.186079554957392e-07, + "loss": 0.9916, + "step": 9449 + }, + { + "epoch": 0.7102059221403878, + "grad_norm": 1.583679470909502, + "learning_rate": 8.182151603732814e-07, + "loss": 1.0363, + "step": 9450 + }, + { + "epoch": 0.7102810762062227, + "grad_norm": 6.346462628159485, + "learning_rate": 8.178224352761008e-07, + "loss": 0.9329, + "step": 9451 + }, + { + "epoch": 0.7103562302720577, + "grad_norm": 1.3159998597547926, + "learning_rate": 8.174297802274668e-07, + "loss": 0.9841, + "step": 9452 + }, + { + "epoch": 0.7104313843378927, + "grad_norm": 1.7660783843879202, + "learning_rate": 8.170371952506483e-07, + "loss": 1.0154, + "step": 9453 + }, + { + "epoch": 0.7105065384037277, + "grad_norm": 1.6808601012373519, + "learning_rate": 8.166446803689045e-07, + "loss": 1.0113, + "step": 9454 + }, + { + "epoch": 0.7105816924695626, + "grad_norm": 1.5799544654996764, + "learning_rate": 8.162522356054952e-07, + "loss": 1.0377, + "step": 9455 + }, + { + "epoch": 0.7106568465353975, + "grad_norm": 1.8061177584803736, + "learning_rate": 8.158598609836733e-07, + "loss": 0.9991, + "step": 9456 + }, + { + "epoch": 0.7107320006012325, + "grad_norm": 2.343293819537494, + "learning_rate": 8.154675565266898e-07, + "loss": 0.9574, + "step": 9457 + }, + { + "epoch": 0.7108071546670675, + "grad_norm": 1.6445690884620365, + "learning_rate": 8.150753222577893e-07, + "loss": 0.9984, + "step": 9458 + }, + { + "epoch": 0.7108823087329025, + "grad_norm": 2.7774190342567695, + "learning_rate": 8.146831582002134e-07, + "loss": 0.9355, + "step": 9459 + }, + { + "epoch": 0.7109574627987374, + "grad_norm": 1.6968251173580307, + "learning_rate": 8.142910643771992e-07, + "loss": 1.0131, + "step": 9460 + }, + { + "epoch": 0.7110326168645724, + "grad_norm": 6.035912169488008, + "learning_rate": 8.13899040811979e-07, + "loss": 0.9627, + "step": 9461 + }, + { + "epoch": 0.7111077709304073, + "grad_norm": 1.4497867920757925, + "learning_rate": 8.135070875277834e-07, + "loss": 1.0915, + "step": 9462 + }, + { + "epoch": 0.7111829249962422, + "grad_norm": 1.8405537460343329, + "learning_rate": 8.131152045478362e-07, + "loss": 0.9783, + "step": 9463 + }, + { + "epoch": 0.7112580790620773, + "grad_norm": 1.9643690624141257, + "learning_rate": 8.12723391895358e-07, + "loss": 0.9893, + "step": 9464 + }, + { + "epoch": 0.7113332331279122, + "grad_norm": 1.8295864556196944, + "learning_rate": 8.123316495935653e-07, + "loss": 0.8795, + "step": 9465 + }, + { + "epoch": 0.7114083871937472, + "grad_norm": 1.3324689845041524, + "learning_rate": 8.119399776656695e-07, + "loss": 1.0685, + "step": 9466 + }, + { + "epoch": 0.7114835412595821, + "grad_norm": 1.8070689420483985, + "learning_rate": 8.115483761348801e-07, + "loss": 1.0515, + "step": 9467 + }, + { + "epoch": 0.7115586953254172, + "grad_norm": 1.6818554748739998, + "learning_rate": 8.111568450244004e-07, + "loss": 1.0062, + "step": 9468 + }, + { + "epoch": 0.7116338493912521, + "grad_norm": 1.5649801026319508, + "learning_rate": 8.107653843574298e-07, + "loss": 0.9869, + "step": 9469 + }, + { + "epoch": 0.711709003457087, + "grad_norm": 1.454135092508187, + "learning_rate": 8.103739941571641e-07, + "loss": 1.0146, + "step": 9470 + }, + { + "epoch": 0.711784157522922, + "grad_norm": 1.8162781372831245, + "learning_rate": 8.099826744467941e-07, + "loss": 0.9989, + "step": 9471 + }, + { + "epoch": 0.7118593115887569, + "grad_norm": 1.8499385617338597, + "learning_rate": 8.095914252495082e-07, + "loss": 0.9259, + "step": 9472 + }, + { + "epoch": 0.711934465654592, + "grad_norm": 2.023539890364588, + "learning_rate": 8.092002465884886e-07, + "loss": 0.9291, + "step": 9473 + }, + { + "epoch": 0.7120096197204269, + "grad_norm": 1.6643182262517113, + "learning_rate": 8.088091384869136e-07, + "loss": 1.0126, + "step": 9474 + }, + { + "epoch": 0.7120847737862618, + "grad_norm": 1.3066722826123385, + "learning_rate": 8.084181009679592e-07, + "loss": 1.0411, + "step": 9475 + }, + { + "epoch": 0.7121599278520968, + "grad_norm": 2.475059365995866, + "learning_rate": 8.080271340547953e-07, + "loss": 0.9873, + "step": 9476 + }, + { + "epoch": 0.7122350819179317, + "grad_norm": 1.66442000439629, + "learning_rate": 8.076362377705881e-07, + "loss": 1.0112, + "step": 9477 + }, + { + "epoch": 0.7123102359837667, + "grad_norm": 1.5023549191064196, + "learning_rate": 8.072454121384988e-07, + "loss": 0.8576, + "step": 9478 + }, + { + "epoch": 0.7123853900496017, + "grad_norm": 1.6144060919496437, + "learning_rate": 8.06854657181687e-07, + "loss": 0.9397, + "step": 9479 + }, + { + "epoch": 0.7124605441154367, + "grad_norm": 2.6443794240713925, + "learning_rate": 8.064639729233056e-07, + "loss": 0.9854, + "step": 9480 + }, + { + "epoch": 0.7125356981812716, + "grad_norm": 1.7265575978481384, + "learning_rate": 8.060733593865041e-07, + "loss": 0.9043, + "step": 9481 + }, + { + "epoch": 0.7126108522471065, + "grad_norm": 2.040940045139196, + "learning_rate": 8.05682816594428e-07, + "loss": 1.0179, + "step": 9482 + }, + { + "epoch": 0.7126860063129415, + "grad_norm": 0.6392310258408794, + "learning_rate": 8.052923445702175e-07, + "loss": 0.786, + "step": 9483 + }, + { + "epoch": 0.7127611603787765, + "grad_norm": 2.1882421052176477, + "learning_rate": 8.049019433370114e-07, + "loss": 1.0388, + "step": 9484 + }, + { + "epoch": 0.7128363144446115, + "grad_norm": 1.731820777421452, + "learning_rate": 8.045116129179412e-07, + "loss": 0.9514, + "step": 9485 + }, + { + "epoch": 0.7129114685104464, + "grad_norm": 1.8840770353008711, + "learning_rate": 8.041213533361359e-07, + "loss": 0.8952, + "step": 9486 + }, + { + "epoch": 0.7129866225762814, + "grad_norm": 1.8023941198353826, + "learning_rate": 8.037311646147198e-07, + "loss": 1.0214, + "step": 9487 + }, + { + "epoch": 0.7130617766421163, + "grad_norm": 1.5862067985212933, + "learning_rate": 8.033410467768122e-07, + "loss": 0.8778, + "step": 9488 + }, + { + "epoch": 0.7131369307079513, + "grad_norm": 2.9033503680015955, + "learning_rate": 8.029509998455308e-07, + "loss": 0.9, + "step": 9489 + }, + { + "epoch": 0.7132120847737863, + "grad_norm": 2.085594349042336, + "learning_rate": 8.025610238439864e-07, + "loss": 0.8754, + "step": 9490 + }, + { + "epoch": 0.7132872388396212, + "grad_norm": 2.540729902478304, + "learning_rate": 8.021711187952864e-07, + "loss": 0.8392, + "step": 9491 + }, + { + "epoch": 0.7133623929054562, + "grad_norm": 3.0773664595188976, + "learning_rate": 8.017812847225347e-07, + "loss": 0.943, + "step": 9492 + }, + { + "epoch": 0.7134375469712911, + "grad_norm": 3.0966076224467023, + "learning_rate": 8.013915216488294e-07, + "loss": 0.9079, + "step": 9493 + }, + { + "epoch": 0.7135127010371262, + "grad_norm": 2.7587767042035147, + "learning_rate": 8.010018295972667e-07, + "loss": 1.027, + "step": 9494 + }, + { + "epoch": 0.7135878551029611, + "grad_norm": 1.2690881513395205, + "learning_rate": 8.00612208590937e-07, + "loss": 0.9112, + "step": 9495 + }, + { + "epoch": 0.713663009168796, + "grad_norm": 1.8238134812097848, + "learning_rate": 8.002226586529261e-07, + "loss": 0.8861, + "step": 9496 + }, + { + "epoch": 0.713738163234631, + "grad_norm": 1.867720843002145, + "learning_rate": 7.998331798063176e-07, + "loss": 0.9814, + "step": 9497 + }, + { + "epoch": 0.7138133173004659, + "grad_norm": 1.8464377258360556, + "learning_rate": 7.994437720741889e-07, + "loss": 1.0306, + "step": 9498 + }, + { + "epoch": 0.713888471366301, + "grad_norm": 1.5781200899859802, + "learning_rate": 7.990544354796139e-07, + "loss": 0.8884, + "step": 9499 + }, + { + "epoch": 0.7139636254321359, + "grad_norm": 1.6440792957775618, + "learning_rate": 7.986651700456617e-07, + "loss": 0.9953, + "step": 9500 + }, + { + "epoch": 0.7140387794979708, + "grad_norm": 2.1344731052801547, + "learning_rate": 7.982759757953989e-07, + "loss": 0.9931, + "step": 9501 + }, + { + "epoch": 0.7141139335638058, + "grad_norm": 1.7597453701534462, + "learning_rate": 7.978868527518864e-07, + "loss": 1.0364, + "step": 9502 + }, + { + "epoch": 0.7141890876296407, + "grad_norm": 1.5051384967939827, + "learning_rate": 7.974978009381812e-07, + "loss": 0.9595, + "step": 9503 + }, + { + "epoch": 0.7142642416954758, + "grad_norm": 1.6470165908431813, + "learning_rate": 7.97108820377336e-07, + "loss": 0.9954, + "step": 9504 + }, + { + "epoch": 0.7143393957613107, + "grad_norm": 1.6985216135711045, + "learning_rate": 7.967199110923983e-07, + "loss": 0.9515, + "step": 9505 + }, + { + "epoch": 0.7144145498271457, + "grad_norm": 1.6657199395982787, + "learning_rate": 7.963310731064146e-07, + "loss": 0.9489, + "step": 9506 + }, + { + "epoch": 0.7144897038929806, + "grad_norm": 1.9230113817930499, + "learning_rate": 7.959423064424238e-07, + "loss": 0.9946, + "step": 9507 + }, + { + "epoch": 0.7145648579588155, + "grad_norm": 2.037703722732872, + "learning_rate": 7.95553611123462e-07, + "loss": 0.9945, + "step": 9508 + }, + { + "epoch": 0.7146400120246506, + "grad_norm": 1.605251548163929, + "learning_rate": 7.95164987172561e-07, + "loss": 0.9087, + "step": 9509 + }, + { + "epoch": 0.7147151660904855, + "grad_norm": 1.770205674479411, + "learning_rate": 7.947764346127472e-07, + "loss": 0.9807, + "step": 9510 + }, + { + "epoch": 0.7147903201563205, + "grad_norm": 1.4075958593326598, + "learning_rate": 7.943879534670457e-07, + "loss": 1.041, + "step": 9511 + }, + { + "epoch": 0.7148654742221554, + "grad_norm": 1.4946273110800863, + "learning_rate": 7.939995437584744e-07, + "loss": 1.0118, + "step": 9512 + }, + { + "epoch": 0.7149406282879904, + "grad_norm": 1.8637923653262947, + "learning_rate": 7.936112055100481e-07, + "loss": 0.9867, + "step": 9513 + }, + { + "epoch": 0.7150157823538253, + "grad_norm": 1.7979071928137886, + "learning_rate": 7.932229387447777e-07, + "loss": 1.0092, + "step": 9514 + }, + { + "epoch": 0.7150909364196603, + "grad_norm": 1.3441699946331365, + "learning_rate": 7.928347434856683e-07, + "loss": 1.0184, + "step": 9515 + }, + { + "epoch": 0.7151660904854953, + "grad_norm": 1.9215217203604587, + "learning_rate": 7.924466197557238e-07, + "loss": 0.9654, + "step": 9516 + }, + { + "epoch": 0.7152412445513302, + "grad_norm": 2.0722844934638163, + "learning_rate": 7.920585675779412e-07, + "loss": 0.9839, + "step": 9517 + }, + { + "epoch": 0.7153163986171652, + "grad_norm": 1.714795852722143, + "learning_rate": 7.916705869753131e-07, + "loss": 0.9825, + "step": 9518 + }, + { + "epoch": 0.7153915526830001, + "grad_norm": 1.6442955736198719, + "learning_rate": 7.912826779708304e-07, + "loss": 0.9255, + "step": 9519 + }, + { + "epoch": 0.7154667067488351, + "grad_norm": 0.8716858703134168, + "learning_rate": 7.908948405874775e-07, + "loss": 0.8314, + "step": 9520 + }, + { + "epoch": 0.7155418608146701, + "grad_norm": 0.7571203079040454, + "learning_rate": 7.905070748482354e-07, + "loss": 0.8981, + "step": 9521 + }, + { + "epoch": 0.715617014880505, + "grad_norm": 2.088053455194691, + "learning_rate": 7.901193807760797e-07, + "loss": 0.999, + "step": 9522 + }, + { + "epoch": 0.71569216894634, + "grad_norm": 1.4326434818863367, + "learning_rate": 7.897317583939846e-07, + "loss": 0.9524, + "step": 9523 + }, + { + "epoch": 0.7157673230121749, + "grad_norm": 1.6803774104351838, + "learning_rate": 7.89344207724917e-07, + "loss": 1.0629, + "step": 9524 + }, + { + "epoch": 0.71584247707801, + "grad_norm": 2.2846702124007203, + "learning_rate": 7.889567287918413e-07, + "loss": 0.9795, + "step": 9525 + }, + { + "epoch": 0.7159176311438449, + "grad_norm": 2.22960414931175, + "learning_rate": 7.885693216177165e-07, + "loss": 1.0047, + "step": 9526 + }, + { + "epoch": 0.7159927852096798, + "grad_norm": 2.2237658689124604, + "learning_rate": 7.881819862254977e-07, + "loss": 1.0222, + "step": 9527 + }, + { + "epoch": 0.7160679392755148, + "grad_norm": 11.669610514936327, + "learning_rate": 7.877947226381372e-07, + "loss": 1.0581, + "step": 9528 + }, + { + "epoch": 0.7161430933413497, + "grad_norm": 1.7731015059346997, + "learning_rate": 7.874075308785813e-07, + "loss": 0.9501, + "step": 9529 + }, + { + "epoch": 0.7162182474071848, + "grad_norm": 1.6633036141170936, + "learning_rate": 7.870204109697724e-07, + "loss": 1.0445, + "step": 9530 + }, + { + "epoch": 0.7162934014730197, + "grad_norm": 1.6707157144346438, + "learning_rate": 7.866333629346491e-07, + "loss": 0.8673, + "step": 9531 + }, + { + "epoch": 0.7163685555388547, + "grad_norm": 1.951563890271879, + "learning_rate": 7.862463867961446e-07, + "loss": 0.8919, + "step": 9532 + }, + { + "epoch": 0.7164437096046896, + "grad_norm": 0.8220813390672498, + "learning_rate": 7.858594825771902e-07, + "loss": 0.8961, + "step": 9533 + }, + { + "epoch": 0.7165188636705245, + "grad_norm": 2.6091037997633313, + "learning_rate": 7.854726503007107e-07, + "loss": 0.8661, + "step": 9534 + }, + { + "epoch": 0.7165940177363596, + "grad_norm": 1.8844660545605525, + "learning_rate": 7.850858899896273e-07, + "loss": 1.009, + "step": 9535 + }, + { + "epoch": 0.7166691718021945, + "grad_norm": 2.1664345201201587, + "learning_rate": 7.846992016668572e-07, + "loss": 1.0709, + "step": 9536 + }, + { + "epoch": 0.7167443258680295, + "grad_norm": 1.6208233343500784, + "learning_rate": 7.843125853553125e-07, + "loss": 1.0507, + "step": 9537 + }, + { + "epoch": 0.7168194799338644, + "grad_norm": 2.2364953923867663, + "learning_rate": 7.839260410779029e-07, + "loss": 1.015, + "step": 9538 + }, + { + "epoch": 0.7168946339996994, + "grad_norm": 0.7093511636233074, + "learning_rate": 7.835395688575319e-07, + "loss": 0.8371, + "step": 9539 + }, + { + "epoch": 0.7169697880655344, + "grad_norm": 1.60098002937056, + "learning_rate": 7.831531687170992e-07, + "loss": 0.8985, + "step": 9540 + }, + { + "epoch": 0.7170449421313693, + "grad_norm": 2.060917912202867, + "learning_rate": 7.827668406795014e-07, + "loss": 0.9797, + "step": 9541 + }, + { + "epoch": 0.7171200961972043, + "grad_norm": 1.4485370852773174, + "learning_rate": 7.823805847676292e-07, + "loss": 0.9756, + "step": 9542 + }, + { + "epoch": 0.7171952502630392, + "grad_norm": 2.011954580761302, + "learning_rate": 7.819944010043702e-07, + "loss": 1.0219, + "step": 9543 + }, + { + "epoch": 0.7172704043288742, + "grad_norm": 3.0246201205737053, + "learning_rate": 7.816082894126061e-07, + "loss": 0.8376, + "step": 9544 + }, + { + "epoch": 0.7173455583947091, + "grad_norm": 1.426333803728785, + "learning_rate": 7.812222500152172e-07, + "loss": 0.9394, + "step": 9545 + }, + { + "epoch": 0.7174207124605441, + "grad_norm": 2.092031542297703, + "learning_rate": 7.80836282835077e-07, + "loss": 0.9816, + "step": 9546 + }, + { + "epoch": 0.7174958665263791, + "grad_norm": 1.5578375872674348, + "learning_rate": 7.804503878950555e-07, + "loss": 0.9603, + "step": 9547 + }, + { + "epoch": 0.717571020592214, + "grad_norm": 3.1741182259244844, + "learning_rate": 7.800645652180184e-07, + "loss": 0.9295, + "step": 9548 + }, + { + "epoch": 0.717646174658049, + "grad_norm": 2.2949716375921283, + "learning_rate": 7.796788148268267e-07, + "loss": 0.8233, + "step": 9549 + }, + { + "epoch": 0.717721328723884, + "grad_norm": 1.3757790179345537, + "learning_rate": 7.79293136744339e-07, + "loss": 1.0423, + "step": 9550 + }, + { + "epoch": 0.717796482789719, + "grad_norm": 2.1982480442641883, + "learning_rate": 7.78907530993407e-07, + "loss": 0.92, + "step": 9551 + }, + { + "epoch": 0.7178716368555539, + "grad_norm": 1.8622879803712762, + "learning_rate": 7.785219975968798e-07, + "loss": 1.0465, + "step": 9552 + }, + { + "epoch": 0.7179467909213888, + "grad_norm": 3.314741276099543, + "learning_rate": 7.781365365776016e-07, + "loss": 0.9166, + "step": 9553 + }, + { + "epoch": 0.7180219449872238, + "grad_norm": 1.6764566976505209, + "learning_rate": 7.777511479584118e-07, + "loss": 0.9924, + "step": 9554 + }, + { + "epoch": 0.7180970990530587, + "grad_norm": 1.8741458218733962, + "learning_rate": 7.773658317621474e-07, + "loss": 0.9556, + "step": 9555 + }, + { + "epoch": 0.7181722531188938, + "grad_norm": 1.8710078018613003, + "learning_rate": 7.769805880116391e-07, + "loss": 1.0561, + "step": 9556 + }, + { + "epoch": 0.7182474071847287, + "grad_norm": 2.3055375939809615, + "learning_rate": 7.765954167297144e-07, + "loss": 0.9463, + "step": 9557 + }, + { + "epoch": 0.7183225612505637, + "grad_norm": 1.762029311725443, + "learning_rate": 7.762103179391961e-07, + "loss": 0.9086, + "step": 9558 + }, + { + "epoch": 0.7183977153163986, + "grad_norm": 2.1818657385816134, + "learning_rate": 7.758252916629017e-07, + "loss": 0.8376, + "step": 9559 + }, + { + "epoch": 0.7184728693822335, + "grad_norm": 1.5242563247937355, + "learning_rate": 7.754403379236474e-07, + "loss": 1.0081, + "step": 9560 + }, + { + "epoch": 0.7185480234480686, + "grad_norm": 2.0751102923760225, + "learning_rate": 7.75055456744242e-07, + "loss": 0.9731, + "step": 9561 + }, + { + "epoch": 0.7186231775139035, + "grad_norm": 1.672324717278906, + "learning_rate": 7.74670648147491e-07, + "loss": 0.9246, + "step": 9562 + }, + { + "epoch": 0.7186983315797385, + "grad_norm": 3.1929405007677283, + "learning_rate": 7.742859121561967e-07, + "loss": 0.9179, + "step": 9563 + }, + { + "epoch": 0.7187734856455734, + "grad_norm": 1.8770610290060254, + "learning_rate": 7.739012487931555e-07, + "loss": 1.005, + "step": 9564 + }, + { + "epoch": 0.7188486397114083, + "grad_norm": 1.801911237433665, + "learning_rate": 7.735166580811607e-07, + "loss": 0.9914, + "step": 9565 + }, + { + "epoch": 0.7189237937772434, + "grad_norm": 1.6390658281290686, + "learning_rate": 7.731321400429995e-07, + "loss": 1.0206, + "step": 9566 + }, + { + "epoch": 0.7189989478430783, + "grad_norm": 1.4816660204279162, + "learning_rate": 7.727476947014578e-07, + "loss": 1.0069, + "step": 9567 + }, + { + "epoch": 0.7190741019089133, + "grad_norm": 1.7821019460927874, + "learning_rate": 7.723633220793146e-07, + "loss": 1.0383, + "step": 9568 + }, + { + "epoch": 0.7191492559747482, + "grad_norm": 1.5096519990038477, + "learning_rate": 7.719790221993456e-07, + "loss": 0.9732, + "step": 9569 + }, + { + "epoch": 0.7192244100405832, + "grad_norm": 1.7847641528885987, + "learning_rate": 7.715947950843218e-07, + "loss": 0.9549, + "step": 9570 + }, + { + "epoch": 0.7192995641064182, + "grad_norm": 2.4234178120485437, + "learning_rate": 7.712106407570096e-07, + "loss": 1.0138, + "step": 9571 + }, + { + "epoch": 0.7193747181722531, + "grad_norm": 1.8390357314554342, + "learning_rate": 7.70826559240173e-07, + "loss": 0.9818, + "step": 9572 + }, + { + "epoch": 0.7194498722380881, + "grad_norm": 1.7840218643850079, + "learning_rate": 7.704425505565697e-07, + "loss": 0.9116, + "step": 9573 + }, + { + "epoch": 0.719525026303923, + "grad_norm": 1.9647628737199103, + "learning_rate": 7.700586147289534e-07, + "loss": 0.978, + "step": 9574 + }, + { + "epoch": 0.719600180369758, + "grad_norm": 1.2586938838326147, + "learning_rate": 7.696747517800742e-07, + "loss": 0.9272, + "step": 9575 + }, + { + "epoch": 0.719675334435593, + "grad_norm": 1.5873690002938423, + "learning_rate": 7.692909617326761e-07, + "loss": 0.9849, + "step": 9576 + }, + { + "epoch": 0.719750488501428, + "grad_norm": 1.9092477459496497, + "learning_rate": 7.689072446095022e-07, + "loss": 1.0306, + "step": 9577 + }, + { + "epoch": 0.7198256425672629, + "grad_norm": 0.6701282594811783, + "learning_rate": 7.685236004332883e-07, + "loss": 0.8182, + "step": 9578 + }, + { + "epoch": 0.7199007966330978, + "grad_norm": 2.1840436604994915, + "learning_rate": 7.681400292267663e-07, + "loss": 1.0628, + "step": 9579 + }, + { + "epoch": 0.7199759506989328, + "grad_norm": 3.5862794933256454, + "learning_rate": 7.67756531012665e-07, + "loss": 1.0234, + "step": 9580 + }, + { + "epoch": 0.7200511047647677, + "grad_norm": 1.594635362230677, + "learning_rate": 7.673731058137072e-07, + "loss": 0.9607, + "step": 9581 + }, + { + "epoch": 0.7201262588306028, + "grad_norm": 1.9113485597554603, + "learning_rate": 7.669897536526133e-07, + "loss": 0.9061, + "step": 9582 + }, + { + "epoch": 0.7202014128964377, + "grad_norm": 1.7622719042203154, + "learning_rate": 7.666064745520982e-07, + "loss": 0.9658, + "step": 9583 + }, + { + "epoch": 0.7202765669622727, + "grad_norm": 2.35910345877007, + "learning_rate": 7.662232685348718e-07, + "loss": 0.8509, + "step": 9584 + }, + { + "epoch": 0.7203517210281076, + "grad_norm": 1.9067496598572138, + "learning_rate": 7.658401356236417e-07, + "loss": 1.0102, + "step": 9585 + }, + { + "epoch": 0.7204268750939425, + "grad_norm": 1.686478764970026, + "learning_rate": 7.654570758411096e-07, + "loss": 1.0153, + "step": 9586 + }, + { + "epoch": 0.7205020291597776, + "grad_norm": 1.4806137617851336, + "learning_rate": 7.650740892099731e-07, + "loss": 1.0147, + "step": 9587 + }, + { + "epoch": 0.7205771832256125, + "grad_norm": 1.761871263622532, + "learning_rate": 7.646911757529251e-07, + "loss": 1.0348, + "step": 9588 + }, + { + "epoch": 0.7206523372914475, + "grad_norm": 2.3394564147202512, + "learning_rate": 7.643083354926558e-07, + "loss": 0.9426, + "step": 9589 + }, + { + "epoch": 0.7207274913572824, + "grad_norm": 2.3943869716869757, + "learning_rate": 7.639255684518495e-07, + "loss": 0.9345, + "step": 9590 + }, + { + "epoch": 0.7208026454231173, + "grad_norm": 2.0515578189282206, + "learning_rate": 7.635428746531865e-07, + "loss": 0.9396, + "step": 9591 + }, + { + "epoch": 0.7208777994889524, + "grad_norm": 2.415111424973565, + "learning_rate": 7.631602541193429e-07, + "loss": 0.9951, + "step": 9592 + }, + { + "epoch": 0.7209529535547873, + "grad_norm": 1.6249489409421285, + "learning_rate": 7.627777068729897e-07, + "loss": 0.9795, + "step": 9593 + }, + { + "epoch": 0.7210281076206223, + "grad_norm": 2.7343157080625167, + "learning_rate": 7.623952329367955e-07, + "loss": 0.9208, + "step": 9594 + }, + { + "epoch": 0.7211032616864572, + "grad_norm": 0.8162668282496021, + "learning_rate": 7.620128323334231e-07, + "loss": 0.9384, + "step": 9595 + }, + { + "epoch": 0.7211784157522922, + "grad_norm": 1.738335426429276, + "learning_rate": 7.616305050855309e-07, + "loss": 0.9447, + "step": 9596 + }, + { + "epoch": 0.7212535698181272, + "grad_norm": 1.7138289314195434, + "learning_rate": 7.612482512157732e-07, + "loss": 0.9295, + "step": 9597 + }, + { + "epoch": 0.7213287238839621, + "grad_norm": 1.488645093602374, + "learning_rate": 7.608660707467996e-07, + "loss": 0.8889, + "step": 9598 + }, + { + "epoch": 0.7214038779497971, + "grad_norm": 0.6619924256332275, + "learning_rate": 7.604839637012568e-07, + "loss": 0.873, + "step": 9599 + }, + { + "epoch": 0.721479032015632, + "grad_norm": 2.3117029001943763, + "learning_rate": 7.601019301017857e-07, + "loss": 0.9752, + "step": 9600 + }, + { + "epoch": 0.721554186081467, + "grad_norm": 1.6128608211203637, + "learning_rate": 7.597199699710224e-07, + "loss": 0.8901, + "step": 9601 + }, + { + "epoch": 0.721629340147302, + "grad_norm": 1.7891655105186721, + "learning_rate": 7.593380833316017e-07, + "loss": 0.9934, + "step": 9602 + }, + { + "epoch": 0.721704494213137, + "grad_norm": 2.1659821627261953, + "learning_rate": 7.589562702061487e-07, + "loss": 0.9494, + "step": 9603 + }, + { + "epoch": 0.7217796482789719, + "grad_norm": 1.5561983061634783, + "learning_rate": 7.585745306172899e-07, + "loss": 0.9354, + "step": 9604 + }, + { + "epoch": 0.7218548023448068, + "grad_norm": 1.656128285092293, + "learning_rate": 7.581928645876439e-07, + "loss": 0.8258, + "step": 9605 + }, + { + "epoch": 0.7219299564106418, + "grad_norm": 2.2543404357646355, + "learning_rate": 7.578112721398251e-07, + "loss": 0.9285, + "step": 9606 + }, + { + "epoch": 0.7220051104764768, + "grad_norm": 1.598619315358989, + "learning_rate": 7.574297532964457e-07, + "loss": 1.0156, + "step": 9607 + }, + { + "epoch": 0.7220802645423118, + "grad_norm": 1.7742061209722602, + "learning_rate": 7.570483080801118e-07, + "loss": 1.105, + "step": 9608 + }, + { + "epoch": 0.7221554186081467, + "grad_norm": 2.4837122579009847, + "learning_rate": 7.56666936513425e-07, + "loss": 0.9898, + "step": 9609 + }, + { + "epoch": 0.7222305726739816, + "grad_norm": 2.5934264199516295, + "learning_rate": 7.562856386189826e-07, + "loss": 0.9526, + "step": 9610 + }, + { + "epoch": 0.7223057267398166, + "grad_norm": 1.3586603237203072, + "learning_rate": 7.559044144193793e-07, + "loss": 0.962, + "step": 9611 + }, + { + "epoch": 0.7223808808056515, + "grad_norm": 2.245603716924871, + "learning_rate": 7.555232639372033e-07, + "loss": 0.9812, + "step": 9612 + }, + { + "epoch": 0.7224560348714866, + "grad_norm": 1.8918664812715997, + "learning_rate": 7.551421871950396e-07, + "loss": 0.9808, + "step": 9613 + }, + { + "epoch": 0.7225311889373215, + "grad_norm": 1.7198882755321958, + "learning_rate": 7.54761184215468e-07, + "loss": 0.9589, + "step": 9614 + }, + { + "epoch": 0.7226063430031565, + "grad_norm": 1.6376429022248775, + "learning_rate": 7.543802550210641e-07, + "loss": 0.8923, + "step": 9615 + }, + { + "epoch": 0.7226814970689914, + "grad_norm": 1.9635915235535775, + "learning_rate": 7.539993996344003e-07, + "loss": 0.8752, + "step": 9616 + }, + { + "epoch": 0.7227566511348263, + "grad_norm": 1.6307315250925096, + "learning_rate": 7.536186180780435e-07, + "loss": 1.0547, + "step": 9617 + }, + { + "epoch": 0.7228318052006614, + "grad_norm": 1.3282859704696948, + "learning_rate": 7.532379103745561e-07, + "loss": 1.0188, + "step": 9618 + }, + { + "epoch": 0.7229069592664963, + "grad_norm": 1.5647163693503885, + "learning_rate": 7.528572765464969e-07, + "loss": 0.9578, + "step": 9619 + }, + { + "epoch": 0.7229821133323313, + "grad_norm": 1.6386663461803332, + "learning_rate": 7.524767166164187e-07, + "loss": 0.9242, + "step": 9620 + }, + { + "epoch": 0.7230572673981662, + "grad_norm": 1.7108708895602347, + "learning_rate": 7.520962306068732e-07, + "loss": 0.9941, + "step": 9621 + }, + { + "epoch": 0.7231324214640013, + "grad_norm": 1.737726221520373, + "learning_rate": 7.517158185404044e-07, + "loss": 1.008, + "step": 9622 + }, + { + "epoch": 0.7232075755298362, + "grad_norm": 1.6960504569384365, + "learning_rate": 7.513354804395527e-07, + "loss": 0.9929, + "step": 9623 + }, + { + "epoch": 0.7232827295956711, + "grad_norm": 1.8049894798585122, + "learning_rate": 7.509552163268564e-07, + "loss": 0.9885, + "step": 9624 + }, + { + "epoch": 0.7233578836615061, + "grad_norm": 2.0555452942234096, + "learning_rate": 7.505750262248453e-07, + "loss": 0.9938, + "step": 9625 + }, + { + "epoch": 0.723433037727341, + "grad_norm": 2.12367162136605, + "learning_rate": 7.50194910156049e-07, + "loss": 1.0237, + "step": 9626 + }, + { + "epoch": 0.723508191793176, + "grad_norm": 1.8429510034060048, + "learning_rate": 7.498148681429893e-07, + "loss": 0.9825, + "step": 9627 + }, + { + "epoch": 0.723583345859011, + "grad_norm": 1.7604946442739011, + "learning_rate": 7.494349002081866e-07, + "loss": 0.9637, + "step": 9628 + }, + { + "epoch": 0.723658499924846, + "grad_norm": 4.16778249422753, + "learning_rate": 7.490550063741548e-07, + "loss": 0.8898, + "step": 9629 + }, + { + "epoch": 0.7237336539906809, + "grad_norm": 1.5389518891684992, + "learning_rate": 7.486751866634043e-07, + "loss": 0.9151, + "step": 9630 + }, + { + "epoch": 0.7238088080565158, + "grad_norm": 1.8339241188558792, + "learning_rate": 7.482954410984403e-07, + "loss": 1.0326, + "step": 9631 + }, + { + "epoch": 0.7238839621223508, + "grad_norm": 1.5696250967220304, + "learning_rate": 7.479157697017639e-07, + "loss": 1.0364, + "step": 9632 + }, + { + "epoch": 0.7239591161881858, + "grad_norm": 1.9739297212189504, + "learning_rate": 7.475361724958735e-07, + "loss": 1.013, + "step": 9633 + }, + { + "epoch": 0.7240342702540208, + "grad_norm": 2.070581497854603, + "learning_rate": 7.471566495032608e-07, + "loss": 0.9581, + "step": 9634 + }, + { + "epoch": 0.7241094243198557, + "grad_norm": 1.5658615573182106, + "learning_rate": 7.467772007464142e-07, + "loss": 0.9989, + "step": 9635 + }, + { + "epoch": 0.7241845783856906, + "grad_norm": 3.330271140850551, + "learning_rate": 7.463978262478172e-07, + "loss": 0.9718, + "step": 9636 + }, + { + "epoch": 0.7242597324515256, + "grad_norm": 0.7455845238442419, + "learning_rate": 7.460185260299488e-07, + "loss": 0.8273, + "step": 9637 + }, + { + "epoch": 0.7243348865173606, + "grad_norm": 1.6863328641604416, + "learning_rate": 7.456393001152849e-07, + "loss": 0.946, + "step": 9638 + }, + { + "epoch": 0.7244100405831956, + "grad_norm": 2.817004940403732, + "learning_rate": 7.452601485262961e-07, + "loss": 1.0641, + "step": 9639 + }, + { + "epoch": 0.7244851946490305, + "grad_norm": 4.48809192547008, + "learning_rate": 7.448810712854483e-07, + "loss": 0.9812, + "step": 9640 + }, + { + "epoch": 0.7245603487148655, + "grad_norm": 22.506037601399875, + "learning_rate": 7.445020684152031e-07, + "loss": 1.0515, + "step": 9641 + }, + { + "epoch": 0.7246355027807004, + "grad_norm": 1.8716357248567517, + "learning_rate": 7.44123139938017e-07, + "loss": 1.0921, + "step": 9642 + }, + { + "epoch": 0.7247106568465354, + "grad_norm": 0.735368705640872, + "learning_rate": 7.437442858763447e-07, + "loss": 0.833, + "step": 9643 + }, + { + "epoch": 0.7247858109123704, + "grad_norm": 2.6554276034743824, + "learning_rate": 7.433655062526343e-07, + "loss": 0.873, + "step": 9644 + }, + { + "epoch": 0.7248609649782053, + "grad_norm": 1.8491378553752456, + "learning_rate": 7.429868010893287e-07, + "loss": 1.0021, + "step": 9645 + }, + { + "epoch": 0.7249361190440403, + "grad_norm": 2.027053959298136, + "learning_rate": 7.426081704088694e-07, + "loss": 0.956, + "step": 9646 + }, + { + "epoch": 0.7250112731098752, + "grad_norm": 1.6821599461940198, + "learning_rate": 7.422296142336908e-07, + "loss": 1.0478, + "step": 9647 + }, + { + "epoch": 0.7250864271757103, + "grad_norm": 1.5996133992909543, + "learning_rate": 7.418511325862238e-07, + "loss": 1.0429, + "step": 9648 + }, + { + "epoch": 0.7251615812415452, + "grad_norm": 1.4660489020368497, + "learning_rate": 7.414727254888944e-07, + "loss": 0.9488, + "step": 9649 + }, + { + "epoch": 0.7252367353073801, + "grad_norm": 1.640942860647064, + "learning_rate": 7.410943929641258e-07, + "loss": 0.9034, + "step": 9650 + }, + { + "epoch": 0.7253118893732151, + "grad_norm": 1.6528010896522203, + "learning_rate": 7.407161350343352e-07, + "loss": 0.9777, + "step": 9651 + }, + { + "epoch": 0.72538704343905, + "grad_norm": 2.0826530744660197, + "learning_rate": 7.403379517219354e-07, + "loss": 1.0338, + "step": 9652 + }, + { + "epoch": 0.725462197504885, + "grad_norm": 1.5783666293543923, + "learning_rate": 7.399598430493359e-07, + "loss": 1.0175, + "step": 9653 + }, + { + "epoch": 0.72553735157072, + "grad_norm": 1.6968929114195532, + "learning_rate": 7.395818090389401e-07, + "loss": 0.9497, + "step": 9654 + }, + { + "epoch": 0.7256125056365549, + "grad_norm": 2.243882020769435, + "learning_rate": 7.39203849713149e-07, + "loss": 1.0384, + "step": 9655 + }, + { + "epoch": 0.7256876597023899, + "grad_norm": 3.1664841415270786, + "learning_rate": 7.388259650943578e-07, + "loss": 0.9992, + "step": 9656 + }, + { + "epoch": 0.7257628137682248, + "grad_norm": 5.0003213779162525, + "learning_rate": 7.384481552049575e-07, + "loss": 1.0281, + "step": 9657 + }, + { + "epoch": 0.7258379678340598, + "grad_norm": 1.797838964216112, + "learning_rate": 7.380704200673349e-07, + "loss": 0.9148, + "step": 9658 + }, + { + "epoch": 0.7259131218998948, + "grad_norm": 3.127380684971126, + "learning_rate": 7.376927597038714e-07, + "loss": 0.9175, + "step": 9659 + }, + { + "epoch": 0.7259882759657298, + "grad_norm": 1.428849438455094, + "learning_rate": 7.373151741369463e-07, + "loss": 0.9546, + "step": 9660 + }, + { + "epoch": 0.7260634300315647, + "grad_norm": 1.9700941311184927, + "learning_rate": 7.369376633889324e-07, + "loss": 0.8488, + "step": 9661 + }, + { + "epoch": 0.7261385840973996, + "grad_norm": 1.6651530386210178, + "learning_rate": 7.365602274821983e-07, + "loss": 1.0414, + "step": 9662 + }, + { + "epoch": 0.7262137381632346, + "grad_norm": 2.1474738161846654, + "learning_rate": 7.361828664391088e-07, + "loss": 0.9978, + "step": 9663 + }, + { + "epoch": 0.7262888922290696, + "grad_norm": 1.6853492594451323, + "learning_rate": 7.358055802820234e-07, + "loss": 0.8668, + "step": 9664 + }, + { + "epoch": 0.7263640462949046, + "grad_norm": 2.21019762148116, + "learning_rate": 7.354283690332987e-07, + "loss": 0.9653, + "step": 9665 + }, + { + "epoch": 0.7264392003607395, + "grad_norm": 1.9201841902797294, + "learning_rate": 7.350512327152858e-07, + "loss": 1.0432, + "step": 9666 + }, + { + "epoch": 0.7265143544265745, + "grad_norm": 1.8203987461336395, + "learning_rate": 7.346741713503304e-07, + "loss": 1.0358, + "step": 9667 + }, + { + "epoch": 0.7265895084924094, + "grad_norm": 1.7357102486565146, + "learning_rate": 7.342971849607762e-07, + "loss": 0.9333, + "step": 9668 + }, + { + "epoch": 0.7266646625582444, + "grad_norm": 1.9901133791575432, + "learning_rate": 7.339202735689606e-07, + "loss": 0.8549, + "step": 9669 + }, + { + "epoch": 0.7267398166240794, + "grad_norm": 1.7272087717307942, + "learning_rate": 7.335434371972169e-07, + "loss": 0.9035, + "step": 9670 + }, + { + "epoch": 0.7268149706899143, + "grad_norm": 1.501167875865778, + "learning_rate": 7.331666758678734e-07, + "loss": 0.9959, + "step": 9671 + }, + { + "epoch": 0.7268901247557493, + "grad_norm": 1.509057277307816, + "learning_rate": 7.327899896032561e-07, + "loss": 1.0287, + "step": 9672 + }, + { + "epoch": 0.7269652788215842, + "grad_norm": 2.295763642164322, + "learning_rate": 7.324133784256846e-07, + "loss": 1.0064, + "step": 9673 + }, + { + "epoch": 0.7270404328874193, + "grad_norm": 0.743389200803938, + "learning_rate": 7.320368423574741e-07, + "loss": 0.798, + "step": 9674 + }, + { + "epoch": 0.7271155869532542, + "grad_norm": 1.8546004341598064, + "learning_rate": 7.316603814209359e-07, + "loss": 0.9681, + "step": 9675 + }, + { + "epoch": 0.7271907410190891, + "grad_norm": 1.663342745581672, + "learning_rate": 7.312839956383765e-07, + "loss": 0.991, + "step": 9676 + }, + { + "epoch": 0.7272658950849241, + "grad_norm": 9.403453322434638, + "learning_rate": 7.309076850320993e-07, + "loss": 0.9385, + "step": 9677 + }, + { + "epoch": 0.727341049150759, + "grad_norm": 1.4614367116614295, + "learning_rate": 7.305314496244012e-07, + "loss": 1.0742, + "step": 9678 + }, + { + "epoch": 0.7274162032165941, + "grad_norm": 2.778787186854704, + "learning_rate": 7.30155289437576e-07, + "loss": 0.9787, + "step": 9679 + }, + { + "epoch": 0.727491357282429, + "grad_norm": 3.895560079909132, + "learning_rate": 7.297792044939124e-07, + "loss": 0.942, + "step": 9680 + }, + { + "epoch": 0.7275665113482639, + "grad_norm": 3.3302274698457954, + "learning_rate": 7.294031948156942e-07, + "loss": 1.0053, + "step": 9681 + }, + { + "epoch": 0.7276416654140989, + "grad_norm": 1.9383020279188499, + "learning_rate": 7.290272604252028e-07, + "loss": 1.0406, + "step": 9682 + }, + { + "epoch": 0.7277168194799338, + "grad_norm": 1.9478943019405834, + "learning_rate": 7.286514013447134e-07, + "loss": 0.9209, + "step": 9683 + }, + { + "epoch": 0.7277919735457689, + "grad_norm": 2.3414776785758176, + "learning_rate": 7.282756175964966e-07, + "loss": 1.0163, + "step": 9684 + }, + { + "epoch": 0.7278671276116038, + "grad_norm": 1.8843743268924353, + "learning_rate": 7.278999092028191e-07, + "loss": 0.9724, + "step": 9685 + }, + { + "epoch": 0.7279422816774388, + "grad_norm": 3.14102656136364, + "learning_rate": 7.275242761859426e-07, + "loss": 0.9623, + "step": 9686 + }, + { + "epoch": 0.7280174357432737, + "grad_norm": 15.035377490902757, + "learning_rate": 7.27148718568126e-07, + "loss": 0.9866, + "step": 9687 + }, + { + "epoch": 0.7280925898091086, + "grad_norm": 2.4033316217099823, + "learning_rate": 7.267732363716219e-07, + "loss": 0.922, + "step": 9688 + }, + { + "epoch": 0.7281677438749437, + "grad_norm": 2.4640717423386036, + "learning_rate": 7.263978296186784e-07, + "loss": 0.9231, + "step": 9689 + }, + { + "epoch": 0.7282428979407786, + "grad_norm": 1.593038886501406, + "learning_rate": 7.260224983315413e-07, + "loss": 0.9359, + "step": 9690 + }, + { + "epoch": 0.7283180520066136, + "grad_norm": 1.8381691862448712, + "learning_rate": 7.256472425324494e-07, + "loss": 1.0459, + "step": 9691 + }, + { + "epoch": 0.7283932060724485, + "grad_norm": 2.0624389639406226, + "learning_rate": 7.252720622436382e-07, + "loss": 0.8991, + "step": 9692 + }, + { + "epoch": 0.7284683601382835, + "grad_norm": 2.5825885396368267, + "learning_rate": 7.248969574873378e-07, + "loss": 1.0188, + "step": 9693 + }, + { + "epoch": 0.7285435142041184, + "grad_norm": 1.6327868737943068, + "learning_rate": 7.245219282857761e-07, + "loss": 0.9624, + "step": 9694 + }, + { + "epoch": 0.7286186682699534, + "grad_norm": 1.7045597717377456, + "learning_rate": 7.241469746611743e-07, + "loss": 1.0041, + "step": 9695 + }, + { + "epoch": 0.7286938223357884, + "grad_norm": 3.409003728272024, + "learning_rate": 7.237720966357499e-07, + "loss": 1.0084, + "step": 9696 + }, + { + "epoch": 0.7287689764016233, + "grad_norm": 6.488919129729386, + "learning_rate": 7.233972942317157e-07, + "loss": 0.8512, + "step": 9697 + }, + { + "epoch": 0.7288441304674583, + "grad_norm": 1.6368646306324488, + "learning_rate": 7.230225674712795e-07, + "loss": 0.9229, + "step": 9698 + }, + { + "epoch": 0.7289192845332932, + "grad_norm": 1.8638561329598238, + "learning_rate": 7.22647916376647e-07, + "loss": 0.9193, + "step": 9699 + }, + { + "epoch": 0.7289944385991282, + "grad_norm": 1.5324995431074797, + "learning_rate": 7.222733409700165e-07, + "loss": 1.0033, + "step": 9700 + }, + { + "epoch": 0.7290695926649632, + "grad_norm": 1.9951478392813373, + "learning_rate": 7.218988412735833e-07, + "loss": 0.951, + "step": 9701 + }, + { + "epoch": 0.7291447467307981, + "grad_norm": 1.724641447535862, + "learning_rate": 7.215244173095381e-07, + "loss": 0.9357, + "step": 9702 + }, + { + "epoch": 0.7292199007966331, + "grad_norm": 1.6555110074492174, + "learning_rate": 7.21150069100066e-07, + "loss": 0.9054, + "step": 9703 + }, + { + "epoch": 0.729295054862468, + "grad_norm": 2.2563060764096514, + "learning_rate": 7.2077579666735e-07, + "loss": 0.9121, + "step": 9704 + }, + { + "epoch": 0.7293702089283031, + "grad_norm": 1.4641652074266318, + "learning_rate": 7.204016000335666e-07, + "loss": 0.9375, + "step": 9705 + }, + { + "epoch": 0.729445362994138, + "grad_norm": 2.3414875554811614, + "learning_rate": 7.200274792208882e-07, + "loss": 0.9169, + "step": 9706 + }, + { + "epoch": 0.7295205170599729, + "grad_norm": 0.6549319700773374, + "learning_rate": 7.196534342514831e-07, + "loss": 0.7812, + "step": 9707 + }, + { + "epoch": 0.7295956711258079, + "grad_norm": 1.7318641427125503, + "learning_rate": 7.192794651475141e-07, + "loss": 0.937, + "step": 9708 + }, + { + "epoch": 0.7296708251916428, + "grad_norm": 1.8435745802366927, + "learning_rate": 7.189055719311416e-07, + "loss": 0.9753, + "step": 9709 + }, + { + "epoch": 0.7297459792574779, + "grad_norm": 2.1250871752640936, + "learning_rate": 7.185317546245198e-07, + "loss": 1.0115, + "step": 9710 + }, + { + "epoch": 0.7298211333233128, + "grad_norm": 1.8723091526332016, + "learning_rate": 7.181580132497978e-07, + "loss": 0.9228, + "step": 9711 + }, + { + "epoch": 0.7298962873891478, + "grad_norm": 3.3045006376783324, + "learning_rate": 7.177843478291225e-07, + "loss": 0.9327, + "step": 9712 + }, + { + "epoch": 0.7299714414549827, + "grad_norm": 2.0263125933910726, + "learning_rate": 7.174107583846348e-07, + "loss": 0.985, + "step": 9713 + }, + { + "epoch": 0.7300465955208176, + "grad_norm": 1.7660772368971867, + "learning_rate": 7.17037244938471e-07, + "loss": 0.9464, + "step": 9714 + }, + { + "epoch": 0.7301217495866527, + "grad_norm": 1.7190268553654924, + "learning_rate": 7.166638075127625e-07, + "loss": 0.9762, + "step": 9715 + }, + { + "epoch": 0.7301969036524876, + "grad_norm": 1.5676097479795312, + "learning_rate": 7.162904461296382e-07, + "loss": 0.925, + "step": 9716 + }, + { + "epoch": 0.7302720577183226, + "grad_norm": 1.8165418348674183, + "learning_rate": 7.159171608112207e-07, + "loss": 1.0198, + "step": 9717 + }, + { + "epoch": 0.7303472117841575, + "grad_norm": 2.1910807049859478, + "learning_rate": 7.155439515796284e-07, + "loss": 0.9941, + "step": 9718 + }, + { + "epoch": 0.7304223658499925, + "grad_norm": 4.049228293282334, + "learning_rate": 7.151708184569758e-07, + "loss": 0.9359, + "step": 9719 + }, + { + "epoch": 0.7304975199158275, + "grad_norm": 3.377570409711483, + "learning_rate": 7.147977614653711e-07, + "loss": 0.9732, + "step": 9720 + }, + { + "epoch": 0.7305726739816624, + "grad_norm": 0.6877066778420953, + "learning_rate": 7.144247806269213e-07, + "loss": 0.8768, + "step": 9721 + }, + { + "epoch": 0.7306478280474974, + "grad_norm": 2.4108580026728066, + "learning_rate": 7.140518759637259e-07, + "loss": 0.9317, + "step": 9722 + }, + { + "epoch": 0.7307229821133323, + "grad_norm": 1.9111078618389177, + "learning_rate": 7.136790474978814e-07, + "loss": 0.9838, + "step": 9723 + }, + { + "epoch": 0.7307981361791673, + "grad_norm": 1.4389628554978628, + "learning_rate": 7.133062952514786e-07, + "loss": 0.9892, + "step": 9724 + }, + { + "epoch": 0.7308732902450022, + "grad_norm": 1.7779109510901, + "learning_rate": 7.129336192466044e-07, + "loss": 0.872, + "step": 9725 + }, + { + "epoch": 0.7309484443108372, + "grad_norm": 1.6064573339690777, + "learning_rate": 7.125610195053424e-07, + "loss": 0.9402, + "step": 9726 + }, + { + "epoch": 0.7310235983766722, + "grad_norm": 1.9134764104883768, + "learning_rate": 7.1218849604977e-07, + "loss": 0.9004, + "step": 9727 + }, + { + "epoch": 0.7310987524425071, + "grad_norm": 2.0245652756478085, + "learning_rate": 7.118160489019605e-07, + "loss": 1.0144, + "step": 9728 + }, + { + "epoch": 0.7311739065083421, + "grad_norm": 3.8039892288280144, + "learning_rate": 7.114436780839827e-07, + "loss": 0.9483, + "step": 9729 + }, + { + "epoch": 0.731249060574177, + "grad_norm": 1.5910963517989751, + "learning_rate": 7.110713836179007e-07, + "loss": 1.0096, + "step": 9730 + }, + { + "epoch": 0.7313242146400121, + "grad_norm": 1.5667520269852853, + "learning_rate": 7.106991655257754e-07, + "loss": 0.9132, + "step": 9731 + }, + { + "epoch": 0.731399368705847, + "grad_norm": 1.049954794864305, + "learning_rate": 7.103270238296619e-07, + "loss": 0.8484, + "step": 9732 + }, + { + "epoch": 0.7314745227716819, + "grad_norm": 2.784703029229245, + "learning_rate": 7.099549585516098e-07, + "loss": 0.9966, + "step": 9733 + }, + { + "epoch": 0.7315496768375169, + "grad_norm": 1.274768251038914, + "learning_rate": 7.095829697136671e-07, + "loss": 1.0051, + "step": 9734 + }, + { + "epoch": 0.7316248309033518, + "grad_norm": 2.030514510586875, + "learning_rate": 7.092110573378747e-07, + "loss": 0.9241, + "step": 9735 + }, + { + "epoch": 0.7316999849691869, + "grad_norm": 1.7900558445919537, + "learning_rate": 7.0883922144627e-07, + "loss": 0.8883, + "step": 9736 + }, + { + "epoch": 0.7317751390350218, + "grad_norm": 1.490572469813361, + "learning_rate": 7.08467462060885e-07, + "loss": 0.9468, + "step": 9737 + }, + { + "epoch": 0.7318502931008568, + "grad_norm": 1.7782321595951915, + "learning_rate": 7.08095779203749e-07, + "loss": 0.9802, + "step": 9738 + }, + { + "epoch": 0.7319254471666917, + "grad_norm": 1.690214375888263, + "learning_rate": 7.077241728968852e-07, + "loss": 1.0473, + "step": 9739 + }, + { + "epoch": 0.7320006012325266, + "grad_norm": 1.5591809689089922, + "learning_rate": 7.073526431623127e-07, + "loss": 0.9747, + "step": 9740 + }, + { + "epoch": 0.7320757552983617, + "grad_norm": 2.074104377768667, + "learning_rate": 7.069811900220458e-07, + "loss": 0.9598, + "step": 9741 + }, + { + "epoch": 0.7321509093641966, + "grad_norm": 0.7578046346039954, + "learning_rate": 7.06609813498094e-07, + "loss": 0.8097, + "step": 9742 + }, + { + "epoch": 0.7322260634300316, + "grad_norm": 1.9316537700394807, + "learning_rate": 7.062385136124642e-07, + "loss": 0.912, + "step": 9743 + }, + { + "epoch": 0.7323012174958665, + "grad_norm": 1.83681645557396, + "learning_rate": 7.058672903871568e-07, + "loss": 0.968, + "step": 9744 + }, + { + "epoch": 0.7323763715617014, + "grad_norm": 2.0738425048283595, + "learning_rate": 7.054961438441678e-07, + "loss": 0.9179, + "step": 9745 + }, + { + "epoch": 0.7324515256275365, + "grad_norm": 1.5156810789450066, + "learning_rate": 7.051250740054892e-07, + "loss": 0.9928, + "step": 9746 + }, + { + "epoch": 0.7325266796933714, + "grad_norm": 3.806568346663412, + "learning_rate": 7.047540808931078e-07, + "loss": 1.0482, + "step": 9747 + }, + { + "epoch": 0.7326018337592064, + "grad_norm": 2.0871292881530623, + "learning_rate": 7.043831645290077e-07, + "loss": 0.9099, + "step": 9748 + }, + { + "epoch": 0.7326769878250413, + "grad_norm": 0.747772804032415, + "learning_rate": 7.040123249351662e-07, + "loss": 0.8664, + "step": 9749 + }, + { + "epoch": 0.7327521418908763, + "grad_norm": 1.5776751793081583, + "learning_rate": 7.036415621335572e-07, + "loss": 0.9323, + "step": 9750 + }, + { + "epoch": 0.7328272959567113, + "grad_norm": 1.3845103126684593, + "learning_rate": 7.032708761461496e-07, + "loss": 0.9402, + "step": 9751 + }, + { + "epoch": 0.7329024500225462, + "grad_norm": 2.237575026804651, + "learning_rate": 7.029002669949075e-07, + "loss": 0.989, + "step": 9752 + }, + { + "epoch": 0.7329776040883812, + "grad_norm": 1.652024186347531, + "learning_rate": 7.025297347017922e-07, + "loss": 0.9291, + "step": 9753 + }, + { + "epoch": 0.7330527581542161, + "grad_norm": 2.4648432663616306, + "learning_rate": 7.021592792887585e-07, + "loss": 0.9373, + "step": 9754 + }, + { + "epoch": 0.7331279122200511, + "grad_norm": 4.053886554969382, + "learning_rate": 7.017889007777566e-07, + "loss": 1.0122, + "step": 9755 + }, + { + "epoch": 0.733203066285886, + "grad_norm": 1.5708217269644071, + "learning_rate": 7.014185991907342e-07, + "loss": 0.8826, + "step": 9756 + }, + { + "epoch": 0.7332782203517211, + "grad_norm": 1.6196375913620014, + "learning_rate": 7.010483745496322e-07, + "loss": 1.0151, + "step": 9757 + }, + { + "epoch": 0.733353374417556, + "grad_norm": 2.102263606658571, + "learning_rate": 7.006782268763885e-07, + "loss": 0.9826, + "step": 9758 + }, + { + "epoch": 0.7334285284833909, + "grad_norm": 1.7167089654593823, + "learning_rate": 7.003081561929342e-07, + "loss": 1.0224, + "step": 9759 + }, + { + "epoch": 0.7335036825492259, + "grad_norm": 1.8417029574286645, + "learning_rate": 6.999381625211993e-07, + "loss": 1.0301, + "step": 9760 + }, + { + "epoch": 0.7335788366150608, + "grad_norm": 36.7216554701548, + "learning_rate": 6.995682458831064e-07, + "loss": 0.9745, + "step": 9761 + }, + { + "epoch": 0.7336539906808959, + "grad_norm": 2.8886271402997425, + "learning_rate": 6.991984063005747e-07, + "loss": 0.9949, + "step": 9762 + }, + { + "epoch": 0.7337291447467308, + "grad_norm": 1.9000876205719646, + "learning_rate": 6.988286437955186e-07, + "loss": 0.9938, + "step": 9763 + }, + { + "epoch": 0.7338042988125658, + "grad_norm": 1.8644214396458774, + "learning_rate": 6.984589583898473e-07, + "loss": 0.9055, + "step": 9764 + }, + { + "epoch": 0.7338794528784007, + "grad_norm": 2.859087986950815, + "learning_rate": 6.98089350105467e-07, + "loss": 0.7445, + "step": 9765 + }, + { + "epoch": 0.7339546069442356, + "grad_norm": 4.82764854672776, + "learning_rate": 6.977198189642783e-07, + "loss": 0.9298, + "step": 9766 + }, + { + "epoch": 0.7340297610100707, + "grad_norm": 2.4848931689872096, + "learning_rate": 6.973503649881769e-07, + "loss": 0.7594, + "step": 9767 + }, + { + "epoch": 0.7341049150759056, + "grad_norm": 1.818783293743771, + "learning_rate": 6.969809881990547e-07, + "loss": 0.9941, + "step": 9768 + }, + { + "epoch": 0.7341800691417406, + "grad_norm": 1.3813723376197278, + "learning_rate": 6.966116886187978e-07, + "loss": 0.9671, + "step": 9769 + }, + { + "epoch": 0.7342552232075755, + "grad_norm": 0.6946208814492352, + "learning_rate": 6.962424662692903e-07, + "loss": 0.8354, + "step": 9770 + }, + { + "epoch": 0.7343303772734104, + "grad_norm": 2.647229934908864, + "learning_rate": 6.958733211724089e-07, + "loss": 1.0827, + "step": 9771 + }, + { + "epoch": 0.7344055313392455, + "grad_norm": 1.5403262144605427, + "learning_rate": 6.955042533500261e-07, + "loss": 0.9424, + "step": 9772 + }, + { + "epoch": 0.7344806854050804, + "grad_norm": 4.005811284602854, + "learning_rate": 6.951352628240133e-07, + "loss": 0.9784, + "step": 9773 + }, + { + "epoch": 0.7345558394709154, + "grad_norm": 0.7167187889869202, + "learning_rate": 6.947663496162313e-07, + "loss": 0.8332, + "step": 9774 + }, + { + "epoch": 0.7346309935367503, + "grad_norm": 1.7413736674913152, + "learning_rate": 6.943975137485418e-07, + "loss": 1.065, + "step": 9775 + }, + { + "epoch": 0.7347061476025853, + "grad_norm": 1.3896540992176478, + "learning_rate": 6.940287552427992e-07, + "loss": 0.9872, + "step": 9776 + }, + { + "epoch": 0.7347813016684203, + "grad_norm": 1.7731835271664365, + "learning_rate": 6.936600741208529e-07, + "loss": 0.9876, + "step": 9777 + }, + { + "epoch": 0.7348564557342552, + "grad_norm": 1.5366472375983382, + "learning_rate": 6.932914704045505e-07, + "loss": 0.9726, + "step": 9778 + }, + { + "epoch": 0.7349316098000902, + "grad_norm": 2.038654856657388, + "learning_rate": 6.929229441157321e-07, + "loss": 1.0175, + "step": 9779 + }, + { + "epoch": 0.7350067638659251, + "grad_norm": 6.163149922050406, + "learning_rate": 6.925544952762341e-07, + "loss": 1.018, + "step": 9780 + }, + { + "epoch": 0.7350819179317601, + "grad_norm": 0.6703153072200583, + "learning_rate": 6.921861239078883e-07, + "loss": 0.8385, + "step": 9781 + }, + { + "epoch": 0.7351570719975951, + "grad_norm": 0.7001232285684039, + "learning_rate": 6.918178300325235e-07, + "loss": 0.8231, + "step": 9782 + }, + { + "epoch": 0.7352322260634301, + "grad_norm": 6.6038486472334, + "learning_rate": 6.914496136719614e-07, + "loss": 0.8927, + "step": 9783 + }, + { + "epoch": 0.735307380129265, + "grad_norm": 1.561737103185015, + "learning_rate": 6.910814748480204e-07, + "loss": 0.8637, + "step": 9784 + }, + { + "epoch": 0.7353825341950999, + "grad_norm": 0.7033163022364234, + "learning_rate": 6.907134135825146e-07, + "loss": 0.8201, + "step": 9785 + }, + { + "epoch": 0.7354576882609349, + "grad_norm": 1.5129124859235832, + "learning_rate": 6.903454298972515e-07, + "loss": 0.9838, + "step": 9786 + }, + { + "epoch": 0.7355328423267699, + "grad_norm": 1.7699523828857955, + "learning_rate": 6.899775238140375e-07, + "loss": 1.067, + "step": 9787 + }, + { + "epoch": 0.7356079963926049, + "grad_norm": 3.120194669665873, + "learning_rate": 6.896096953546717e-07, + "loss": 0.9826, + "step": 9788 + }, + { + "epoch": 0.7356831504584398, + "grad_norm": 1.9039917373718365, + "learning_rate": 6.892419445409492e-07, + "loss": 0.9529, + "step": 9789 + }, + { + "epoch": 0.7357583045242747, + "grad_norm": 1.3981803785533824, + "learning_rate": 6.88874271394661e-07, + "loss": 0.9997, + "step": 9790 + }, + { + "epoch": 0.7358334585901097, + "grad_norm": 1.5168577077388417, + "learning_rate": 6.885066759375917e-07, + "loss": 1.0518, + "step": 9791 + }, + { + "epoch": 0.7359086126559446, + "grad_norm": 1.5569665201516205, + "learning_rate": 6.881391581915248e-07, + "loss": 0.9625, + "step": 9792 + }, + { + "epoch": 0.7359837667217797, + "grad_norm": 1.6809741393034088, + "learning_rate": 6.877717181782363e-07, + "loss": 0.9578, + "step": 9793 + }, + { + "epoch": 0.7360589207876146, + "grad_norm": 1.586879657078956, + "learning_rate": 6.874043559194976e-07, + "loss": 0.9364, + "step": 9794 + }, + { + "epoch": 0.7361340748534496, + "grad_norm": 0.5810267081055721, + "learning_rate": 6.870370714370784e-07, + "loss": 0.7863, + "step": 9795 + }, + { + "epoch": 0.7362092289192845, + "grad_norm": 1.3336401477110424, + "learning_rate": 6.866698647527391e-07, + "loss": 0.9048, + "step": 9796 + }, + { + "epoch": 0.7362843829851194, + "grad_norm": 1.6237370645126756, + "learning_rate": 6.8630273588824e-07, + "loss": 1.0027, + "step": 9797 + }, + { + "epoch": 0.7363595370509545, + "grad_norm": 1.6996284078743553, + "learning_rate": 6.859356848653344e-07, + "loss": 0.9286, + "step": 9798 + }, + { + "epoch": 0.7364346911167894, + "grad_norm": 2.1875920957524695, + "learning_rate": 6.855687117057707e-07, + "loss": 1.0195, + "step": 9799 + }, + { + "epoch": 0.7365098451826244, + "grad_norm": 1.9531659541604696, + "learning_rate": 6.852018164312947e-07, + "loss": 0.9999, + "step": 9800 + }, + { + "epoch": 0.7365849992484593, + "grad_norm": 1.9828411993019435, + "learning_rate": 6.848349990636457e-07, + "loss": 0.8854, + "step": 9801 + }, + { + "epoch": 0.7366601533142944, + "grad_norm": 1.9750500346738895, + "learning_rate": 6.844682596245592e-07, + "loss": 0.967, + "step": 9802 + }, + { + "epoch": 0.7367353073801293, + "grad_norm": 1.4078486786151263, + "learning_rate": 6.841015981357652e-07, + "loss": 0.9647, + "step": 9803 + }, + { + "epoch": 0.7368104614459642, + "grad_norm": 1.9997505390039179, + "learning_rate": 6.837350146189909e-07, + "loss": 0.9607, + "step": 9804 + }, + { + "epoch": 0.7368856155117992, + "grad_norm": 1.6325412506686892, + "learning_rate": 6.833685090959575e-07, + "loss": 0.9496, + "step": 9805 + }, + { + "epoch": 0.7369607695776341, + "grad_norm": 1.863745097838926, + "learning_rate": 6.830020815883815e-07, + "loss": 1.0626, + "step": 9806 + }, + { + "epoch": 0.7370359236434691, + "grad_norm": 1.8113391710528242, + "learning_rate": 6.826357321179754e-07, + "loss": 0.9892, + "step": 9807 + }, + { + "epoch": 0.7371110777093041, + "grad_norm": 1.760343501208936, + "learning_rate": 6.822694607064461e-07, + "loss": 0.8769, + "step": 9808 + }, + { + "epoch": 0.7371862317751391, + "grad_norm": 1.399198825679415, + "learning_rate": 6.819032673754976e-07, + "loss": 1.0283, + "step": 9809 + }, + { + "epoch": 0.737261385840974, + "grad_norm": 2.4438186323669875, + "learning_rate": 6.81537152146828e-07, + "loss": 0.9332, + "step": 9810 + }, + { + "epoch": 0.7373365399068089, + "grad_norm": 3.2183858424729426, + "learning_rate": 6.811711150421309e-07, + "loss": 0.9649, + "step": 9811 + }, + { + "epoch": 0.737411693972644, + "grad_norm": 3.1861249164199803, + "learning_rate": 6.808051560830954e-07, + "loss": 0.9792, + "step": 9812 + }, + { + "epoch": 0.7374868480384789, + "grad_norm": 2.6885780124111704, + "learning_rate": 6.804392752914052e-07, + "loss": 0.9805, + "step": 9813 + }, + { + "epoch": 0.7375620021043139, + "grad_norm": 1.7412509196895847, + "learning_rate": 6.800734726887416e-07, + "loss": 1.0077, + "step": 9814 + }, + { + "epoch": 0.7376371561701488, + "grad_norm": 2.1546407029213253, + "learning_rate": 6.79707748296779e-07, + "loss": 0.9582, + "step": 9815 + }, + { + "epoch": 0.7377123102359837, + "grad_norm": 2.1306766533066543, + "learning_rate": 6.793421021371872e-07, + "loss": 0.9687, + "step": 9816 + }, + { + "epoch": 0.7377874643018187, + "grad_norm": 1.7270787360571525, + "learning_rate": 6.789765342316341e-07, + "loss": 1.0332, + "step": 9817 + }, + { + "epoch": 0.7378626183676537, + "grad_norm": 2.1044270445616706, + "learning_rate": 6.786110446017794e-07, + "loss": 0.9593, + "step": 9818 + }, + { + "epoch": 0.7379377724334887, + "grad_norm": 2.964955203407858, + "learning_rate": 6.782456332692805e-07, + "loss": 0.9794, + "step": 9819 + }, + { + "epoch": 0.7380129264993236, + "grad_norm": 1.9741802711495766, + "learning_rate": 6.778803002557891e-07, + "loss": 1.0012, + "step": 9820 + }, + { + "epoch": 0.7380880805651586, + "grad_norm": 2.18499470707931, + "learning_rate": 6.775150455829521e-07, + "loss": 1.0071, + "step": 9821 + }, + { + "epoch": 0.7381632346309935, + "grad_norm": 3.2373029199655807, + "learning_rate": 6.771498692724133e-07, + "loss": 0.9436, + "step": 9822 + }, + { + "epoch": 0.7382383886968285, + "grad_norm": 1.7971059899509076, + "learning_rate": 6.767847713458104e-07, + "loss": 0.9005, + "step": 9823 + }, + { + "epoch": 0.7383135427626635, + "grad_norm": 1.7116371013436718, + "learning_rate": 6.764197518247767e-07, + "loss": 1.0112, + "step": 9824 + }, + { + "epoch": 0.7383886968284984, + "grad_norm": 1.8056877633948516, + "learning_rate": 6.760548107309403e-07, + "loss": 1.0628, + "step": 9825 + }, + { + "epoch": 0.7384638508943334, + "grad_norm": 1.7132680688047441, + "learning_rate": 6.756899480859268e-07, + "loss": 1.0081, + "step": 9826 + }, + { + "epoch": 0.7385390049601683, + "grad_norm": 1.4718766716625657, + "learning_rate": 6.753251639113551e-07, + "loss": 0.9099, + "step": 9827 + }, + { + "epoch": 0.7386141590260034, + "grad_norm": 1.3941116342819653, + "learning_rate": 6.749604582288397e-07, + "loss": 0.9817, + "step": 9828 + }, + { + "epoch": 0.7386893130918383, + "grad_norm": 0.6368746888344317, + "learning_rate": 6.745958310599913e-07, + "loss": 0.738, + "step": 9829 + }, + { + "epoch": 0.7387644671576732, + "grad_norm": 1.8553104373251545, + "learning_rate": 6.742312824264145e-07, + "loss": 1.0352, + "step": 9830 + }, + { + "epoch": 0.7388396212235082, + "grad_norm": 1.7327577078354348, + "learning_rate": 6.738668123497115e-07, + "loss": 0.9227, + "step": 9831 + }, + { + "epoch": 0.7389147752893431, + "grad_norm": 1.7784919132266521, + "learning_rate": 6.735024208514782e-07, + "loss": 1.1008, + "step": 9832 + }, + { + "epoch": 0.7389899293551782, + "grad_norm": 1.8987214931736371, + "learning_rate": 6.731381079533056e-07, + "loss": 0.9301, + "step": 9833 + }, + { + "epoch": 0.7390650834210131, + "grad_norm": 1.9942098487847235, + "learning_rate": 6.727738736767812e-07, + "loss": 0.9318, + "step": 9834 + }, + { + "epoch": 0.739140237486848, + "grad_norm": 1.7150442227814977, + "learning_rate": 6.724097180434865e-07, + "loss": 0.9301, + "step": 9835 + }, + { + "epoch": 0.739215391552683, + "grad_norm": 1.9390694814059894, + "learning_rate": 6.720456410750002e-07, + "loss": 0.9237, + "step": 9836 + }, + { + "epoch": 0.7392905456185179, + "grad_norm": 1.8327265804669282, + "learning_rate": 6.716816427928949e-07, + "loss": 0.9768, + "step": 9837 + }, + { + "epoch": 0.739365699684353, + "grad_norm": 3.248730925224759, + "learning_rate": 6.71317723218738e-07, + "loss": 0.8273, + "step": 9838 + }, + { + "epoch": 0.7394408537501879, + "grad_norm": 1.6953321183295678, + "learning_rate": 6.709538823740943e-07, + "loss": 1.0003, + "step": 9839 + }, + { + "epoch": 0.7395160078160229, + "grad_norm": 3.294609122818026, + "learning_rate": 6.705901202805226e-07, + "loss": 1.0011, + "step": 9840 + }, + { + "epoch": 0.7395911618818578, + "grad_norm": 2.0366849521163664, + "learning_rate": 6.702264369595767e-07, + "loss": 0.9135, + "step": 9841 + }, + { + "epoch": 0.7396663159476927, + "grad_norm": 1.707834923152879, + "learning_rate": 6.698628324328066e-07, + "loss": 0.9457, + "step": 9842 + }, + { + "epoch": 0.7397414700135277, + "grad_norm": 2.2902548747257003, + "learning_rate": 6.694993067217565e-07, + "loss": 0.948, + "step": 9843 + }, + { + "epoch": 0.7398166240793627, + "grad_norm": 1.8325594728194465, + "learning_rate": 6.691358598479679e-07, + "loss": 1.0209, + "step": 9844 + }, + { + "epoch": 0.7398917781451977, + "grad_norm": 1.703940729896902, + "learning_rate": 6.687724918329758e-07, + "loss": 0.9611, + "step": 9845 + }, + { + "epoch": 0.7399669322110326, + "grad_norm": 2.079756934929231, + "learning_rate": 6.684092026983113e-07, + "loss": 0.9759, + "step": 9846 + }, + { + "epoch": 0.7400420862768676, + "grad_norm": 1.67835541178713, + "learning_rate": 6.680459924654997e-07, + "loss": 1.0077, + "step": 9847 + }, + { + "epoch": 0.7401172403427025, + "grad_norm": 0.7497170629869938, + "learning_rate": 6.676828611560643e-07, + "loss": 0.8387, + "step": 9848 + }, + { + "epoch": 0.7401923944085375, + "grad_norm": 1.7450688915262014, + "learning_rate": 6.673198087915211e-07, + "loss": 0.9972, + "step": 9849 + }, + { + "epoch": 0.7402675484743725, + "grad_norm": 2.216525614942624, + "learning_rate": 6.669568353933824e-07, + "loss": 1.0433, + "step": 9850 + }, + { + "epoch": 0.7403427025402074, + "grad_norm": 0.8459097453462329, + "learning_rate": 6.665939409831556e-07, + "loss": 0.8642, + "step": 9851 + }, + { + "epoch": 0.7404178566060424, + "grad_norm": 1.4731988286780435, + "learning_rate": 6.662311255823432e-07, + "loss": 0.9006, + "step": 9852 + }, + { + "epoch": 0.7404930106718773, + "grad_norm": 1.4191010633183871, + "learning_rate": 6.658683892124446e-07, + "loss": 1.043, + "step": 9853 + }, + { + "epoch": 0.7405681647377124, + "grad_norm": 1.9881125268002238, + "learning_rate": 6.655057318949526e-07, + "loss": 0.9442, + "step": 9854 + }, + { + "epoch": 0.7406433188035473, + "grad_norm": 1.6746578407971242, + "learning_rate": 6.651431536513563e-07, + "loss": 0.8825, + "step": 9855 + }, + { + "epoch": 0.7407184728693822, + "grad_norm": 2.179414574821994, + "learning_rate": 6.647806545031396e-07, + "loss": 1.0122, + "step": 9856 + }, + { + "epoch": 0.7407936269352172, + "grad_norm": 2.4156244157047726, + "learning_rate": 6.644182344717813e-07, + "loss": 0.9378, + "step": 9857 + }, + { + "epoch": 0.7408687810010521, + "grad_norm": 1.9153064033459817, + "learning_rate": 6.640558935787575e-07, + "loss": 0.9628, + "step": 9858 + }, + { + "epoch": 0.7409439350668872, + "grad_norm": 2.1814780042345108, + "learning_rate": 6.636936318455377e-07, + "loss": 0.9271, + "step": 9859 + }, + { + "epoch": 0.7410190891327221, + "grad_norm": 2.2062656391227087, + "learning_rate": 6.633314492935866e-07, + "loss": 0.8872, + "step": 9860 + }, + { + "epoch": 0.741094243198557, + "grad_norm": 2.0988488811671417, + "learning_rate": 6.629693459443664e-07, + "loss": 0.9428, + "step": 9861 + }, + { + "epoch": 0.741169397264392, + "grad_norm": 1.8451769204847008, + "learning_rate": 6.62607321819332e-07, + "loss": 0.9659, + "step": 9862 + }, + { + "epoch": 0.7412445513302269, + "grad_norm": 1.5520038541716472, + "learning_rate": 6.622453769399353e-07, + "loss": 0.9858, + "step": 9863 + }, + { + "epoch": 0.741319705396062, + "grad_norm": 1.7006181939271712, + "learning_rate": 6.618835113276225e-07, + "loss": 0.9622, + "step": 9864 + }, + { + "epoch": 0.7413948594618969, + "grad_norm": 2.0756905429304693, + "learning_rate": 6.61521725003835e-07, + "loss": 0.9221, + "step": 9865 + }, + { + "epoch": 0.7414700135277319, + "grad_norm": 2.3864387310482367, + "learning_rate": 6.611600179900112e-07, + "loss": 1.0054, + "step": 9866 + }, + { + "epoch": 0.7415451675935668, + "grad_norm": 3.787458383060858, + "learning_rate": 6.607983903075832e-07, + "loss": 1.0327, + "step": 9867 + }, + { + "epoch": 0.7416203216594017, + "grad_norm": 1.6458482701392352, + "learning_rate": 6.604368419779787e-07, + "loss": 0.9824, + "step": 9868 + }, + { + "epoch": 0.7416954757252368, + "grad_norm": 1.8551494122815089, + "learning_rate": 6.600753730226203e-07, + "loss": 0.981, + "step": 9869 + }, + { + "epoch": 0.7417706297910717, + "grad_norm": 1.5004742985131925, + "learning_rate": 6.597139834629275e-07, + "loss": 0.9685, + "step": 9870 + }, + { + "epoch": 0.7418457838569067, + "grad_norm": 1.6569662164986338, + "learning_rate": 6.593526733203134e-07, + "loss": 0.9744, + "step": 9871 + }, + { + "epoch": 0.7419209379227416, + "grad_norm": 1.8334787340015397, + "learning_rate": 6.589914426161871e-07, + "loss": 1.0499, + "step": 9872 + }, + { + "epoch": 0.7419960919885766, + "grad_norm": 1.9119321840382006, + "learning_rate": 6.58630291371953e-07, + "loss": 0.9852, + "step": 9873 + }, + { + "epoch": 0.7420712460544115, + "grad_norm": 3.9180131721450415, + "learning_rate": 6.582692196090101e-07, + "loss": 1.048, + "step": 9874 + }, + { + "epoch": 0.7421464001202465, + "grad_norm": 1.7392197036637738, + "learning_rate": 6.579082273487541e-07, + "loss": 1.0166, + "step": 9875 + }, + { + "epoch": 0.7422215541860815, + "grad_norm": 2.34944265127646, + "learning_rate": 6.575473146125749e-07, + "loss": 1.0578, + "step": 9876 + }, + { + "epoch": 0.7422967082519164, + "grad_norm": 1.7259494159044715, + "learning_rate": 6.57186481421858e-07, + "loss": 0.9763, + "step": 9877 + }, + { + "epoch": 0.7423718623177514, + "grad_norm": 1.5846423042124695, + "learning_rate": 6.568257277979841e-07, + "loss": 0.9735, + "step": 9878 + }, + { + "epoch": 0.7424470163835863, + "grad_norm": 1.5509349064648672, + "learning_rate": 6.564650537623284e-07, + "loss": 0.8166, + "step": 9879 + }, + { + "epoch": 0.7425221704494213, + "grad_norm": 2.266753553613856, + "learning_rate": 6.561044593362636e-07, + "loss": 1.0625, + "step": 9880 + }, + { + "epoch": 0.7425973245152563, + "grad_norm": 1.9400725359973092, + "learning_rate": 6.557439445411559e-07, + "loss": 0.9763, + "step": 9881 + }, + { + "epoch": 0.7426724785810912, + "grad_norm": 3.7019464734049796, + "learning_rate": 6.553835093983662e-07, + "loss": 0.9783, + "step": 9882 + }, + { + "epoch": 0.7427476326469262, + "grad_norm": 2.4705244037385614, + "learning_rate": 6.550231539292533e-07, + "loss": 0.9297, + "step": 9883 + }, + { + "epoch": 0.7428227867127611, + "grad_norm": 1.7505063959139424, + "learning_rate": 6.546628781551687e-07, + "loss": 0.9695, + "step": 9884 + }, + { + "epoch": 0.7428979407785962, + "grad_norm": 1.343300766483072, + "learning_rate": 6.543026820974599e-07, + "loss": 1.0291, + "step": 9885 + }, + { + "epoch": 0.7429730948444311, + "grad_norm": 1.592356053528386, + "learning_rate": 6.539425657774706e-07, + "loss": 1.0825, + "step": 9886 + }, + { + "epoch": 0.743048248910266, + "grad_norm": 2.0886836459511455, + "learning_rate": 6.535825292165377e-07, + "loss": 0.9632, + "step": 9887 + }, + { + "epoch": 0.743123402976101, + "grad_norm": 2.362728943403699, + "learning_rate": 6.532225724359967e-07, + "loss": 0.9138, + "step": 9888 + }, + { + "epoch": 0.7431985570419359, + "grad_norm": 1.8887418716225464, + "learning_rate": 6.528626954571753e-07, + "loss": 1.0557, + "step": 9889 + }, + { + "epoch": 0.743273711107771, + "grad_norm": 1.6239158241461757, + "learning_rate": 6.525028983013976e-07, + "loss": 0.9255, + "step": 9890 + }, + { + "epoch": 0.7433488651736059, + "grad_norm": 1.5415390924130015, + "learning_rate": 6.521431809899827e-07, + "loss": 0.849, + "step": 9891 + }, + { + "epoch": 0.7434240192394409, + "grad_norm": 2.290262890514103, + "learning_rate": 6.517835435442461e-07, + "loss": 0.9144, + "step": 9892 + }, + { + "epoch": 0.7434991733052758, + "grad_norm": 1.6836965578202974, + "learning_rate": 6.514239859854973e-07, + "loss": 1.0274, + "step": 9893 + }, + { + "epoch": 0.7435743273711107, + "grad_norm": 3.2419870337592327, + "learning_rate": 6.510645083350412e-07, + "loss": 0.8894, + "step": 9894 + }, + { + "epoch": 0.7436494814369458, + "grad_norm": 2.0346779874550855, + "learning_rate": 6.507051106141786e-07, + "loss": 0.9278, + "step": 9895 + }, + { + "epoch": 0.7437246355027807, + "grad_norm": 2.777266403387425, + "learning_rate": 6.503457928442042e-07, + "loss": 0.9275, + "step": 9896 + }, + { + "epoch": 0.7437997895686157, + "grad_norm": 2.426647578293618, + "learning_rate": 6.499865550464103e-07, + "loss": 1.0393, + "step": 9897 + }, + { + "epoch": 0.7438749436344506, + "grad_norm": 6.878737387461038, + "learning_rate": 6.496273972420827e-07, + "loss": 1.0085, + "step": 9898 + }, + { + "epoch": 0.7439500977002856, + "grad_norm": 1.3923205636394547, + "learning_rate": 6.492683194525028e-07, + "loss": 0.9102, + "step": 9899 + }, + { + "epoch": 0.7440252517661206, + "grad_norm": 2.0906621453559087, + "learning_rate": 6.489093216989472e-07, + "loss": 1.022, + "step": 9900 + }, + { + "epoch": 0.7441004058319555, + "grad_norm": 1.808582347067691, + "learning_rate": 6.485504040026872e-07, + "loss": 1.0053, + "step": 9901 + }, + { + "epoch": 0.7441755598977905, + "grad_norm": 1.7941126280763147, + "learning_rate": 6.481915663849917e-07, + "loss": 0.9743, + "step": 9902 + }, + { + "epoch": 0.7442507139636254, + "grad_norm": 1.6257394428734582, + "learning_rate": 6.478328088671221e-07, + "loss": 0.9047, + "step": 9903 + }, + { + "epoch": 0.7443258680294604, + "grad_norm": 2.37652689388571, + "learning_rate": 6.474741314703358e-07, + "loss": 0.9487, + "step": 9904 + }, + { + "epoch": 0.7444010220952954, + "grad_norm": 1.2712063111756544, + "learning_rate": 6.471155342158871e-07, + "loss": 0.9931, + "step": 9905 + }, + { + "epoch": 0.7444761761611303, + "grad_norm": 1.586952748927482, + "learning_rate": 6.467570171250234e-07, + "loss": 0.9547, + "step": 9906 + }, + { + "epoch": 0.7445513302269653, + "grad_norm": 1.6917918692924099, + "learning_rate": 6.463985802189884e-07, + "loss": 0.8114, + "step": 9907 + }, + { + "epoch": 0.7446264842928002, + "grad_norm": 0.8637118754274629, + "learning_rate": 6.46040223519021e-07, + "loss": 0.9404, + "step": 9908 + }, + { + "epoch": 0.7447016383586352, + "grad_norm": 1.834018824764638, + "learning_rate": 6.456819470463542e-07, + "loss": 0.8384, + "step": 9909 + }, + { + "epoch": 0.7447767924244701, + "grad_norm": 1.9154761251875028, + "learning_rate": 6.453237508222186e-07, + "loss": 0.9047, + "step": 9910 + }, + { + "epoch": 0.7448519464903052, + "grad_norm": 2.1152069850849577, + "learning_rate": 6.449656348678383e-07, + "loss": 0.9757, + "step": 9911 + }, + { + "epoch": 0.7449271005561401, + "grad_norm": 2.0216440626087837, + "learning_rate": 6.446075992044329e-07, + "loss": 0.8937, + "step": 9912 + }, + { + "epoch": 0.745002254621975, + "grad_norm": 2.5084092805671583, + "learning_rate": 6.442496438532168e-07, + "loss": 0.9528, + "step": 9913 + }, + { + "epoch": 0.74507740868781, + "grad_norm": 0.8557952536481416, + "learning_rate": 6.438917688354013e-07, + "loss": 0.9973, + "step": 9914 + }, + { + "epoch": 0.7451525627536449, + "grad_norm": 2.0879118644411974, + "learning_rate": 6.435339741721915e-07, + "loss": 0.9889, + "step": 9915 + }, + { + "epoch": 0.74522771681948, + "grad_norm": 1.6932755991838009, + "learning_rate": 6.431762598847879e-07, + "loss": 0.9095, + "step": 9916 + }, + { + "epoch": 0.7453028708853149, + "grad_norm": 1.5682258216554494, + "learning_rate": 6.428186259943866e-07, + "loss": 0.9497, + "step": 9917 + }, + { + "epoch": 0.7453780249511499, + "grad_norm": 0.6909609723945825, + "learning_rate": 6.424610725221779e-07, + "loss": 0.8177, + "step": 9918 + }, + { + "epoch": 0.7454531790169848, + "grad_norm": 1.9586551956380123, + "learning_rate": 6.4210359948935e-07, + "loss": 1.034, + "step": 9919 + }, + { + "epoch": 0.7455283330828197, + "grad_norm": 3.1078429161643184, + "learning_rate": 6.417462069170834e-07, + "loss": 1.0126, + "step": 9920 + }, + { + "epoch": 0.7456034871486548, + "grad_norm": 1.7384815572133971, + "learning_rate": 6.413888948265551e-07, + "loss": 0.9828, + "step": 9921 + }, + { + "epoch": 0.7456786412144897, + "grad_norm": 3.5143286455392047, + "learning_rate": 6.410316632389372e-07, + "loss": 1.1001, + "step": 9922 + }, + { + "epoch": 0.7457537952803247, + "grad_norm": 0.6926788213844116, + "learning_rate": 6.406745121753964e-07, + "loss": 0.8263, + "step": 9923 + }, + { + "epoch": 0.7458289493461596, + "grad_norm": 2.15053874518253, + "learning_rate": 6.403174416570967e-07, + "loss": 0.895, + "step": 9924 + }, + { + "epoch": 0.7459041034119945, + "grad_norm": 4.006434510455932, + "learning_rate": 6.399604517051953e-07, + "loss": 1.0162, + "step": 9925 + }, + { + "epoch": 0.7459792574778296, + "grad_norm": 4.210233781716177, + "learning_rate": 6.396035423408442e-07, + "loss": 0.9563, + "step": 9926 + }, + { + "epoch": 0.7460544115436645, + "grad_norm": 2.7503564776942153, + "learning_rate": 6.392467135851934e-07, + "loss": 0.9417, + "step": 9927 + }, + { + "epoch": 0.7461295656094995, + "grad_norm": 3.0686482585504304, + "learning_rate": 6.388899654593853e-07, + "loss": 0.9307, + "step": 9928 + }, + { + "epoch": 0.7462047196753344, + "grad_norm": 13.63798757882538, + "learning_rate": 6.385332979845588e-07, + "loss": 0.9356, + "step": 9929 + }, + { + "epoch": 0.7462798737411694, + "grad_norm": 1.893178492256631, + "learning_rate": 6.381767111818479e-07, + "loss": 1.0152, + "step": 9930 + }, + { + "epoch": 0.7463550278070044, + "grad_norm": 2.25508010417482, + "learning_rate": 6.378202050723809e-07, + "loss": 0.87, + "step": 9931 + }, + { + "epoch": 0.7464301818728393, + "grad_norm": 1.390757307980743, + "learning_rate": 6.374637796772835e-07, + "loss": 1.0197, + "step": 9932 + }, + { + "epoch": 0.7465053359386743, + "grad_norm": 1.5360718925410248, + "learning_rate": 6.371074350176746e-07, + "loss": 0.976, + "step": 9933 + }, + { + "epoch": 0.7465804900045092, + "grad_norm": 1.773946142985432, + "learning_rate": 6.367511711146691e-07, + "loss": 0.9812, + "step": 9934 + }, + { + "epoch": 0.7466556440703442, + "grad_norm": 2.0238305376602104, + "learning_rate": 6.363949879893764e-07, + "loss": 1.0305, + "step": 9935 + }, + { + "epoch": 0.7467307981361792, + "grad_norm": 1.3800910851922505, + "learning_rate": 6.360388856629029e-07, + "loss": 0.9464, + "step": 9936 + }, + { + "epoch": 0.7468059522020142, + "grad_norm": 2.5011899976954037, + "learning_rate": 6.356828641563483e-07, + "loss": 0.9628, + "step": 9937 + }, + { + "epoch": 0.7468811062678491, + "grad_norm": 2.2148629108656417, + "learning_rate": 6.353269234908083e-07, + "loss": 0.9011, + "step": 9938 + }, + { + "epoch": 0.746956260333684, + "grad_norm": 2.5649310419877933, + "learning_rate": 6.349710636873739e-07, + "loss": 1.0224, + "step": 9939 + }, + { + "epoch": 0.747031414399519, + "grad_norm": 2.4498890715418336, + "learning_rate": 6.346152847671302e-07, + "loss": 0.9603, + "step": 9940 + }, + { + "epoch": 0.747106568465354, + "grad_norm": 0.7685072228632712, + "learning_rate": 6.3425958675116e-07, + "loss": 0.8395, + "step": 9941 + }, + { + "epoch": 0.747181722531189, + "grad_norm": 1.7267727961136616, + "learning_rate": 6.33903969660539e-07, + "loss": 1.0554, + "step": 9942 + }, + { + "epoch": 0.7472568765970239, + "grad_norm": 2.5348883018488886, + "learning_rate": 6.335484335163384e-07, + "loss": 0.929, + "step": 9943 + }, + { + "epoch": 0.7473320306628589, + "grad_norm": 2.2418393386459017, + "learning_rate": 6.331929783396268e-07, + "loss": 0.9513, + "step": 9944 + }, + { + "epoch": 0.7474071847286938, + "grad_norm": 2.0110020814616445, + "learning_rate": 6.32837604151464e-07, + "loss": 0.9885, + "step": 9945 + }, + { + "epoch": 0.7474823387945287, + "grad_norm": 1.8562640282714145, + "learning_rate": 6.324823109729087e-07, + "loss": 1.0399, + "step": 9946 + }, + { + "epoch": 0.7475574928603638, + "grad_norm": 1.6165337462036964, + "learning_rate": 6.321270988250134e-07, + "loss": 0.9463, + "step": 9947 + }, + { + "epoch": 0.7476326469261987, + "grad_norm": 1.3657374948916805, + "learning_rate": 6.317719677288245e-07, + "loss": 0.9988, + "step": 9948 + }, + { + "epoch": 0.7477078009920337, + "grad_norm": 11.16343615665582, + "learning_rate": 6.314169177053866e-07, + "loss": 0.9578, + "step": 9949 + }, + { + "epoch": 0.7477829550578686, + "grad_norm": 2.1097920323431807, + "learning_rate": 6.310619487757369e-07, + "loss": 0.8714, + "step": 9950 + }, + { + "epoch": 0.7478581091237035, + "grad_norm": 1.6611643165323489, + "learning_rate": 6.307070609609086e-07, + "loss": 0.953, + "step": 9951 + }, + { + "epoch": 0.7479332631895386, + "grad_norm": 1.7777726881961022, + "learning_rate": 6.303522542819306e-07, + "loss": 0.9985, + "step": 9952 + }, + { + "epoch": 0.7480084172553735, + "grad_norm": 1.8982458449586639, + "learning_rate": 6.299975287598255e-07, + "loss": 0.9896, + "step": 9953 + }, + { + "epoch": 0.7480835713212085, + "grad_norm": 3.562277134399173, + "learning_rate": 6.296428844156137e-07, + "loss": 1.0407, + "step": 9954 + }, + { + "epoch": 0.7481587253870434, + "grad_norm": 1.3480022483829572, + "learning_rate": 6.292883212703082e-07, + "loss": 0.956, + "step": 9955 + }, + { + "epoch": 0.7482338794528784, + "grad_norm": 1.9983871632575259, + "learning_rate": 6.289338393449187e-07, + "loss": 0.9712, + "step": 9956 + }, + { + "epoch": 0.7483090335187134, + "grad_norm": 1.8688719744728943, + "learning_rate": 6.285794386604484e-07, + "loss": 0.9053, + "step": 9957 + }, + { + "epoch": 0.7483841875845483, + "grad_norm": 1.6530888457932869, + "learning_rate": 6.282251192378987e-07, + "loss": 0.9966, + "step": 9958 + }, + { + "epoch": 0.7484593416503833, + "grad_norm": 1.8444437564455356, + "learning_rate": 6.278708810982635e-07, + "loss": 0.8956, + "step": 9959 + }, + { + "epoch": 0.7485344957162182, + "grad_norm": 1.6301183787462175, + "learning_rate": 6.27516724262533e-07, + "loss": 0.9723, + "step": 9960 + }, + { + "epoch": 0.7486096497820532, + "grad_norm": 1.923654802134538, + "learning_rate": 6.271626487516921e-07, + "loss": 0.8946, + "step": 9961 + }, + { + "epoch": 0.7486848038478882, + "grad_norm": 1.3942823428321498, + "learning_rate": 6.268086545867206e-07, + "loss": 0.9911, + "step": 9962 + }, + { + "epoch": 0.7487599579137232, + "grad_norm": 1.7656699655934236, + "learning_rate": 6.264547417885953e-07, + "loss": 0.9175, + "step": 9963 + }, + { + "epoch": 0.7488351119795581, + "grad_norm": 1.4271279859402184, + "learning_rate": 6.261009103782861e-07, + "loss": 0.991, + "step": 9964 + }, + { + "epoch": 0.748910266045393, + "grad_norm": 1.6054762494359895, + "learning_rate": 6.257471603767583e-07, + "loss": 0.9955, + "step": 9965 + }, + { + "epoch": 0.748985420111228, + "grad_norm": 1.8785782684396428, + "learning_rate": 6.25393491804975e-07, + "loss": 1.0314, + "step": 9966 + }, + { + "epoch": 0.749060574177063, + "grad_norm": 0.8244212837327735, + "learning_rate": 6.250399046838897e-07, + "loss": 0.7994, + "step": 9967 + }, + { + "epoch": 0.749135728242898, + "grad_norm": 3.557832571583699, + "learning_rate": 6.246863990344557e-07, + "loss": 1.0004, + "step": 9968 + }, + { + "epoch": 0.7492108823087329, + "grad_norm": 2.7832283955815478, + "learning_rate": 6.243329748776192e-07, + "loss": 0.9767, + "step": 9969 + }, + { + "epoch": 0.7492860363745678, + "grad_norm": 1.8583507120776719, + "learning_rate": 6.23979632234321e-07, + "loss": 1.0605, + "step": 9970 + }, + { + "epoch": 0.7493611904404028, + "grad_norm": 1.7715915702174836, + "learning_rate": 6.236263711254993e-07, + "loss": 0.92, + "step": 9971 + }, + { + "epoch": 0.7494363445062378, + "grad_norm": 1.8171613538834033, + "learning_rate": 6.232731915720855e-07, + "loss": 0.9059, + "step": 9972 + }, + { + "epoch": 0.7495114985720728, + "grad_norm": 0.7847811433920171, + "learning_rate": 6.22920093595007e-07, + "loss": 0.8337, + "step": 9973 + }, + { + "epoch": 0.7495866526379077, + "grad_norm": 5.31889012798756, + "learning_rate": 6.225670772151861e-07, + "loss": 0.8942, + "step": 9974 + }, + { + "epoch": 0.7496618067037427, + "grad_norm": 1.4251010055884286, + "learning_rate": 6.222141424535399e-07, + "loss": 0.9434, + "step": 9975 + }, + { + "epoch": 0.7497369607695776, + "grad_norm": 1.4828566967695485, + "learning_rate": 6.218612893309823e-07, + "loss": 0.9511, + "step": 9976 + }, + { + "epoch": 0.7498121148354125, + "grad_norm": 2.419586872746177, + "learning_rate": 6.215085178684205e-07, + "loss": 0.9592, + "step": 9977 + }, + { + "epoch": 0.7498872689012476, + "grad_norm": 0.7971305437426137, + "learning_rate": 6.211558280867575e-07, + "loss": 0.8331, + "step": 9978 + }, + { + "epoch": 0.7499624229670825, + "grad_norm": 1.771021335570219, + "learning_rate": 6.208032200068911e-07, + "loss": 0.8637, + "step": 9979 + }, + { + "epoch": 0.7500375770329175, + "grad_norm": 1.9127726042696584, + "learning_rate": 6.20450693649716e-07, + "loss": 0.9928, + "step": 9980 + }, + { + "epoch": 0.7501127310987524, + "grad_norm": 2.4776587722360364, + "learning_rate": 6.200982490361197e-07, + "loss": 1.0154, + "step": 9981 + }, + { + "epoch": 0.7501878851645875, + "grad_norm": 3.6269740287412557, + "learning_rate": 6.197458861869862e-07, + "loss": 0.9905, + "step": 9982 + }, + { + "epoch": 0.7502630392304224, + "grad_norm": 1.98535450467456, + "learning_rate": 6.193936051231945e-07, + "loss": 0.9122, + "step": 9983 + }, + { + "epoch": 0.7503381932962573, + "grad_norm": 4.369747387461271, + "learning_rate": 6.190414058656175e-07, + "loss": 0.9356, + "step": 9984 + }, + { + "epoch": 0.7504133473620923, + "grad_norm": 1.5164285476358057, + "learning_rate": 6.18689288435126e-07, + "loss": 0.9648, + "step": 9985 + }, + { + "epoch": 0.7504885014279272, + "grad_norm": 1.8191141924044583, + "learning_rate": 6.183372528525834e-07, + "loss": 0.8714, + "step": 9986 + }, + { + "epoch": 0.7505636554937622, + "grad_norm": 2.3866712999387696, + "learning_rate": 6.17985299138849e-07, + "loss": 0.9539, + "step": 9987 + }, + { + "epoch": 0.7506388095595972, + "grad_norm": 1.6582517942513488, + "learning_rate": 6.17633427314778e-07, + "loss": 0.863, + "step": 9988 + }, + { + "epoch": 0.7507139636254322, + "grad_norm": 3.8297155695944496, + "learning_rate": 6.1728163740122e-07, + "loss": 0.9666, + "step": 9989 + }, + { + "epoch": 0.7507891176912671, + "grad_norm": 0.7434346211567284, + "learning_rate": 6.169299294190198e-07, + "loss": 0.8664, + "step": 9990 + }, + { + "epoch": 0.750864271757102, + "grad_norm": 1.655284168129857, + "learning_rate": 6.165783033890175e-07, + "loss": 0.9586, + "step": 9991 + }, + { + "epoch": 0.750939425822937, + "grad_norm": 12.12779156512307, + "learning_rate": 6.162267593320474e-07, + "loss": 1.0756, + "step": 9992 + }, + { + "epoch": 0.751014579888772, + "grad_norm": 2.0306183052833218, + "learning_rate": 6.158752972689414e-07, + "loss": 1.0258, + "step": 9993 + }, + { + "epoch": 0.751089733954607, + "grad_norm": 1.5129680350760035, + "learning_rate": 6.15523917220524e-07, + "loss": 0.9263, + "step": 9994 + }, + { + "epoch": 0.7511648880204419, + "grad_norm": 1.6570333391794203, + "learning_rate": 6.151726192076161e-07, + "loss": 0.8893, + "step": 9995 + }, + { + "epoch": 0.7512400420862768, + "grad_norm": 1.680604267552841, + "learning_rate": 6.148214032510335e-07, + "loss": 1.0206, + "step": 9996 + }, + { + "epoch": 0.7513151961521118, + "grad_norm": 2.1115061978631515, + "learning_rate": 6.144702693715862e-07, + "loss": 1.0175, + "step": 9997 + }, + { + "epoch": 0.7513903502179468, + "grad_norm": 1.4531756976728192, + "learning_rate": 6.141192175900818e-07, + "loss": 0.9689, + "step": 9998 + }, + { + "epoch": 0.7514655042837818, + "grad_norm": 7.959363487903311, + "learning_rate": 6.137682479273205e-07, + "loss": 0.9353, + "step": 9999 + }, + { + "epoch": 0.7515406583496167, + "grad_norm": 1.844162716982692, + "learning_rate": 6.134173604040987e-07, + "loss": 1.0074, + "step": 10000 + }, + { + "epoch": 0.7516158124154517, + "grad_norm": 2.044708969560861, + "learning_rate": 6.130665550412073e-07, + "loss": 0.9863, + "step": 10001 + }, + { + "epoch": 0.7516909664812866, + "grad_norm": 1.949384968919572, + "learning_rate": 6.127158318594341e-07, + "loss": 1.0404, + "step": 10002 + }, + { + "epoch": 0.7517661205471216, + "grad_norm": 1.7130291142258216, + "learning_rate": 6.1236519087956e-07, + "loss": 0.9872, + "step": 10003 + }, + { + "epoch": 0.7518412746129566, + "grad_norm": 2.5258769712472047, + "learning_rate": 6.12014632122362e-07, + "loss": 0.9825, + "step": 10004 + }, + { + "epoch": 0.7519164286787915, + "grad_norm": 2.761617828640238, + "learning_rate": 6.116641556086122e-07, + "loss": 0.9535, + "step": 10005 + }, + { + "epoch": 0.7519915827446265, + "grad_norm": 1.880876614342075, + "learning_rate": 6.113137613590767e-07, + "loss": 1.0155, + "step": 10006 + }, + { + "epoch": 0.7520667368104614, + "grad_norm": 1.5282900481228707, + "learning_rate": 6.109634493945191e-07, + "loss": 0.9329, + "step": 10007 + }, + { + "epoch": 0.7521418908762965, + "grad_norm": 1.9539519123536293, + "learning_rate": 6.106132197356959e-07, + "loss": 0.9726, + "step": 10008 + }, + { + "epoch": 0.7522170449421314, + "grad_norm": 1.5290442948745717, + "learning_rate": 6.102630724033593e-07, + "loss": 1.0818, + "step": 10009 + }, + { + "epoch": 0.7522921990079663, + "grad_norm": 2.433077296630375, + "learning_rate": 6.099130074182581e-07, + "loss": 0.9824, + "step": 10010 + }, + { + "epoch": 0.7523673530738013, + "grad_norm": 1.7312137558651457, + "learning_rate": 6.095630248011341e-07, + "loss": 0.93, + "step": 10011 + }, + { + "epoch": 0.7524425071396362, + "grad_norm": 1.5774023082003823, + "learning_rate": 6.09213124572725e-07, + "loss": 0.9785, + "step": 10012 + }, + { + "epoch": 0.7525176612054713, + "grad_norm": 1.5221444927257166, + "learning_rate": 6.088633067537643e-07, + "loss": 1.0126, + "step": 10013 + }, + { + "epoch": 0.7525928152713062, + "grad_norm": 3.2061628125328805, + "learning_rate": 6.085135713649787e-07, + "loss": 0.999, + "step": 10014 + }, + { + "epoch": 0.7526679693371411, + "grad_norm": 1.950637820665248, + "learning_rate": 6.081639184270932e-07, + "loss": 0.9789, + "step": 10015 + }, + { + "epoch": 0.7527431234029761, + "grad_norm": 1.6955834800622787, + "learning_rate": 6.078143479608253e-07, + "loss": 1.0177, + "step": 10016 + }, + { + "epoch": 0.752818277468811, + "grad_norm": 2.100720431865036, + "learning_rate": 6.074648599868884e-07, + "loss": 1.053, + "step": 10017 + }, + { + "epoch": 0.752893431534646, + "grad_norm": 2.0398040241652464, + "learning_rate": 6.071154545259907e-07, + "loss": 0.9456, + "step": 10018 + }, + { + "epoch": 0.752968585600481, + "grad_norm": 1.9050003708566219, + "learning_rate": 6.067661315988353e-07, + "loss": 1.0607, + "step": 10019 + }, + { + "epoch": 0.753043739666316, + "grad_norm": 0.6297385826877966, + "learning_rate": 6.064168912261225e-07, + "loss": 0.8118, + "step": 10020 + }, + { + "epoch": 0.7531188937321509, + "grad_norm": 0.8365639451662448, + "learning_rate": 6.060677334285452e-07, + "loss": 0.8822, + "step": 10021 + }, + { + "epoch": 0.7531940477979858, + "grad_norm": 1.809442176955372, + "learning_rate": 6.057186582267923e-07, + "loss": 0.9985, + "step": 10022 + }, + { + "epoch": 0.7532692018638208, + "grad_norm": 2.723503429776955, + "learning_rate": 6.053696656415474e-07, + "loss": 1.0079, + "step": 10023 + }, + { + "epoch": 0.7533443559296558, + "grad_norm": 1.6185115240875259, + "learning_rate": 6.05020755693491e-07, + "loss": 1.0437, + "step": 10024 + }, + { + "epoch": 0.7534195099954908, + "grad_norm": 1.6045255465995398, + "learning_rate": 6.046719284032963e-07, + "loss": 0.8209, + "step": 10025 + }, + { + "epoch": 0.7534946640613257, + "grad_norm": 1.3421220010711683, + "learning_rate": 6.043231837916332e-07, + "loss": 1.0057, + "step": 10026 + }, + { + "epoch": 0.7535698181271607, + "grad_norm": 1.9278545863924526, + "learning_rate": 6.039745218791658e-07, + "loss": 0.9081, + "step": 10027 + }, + { + "epoch": 0.7536449721929956, + "grad_norm": 1.7738896268336048, + "learning_rate": 6.036259426865531e-07, + "loss": 1.0098, + "step": 10028 + }, + { + "epoch": 0.7537201262588306, + "grad_norm": 1.6884260109371627, + "learning_rate": 6.032774462344507e-07, + "loss": 0.9417, + "step": 10029 + }, + { + "epoch": 0.7537952803246656, + "grad_norm": 1.4576630641480708, + "learning_rate": 6.029290325435084e-07, + "loss": 0.8459, + "step": 10030 + }, + { + "epoch": 0.7538704343905005, + "grad_norm": 1.5122615815455593, + "learning_rate": 6.025807016343698e-07, + "loss": 1.0217, + "step": 10031 + }, + { + "epoch": 0.7539455884563355, + "grad_norm": 1.4298663418335376, + "learning_rate": 6.022324535276763e-07, + "loss": 1.0518, + "step": 10032 + }, + { + "epoch": 0.7540207425221704, + "grad_norm": 2.113322090135063, + "learning_rate": 6.018842882440625e-07, + "loss": 0.8836, + "step": 10033 + }, + { + "epoch": 0.7540958965880055, + "grad_norm": 0.7710357520525869, + "learning_rate": 6.015362058041584e-07, + "loss": 0.8691, + "step": 10034 + }, + { + "epoch": 0.7541710506538404, + "grad_norm": 1.8269141335982755, + "learning_rate": 6.011882062285892e-07, + "loss": 1.1002, + "step": 10035 + }, + { + "epoch": 0.7542462047196753, + "grad_norm": 1.8882069533239552, + "learning_rate": 6.008402895379743e-07, + "loss": 0.9897, + "step": 10036 + }, + { + "epoch": 0.7543213587855103, + "grad_norm": 2.068446055785202, + "learning_rate": 6.004924557529307e-07, + "loss": 1.1235, + "step": 10037 + }, + { + "epoch": 0.7543965128513452, + "grad_norm": 1.7996260439989014, + "learning_rate": 6.001447048940682e-07, + "loss": 0.9638, + "step": 10038 + }, + { + "epoch": 0.7544716669171803, + "grad_norm": 1.7226237944921299, + "learning_rate": 5.997970369819925e-07, + "loss": 0.8614, + "step": 10039 + }, + { + "epoch": 0.7545468209830152, + "grad_norm": 1.9518595754189236, + "learning_rate": 5.994494520373039e-07, + "loss": 1.0039, + "step": 10040 + }, + { + "epoch": 0.7546219750488501, + "grad_norm": 1.8431412289882558, + "learning_rate": 5.991019500805976e-07, + "loss": 1.022, + "step": 10041 + }, + { + "epoch": 0.7546971291146851, + "grad_norm": 1.6970921330075923, + "learning_rate": 5.98754531132466e-07, + "loss": 0.9916, + "step": 10042 + }, + { + "epoch": 0.75477228318052, + "grad_norm": 1.8660756395961262, + "learning_rate": 5.984071952134941e-07, + "loss": 1.0795, + "step": 10043 + }, + { + "epoch": 0.7548474372463551, + "grad_norm": 1.8882783560677425, + "learning_rate": 5.98059942344263e-07, + "loss": 0.9561, + "step": 10044 + }, + { + "epoch": 0.75492259131219, + "grad_norm": 1.8400997299657682, + "learning_rate": 5.977127725453482e-07, + "loss": 0.938, + "step": 10045 + }, + { + "epoch": 0.754997745378025, + "grad_norm": 2.8211822436727054, + "learning_rate": 5.973656858373217e-07, + "loss": 0.9033, + "step": 10046 + }, + { + "epoch": 0.7550728994438599, + "grad_norm": 2.644870122161814, + "learning_rate": 5.970186822407495e-07, + "loss": 0.9836, + "step": 10047 + }, + { + "epoch": 0.7551480535096948, + "grad_norm": 1.9054644482553496, + "learning_rate": 5.966717617761925e-07, + "loss": 0.8571, + "step": 10048 + }, + { + "epoch": 0.7552232075755299, + "grad_norm": 1.758442676754184, + "learning_rate": 5.963249244642077e-07, + "loss": 1.0185, + "step": 10049 + }, + { + "epoch": 0.7552983616413648, + "grad_norm": 1.7612228879879133, + "learning_rate": 5.959781703253452e-07, + "loss": 0.9491, + "step": 10050 + }, + { + "epoch": 0.7553735157071998, + "grad_norm": 2.0729476568566865, + "learning_rate": 5.956314993801532e-07, + "loss": 0.7826, + "step": 10051 + }, + { + "epoch": 0.7554486697730347, + "grad_norm": 1.7726003544730145, + "learning_rate": 5.952849116491728e-07, + "loss": 0.9492, + "step": 10052 + }, + { + "epoch": 0.7555238238388697, + "grad_norm": 1.5194393022269286, + "learning_rate": 5.949384071529395e-07, + "loss": 0.9015, + "step": 10053 + }, + { + "epoch": 0.7555989779047046, + "grad_norm": 1.8310690754643049, + "learning_rate": 5.945919859119865e-07, + "loss": 0.8887, + "step": 10054 + }, + { + "epoch": 0.7556741319705396, + "grad_norm": 1.8690779943546265, + "learning_rate": 5.942456479468401e-07, + "loss": 0.9948, + "step": 10055 + }, + { + "epoch": 0.7557492860363746, + "grad_norm": 0.628835022488383, + "learning_rate": 5.938993932780221e-07, + "loss": 0.8274, + "step": 10056 + }, + { + "epoch": 0.7558244401022095, + "grad_norm": 1.8213477731195609, + "learning_rate": 5.935532219260493e-07, + "loss": 0.9386, + "step": 10057 + }, + { + "epoch": 0.7558995941680445, + "grad_norm": 1.3888437242581964, + "learning_rate": 5.932071339114331e-07, + "loss": 1.0246, + "step": 10058 + }, + { + "epoch": 0.7559747482338794, + "grad_norm": 1.5499645167565437, + "learning_rate": 5.928611292546819e-07, + "loss": 1.0699, + "step": 10059 + }, + { + "epoch": 0.7560499022997144, + "grad_norm": 1.773105405390635, + "learning_rate": 5.92515207976297e-07, + "loss": 0.9238, + "step": 10060 + }, + { + "epoch": 0.7561250563655494, + "grad_norm": 3.674787299819348, + "learning_rate": 5.921693700967758e-07, + "loss": 0.9443, + "step": 10061 + }, + { + "epoch": 0.7562002104313843, + "grad_norm": 1.809351060145625, + "learning_rate": 5.918236156366101e-07, + "loss": 0.9958, + "step": 10062 + }, + { + "epoch": 0.7562753644972193, + "grad_norm": 1.6310682699766013, + "learning_rate": 5.91477944616287e-07, + "loss": 0.9877, + "step": 10063 + }, + { + "epoch": 0.7563505185630542, + "grad_norm": 2.5555438085184687, + "learning_rate": 5.911323570562898e-07, + "loss": 0.8793, + "step": 10064 + }, + { + "epoch": 0.7564256726288893, + "grad_norm": 1.8035649362903166, + "learning_rate": 5.907868529770957e-07, + "loss": 0.9377, + "step": 10065 + }, + { + "epoch": 0.7565008266947242, + "grad_norm": 2.114781888713021, + "learning_rate": 5.904414323991764e-07, + "loss": 0.91, + "step": 10066 + }, + { + "epoch": 0.7565759807605591, + "grad_norm": 1.585442602051561, + "learning_rate": 5.900960953429992e-07, + "loss": 0.9406, + "step": 10067 + }, + { + "epoch": 0.7566511348263941, + "grad_norm": 0.7763187033430979, + "learning_rate": 5.89750841829028e-07, + "loss": 0.8748, + "step": 10068 + }, + { + "epoch": 0.756726288892229, + "grad_norm": 3.2704869631595064, + "learning_rate": 5.894056718777196e-07, + "loss": 0.9197, + "step": 10069 + }, + { + "epoch": 0.7568014429580641, + "grad_norm": 1.4976631876785305, + "learning_rate": 5.890605855095265e-07, + "loss": 1.011, + "step": 10070 + }, + { + "epoch": 0.756876597023899, + "grad_norm": 0.7317309973972898, + "learning_rate": 5.887155827448968e-07, + "loss": 0.8275, + "step": 10071 + }, + { + "epoch": 0.756951751089734, + "grad_norm": 1.8438235122767141, + "learning_rate": 5.883706636042722e-07, + "loss": 0.9604, + "step": 10072 + }, + { + "epoch": 0.7570269051555689, + "grad_norm": 2.390355013891816, + "learning_rate": 5.880258281080921e-07, + "loss": 1.0011, + "step": 10073 + }, + { + "epoch": 0.7571020592214038, + "grad_norm": 1.6021044302726548, + "learning_rate": 5.876810762767883e-07, + "loss": 1.0362, + "step": 10074 + }, + { + "epoch": 0.7571772132872389, + "grad_norm": 1.730932582580336, + "learning_rate": 5.873364081307884e-07, + "loss": 0.8468, + "step": 10075 + }, + { + "epoch": 0.7572523673530738, + "grad_norm": 2.2834844228314686, + "learning_rate": 5.869918236905162e-07, + "loss": 0.9298, + "step": 10076 + }, + { + "epoch": 0.7573275214189088, + "grad_norm": 1.8182877179212955, + "learning_rate": 5.866473229763893e-07, + "loss": 0.9181, + "step": 10077 + }, + { + "epoch": 0.7574026754847437, + "grad_norm": 1.8111584401073582, + "learning_rate": 5.863029060088205e-07, + "loss": 1.0211, + "step": 10078 + }, + { + "epoch": 0.7574778295505787, + "grad_norm": 1.6468351992696375, + "learning_rate": 5.859585728082181e-07, + "loss": 0.9264, + "step": 10079 + }, + { + "epoch": 0.7575529836164137, + "grad_norm": 1.7291709562807611, + "learning_rate": 5.856143233949844e-07, + "loss": 0.9439, + "step": 10080 + }, + { + "epoch": 0.7576281376822486, + "grad_norm": 2.668204211991359, + "learning_rate": 5.852701577895184e-07, + "loss": 0.9246, + "step": 10081 + }, + { + "epoch": 0.7577032917480836, + "grad_norm": 1.8559657665951361, + "learning_rate": 5.849260760122132e-07, + "loss": 0.9874, + "step": 10082 + }, + { + "epoch": 0.7577784458139185, + "grad_norm": 1.4655678548835829, + "learning_rate": 5.845820780834568e-07, + "loss": 1.038, + "step": 10083 + }, + { + "epoch": 0.7578535998797535, + "grad_norm": 2.423775204335151, + "learning_rate": 5.842381640236318e-07, + "loss": 0.9035, + "step": 10084 + }, + { + "epoch": 0.7579287539455885, + "grad_norm": 1.7242992636407524, + "learning_rate": 5.838943338531166e-07, + "loss": 1.0053, + "step": 10085 + }, + { + "epoch": 0.7580039080114234, + "grad_norm": 1.6030394652155555, + "learning_rate": 5.835505875922853e-07, + "loss": 1.0, + "step": 10086 + }, + { + "epoch": 0.7580790620772584, + "grad_norm": 0.7184574112159395, + "learning_rate": 5.832069252615058e-07, + "loss": 0.826, + "step": 10087 + }, + { + "epoch": 0.7581542161430933, + "grad_norm": 1.6191814863803229, + "learning_rate": 5.82863346881141e-07, + "loss": 1.0024, + "step": 10088 + }, + { + "epoch": 0.7582293702089283, + "grad_norm": 1.4329047353295281, + "learning_rate": 5.825198524715489e-07, + "loss": 0.9857, + "step": 10089 + }, + { + "epoch": 0.7583045242747632, + "grad_norm": 0.7837024723590785, + "learning_rate": 5.821764420530842e-07, + "loss": 0.864, + "step": 10090 + }, + { + "epoch": 0.7583796783405983, + "grad_norm": 1.6302596582817783, + "learning_rate": 5.818331156460943e-07, + "loss": 1.0689, + "step": 10091 + }, + { + "epoch": 0.7584548324064332, + "grad_norm": 0.7022593573764695, + "learning_rate": 5.814898732709228e-07, + "loss": 0.8422, + "step": 10092 + }, + { + "epoch": 0.7585299864722681, + "grad_norm": 2.180106720893187, + "learning_rate": 5.811467149479083e-07, + "loss": 0.9673, + "step": 10093 + }, + { + "epoch": 0.7586051405381031, + "grad_norm": 1.792767814729648, + "learning_rate": 5.808036406973835e-07, + "loss": 0.9173, + "step": 10094 + }, + { + "epoch": 0.758680294603938, + "grad_norm": 1.667881125799613, + "learning_rate": 5.804606505396781e-07, + "loss": 1.0107, + "step": 10095 + }, + { + "epoch": 0.7587554486697731, + "grad_norm": 1.975867891957328, + "learning_rate": 5.801177444951148e-07, + "loss": 0.8222, + "step": 10096 + }, + { + "epoch": 0.758830602735608, + "grad_norm": 1.7117618334252396, + "learning_rate": 5.797749225840117e-07, + "loss": 1.0346, + "step": 10097 + }, + { + "epoch": 0.758905756801443, + "grad_norm": 1.366052996086336, + "learning_rate": 5.794321848266835e-07, + "loss": 0.9693, + "step": 10098 + }, + { + "epoch": 0.7589809108672779, + "grad_norm": 1.64112831751238, + "learning_rate": 5.790895312434378e-07, + "loss": 0.928, + "step": 10099 + }, + { + "epoch": 0.7590560649331128, + "grad_norm": 2.39106377458258, + "learning_rate": 5.787469618545786e-07, + "loss": 0.9344, + "step": 10100 + }, + { + "epoch": 0.7591312189989479, + "grad_norm": 2.388017316765035, + "learning_rate": 5.784044766804044e-07, + "loss": 0.9703, + "step": 10101 + }, + { + "epoch": 0.7592063730647828, + "grad_norm": 1.815776034719345, + "learning_rate": 5.780620757412078e-07, + "loss": 0.9958, + "step": 10102 + }, + { + "epoch": 0.7592815271306178, + "grad_norm": 1.9823292076363401, + "learning_rate": 5.777197590572789e-07, + "loss": 1.0765, + "step": 10103 + }, + { + "epoch": 0.7593566811964527, + "grad_norm": 1.9418538304879471, + "learning_rate": 5.773775266489005e-07, + "loss": 0.9878, + "step": 10104 + }, + { + "epoch": 0.7594318352622876, + "grad_norm": 1.4427419536910215, + "learning_rate": 5.770353785363511e-07, + "loss": 0.9509, + "step": 10105 + }, + { + "epoch": 0.7595069893281227, + "grad_norm": 1.8038594381504258, + "learning_rate": 5.766933147399045e-07, + "loss": 1.0253, + "step": 10106 + }, + { + "epoch": 0.7595821433939576, + "grad_norm": 0.7635117775209148, + "learning_rate": 5.763513352798286e-07, + "loss": 0.828, + "step": 10107 + }, + { + "epoch": 0.7596572974597926, + "grad_norm": 2.1938206500987736, + "learning_rate": 5.760094401763884e-07, + "loss": 0.9881, + "step": 10108 + }, + { + "epoch": 0.7597324515256275, + "grad_norm": 1.596865731229216, + "learning_rate": 5.756676294498415e-07, + "loss": 1.0214, + "step": 10109 + }, + { + "epoch": 0.7598076055914625, + "grad_norm": 1.578034804382195, + "learning_rate": 5.753259031204416e-07, + "loss": 0.9547, + "step": 10110 + }, + { + "epoch": 0.7598827596572975, + "grad_norm": 1.8948931476190298, + "learning_rate": 5.74984261208437e-07, + "loss": 0.8985, + "step": 10111 + }, + { + "epoch": 0.7599579137231324, + "grad_norm": 0.6201532788925141, + "learning_rate": 5.746427037340722e-07, + "loss": 0.8126, + "step": 10112 + }, + { + "epoch": 0.7600330677889674, + "grad_norm": 1.4146110355921315, + "learning_rate": 5.743012307175852e-07, + "loss": 1.0099, + "step": 10113 + }, + { + "epoch": 0.7601082218548023, + "grad_norm": 2.307912864904606, + "learning_rate": 5.739598421792091e-07, + "loss": 0.9081, + "step": 10114 + }, + { + "epoch": 0.7601833759206373, + "grad_norm": 1.5608622550966424, + "learning_rate": 5.736185381391743e-07, + "loss": 1.0166, + "step": 10115 + }, + { + "epoch": 0.7602585299864723, + "grad_norm": 2.513418141847437, + "learning_rate": 5.732773186177016e-07, + "loss": 0.996, + "step": 10116 + }, + { + "epoch": 0.7603336840523073, + "grad_norm": 1.7756693236241203, + "learning_rate": 5.729361836350119e-07, + "loss": 0.8982, + "step": 10117 + }, + { + "epoch": 0.7604088381181422, + "grad_norm": 1.8495536239906556, + "learning_rate": 5.725951332113179e-07, + "loss": 1.0223, + "step": 10118 + }, + { + "epoch": 0.7604839921839771, + "grad_norm": 1.917111393569026, + "learning_rate": 5.722541673668275e-07, + "loss": 0.9886, + "step": 10119 + }, + { + "epoch": 0.7605591462498121, + "grad_norm": 1.5903461234311926, + "learning_rate": 5.719132861217455e-07, + "loss": 1.0231, + "step": 10120 + }, + { + "epoch": 0.760634300315647, + "grad_norm": 2.2780975088964657, + "learning_rate": 5.715724894962699e-07, + "loss": 0.998, + "step": 10121 + }, + { + "epoch": 0.7607094543814821, + "grad_norm": 1.5918856168946314, + "learning_rate": 5.712317775105939e-07, + "loss": 0.9175, + "step": 10122 + }, + { + "epoch": 0.760784608447317, + "grad_norm": 1.6390352812362543, + "learning_rate": 5.708911501849065e-07, + "loss": 1.0222, + "step": 10123 + }, + { + "epoch": 0.760859762513152, + "grad_norm": 1.599637991960139, + "learning_rate": 5.7055060753939e-07, + "loss": 1.019, + "step": 10124 + }, + { + "epoch": 0.7609349165789869, + "grad_norm": 1.7620837727916954, + "learning_rate": 5.702101495942245e-07, + "loss": 0.8587, + "step": 10125 + }, + { + "epoch": 0.7610100706448218, + "grad_norm": 1.724139831147298, + "learning_rate": 5.698697763695826e-07, + "loss": 1.006, + "step": 10126 + }, + { + "epoch": 0.7610852247106569, + "grad_norm": 1.4102244268194009, + "learning_rate": 5.695294878856327e-07, + "loss": 0.9359, + "step": 10127 + }, + { + "epoch": 0.7611603787764918, + "grad_norm": 4.133282755470514, + "learning_rate": 5.691892841625385e-07, + "loss": 0.8523, + "step": 10128 + }, + { + "epoch": 0.7612355328423268, + "grad_norm": 1.9349561419186434, + "learning_rate": 5.688491652204573e-07, + "loss": 0.9159, + "step": 10129 + }, + { + "epoch": 0.7613106869081617, + "grad_norm": 1.720244919283637, + "learning_rate": 5.68509131079544e-07, + "loss": 0.8373, + "step": 10130 + }, + { + "epoch": 0.7613858409739966, + "grad_norm": 2.2615858180588835, + "learning_rate": 5.681691817599463e-07, + "loss": 0.9675, + "step": 10131 + }, + { + "epoch": 0.7614609950398317, + "grad_norm": 2.0731117757353035, + "learning_rate": 5.678293172818074e-07, + "loss": 0.8598, + "step": 10132 + }, + { + "epoch": 0.7615361491056666, + "grad_norm": 1.426618523557865, + "learning_rate": 5.674895376652649e-07, + "loss": 0.9146, + "step": 10133 + }, + { + "epoch": 0.7616113031715016, + "grad_norm": 2.3303306423574233, + "learning_rate": 5.671498429304535e-07, + "loss": 1.1051, + "step": 10134 + }, + { + "epoch": 0.7616864572373365, + "grad_norm": 1.6912081208157699, + "learning_rate": 5.668102330975007e-07, + "loss": 1.0305, + "step": 10135 + }, + { + "epoch": 0.7617616113031715, + "grad_norm": 1.5819997525317435, + "learning_rate": 5.664707081865288e-07, + "loss": 1.0541, + "step": 10136 + }, + { + "epoch": 0.7618367653690065, + "grad_norm": 1.4549794822897941, + "learning_rate": 5.661312682176582e-07, + "loss": 0.9282, + "step": 10137 + }, + { + "epoch": 0.7619119194348414, + "grad_norm": 1.7375320706049184, + "learning_rate": 5.657919132109991e-07, + "loss": 0.8443, + "step": 10138 + }, + { + "epoch": 0.7619870735006764, + "grad_norm": 0.6941590331925995, + "learning_rate": 5.65452643186662e-07, + "loss": 0.8382, + "step": 10139 + }, + { + "epoch": 0.7620622275665113, + "grad_norm": 1.6404542379975466, + "learning_rate": 5.65113458164749e-07, + "loss": 1.0063, + "step": 10140 + }, + { + "epoch": 0.7621373816323463, + "grad_norm": 1.5060847369819843, + "learning_rate": 5.647743581653575e-07, + "loss": 1.0199, + "step": 10141 + }, + { + "epoch": 0.7622125356981813, + "grad_norm": 0.9930628722187338, + "learning_rate": 5.644353432085818e-07, + "loss": 0.9027, + "step": 10142 + }, + { + "epoch": 0.7622876897640163, + "grad_norm": 1.4586744272878744, + "learning_rate": 5.64096413314509e-07, + "loss": 0.9781, + "step": 10143 + }, + { + "epoch": 0.7623628438298512, + "grad_norm": 2.1289572053463277, + "learning_rate": 5.637575685032223e-07, + "loss": 0.9339, + "step": 10144 + }, + { + "epoch": 0.7624379978956861, + "grad_norm": 3.0392234779224254, + "learning_rate": 5.634188087947993e-07, + "loss": 0.9965, + "step": 10145 + }, + { + "epoch": 0.7625131519615211, + "grad_norm": 1.7006617940884996, + "learning_rate": 5.630801342093123e-07, + "loss": 1.0231, + "step": 10146 + }, + { + "epoch": 0.762588306027356, + "grad_norm": 1.6729702793270043, + "learning_rate": 5.627415447668304e-07, + "loss": 0.9346, + "step": 10147 + }, + { + "epoch": 0.7626634600931911, + "grad_norm": 1.7331255974753086, + "learning_rate": 5.624030404874154e-07, + "loss": 1.0354, + "step": 10148 + }, + { + "epoch": 0.762738614159026, + "grad_norm": 1.6533843386893865, + "learning_rate": 5.620646213911253e-07, + "loss": 0.8196, + "step": 10149 + }, + { + "epoch": 0.7628137682248609, + "grad_norm": 1.4553989952682584, + "learning_rate": 5.617262874980122e-07, + "loss": 0.9183, + "step": 10150 + }, + { + "epoch": 0.7628889222906959, + "grad_norm": 1.671037410662021, + "learning_rate": 5.613880388281245e-07, + "loss": 0.9986, + "step": 10151 + }, + { + "epoch": 0.7629640763565309, + "grad_norm": 2.5531666746438555, + "learning_rate": 5.610498754015043e-07, + "loss": 1.0047, + "step": 10152 + }, + { + "epoch": 0.7630392304223659, + "grad_norm": 1.6853675083807875, + "learning_rate": 5.607117972381892e-07, + "loss": 1.043, + "step": 10153 + }, + { + "epoch": 0.7631143844882008, + "grad_norm": 2.1456183785731673, + "learning_rate": 5.603738043582113e-07, + "loss": 0.9274, + "step": 10154 + }, + { + "epoch": 0.7631895385540358, + "grad_norm": 1.6479045132825052, + "learning_rate": 5.600358967815977e-07, + "loss": 0.9264, + "step": 10155 + }, + { + "epoch": 0.7632646926198707, + "grad_norm": 1.6742496916616998, + "learning_rate": 5.59698074528372e-07, + "loss": 1.0361, + "step": 10156 + }, + { + "epoch": 0.7633398466857056, + "grad_norm": 2.308350009605498, + "learning_rate": 5.593603376185503e-07, + "loss": 0.847, + "step": 10157 + }, + { + "epoch": 0.7634150007515407, + "grad_norm": 2.082009492513544, + "learning_rate": 5.590226860721447e-07, + "loss": 1.0315, + "step": 10158 + }, + { + "epoch": 0.7634901548173756, + "grad_norm": 1.4279243979436291, + "learning_rate": 5.586851199091635e-07, + "loss": 0.9969, + "step": 10159 + }, + { + "epoch": 0.7635653088832106, + "grad_norm": 1.5991946697052768, + "learning_rate": 5.583476391496083e-07, + "loss": 1.019, + "step": 10160 + }, + { + "epoch": 0.7636404629490455, + "grad_norm": 1.7388343873926506, + "learning_rate": 5.580102438134755e-07, + "loss": 1.0299, + "step": 10161 + }, + { + "epoch": 0.7637156170148806, + "grad_norm": 1.9247861693878676, + "learning_rate": 5.576729339207578e-07, + "loss": 0.9547, + "step": 10162 + }, + { + "epoch": 0.7637907710807155, + "grad_norm": 2.9934981142608073, + "learning_rate": 5.573357094914413e-07, + "loss": 0.9368, + "step": 10163 + }, + { + "epoch": 0.7638659251465504, + "grad_norm": 1.829436198762785, + "learning_rate": 5.569985705455087e-07, + "loss": 1.0637, + "step": 10164 + }, + { + "epoch": 0.7639410792123854, + "grad_norm": 2.115309441838003, + "learning_rate": 5.566615171029365e-07, + "loss": 1.0203, + "step": 10165 + }, + { + "epoch": 0.7640162332782203, + "grad_norm": 2.252259920897617, + "learning_rate": 5.563245491836963e-07, + "loss": 0.9241, + "step": 10166 + }, + { + "epoch": 0.7640913873440554, + "grad_norm": 13.737559846534223, + "learning_rate": 5.559876668077548e-07, + "loss": 0.8813, + "step": 10167 + }, + { + "epoch": 0.7641665414098903, + "grad_norm": 2.1017216618618195, + "learning_rate": 5.556508699950728e-07, + "loss": 1.003, + "step": 10168 + }, + { + "epoch": 0.7642416954757253, + "grad_norm": 0.7662700738191015, + "learning_rate": 5.553141587656083e-07, + "loss": 0.7998, + "step": 10169 + }, + { + "epoch": 0.7643168495415602, + "grad_norm": 1.6695842557783493, + "learning_rate": 5.549775331393118e-07, + "loss": 0.9809, + "step": 10170 + }, + { + "epoch": 0.7643920036073951, + "grad_norm": 1.4848693024633344, + "learning_rate": 5.546409931361299e-07, + "loss": 0.9361, + "step": 10171 + }, + { + "epoch": 0.7644671576732301, + "grad_norm": 1.7662407805572728, + "learning_rate": 5.543045387760035e-07, + "loss": 1.0826, + "step": 10172 + }, + { + "epoch": 0.7645423117390651, + "grad_norm": 1.8189631360152672, + "learning_rate": 5.539681700788694e-07, + "loss": 0.903, + "step": 10173 + }, + { + "epoch": 0.7646174658049001, + "grad_norm": 1.57467357038093, + "learning_rate": 5.536318870646586e-07, + "loss": 0.9857, + "step": 10174 + }, + { + "epoch": 0.764692619870735, + "grad_norm": 2.1481170970248615, + "learning_rate": 5.532956897532968e-07, + "loss": 0.9543, + "step": 10175 + }, + { + "epoch": 0.7647677739365699, + "grad_norm": 1.9629667808221014, + "learning_rate": 5.529595781647054e-07, + "loss": 1.0102, + "step": 10176 + }, + { + "epoch": 0.7648429280024049, + "grad_norm": 0.7903003991265142, + "learning_rate": 5.526235523187992e-07, + "loss": 0.8358, + "step": 10177 + }, + { + "epoch": 0.7649180820682399, + "grad_norm": 1.3608691017665675, + "learning_rate": 5.522876122354907e-07, + "loss": 1.025, + "step": 10178 + }, + { + "epoch": 0.7649932361340749, + "grad_norm": 2.170553450257797, + "learning_rate": 5.519517579346849e-07, + "loss": 0.9986, + "step": 10179 + }, + { + "epoch": 0.7650683901999098, + "grad_norm": 2.340477146016729, + "learning_rate": 5.516159894362817e-07, + "loss": 0.8468, + "step": 10180 + }, + { + "epoch": 0.7651435442657448, + "grad_norm": 1.8543446558985461, + "learning_rate": 5.512803067601779e-07, + "loss": 0.9076, + "step": 10181 + }, + { + "epoch": 0.7652186983315797, + "grad_norm": 12.144864200164186, + "learning_rate": 5.509447099262636e-07, + "loss": 1.0474, + "step": 10182 + }, + { + "epoch": 0.7652938523974147, + "grad_norm": 2.2621087510521933, + "learning_rate": 5.506091989544239e-07, + "loss": 0.8878, + "step": 10183 + }, + { + "epoch": 0.7653690064632497, + "grad_norm": 2.5196642936651297, + "learning_rate": 5.502737738645393e-07, + "loss": 0.9343, + "step": 10184 + }, + { + "epoch": 0.7654441605290846, + "grad_norm": 1.3631652083100914, + "learning_rate": 5.499384346764843e-07, + "loss": 0.9782, + "step": 10185 + }, + { + "epoch": 0.7655193145949196, + "grad_norm": 1.705442828734135, + "learning_rate": 5.496031814101303e-07, + "loss": 0.973, + "step": 10186 + }, + { + "epoch": 0.7655944686607545, + "grad_norm": 1.6879396219167107, + "learning_rate": 5.492680140853418e-07, + "loss": 0.8777, + "step": 10187 + }, + { + "epoch": 0.7656696227265896, + "grad_norm": 1.590567908920642, + "learning_rate": 5.489329327219787e-07, + "loss": 0.9959, + "step": 10188 + }, + { + "epoch": 0.7657447767924245, + "grad_norm": 2.1309468699813015, + "learning_rate": 5.485979373398959e-07, + "loss": 0.9455, + "step": 10189 + }, + { + "epoch": 0.7658199308582594, + "grad_norm": 1.6966725883805234, + "learning_rate": 5.482630279589424e-07, + "loss": 0.9092, + "step": 10190 + }, + { + "epoch": 0.7658950849240944, + "grad_norm": 0.6752181601307049, + "learning_rate": 5.479282045989644e-07, + "loss": 0.8349, + "step": 10191 + }, + { + "epoch": 0.7659702389899293, + "grad_norm": 4.197494195294527, + "learning_rate": 5.475934672798004e-07, + "loss": 0.9969, + "step": 10192 + }, + { + "epoch": 0.7660453930557644, + "grad_norm": 2.3955443387642807, + "learning_rate": 5.47258816021285e-07, + "loss": 0.9624, + "step": 10193 + }, + { + "epoch": 0.7661205471215993, + "grad_norm": 1.8491000770359642, + "learning_rate": 5.469242508432472e-07, + "loss": 0.9997, + "step": 10194 + }, + { + "epoch": 0.7661957011874342, + "grad_norm": 3.160941235120059, + "learning_rate": 5.465897717655123e-07, + "loss": 1.0099, + "step": 10195 + }, + { + "epoch": 0.7662708552532692, + "grad_norm": 3.727588440365904, + "learning_rate": 5.462553788078992e-07, + "loss": 0.9018, + "step": 10196 + }, + { + "epoch": 0.7663460093191041, + "grad_norm": 4.785956964257065, + "learning_rate": 5.459210719902211e-07, + "loss": 0.9925, + "step": 10197 + }, + { + "epoch": 0.7664211633849392, + "grad_norm": 1.601872930447257, + "learning_rate": 5.455868513322881e-07, + "loss": 1.0003, + "step": 10198 + }, + { + "epoch": 0.7664963174507741, + "grad_norm": 2.1158742738088003, + "learning_rate": 5.452527168539025e-07, + "loss": 0.9329, + "step": 10199 + }, + { + "epoch": 0.7665714715166091, + "grad_norm": 2.1992007537701737, + "learning_rate": 5.449186685748648e-07, + "loss": 0.8503, + "step": 10200 + }, + { + "epoch": 0.766646625582444, + "grad_norm": 1.6388569342017778, + "learning_rate": 5.445847065149678e-07, + "loss": 1.1061, + "step": 10201 + }, + { + "epoch": 0.7667217796482789, + "grad_norm": 1.823599144472732, + "learning_rate": 5.442508306939995e-07, + "loss": 0.9284, + "step": 10202 + }, + { + "epoch": 0.766796933714114, + "grad_norm": 1.711554498770446, + "learning_rate": 5.439170411317446e-07, + "loss": 0.9544, + "step": 10203 + }, + { + "epoch": 0.7668720877799489, + "grad_norm": 2.215911123634953, + "learning_rate": 5.435833378479807e-07, + "loss": 0.886, + "step": 10204 + }, + { + "epoch": 0.7669472418457839, + "grad_norm": 3.5058078262662193, + "learning_rate": 5.432497208624809e-07, + "loss": 0.9575, + "step": 10205 + }, + { + "epoch": 0.7670223959116188, + "grad_norm": 1.5431947952993534, + "learning_rate": 5.429161901950134e-07, + "loss": 1.0492, + "step": 10206 + }, + { + "epoch": 0.7670975499774538, + "grad_norm": 2.2122129394172143, + "learning_rate": 5.425827458653407e-07, + "loss": 0.8313, + "step": 10207 + }, + { + "epoch": 0.7671727040432887, + "grad_norm": 2.217359080003248, + "learning_rate": 5.422493878932217e-07, + "loss": 0.9714, + "step": 10208 + }, + { + "epoch": 0.7672478581091237, + "grad_norm": 1.8035718103108953, + "learning_rate": 5.419161162984085e-07, + "loss": 1.0315, + "step": 10209 + }, + { + "epoch": 0.7673230121749587, + "grad_norm": 5.9191127968023665, + "learning_rate": 5.415829311006487e-07, + "loss": 1.0058, + "step": 10210 + }, + { + "epoch": 0.7673981662407936, + "grad_norm": 2.6758461157739872, + "learning_rate": 5.412498323196852e-07, + "loss": 0.9993, + "step": 10211 + }, + { + "epoch": 0.7674733203066286, + "grad_norm": 1.5576640250599036, + "learning_rate": 5.409168199752543e-07, + "loss": 0.9803, + "step": 10212 + }, + { + "epoch": 0.7675484743724635, + "grad_norm": 1.9520867601796799, + "learning_rate": 5.405838940870897e-07, + "loss": 1.024, + "step": 10213 + }, + { + "epoch": 0.7676236284382986, + "grad_norm": 1.7149970262017895, + "learning_rate": 5.402510546749177e-07, + "loss": 1.0025, + "step": 10214 + }, + { + "epoch": 0.7676987825041335, + "grad_norm": 1.5563720318238443, + "learning_rate": 5.399183017584605e-07, + "loss": 1.0059, + "step": 10215 + }, + { + "epoch": 0.7677739365699684, + "grad_norm": 1.7452277781961516, + "learning_rate": 5.395856353574344e-07, + "loss": 0.9857, + "step": 10216 + }, + { + "epoch": 0.7678490906358034, + "grad_norm": 0.8441447288346349, + "learning_rate": 5.392530554915522e-07, + "loss": 0.8421, + "step": 10217 + }, + { + "epoch": 0.7679242447016383, + "grad_norm": 1.6157785073406365, + "learning_rate": 5.389205621805202e-07, + "loss": 1.0099, + "step": 10218 + }, + { + "epoch": 0.7679993987674734, + "grad_norm": 2.5900253761550673, + "learning_rate": 5.385881554440397e-07, + "loss": 0.9611, + "step": 10219 + }, + { + "epoch": 0.7680745528333083, + "grad_norm": 1.683516285966229, + "learning_rate": 5.382558353018069e-07, + "loss": 0.9767, + "step": 10220 + }, + { + "epoch": 0.7681497068991432, + "grad_norm": 1.0108578233406982, + "learning_rate": 5.37923601773513e-07, + "loss": 0.9825, + "step": 10221 + }, + { + "epoch": 0.7682248609649782, + "grad_norm": 1.725250297852598, + "learning_rate": 5.375914548788447e-07, + "loss": 0.9739, + "step": 10222 + }, + { + "epoch": 0.7683000150308131, + "grad_norm": 3.144983408388363, + "learning_rate": 5.372593946374826e-07, + "loss": 0.9582, + "step": 10223 + }, + { + "epoch": 0.7683751690966482, + "grad_norm": 1.9946729526769218, + "learning_rate": 5.369274210691022e-07, + "loss": 1.0547, + "step": 10224 + }, + { + "epoch": 0.7684503231624831, + "grad_norm": 1.6227579054674421, + "learning_rate": 5.36595534193375e-07, + "loss": 1.0072, + "step": 10225 + }, + { + "epoch": 0.7685254772283181, + "grad_norm": 1.5736828215958385, + "learning_rate": 5.362637340299662e-07, + "loss": 1.0468, + "step": 10226 + }, + { + "epoch": 0.768600631294153, + "grad_norm": 2.0070793505436355, + "learning_rate": 5.359320205985363e-07, + "loss": 0.9783, + "step": 10227 + }, + { + "epoch": 0.7686757853599879, + "grad_norm": 1.617846589608902, + "learning_rate": 5.356003939187402e-07, + "loss": 0.9526, + "step": 10228 + }, + { + "epoch": 0.768750939425823, + "grad_norm": 1.9203679448466016, + "learning_rate": 5.352688540102279e-07, + "loss": 0.8314, + "step": 10229 + }, + { + "epoch": 0.7688260934916579, + "grad_norm": 1.7901972874015377, + "learning_rate": 5.349374008926454e-07, + "loss": 0.9765, + "step": 10230 + }, + { + "epoch": 0.7689012475574929, + "grad_norm": 1.6153682482592415, + "learning_rate": 5.34606034585632e-07, + "loss": 0.9401, + "step": 10231 + }, + { + "epoch": 0.7689764016233278, + "grad_norm": 1.8313824576427227, + "learning_rate": 5.342747551088225e-07, + "loss": 0.9208, + "step": 10232 + }, + { + "epoch": 0.7690515556891628, + "grad_norm": 3.9819401500654945, + "learning_rate": 5.339435624818463e-07, + "loss": 0.9138, + "step": 10233 + }, + { + "epoch": 0.7691267097549978, + "grad_norm": 1.6502278083981605, + "learning_rate": 5.336124567243275e-07, + "loss": 0.9369, + "step": 10234 + }, + { + "epoch": 0.7692018638208327, + "grad_norm": 1.5186562654430988, + "learning_rate": 5.332814378558861e-07, + "loss": 0.9872, + "step": 10235 + }, + { + "epoch": 0.7692770178866677, + "grad_norm": 2.223560652066678, + "learning_rate": 5.329505058961361e-07, + "loss": 0.9862, + "step": 10236 + }, + { + "epoch": 0.7693521719525026, + "grad_norm": 1.6961312140953209, + "learning_rate": 5.326196608646862e-07, + "loss": 0.9398, + "step": 10237 + }, + { + "epoch": 0.7694273260183376, + "grad_norm": 2.1977690395536795, + "learning_rate": 5.322889027811402e-07, + "loss": 0.9186, + "step": 10238 + }, + { + "epoch": 0.7695024800841725, + "grad_norm": 2.3328864032871572, + "learning_rate": 5.31958231665097e-07, + "loss": 0.9737, + "step": 10239 + }, + { + "epoch": 0.7695776341500075, + "grad_norm": 2.1506612470566036, + "learning_rate": 5.316276475361505e-07, + "loss": 0.8553, + "step": 10240 + }, + { + "epoch": 0.7696527882158425, + "grad_norm": 2.4900405868225195, + "learning_rate": 5.312971504138883e-07, + "loss": 0.9685, + "step": 10241 + }, + { + "epoch": 0.7697279422816774, + "grad_norm": 3.237512071505985, + "learning_rate": 5.309667403178944e-07, + "loss": 0.8902, + "step": 10242 + }, + { + "epoch": 0.7698030963475124, + "grad_norm": 1.4070639903544795, + "learning_rate": 5.306364172677455e-07, + "loss": 1.0458, + "step": 10243 + }, + { + "epoch": 0.7698782504133473, + "grad_norm": 0.7233963759002733, + "learning_rate": 5.303061812830163e-07, + "loss": 0.8164, + "step": 10244 + }, + { + "epoch": 0.7699534044791824, + "grad_norm": 1.8512760375340644, + "learning_rate": 5.299760323832734e-07, + "loss": 0.9884, + "step": 10245 + }, + { + "epoch": 0.7700285585450173, + "grad_norm": 2.541746065327516, + "learning_rate": 5.296459705880794e-07, + "loss": 0.8792, + "step": 10246 + }, + { + "epoch": 0.7701037126108522, + "grad_norm": 1.5715584438087304, + "learning_rate": 5.293159959169924e-07, + "loss": 0.9209, + "step": 10247 + }, + { + "epoch": 0.7701788666766872, + "grad_norm": 2.252381336164504, + "learning_rate": 5.289861083895642e-07, + "loss": 0.9628, + "step": 10248 + }, + { + "epoch": 0.7702540207425221, + "grad_norm": 1.351019557320015, + "learning_rate": 5.286563080253421e-07, + "loss": 0.9319, + "step": 10249 + }, + { + "epoch": 0.7703291748083572, + "grad_norm": 2.325483116080409, + "learning_rate": 5.283265948438678e-07, + "loss": 0.9271, + "step": 10250 + }, + { + "epoch": 0.7704043288741921, + "grad_norm": 1.965318990154714, + "learning_rate": 5.279969688646777e-07, + "loss": 0.9838, + "step": 10251 + }, + { + "epoch": 0.7704794829400271, + "grad_norm": 1.4595846348236812, + "learning_rate": 5.276674301073045e-07, + "loss": 1.0087, + "step": 10252 + }, + { + "epoch": 0.770554637005862, + "grad_norm": 1.5801768042231794, + "learning_rate": 5.273379785912739e-07, + "loss": 0.9407, + "step": 10253 + }, + { + "epoch": 0.7706297910716969, + "grad_norm": 2.1108086129462995, + "learning_rate": 5.270086143361072e-07, + "loss": 0.8694, + "step": 10254 + }, + { + "epoch": 0.770704945137532, + "grad_norm": 2.953312297714556, + "learning_rate": 5.266793373613207e-07, + "loss": 0.9652, + "step": 10255 + }, + { + "epoch": 0.7707800992033669, + "grad_norm": 1.6412793988792849, + "learning_rate": 5.263501476864245e-07, + "loss": 0.9428, + "step": 10256 + }, + { + "epoch": 0.7708552532692019, + "grad_norm": 1.7590916027617034, + "learning_rate": 5.260210453309257e-07, + "loss": 0.9034, + "step": 10257 + }, + { + "epoch": 0.7709304073350368, + "grad_norm": 1.7046577485982215, + "learning_rate": 5.256920303143242e-07, + "loss": 0.9575, + "step": 10258 + }, + { + "epoch": 0.7710055614008718, + "grad_norm": 2.5902392975713386, + "learning_rate": 5.253631026561154e-07, + "loss": 0.9405, + "step": 10259 + }, + { + "epoch": 0.7710807154667068, + "grad_norm": 1.6717934276609978, + "learning_rate": 5.250342623757889e-07, + "loss": 0.8502, + "step": 10260 + }, + { + "epoch": 0.7711558695325417, + "grad_norm": 2.201968599745218, + "learning_rate": 5.24705509492831e-07, + "loss": 0.8701, + "step": 10261 + }, + { + "epoch": 0.7712310235983767, + "grad_norm": 1.9202860024147306, + "learning_rate": 5.243768440267209e-07, + "loss": 0.8418, + "step": 10262 + }, + { + "epoch": 0.7713061776642116, + "grad_norm": 2.2017402355176383, + "learning_rate": 5.240482659969332e-07, + "loss": 0.9663, + "step": 10263 + }, + { + "epoch": 0.7713813317300466, + "grad_norm": 1.8340249346553334, + "learning_rate": 5.237197754229376e-07, + "loss": 0.9033, + "step": 10264 + }, + { + "epoch": 0.7714564857958816, + "grad_norm": 1.7279331066243935, + "learning_rate": 5.23391372324198e-07, + "loss": 0.933, + "step": 10265 + }, + { + "epoch": 0.7715316398617165, + "grad_norm": 2.710046363377663, + "learning_rate": 5.230630567201744e-07, + "loss": 0.8309, + "step": 10266 + }, + { + "epoch": 0.7716067939275515, + "grad_norm": 1.6558432889401329, + "learning_rate": 5.227348286303201e-07, + "loss": 0.9145, + "step": 10267 + }, + { + "epoch": 0.7716819479933864, + "grad_norm": 1.9082544138330213, + "learning_rate": 5.224066880740836e-07, + "loss": 0.9928, + "step": 10268 + }, + { + "epoch": 0.7717571020592214, + "grad_norm": 1.4756848991933045, + "learning_rate": 5.220786350709094e-07, + "loss": 0.9583, + "step": 10269 + }, + { + "epoch": 0.7718322561250563, + "grad_norm": 2.6275331899069325, + "learning_rate": 5.217506696402354e-07, + "loss": 0.956, + "step": 10270 + }, + { + "epoch": 0.7719074101908914, + "grad_norm": 1.9686035828413329, + "learning_rate": 5.214227918014951e-07, + "loss": 1.0503, + "step": 10271 + }, + { + "epoch": 0.7719825642567263, + "grad_norm": 1.6707969826464433, + "learning_rate": 5.21095001574116e-07, + "loss": 0.8861, + "step": 10272 + }, + { + "epoch": 0.7720577183225612, + "grad_norm": 1.7416120870455523, + "learning_rate": 5.207672989775205e-07, + "loss": 0.9708, + "step": 10273 + }, + { + "epoch": 0.7721328723883962, + "grad_norm": 1.7028702011595218, + "learning_rate": 5.204396840311276e-07, + "loss": 1.0153, + "step": 10274 + }, + { + "epoch": 0.7722080264542311, + "grad_norm": 2.5043345068753613, + "learning_rate": 5.20112156754349e-07, + "loss": 0.9999, + "step": 10275 + }, + { + "epoch": 0.7722831805200662, + "grad_norm": 1.6721370170841856, + "learning_rate": 5.197847171665919e-07, + "loss": 0.9579, + "step": 10276 + }, + { + "epoch": 0.7723583345859011, + "grad_norm": 1.744404976729703, + "learning_rate": 5.194573652872585e-07, + "loss": 1.0171, + "step": 10277 + }, + { + "epoch": 0.7724334886517361, + "grad_norm": 1.6261462057242446, + "learning_rate": 5.191301011357451e-07, + "loss": 0.9322, + "step": 10278 + }, + { + "epoch": 0.772508642717571, + "grad_norm": 1.822990837404763, + "learning_rate": 5.188029247314442e-07, + "loss": 0.8914, + "step": 10279 + }, + { + "epoch": 0.7725837967834059, + "grad_norm": 5.558934129599537, + "learning_rate": 5.184758360937422e-07, + "loss": 1.011, + "step": 10280 + }, + { + "epoch": 0.772658950849241, + "grad_norm": 1.7110781677110651, + "learning_rate": 5.181488352420198e-07, + "loss": 0.9708, + "step": 10281 + }, + { + "epoch": 0.7727341049150759, + "grad_norm": 1.775945830532863, + "learning_rate": 5.178219221956528e-07, + "loss": 1.0453, + "step": 10282 + }, + { + "epoch": 0.7728092589809109, + "grad_norm": 1.5367888876303222, + "learning_rate": 5.17495096974013e-07, + "loss": 0.9319, + "step": 10283 + }, + { + "epoch": 0.7728844130467458, + "grad_norm": 36.857036805871616, + "learning_rate": 5.17168359596466e-07, + "loss": 1.0119, + "step": 10284 + }, + { + "epoch": 0.7729595671125807, + "grad_norm": 2.020959699422394, + "learning_rate": 5.168417100823707e-07, + "loss": 0.8798, + "step": 10285 + }, + { + "epoch": 0.7730347211784158, + "grad_norm": 1.870231158814959, + "learning_rate": 5.16515148451085e-07, + "loss": 0.999, + "step": 10286 + }, + { + "epoch": 0.7731098752442507, + "grad_norm": 1.582830382657495, + "learning_rate": 5.161886747219562e-07, + "loss": 0.9073, + "step": 10287 + }, + { + "epoch": 0.7731850293100857, + "grad_norm": 1.8467246160584292, + "learning_rate": 5.158622889143309e-07, + "loss": 0.9428, + "step": 10288 + }, + { + "epoch": 0.7732601833759206, + "grad_norm": 1.7943013875617797, + "learning_rate": 5.155359910475481e-07, + "loss": 0.9946, + "step": 10289 + }, + { + "epoch": 0.7733353374417556, + "grad_norm": 2.3290632776596634, + "learning_rate": 5.152097811409415e-07, + "loss": 0.9071, + "step": 10290 + }, + { + "epoch": 0.7734104915075906, + "grad_norm": 2.067403801337127, + "learning_rate": 5.148836592138417e-07, + "loss": 0.9382, + "step": 10291 + }, + { + "epoch": 0.7734856455734255, + "grad_norm": 0.9819377821307056, + "learning_rate": 5.145576252855719e-07, + "loss": 0.8758, + "step": 10292 + }, + { + "epoch": 0.7735607996392605, + "grad_norm": 5.537426963996166, + "learning_rate": 5.142316793754511e-07, + "loss": 0.9696, + "step": 10293 + }, + { + "epoch": 0.7736359537050954, + "grad_norm": 1.4084007769956561, + "learning_rate": 5.139058215027927e-07, + "loss": 0.9136, + "step": 10294 + }, + { + "epoch": 0.7737111077709304, + "grad_norm": 1.7905109655475366, + "learning_rate": 5.135800516869042e-07, + "loss": 0.9858, + "step": 10295 + }, + { + "epoch": 0.7737862618367654, + "grad_norm": 2.2002213193369755, + "learning_rate": 5.132543699470904e-07, + "loss": 0.9263, + "step": 10296 + }, + { + "epoch": 0.7738614159026004, + "grad_norm": 1.6032005307137682, + "learning_rate": 5.129287763026479e-07, + "loss": 0.9365, + "step": 10297 + }, + { + "epoch": 0.7739365699684353, + "grad_norm": 2.2793529408807145, + "learning_rate": 5.126032707728702e-07, + "loss": 0.9255, + "step": 10298 + }, + { + "epoch": 0.7740117240342702, + "grad_norm": 1.5657152283151552, + "learning_rate": 5.122778533770442e-07, + "loss": 0.9481, + "step": 10299 + }, + { + "epoch": 0.7740868781001052, + "grad_norm": 1.486533512957947, + "learning_rate": 5.119525241344515e-07, + "loss": 1.0031, + "step": 10300 + }, + { + "epoch": 0.7741620321659402, + "grad_norm": 2.3660620100548515, + "learning_rate": 5.116272830643707e-07, + "loss": 0.9844, + "step": 10301 + }, + { + "epoch": 0.7742371862317752, + "grad_norm": 1.7018891496521908, + "learning_rate": 5.113021301860725e-07, + "loss": 0.9675, + "step": 10302 + }, + { + "epoch": 0.7743123402976101, + "grad_norm": 3.511596405987237, + "learning_rate": 5.109770655188236e-07, + "loss": 0.9228, + "step": 10303 + }, + { + "epoch": 0.7743874943634451, + "grad_norm": 2.1942148960413532, + "learning_rate": 5.106520890818853e-07, + "loss": 0.9958, + "step": 10304 + }, + { + "epoch": 0.77446264842928, + "grad_norm": 1.78333535179428, + "learning_rate": 5.103272008945141e-07, + "loss": 1.0079, + "step": 10305 + }, + { + "epoch": 0.774537802495115, + "grad_norm": 0.8031334546308389, + "learning_rate": 5.100024009759605e-07, + "loss": 0.9007, + "step": 10306 + }, + { + "epoch": 0.77461295656095, + "grad_norm": 1.5152434035816749, + "learning_rate": 5.096776893454697e-07, + "loss": 1.0247, + "step": 10307 + }, + { + "epoch": 0.7746881106267849, + "grad_norm": 2.4925018399307364, + "learning_rate": 5.09353066022284e-07, + "loss": 0.9586, + "step": 10308 + }, + { + "epoch": 0.7747632646926199, + "grad_norm": 5.261878245541077, + "learning_rate": 5.090285310256359e-07, + "loss": 1.0276, + "step": 10309 + }, + { + "epoch": 0.7748384187584548, + "grad_norm": 1.636106317329976, + "learning_rate": 5.087040843747572e-07, + "loss": 0.9019, + "step": 10310 + }, + { + "epoch": 0.7749135728242897, + "grad_norm": 2.4442726337375564, + "learning_rate": 5.08379726088872e-07, + "loss": 0.9326, + "step": 10311 + }, + { + "epoch": 0.7749887268901248, + "grad_norm": 1.4780629925292912, + "learning_rate": 5.080554561871995e-07, + "loss": 0.8685, + "step": 10312 + }, + { + "epoch": 0.7750638809559597, + "grad_norm": 2.113573431381391, + "learning_rate": 5.077312746889547e-07, + "loss": 0.9504, + "step": 10313 + }, + { + "epoch": 0.7751390350217947, + "grad_norm": 1.807514305216424, + "learning_rate": 5.074071816133461e-07, + "loss": 0.9363, + "step": 10314 + }, + { + "epoch": 0.7752141890876296, + "grad_norm": 2.025679476675449, + "learning_rate": 5.070831769795773e-07, + "loss": 0.9502, + "step": 10315 + }, + { + "epoch": 0.7752893431534646, + "grad_norm": 1.6856228312929973, + "learning_rate": 5.067592608068474e-07, + "loss": 1.0588, + "step": 10316 + }, + { + "epoch": 0.7753644972192996, + "grad_norm": 4.8389195136234875, + "learning_rate": 5.064354331143485e-07, + "loss": 1.0143, + "step": 10317 + }, + { + "epoch": 0.7754396512851345, + "grad_norm": 1.6581091612667966, + "learning_rate": 5.061116939212702e-07, + "loss": 0.9786, + "step": 10318 + }, + { + "epoch": 0.7755148053509695, + "grad_norm": 2.676735901212308, + "learning_rate": 5.057880432467943e-07, + "loss": 1.0244, + "step": 10319 + }, + { + "epoch": 0.7755899594168044, + "grad_norm": 0.7204111639871332, + "learning_rate": 5.054644811100986e-07, + "loss": 0.8861, + "step": 10320 + }, + { + "epoch": 0.7756651134826394, + "grad_norm": 2.275830769024688, + "learning_rate": 5.051410075303555e-07, + "loss": 0.9306, + "step": 10321 + }, + { + "epoch": 0.7757402675484744, + "grad_norm": 1.498434521430499, + "learning_rate": 5.048176225267311e-07, + "loss": 0.968, + "step": 10322 + }, + { + "epoch": 0.7758154216143094, + "grad_norm": 2.7694416327564793, + "learning_rate": 5.044943261183887e-07, + "loss": 1.0853, + "step": 10323 + }, + { + "epoch": 0.7758905756801443, + "grad_norm": 2.1821184719311675, + "learning_rate": 5.041711183244842e-07, + "loss": 0.9598, + "step": 10324 + }, + { + "epoch": 0.7759657297459792, + "grad_norm": 3.003276783802071, + "learning_rate": 5.038479991641689e-07, + "loss": 1.0407, + "step": 10325 + }, + { + "epoch": 0.7760408838118142, + "grad_norm": 2.132295427069045, + "learning_rate": 5.035249686565881e-07, + "loss": 0.9769, + "step": 10326 + }, + { + "epoch": 0.7761160378776492, + "grad_norm": 2.5714312962108887, + "learning_rate": 5.032020268208838e-07, + "loss": 0.9819, + "step": 10327 + }, + { + "epoch": 0.7761911919434842, + "grad_norm": 5.820947558684671, + "learning_rate": 5.028791736761913e-07, + "loss": 1.0253, + "step": 10328 + }, + { + "epoch": 0.7762663460093191, + "grad_norm": 1.7626131413716013, + "learning_rate": 5.025564092416397e-07, + "loss": 1.0445, + "step": 10329 + }, + { + "epoch": 0.776341500075154, + "grad_norm": 1.6532815927566968, + "learning_rate": 5.022337335363558e-07, + "loss": 1.0376, + "step": 10330 + }, + { + "epoch": 0.776416654140989, + "grad_norm": 1.6113432172588658, + "learning_rate": 5.019111465794583e-07, + "loss": 0.9666, + "step": 10331 + }, + { + "epoch": 0.776491808206824, + "grad_norm": 1.6745617392803107, + "learning_rate": 5.01588648390062e-07, + "loss": 0.9653, + "step": 10332 + }, + { + "epoch": 0.776566962272659, + "grad_norm": 2.047247932763779, + "learning_rate": 5.012662389872762e-07, + "loss": 1.0235, + "step": 10333 + }, + { + "epoch": 0.7766421163384939, + "grad_norm": 2.039571998052949, + "learning_rate": 5.009439183902043e-07, + "loss": 0.9326, + "step": 10334 + }, + { + "epoch": 0.7767172704043289, + "grad_norm": 2.179259227393998, + "learning_rate": 5.006216866179458e-07, + "loss": 0.9372, + "step": 10335 + }, + { + "epoch": 0.7767924244701638, + "grad_norm": 1.5478608043888131, + "learning_rate": 5.002995436895938e-07, + "loss": 0.9932, + "step": 10336 + }, + { + "epoch": 0.7768675785359987, + "grad_norm": 1.8102769043430758, + "learning_rate": 4.999774896242368e-07, + "loss": 1.0471, + "step": 10337 + }, + { + "epoch": 0.7769427326018338, + "grad_norm": 1.5789895286285323, + "learning_rate": 4.996555244409575e-07, + "loss": 1.0262, + "step": 10338 + }, + { + "epoch": 0.7770178866676687, + "grad_norm": 0.9893207134648381, + "learning_rate": 4.993336481588331e-07, + "loss": 0.9649, + "step": 10339 + }, + { + "epoch": 0.7770930407335037, + "grad_norm": 1.8945051997645228, + "learning_rate": 4.990118607969367e-07, + "loss": 0.9223, + "step": 10340 + }, + { + "epoch": 0.7771681947993386, + "grad_norm": 2.547810295007641, + "learning_rate": 4.986901623743356e-07, + "loss": 0.9525, + "step": 10341 + }, + { + "epoch": 0.7772433488651737, + "grad_norm": 3.5335824740641306, + "learning_rate": 4.98368552910091e-07, + "loss": 0.9761, + "step": 10342 + }, + { + "epoch": 0.7773185029310086, + "grad_norm": 2.220998457542684, + "learning_rate": 4.9804703242326e-07, + "loss": 0.8736, + "step": 10343 + }, + { + "epoch": 0.7773936569968435, + "grad_norm": 1.7921931432154574, + "learning_rate": 4.97725600932893e-07, + "loss": 0.951, + "step": 10344 + }, + { + "epoch": 0.7774688110626785, + "grad_norm": 2.295400029143662, + "learning_rate": 4.974042584580372e-07, + "loss": 0.9827, + "step": 10345 + }, + { + "epoch": 0.7775439651285134, + "grad_norm": 2.2187080916624207, + "learning_rate": 4.97083005017733e-07, + "loss": 0.9433, + "step": 10346 + }, + { + "epoch": 0.7776191191943485, + "grad_norm": 1.685747154598898, + "learning_rate": 4.967618406310158e-07, + "loss": 0.9065, + "step": 10347 + }, + { + "epoch": 0.7776942732601834, + "grad_norm": 1.7181944209256343, + "learning_rate": 4.964407653169154e-07, + "loss": 1.0111, + "step": 10348 + }, + { + "epoch": 0.7777694273260184, + "grad_norm": 1.7520107570256995, + "learning_rate": 4.961197790944576e-07, + "loss": 0.8287, + "step": 10349 + }, + { + "epoch": 0.7778445813918533, + "grad_norm": 9.836636737332842, + "learning_rate": 4.957988819826617e-07, + "loss": 0.8894, + "step": 10350 + }, + { + "epoch": 0.7779197354576882, + "grad_norm": 1.517807401915972, + "learning_rate": 4.954780740005413e-07, + "loss": 0.9702, + "step": 10351 + }, + { + "epoch": 0.7779948895235232, + "grad_norm": 1.9818071586765242, + "learning_rate": 4.951573551671069e-07, + "loss": 0.9494, + "step": 10352 + }, + { + "epoch": 0.7780700435893582, + "grad_norm": 1.3370100924525388, + "learning_rate": 4.948367255013617e-07, + "loss": 0.9258, + "step": 10353 + }, + { + "epoch": 0.7781451976551932, + "grad_norm": 1.3431731915603997, + "learning_rate": 4.945161850223041e-07, + "loss": 0.9945, + "step": 10354 + }, + { + "epoch": 0.7782203517210281, + "grad_norm": 1.6540551667705823, + "learning_rate": 4.941957337489273e-07, + "loss": 1.0543, + "step": 10355 + }, + { + "epoch": 0.778295505786863, + "grad_norm": 1.843437232951772, + "learning_rate": 4.938753717002189e-07, + "loss": 0.9105, + "step": 10356 + }, + { + "epoch": 0.778370659852698, + "grad_norm": 1.6043427688576908, + "learning_rate": 4.935550988951627e-07, + "loss": 0.9987, + "step": 10357 + }, + { + "epoch": 0.778445813918533, + "grad_norm": 2.3757697414030385, + "learning_rate": 4.932349153527353e-07, + "loss": 0.9523, + "step": 10358 + }, + { + "epoch": 0.778520967984368, + "grad_norm": 1.7753582606287566, + "learning_rate": 4.929148210919092e-07, + "loss": 0.9076, + "step": 10359 + }, + { + "epoch": 0.7785961220502029, + "grad_norm": 4.950969431573773, + "learning_rate": 4.925948161316506e-07, + "loss": 0.9505, + "step": 10360 + }, + { + "epoch": 0.7786712761160379, + "grad_norm": 1.5620459850639308, + "learning_rate": 4.922749004909213e-07, + "loss": 0.8262, + "step": 10361 + }, + { + "epoch": 0.7787464301818728, + "grad_norm": 1.643354778859154, + "learning_rate": 4.919550741886777e-07, + "loss": 0.9965, + "step": 10362 + }, + { + "epoch": 0.7788215842477078, + "grad_norm": 2.704673808485755, + "learning_rate": 4.916353372438711e-07, + "loss": 1.0201, + "step": 10363 + }, + { + "epoch": 0.7788967383135428, + "grad_norm": 1.7076257155064845, + "learning_rate": 4.913156896754462e-07, + "loss": 0.9423, + "step": 10364 + }, + { + "epoch": 0.7789718923793777, + "grad_norm": 2.363804480269289, + "learning_rate": 4.909961315023441e-07, + "loss": 1.0415, + "step": 10365 + }, + { + "epoch": 0.7790470464452127, + "grad_norm": 2.1383490053089695, + "learning_rate": 4.90676662743499e-07, + "loss": 0.848, + "step": 10366 + }, + { + "epoch": 0.7791222005110476, + "grad_norm": 2.379036334510343, + "learning_rate": 4.903572834178417e-07, + "loss": 0.9112, + "step": 10367 + }, + { + "epoch": 0.7791973545768827, + "grad_norm": 1.655232314777717, + "learning_rate": 4.900379935442964e-07, + "loss": 0.9609, + "step": 10368 + }, + { + "epoch": 0.7792725086427176, + "grad_norm": 1.4959889666563533, + "learning_rate": 4.897187931417817e-07, + "loss": 0.9841, + "step": 10369 + }, + { + "epoch": 0.7793476627085525, + "grad_norm": 1.5835286906672261, + "learning_rate": 4.893996822292115e-07, + "loss": 0.9525, + "step": 10370 + }, + { + "epoch": 0.7794228167743875, + "grad_norm": 1.8380252736110854, + "learning_rate": 4.89080660825495e-07, + "loss": 0.9759, + "step": 10371 + }, + { + "epoch": 0.7794979708402224, + "grad_norm": 2.376887826014782, + "learning_rate": 4.887617289495349e-07, + "loss": 0.9356, + "step": 10372 + }, + { + "epoch": 0.7795731249060575, + "grad_norm": 1.7965739122321234, + "learning_rate": 4.884428866202288e-07, + "loss": 0.9592, + "step": 10373 + }, + { + "epoch": 0.7796482789718924, + "grad_norm": 1.8496189781833925, + "learning_rate": 4.881241338564706e-07, + "loss": 1.058, + "step": 10374 + }, + { + "epoch": 0.7797234330377273, + "grad_norm": 0.7510521581572172, + "learning_rate": 4.878054706771466e-07, + "loss": 0.8532, + "step": 10375 + }, + { + "epoch": 0.7797985871035623, + "grad_norm": 2.030257466718089, + "learning_rate": 4.87486897101139e-07, + "loss": 0.9835, + "step": 10376 + }, + { + "epoch": 0.7798737411693972, + "grad_norm": 2.70824718827403, + "learning_rate": 4.871684131473246e-07, + "loss": 0.9917, + "step": 10377 + }, + { + "epoch": 0.7799488952352323, + "grad_norm": 2.161422275717377, + "learning_rate": 4.868500188345744e-07, + "loss": 0.9642, + "step": 10378 + }, + { + "epoch": 0.7800240493010672, + "grad_norm": 2.0235403616546326, + "learning_rate": 4.865317141817551e-07, + "loss": 0.9655, + "step": 10379 + }, + { + "epoch": 0.7800992033669022, + "grad_norm": 1.8733632255128687, + "learning_rate": 4.862134992077274e-07, + "loss": 1.0212, + "step": 10380 + }, + { + "epoch": 0.7801743574327371, + "grad_norm": 22.61655736384712, + "learning_rate": 4.858953739313463e-07, + "loss": 1.0008, + "step": 10381 + }, + { + "epoch": 0.780249511498572, + "grad_norm": 2.5108050498378036, + "learning_rate": 4.855773383714623e-07, + "loss": 0.9359, + "step": 10382 + }, + { + "epoch": 0.780324665564407, + "grad_norm": 2.469897812185095, + "learning_rate": 4.852593925469198e-07, + "loss": 1.0276, + "step": 10383 + }, + { + "epoch": 0.780399819630242, + "grad_norm": 0.6954810334738318, + "learning_rate": 4.849415364765587e-07, + "loss": 0.8111, + "step": 10384 + }, + { + "epoch": 0.780474973696077, + "grad_norm": 2.9807685511283006, + "learning_rate": 4.846237701792136e-07, + "loss": 0.8816, + "step": 10385 + }, + { + "epoch": 0.7805501277619119, + "grad_norm": 1.3125445494584362, + "learning_rate": 4.843060936737125e-07, + "loss": 0.9703, + "step": 10386 + }, + { + "epoch": 0.7806252818277469, + "grad_norm": 2.422462736616733, + "learning_rate": 4.839885069788796e-07, + "loss": 1.0059, + "step": 10387 + }, + { + "epoch": 0.7807004358935818, + "grad_norm": 1.8891915806542536, + "learning_rate": 4.836710101135322e-07, + "loss": 1.0113, + "step": 10388 + }, + { + "epoch": 0.7807755899594168, + "grad_norm": 1.9571662901926425, + "learning_rate": 4.833536030964842e-07, + "loss": 0.8181, + "step": 10389 + }, + { + "epoch": 0.7808507440252518, + "grad_norm": 2.717803790422941, + "learning_rate": 4.830362859465431e-07, + "loss": 0.954, + "step": 10390 + }, + { + "epoch": 0.7809258980910867, + "grad_norm": 1.5283826333889037, + "learning_rate": 4.827190586825109e-07, + "loss": 1.0414, + "step": 10391 + }, + { + "epoch": 0.7810010521569217, + "grad_norm": 1.5953373203719283, + "learning_rate": 4.824019213231838e-07, + "loss": 1.0519, + "step": 10392 + }, + { + "epoch": 0.7810762062227566, + "grad_norm": 1.365724140121423, + "learning_rate": 4.820848738873549e-07, + "loss": 0.8529, + "step": 10393 + }, + { + "epoch": 0.7811513602885917, + "grad_norm": 2.011627847043588, + "learning_rate": 4.817679163938095e-07, + "loss": 1.0039, + "step": 10394 + }, + { + "epoch": 0.7812265143544266, + "grad_norm": 1.9733373089483954, + "learning_rate": 4.814510488613284e-07, + "loss": 1.0264, + "step": 10395 + }, + { + "epoch": 0.7813016684202615, + "grad_norm": 2.011780969342741, + "learning_rate": 4.811342713086881e-07, + "loss": 1.0043, + "step": 10396 + }, + { + "epoch": 0.7813768224860965, + "grad_norm": 1.5070553000010825, + "learning_rate": 4.808175837546582e-07, + "loss": 1.0135, + "step": 10397 + }, + { + "epoch": 0.7814519765519314, + "grad_norm": 1.9894739556996228, + "learning_rate": 4.805009862180038e-07, + "loss": 1.0604, + "step": 10398 + }, + { + "epoch": 0.7815271306177665, + "grad_norm": 1.5191141453360437, + "learning_rate": 4.801844787174847e-07, + "loss": 0.9068, + "step": 10399 + }, + { + "epoch": 0.7816022846836014, + "grad_norm": 0.8222543892353268, + "learning_rate": 4.798680612718544e-07, + "loss": 0.8556, + "step": 10400 + }, + { + "epoch": 0.7816774387494363, + "grad_norm": 3.4641587398445397, + "learning_rate": 4.795517338998629e-07, + "loss": 0.8823, + "step": 10401 + }, + { + "epoch": 0.7817525928152713, + "grad_norm": 1.4144247023226968, + "learning_rate": 4.792354966202534e-07, + "loss": 1.078, + "step": 10402 + }, + { + "epoch": 0.7818277468811062, + "grad_norm": 2.059767097343846, + "learning_rate": 4.78919349451764e-07, + "loss": 0.9649, + "step": 10403 + }, + { + "epoch": 0.7819029009469413, + "grad_norm": 4.597455664499238, + "learning_rate": 4.78603292413128e-07, + "loss": 0.8771, + "step": 10404 + }, + { + "epoch": 0.7819780550127762, + "grad_norm": 1.681212119203969, + "learning_rate": 4.78287325523072e-07, + "loss": 0.8812, + "step": 10405 + }, + { + "epoch": 0.7820532090786112, + "grad_norm": 1.9634612966695542, + "learning_rate": 4.779714488003197e-07, + "loss": 0.996, + "step": 10406 + }, + { + "epoch": 0.7821283631444461, + "grad_norm": 1.7134452800445392, + "learning_rate": 4.776556622635872e-07, + "loss": 0.9419, + "step": 10407 + }, + { + "epoch": 0.782203517210281, + "grad_norm": 1.875482179313492, + "learning_rate": 4.77339965931586e-07, + "loss": 0.9567, + "step": 10408 + }, + { + "epoch": 0.782278671276116, + "grad_norm": 2.476749160422994, + "learning_rate": 4.770243598230228e-07, + "loss": 1.0251, + "step": 10409 + }, + { + "epoch": 0.782353825341951, + "grad_norm": 1.6968828654238677, + "learning_rate": 4.7670884395659737e-07, + "loss": 1.009, + "step": 10410 + }, + { + "epoch": 0.782428979407786, + "grad_norm": 0.7548938666517614, + "learning_rate": 4.7639341835100654e-07, + "loss": 0.8904, + "step": 10411 + }, + { + "epoch": 0.7825041334736209, + "grad_norm": 3.8465807655787763, + "learning_rate": 4.7607808302494003e-07, + "loss": 0.8888, + "step": 10412 + }, + { + "epoch": 0.7825792875394559, + "grad_norm": 1.7402550624384143, + "learning_rate": 4.757628379970826e-07, + "loss": 0.9125, + "step": 10413 + }, + { + "epoch": 0.7826544416052909, + "grad_norm": 1.4960039475383797, + "learning_rate": 4.7544768328611317e-07, + "loss": 0.9649, + "step": 10414 + }, + { + "epoch": 0.7827295956711258, + "grad_norm": 1.6400350327355369, + "learning_rate": 4.7513261891070676e-07, + "loss": 0.9337, + "step": 10415 + }, + { + "epoch": 0.7828047497369608, + "grad_norm": 1.666917742255205, + "learning_rate": 4.74817644889532e-07, + "loss": 0.9151, + "step": 10416 + }, + { + "epoch": 0.7828799038027957, + "grad_norm": 1.831821388962079, + "learning_rate": 4.7450276124125153e-07, + "loss": 0.9836, + "step": 10417 + }, + { + "epoch": 0.7829550578686307, + "grad_norm": 8.866231105498732, + "learning_rate": 4.741879679845244e-07, + "loss": 0.9841, + "step": 10418 + }, + { + "epoch": 0.7830302119344656, + "grad_norm": 1.8339663048647188, + "learning_rate": 4.738732651380031e-07, + "loss": 0.9332, + "step": 10419 + }, + { + "epoch": 0.7831053660003006, + "grad_norm": 1.8343928555556996, + "learning_rate": 4.7355865272033455e-07, + "loss": 0.9266, + "step": 10420 + }, + { + "epoch": 0.7831805200661356, + "grad_norm": 2.0103174162036206, + "learning_rate": 4.7324413075016114e-07, + "loss": 0.976, + "step": 10421 + }, + { + "epoch": 0.7832556741319705, + "grad_norm": 1.7796966070221814, + "learning_rate": 4.729296992461187e-07, + "loss": 0.946, + "step": 10422 + }, + { + "epoch": 0.7833308281978055, + "grad_norm": 1.4260198524131136, + "learning_rate": 4.726153582268397e-07, + "loss": 0.9567, + "step": 10423 + }, + { + "epoch": 0.7834059822636404, + "grad_norm": 3.8093259762871075, + "learning_rate": 4.7230110771094933e-07, + "loss": 1.0174, + "step": 10424 + }, + { + "epoch": 0.7834811363294755, + "grad_norm": 1.4203362994828161, + "learning_rate": 4.7198694771706836e-07, + "loss": 0.9252, + "step": 10425 + }, + { + "epoch": 0.7835562903953104, + "grad_norm": 2.63177515076485, + "learning_rate": 4.7167287826381196e-07, + "loss": 1.0033, + "step": 10426 + }, + { + "epoch": 0.7836314444611453, + "grad_norm": 1.6464556310278262, + "learning_rate": 4.713588993697892e-07, + "loss": 0.9971, + "step": 10427 + }, + { + "epoch": 0.7837065985269803, + "grad_norm": 1.5550462318490297, + "learning_rate": 4.7104501105360594e-07, + "loss": 1.019, + "step": 10428 + }, + { + "epoch": 0.7837817525928152, + "grad_norm": 1.571357948400667, + "learning_rate": 4.7073121333386056e-07, + "loss": 1.0859, + "step": 10429 + }, + { + "epoch": 0.7838569066586503, + "grad_norm": 2.025964756725942, + "learning_rate": 4.7041750622914645e-07, + "loss": 0.9319, + "step": 10430 + }, + { + "epoch": 0.7839320607244852, + "grad_norm": 6.653692336147018, + "learning_rate": 4.701038897580525e-07, + "loss": 0.9726, + "step": 10431 + }, + { + "epoch": 0.7840072147903202, + "grad_norm": 2.038987666251319, + "learning_rate": 4.6979036393916093e-07, + "loss": 0.9823, + "step": 10432 + }, + { + "epoch": 0.7840823688561551, + "grad_norm": 2.7068953486866563, + "learning_rate": 4.694769287910503e-07, + "loss": 0.9354, + "step": 10433 + }, + { + "epoch": 0.78415752292199, + "grad_norm": 27.50325797462457, + "learning_rate": 4.6916358433229233e-07, + "loss": 0.9731, + "step": 10434 + }, + { + "epoch": 0.7842326769878251, + "grad_norm": 1.8088833471994912, + "learning_rate": 4.688503305814542e-07, + "loss": 1.0039, + "step": 10435 + }, + { + "epoch": 0.78430783105366, + "grad_norm": 1.6700944777849496, + "learning_rate": 4.6853716755709635e-07, + "loss": 1.0095, + "step": 10436 + }, + { + "epoch": 0.784382985119495, + "grad_norm": 2.312467213991085, + "learning_rate": 4.682240952777763e-07, + "loss": 0.858, + "step": 10437 + }, + { + "epoch": 0.7844581391853299, + "grad_norm": 1.4907187378397977, + "learning_rate": 4.679111137620442e-07, + "loss": 1.0043, + "step": 10438 + }, + { + "epoch": 0.7845332932511649, + "grad_norm": 1.9590976792616615, + "learning_rate": 4.675982230284448e-07, + "loss": 0.9869, + "step": 10439 + }, + { + "epoch": 0.7846084473169999, + "grad_norm": 3.457948241950181, + "learning_rate": 4.6728542309551923e-07, + "loss": 0.9498, + "step": 10440 + }, + { + "epoch": 0.7846836013828348, + "grad_norm": 1.8745004624279447, + "learning_rate": 4.669727139818014e-07, + "loss": 0.9462, + "step": 10441 + }, + { + "epoch": 0.7847587554486698, + "grad_norm": 3.3091080771371484, + "learning_rate": 4.6666009570582064e-07, + "loss": 0.9094, + "step": 10442 + }, + { + "epoch": 0.7848339095145047, + "grad_norm": 1.764672419053115, + "learning_rate": 4.663475682861009e-07, + "loss": 0.991, + "step": 10443 + }, + { + "epoch": 0.7849090635803397, + "grad_norm": 1.54396700858906, + "learning_rate": 4.6603513174115973e-07, + "loss": 1.0413, + "step": 10444 + }, + { + "epoch": 0.7849842176461747, + "grad_norm": 5.059890828217685, + "learning_rate": 4.6572278608951165e-07, + "loss": 0.9142, + "step": 10445 + }, + { + "epoch": 0.7850593717120096, + "grad_norm": 1.6842795831551833, + "learning_rate": 4.654105313496637e-07, + "loss": 0.9936, + "step": 10446 + }, + { + "epoch": 0.7851345257778446, + "grad_norm": 2.4402803067714283, + "learning_rate": 4.6509836754011787e-07, + "loss": 0.9851, + "step": 10447 + }, + { + "epoch": 0.7852096798436795, + "grad_norm": 1.7469694599960628, + "learning_rate": 4.647862946793715e-07, + "loss": 0.9822, + "step": 10448 + }, + { + "epoch": 0.7852848339095145, + "grad_norm": 2.247893300846859, + "learning_rate": 4.644743127859152e-07, + "loss": 0.9195, + "step": 10449 + }, + { + "epoch": 0.7853599879753494, + "grad_norm": 1.7812885481026264, + "learning_rate": 4.641624218782365e-07, + "loss": 0.9614, + "step": 10450 + }, + { + "epoch": 0.7854351420411845, + "grad_norm": 1.6445900369679411, + "learning_rate": 4.6385062197481527e-07, + "loss": 0.8524, + "step": 10451 + }, + { + "epoch": 0.7855102961070194, + "grad_norm": 2.6250669380009333, + "learning_rate": 4.635389130941272e-07, + "loss": 0.9144, + "step": 10452 + }, + { + "epoch": 0.7855854501728543, + "grad_norm": 2.1247740232700933, + "learning_rate": 4.6322729525464185e-07, + "loss": 0.9656, + "step": 10453 + }, + { + "epoch": 0.7856606042386893, + "grad_norm": 2.1659308675791857, + "learning_rate": 4.629157684748233e-07, + "loss": 0.9847, + "step": 10454 + }, + { + "epoch": 0.7857357583045242, + "grad_norm": 2.9930200439454917, + "learning_rate": 4.6260433277313215e-07, + "loss": 0.994, + "step": 10455 + }, + { + "epoch": 0.7858109123703593, + "grad_norm": 1.8516723986496035, + "learning_rate": 4.6229298816802066e-07, + "loss": 0.9653, + "step": 10456 + }, + { + "epoch": 0.7858860664361942, + "grad_norm": 1.4356759149592881, + "learning_rate": 4.619817346779391e-07, + "loss": 0.8976, + "step": 10457 + }, + { + "epoch": 0.7859612205020292, + "grad_norm": 2.232519379603222, + "learning_rate": 4.6167057232132787e-07, + "loss": 0.971, + "step": 10458 + }, + { + "epoch": 0.7860363745678641, + "grad_norm": 1.3782289046304899, + "learning_rate": 4.613595011166267e-07, + "loss": 0.8955, + "step": 10459 + }, + { + "epoch": 0.786111528633699, + "grad_norm": 2.513631563967516, + "learning_rate": 4.61048521082267e-07, + "loss": 0.9534, + "step": 10460 + }, + { + "epoch": 0.7861866826995341, + "grad_norm": 1.7327021186145055, + "learning_rate": 4.6073763223667474e-07, + "loss": 0.9452, + "step": 10461 + }, + { + "epoch": 0.786261836765369, + "grad_norm": 4.449801948512029, + "learning_rate": 4.6042683459827245e-07, + "loss": 0.999, + "step": 10462 + }, + { + "epoch": 0.786336990831204, + "grad_norm": 1.427249601753301, + "learning_rate": 4.6011612818547597e-07, + "loss": 0.9166, + "step": 10463 + }, + { + "epoch": 0.7864121448970389, + "grad_norm": 1.7396415274322106, + "learning_rate": 4.5980551301669535e-07, + "loss": 0.9971, + "step": 10464 + }, + { + "epoch": 0.7864872989628738, + "grad_norm": 2.5120528549173557, + "learning_rate": 4.5949498911033566e-07, + "loss": 1.0347, + "step": 10465 + }, + { + "epoch": 0.7865624530287089, + "grad_norm": 2.1797076261951647, + "learning_rate": 4.5918455648479647e-07, + "loss": 1.0008, + "step": 10466 + }, + { + "epoch": 0.7866376070945438, + "grad_norm": 1.6881647566711324, + "learning_rate": 4.58874215158473e-07, + "loss": 0.9973, + "step": 10467 + }, + { + "epoch": 0.7867127611603788, + "grad_norm": 2.329783111595673, + "learning_rate": 4.585639651497539e-07, + "loss": 0.882, + "step": 10468 + }, + { + "epoch": 0.7867879152262137, + "grad_norm": 1.7639476269064085, + "learning_rate": 4.5825380647702207e-07, + "loss": 1.0399, + "step": 10469 + }, + { + "epoch": 0.7868630692920487, + "grad_norm": 0.6575300584384073, + "learning_rate": 4.5794373915865625e-07, + "loss": 0.8351, + "step": 10470 + }, + { + "epoch": 0.7869382233578837, + "grad_norm": 1.845538032050974, + "learning_rate": 4.5763376321302804e-07, + "loss": 1.0136, + "step": 10471 + }, + { + "epoch": 0.7870133774237186, + "grad_norm": 1.8148696799033115, + "learning_rate": 4.573238786585061e-07, + "loss": 0.9513, + "step": 10472 + }, + { + "epoch": 0.7870885314895536, + "grad_norm": 1.6388541701064328, + "learning_rate": 4.5701408551345166e-07, + "loss": 0.9799, + "step": 10473 + }, + { + "epoch": 0.7871636855553885, + "grad_norm": 1.9048340338151812, + "learning_rate": 4.56704383796221e-07, + "loss": 0.867, + "step": 10474 + }, + { + "epoch": 0.7872388396212235, + "grad_norm": 1.8969898120001265, + "learning_rate": 4.5639477352516543e-07, + "loss": 0.9606, + "step": 10475 + }, + { + "epoch": 0.7873139936870585, + "grad_norm": 1.8693975034919705, + "learning_rate": 4.560852547186298e-07, + "loss": 0.9819, + "step": 10476 + }, + { + "epoch": 0.7873891477528935, + "grad_norm": 1.6259580868882761, + "learning_rate": 4.5577582739495545e-07, + "loss": 0.9736, + "step": 10477 + }, + { + "epoch": 0.7874643018187284, + "grad_norm": 1.9716366994176073, + "learning_rate": 4.5546649157247597e-07, + "loss": 0.984, + "step": 10478 + }, + { + "epoch": 0.7875394558845633, + "grad_norm": 1.5652572717745992, + "learning_rate": 4.551572472695224e-07, + "loss": 1.0033, + "step": 10479 + }, + { + "epoch": 0.7876146099503983, + "grad_norm": 2.566294582033579, + "learning_rate": 4.548480945044164e-07, + "loss": 0.8761, + "step": 10480 + }, + { + "epoch": 0.7876897640162333, + "grad_norm": 2.6857159037715594, + "learning_rate": 4.5453903329547816e-07, + "loss": 0.9716, + "step": 10481 + }, + { + "epoch": 0.7877649180820683, + "grad_norm": 2.7058675447288065, + "learning_rate": 4.5423006366102015e-07, + "loss": 1.086, + "step": 10482 + }, + { + "epoch": 0.7878400721479032, + "grad_norm": 1.5133888029163969, + "learning_rate": 4.539211856193494e-07, + "loss": 0.9584, + "step": 10483 + }, + { + "epoch": 0.7879152262137382, + "grad_norm": 1.621257580671036, + "learning_rate": 4.5361239918876946e-07, + "loss": 0.9983, + "step": 10484 + }, + { + "epoch": 0.7879903802795731, + "grad_norm": 1.799867079383499, + "learning_rate": 4.5330370438757624e-07, + "loss": 0.8962, + "step": 10485 + }, + { + "epoch": 0.788065534345408, + "grad_norm": 1.49303735628298, + "learning_rate": 4.5299510123406115e-07, + "loss": 0.9702, + "step": 10486 + }, + { + "epoch": 0.7881406884112431, + "grad_norm": 1.6810171143286352, + "learning_rate": 4.5268658974651044e-07, + "loss": 0.9794, + "step": 10487 + }, + { + "epoch": 0.788215842477078, + "grad_norm": 0.7971010448510698, + "learning_rate": 4.5237816994320365e-07, + "loss": 0.8343, + "step": 10488 + }, + { + "epoch": 0.788290996542913, + "grad_norm": 1.948816536971962, + "learning_rate": 4.5206984184241715e-07, + "loss": 0.984, + "step": 10489 + }, + { + "epoch": 0.7883661506087479, + "grad_norm": 1.8740345694652112, + "learning_rate": 4.517616054624198e-07, + "loss": 0.9124, + "step": 10490 + }, + { + "epoch": 0.7884413046745828, + "grad_norm": 1.5344716649576366, + "learning_rate": 4.5145346082147594e-07, + "loss": 0.9008, + "step": 10491 + }, + { + "epoch": 0.7885164587404179, + "grad_norm": 1.6131248860172556, + "learning_rate": 4.511454079378445e-07, + "loss": 0.9331, + "step": 10492 + }, + { + "epoch": 0.7885916128062528, + "grad_norm": 1.6432950771874837, + "learning_rate": 4.5083744682977775e-07, + "loss": 0.9438, + "step": 10493 + }, + { + "epoch": 0.7886667668720878, + "grad_norm": 1.7464212882725612, + "learning_rate": 4.505295775155251e-07, + "loss": 0.9006, + "step": 10494 + }, + { + "epoch": 0.7887419209379227, + "grad_norm": 1.6333552284459956, + "learning_rate": 4.502218000133284e-07, + "loss": 0.8736, + "step": 10495 + }, + { + "epoch": 0.7888170750037578, + "grad_norm": 2.4352307638370596, + "learning_rate": 4.4991411434142445e-07, + "loss": 0.8727, + "step": 10496 + }, + { + "epoch": 0.7888922290695927, + "grad_norm": 2.019682238485031, + "learning_rate": 4.49606520518045e-07, + "loss": 0.8926, + "step": 10497 + }, + { + "epoch": 0.7889673831354276, + "grad_norm": 6.630126121552689, + "learning_rate": 4.492990185614154e-07, + "loss": 0.8582, + "step": 10498 + }, + { + "epoch": 0.7890425372012626, + "grad_norm": 4.525173559074402, + "learning_rate": 4.489916084897576e-07, + "loss": 1.079, + "step": 10499 + }, + { + "epoch": 0.7891176912670975, + "grad_norm": 1.5757374596173168, + "learning_rate": 4.4868429032128575e-07, + "loss": 0.9606, + "step": 10500 + }, + { + "epoch": 0.7891928453329325, + "grad_norm": 1.6427530202389373, + "learning_rate": 4.483770640742104e-07, + "loss": 0.9926, + "step": 10501 + }, + { + "epoch": 0.7892679993987675, + "grad_norm": 2.718999741582658, + "learning_rate": 4.480699297667356e-07, + "loss": 0.9481, + "step": 10502 + }, + { + "epoch": 0.7893431534646025, + "grad_norm": 2.10207261485104, + "learning_rate": 4.4776288741706047e-07, + "loss": 0.9849, + "step": 10503 + }, + { + "epoch": 0.7894183075304374, + "grad_norm": 1.438145865596116, + "learning_rate": 4.474559370433779e-07, + "loss": 0.9328, + "step": 10504 + }, + { + "epoch": 0.7894934615962723, + "grad_norm": 3.6955122460219294, + "learning_rate": 4.4714907866387565e-07, + "loss": 0.9993, + "step": 10505 + }, + { + "epoch": 0.7895686156621073, + "grad_norm": 1.5620091239428275, + "learning_rate": 4.468423122967373e-07, + "loss": 0.8817, + "step": 10506 + }, + { + "epoch": 0.7896437697279423, + "grad_norm": 3.7874986705211158, + "learning_rate": 4.465356379601395e-07, + "loss": 0.9697, + "step": 10507 + }, + { + "epoch": 0.7897189237937773, + "grad_norm": 0.8737664724060622, + "learning_rate": 4.462290556722537e-07, + "loss": 0.9289, + "step": 10508 + }, + { + "epoch": 0.7897940778596122, + "grad_norm": 1.4157707897170762, + "learning_rate": 4.4592256545124616e-07, + "loss": 1.0218, + "step": 10509 + }, + { + "epoch": 0.7898692319254471, + "grad_norm": 1.5999616856755923, + "learning_rate": 4.4561616731527695e-07, + "loss": 0.8812, + "step": 10510 + }, + { + "epoch": 0.7899443859912821, + "grad_norm": 1.518317122444406, + "learning_rate": 4.4530986128250257e-07, + "loss": 0.9725, + "step": 10511 + }, + { + "epoch": 0.790019540057117, + "grad_norm": 1.4695005731052606, + "learning_rate": 4.450036473710721e-07, + "loss": 1.0009, + "step": 10512 + }, + { + "epoch": 0.7900946941229521, + "grad_norm": 1.5517528958779883, + "learning_rate": 4.446975255991301e-07, + "loss": 0.9618, + "step": 10513 + }, + { + "epoch": 0.790169848188787, + "grad_norm": 1.8472717401214138, + "learning_rate": 4.443914959848154e-07, + "loss": 0.9686, + "step": 10514 + }, + { + "epoch": 0.790245002254622, + "grad_norm": 1.7669165284627024, + "learning_rate": 4.4408555854626085e-07, + "loss": 0.9044, + "step": 10515 + }, + { + "epoch": 0.7903201563204569, + "grad_norm": 2.095454874115777, + "learning_rate": 4.437797133015955e-07, + "loss": 0.998, + "step": 10516 + }, + { + "epoch": 0.7903953103862918, + "grad_norm": 16.08425842681673, + "learning_rate": 4.434739602689412e-07, + "loss": 0.918, + "step": 10517 + }, + { + "epoch": 0.7904704644521269, + "grad_norm": 1.462932835620578, + "learning_rate": 4.43168299466415e-07, + "loss": 0.9346, + "step": 10518 + }, + { + "epoch": 0.7905456185179618, + "grad_norm": 2.2141622473754348, + "learning_rate": 4.428627309121287e-07, + "loss": 0.9927, + "step": 10519 + }, + { + "epoch": 0.7906207725837968, + "grad_norm": 2.2246888961176747, + "learning_rate": 4.425572546241878e-07, + "loss": 0.9921, + "step": 10520 + }, + { + "epoch": 0.7906959266496317, + "grad_norm": 2.511935069022507, + "learning_rate": 4.422518706206939e-07, + "loss": 1.0006, + "step": 10521 + }, + { + "epoch": 0.7907710807154668, + "grad_norm": 2.171098748771346, + "learning_rate": 4.4194657891974097e-07, + "loss": 1.0018, + "step": 10522 + }, + { + "epoch": 0.7908462347813017, + "grad_norm": 0.744790141675406, + "learning_rate": 4.416413795394203e-07, + "loss": 0.8191, + "step": 10523 + }, + { + "epoch": 0.7909213888471366, + "grad_norm": 1.8870835988095158, + "learning_rate": 4.413362724978149e-07, + "loss": 0.9259, + "step": 10524 + }, + { + "epoch": 0.7909965429129716, + "grad_norm": 1.4711423018066325, + "learning_rate": 4.41031257813004e-07, + "loss": 0.9403, + "step": 10525 + }, + { + "epoch": 0.7910716969788065, + "grad_norm": 2.842796920527364, + "learning_rate": 4.407263355030608e-07, + "loss": 1.0005, + "step": 10526 + }, + { + "epoch": 0.7911468510446416, + "grad_norm": 1.7760834302121673, + "learning_rate": 4.404215055860525e-07, + "loss": 1.0211, + "step": 10527 + }, + { + "epoch": 0.7912220051104765, + "grad_norm": 3.0141692922290875, + "learning_rate": 4.4011676808004263e-07, + "loss": 0.9198, + "step": 10528 + }, + { + "epoch": 0.7912971591763115, + "grad_norm": 2.103474481121821, + "learning_rate": 4.398121230030876e-07, + "loss": 0.8858, + "step": 10529 + }, + { + "epoch": 0.7913723132421464, + "grad_norm": 1.5508807940688087, + "learning_rate": 4.3950757037323826e-07, + "loss": 0.9757, + "step": 10530 + }, + { + "epoch": 0.7914474673079813, + "grad_norm": 2.522090587831361, + "learning_rate": 4.3920311020854117e-07, + "loss": 1.0495, + "step": 10531 + }, + { + "epoch": 0.7915226213738163, + "grad_norm": 1.5786055031657669, + "learning_rate": 4.3889874252703585e-07, + "loss": 0.9804, + "step": 10532 + }, + { + "epoch": 0.7915977754396513, + "grad_norm": 1.658650368269014, + "learning_rate": 4.385944673467585e-07, + "loss": 0.9916, + "step": 10533 + }, + { + "epoch": 0.7916729295054863, + "grad_norm": 2.596720637428612, + "learning_rate": 4.3829028468573793e-07, + "loss": 0.8604, + "step": 10534 + }, + { + "epoch": 0.7917480835713212, + "grad_norm": 2.27362468774722, + "learning_rate": 4.3798619456199803e-07, + "loss": 0.9776, + "step": 10535 + }, + { + "epoch": 0.7918232376371561, + "grad_norm": 1.7881366716288636, + "learning_rate": 4.376821969935578e-07, + "loss": 0.9813, + "step": 10536 + }, + { + "epoch": 0.7918983917029911, + "grad_norm": 1.8130814507562576, + "learning_rate": 4.3737829199842903e-07, + "loss": 0.8224, + "step": 10537 + }, + { + "epoch": 0.7919735457688261, + "grad_norm": 1.5779260944373237, + "learning_rate": 4.3707447959462087e-07, + "loss": 1.0023, + "step": 10538 + }, + { + "epoch": 0.7920486998346611, + "grad_norm": 1.7522139169093944, + "learning_rate": 4.3677075980013465e-07, + "loss": 0.9644, + "step": 10539 + }, + { + "epoch": 0.792123853900496, + "grad_norm": 1.7749721927882445, + "learning_rate": 4.3646713263296677e-07, + "loss": 0.9566, + "step": 10540 + }, + { + "epoch": 0.792199007966331, + "grad_norm": 1.8594822572178087, + "learning_rate": 4.3616359811110847e-07, + "loss": 0.9161, + "step": 10541 + }, + { + "epoch": 0.7922741620321659, + "grad_norm": 1.806707531489145, + "learning_rate": 4.35860156252545e-07, + "loss": 0.9652, + "step": 10542 + }, + { + "epoch": 0.7923493160980009, + "grad_norm": 5.405598331049986, + "learning_rate": 4.355568070752571e-07, + "loss": 1.0129, + "step": 10543 + }, + { + "epoch": 0.7924244701638359, + "grad_norm": 1.6528219338225538, + "learning_rate": 4.352535505972186e-07, + "loss": 0.9891, + "step": 10544 + }, + { + "epoch": 0.7924996242296708, + "grad_norm": 1.5904237781046193, + "learning_rate": 4.349503868363993e-07, + "loss": 0.9898, + "step": 10545 + }, + { + "epoch": 0.7925747782955058, + "grad_norm": 0.8264140322054301, + "learning_rate": 4.346473158107629e-07, + "loss": 0.8708, + "step": 10546 + }, + { + "epoch": 0.7926499323613407, + "grad_norm": 1.420042009269518, + "learning_rate": 4.3434433753826696e-07, + "loss": 1.0166, + "step": 10547 + }, + { + "epoch": 0.7927250864271758, + "grad_norm": 2.1486868418875593, + "learning_rate": 4.340414520368645e-07, + "loss": 0.9215, + "step": 10548 + }, + { + "epoch": 0.7928002404930107, + "grad_norm": 1.706216019512145, + "learning_rate": 4.3373865932450184e-07, + "loss": 0.968, + "step": 10549 + }, + { + "epoch": 0.7928753945588456, + "grad_norm": 2.0929525194610274, + "learning_rate": 4.334359594191217e-07, + "loss": 0.9358, + "step": 10550 + }, + { + "epoch": 0.7929505486246806, + "grad_norm": 0.6115944864039514, + "learning_rate": 4.3313335233865976e-07, + "loss": 0.7589, + "step": 10551 + }, + { + "epoch": 0.7930257026905155, + "grad_norm": 2.08449821965834, + "learning_rate": 4.328308381010466e-07, + "loss": 0.9955, + "step": 10552 + }, + { + "epoch": 0.7931008567563506, + "grad_norm": 1.7110659059041338, + "learning_rate": 4.325284167242076e-07, + "loss": 0.91, + "step": 10553 + }, + { + "epoch": 0.7931760108221855, + "grad_norm": 3.4925880970421614, + "learning_rate": 4.3222608822606134e-07, + "loss": 0.8987, + "step": 10554 + }, + { + "epoch": 0.7932511648880204, + "grad_norm": 1.6850533586198018, + "learning_rate": 4.3192385262452344e-07, + "loss": 0.9478, + "step": 10555 + }, + { + "epoch": 0.7933263189538554, + "grad_norm": 1.722927911145912, + "learning_rate": 4.316217099375017e-07, + "loss": 0.9657, + "step": 10556 + }, + { + "epoch": 0.7934014730196903, + "grad_norm": 1.6853341225811362, + "learning_rate": 4.3131966018289946e-07, + "loss": 0.9865, + "step": 10557 + }, + { + "epoch": 0.7934766270855254, + "grad_norm": 4.337651424348249, + "learning_rate": 4.3101770337861445e-07, + "loss": 0.9848, + "step": 10558 + }, + { + "epoch": 0.7935517811513603, + "grad_norm": 6.300196375132633, + "learning_rate": 4.3071583954253765e-07, + "loss": 0.9007, + "step": 10559 + }, + { + "epoch": 0.7936269352171953, + "grad_norm": 1.6336249561512672, + "learning_rate": 4.3041406869255726e-07, + "loss": 0.9806, + "step": 10560 + }, + { + "epoch": 0.7937020892830302, + "grad_norm": 1.4267141138339225, + "learning_rate": 4.301123908465536e-07, + "loss": 0.9107, + "step": 10561 + }, + { + "epoch": 0.7937772433488651, + "grad_norm": 1.5433801030239693, + "learning_rate": 4.298108060224024e-07, + "loss": 0.9358, + "step": 10562 + }, + { + "epoch": 0.7938523974147002, + "grad_norm": 1.5107783268241286, + "learning_rate": 4.295093142379735e-07, + "loss": 0.9592, + "step": 10563 + }, + { + "epoch": 0.7939275514805351, + "grad_norm": 1.4486668607802655, + "learning_rate": 4.29207915511131e-07, + "loss": 1.0129, + "step": 10564 + }, + { + "epoch": 0.7940027055463701, + "grad_norm": 2.2249705194813334, + "learning_rate": 4.289066098597349e-07, + "loss": 1.0065, + "step": 10565 + }, + { + "epoch": 0.794077859612205, + "grad_norm": 2.1429277272179736, + "learning_rate": 4.286053973016379e-07, + "loss": 1.0391, + "step": 10566 + }, + { + "epoch": 0.79415301367804, + "grad_norm": 0.6723408303983092, + "learning_rate": 4.28304277854689e-07, + "loss": 0.8416, + "step": 10567 + }, + { + "epoch": 0.794228167743875, + "grad_norm": 1.5047962280159237, + "learning_rate": 4.2800325153673e-07, + "loss": 0.953, + "step": 10568 + }, + { + "epoch": 0.7943033218097099, + "grad_norm": 1.8893975931697604, + "learning_rate": 4.277023183655977e-07, + "loss": 0.8647, + "step": 10569 + }, + { + "epoch": 0.7943784758755449, + "grad_norm": 1.7058976728378175, + "learning_rate": 4.27401478359124e-07, + "loss": 0.8594, + "step": 10570 + }, + { + "epoch": 0.7944536299413798, + "grad_norm": 2.650550202166815, + "learning_rate": 4.2710073153513404e-07, + "loss": 1.0364, + "step": 10571 + }, + { + "epoch": 0.7945287840072148, + "grad_norm": 2.924312288998291, + "learning_rate": 4.268000779114491e-07, + "loss": 0.9665, + "step": 10572 + }, + { + "epoch": 0.7946039380730497, + "grad_norm": 2.6580523490888637, + "learning_rate": 4.264995175058841e-07, + "loss": 0.9849, + "step": 10573 + }, + { + "epoch": 0.7946790921388848, + "grad_norm": 0.7838187901486543, + "learning_rate": 4.261990503362478e-07, + "loss": 0.8816, + "step": 10574 + }, + { + "epoch": 0.7947542462047197, + "grad_norm": 2.1709212809470655, + "learning_rate": 4.2589867642034427e-07, + "loss": 0.9586, + "step": 10575 + }, + { + "epoch": 0.7948294002705546, + "grad_norm": 3.8230161870463557, + "learning_rate": 4.255983957759712e-07, + "loss": 1.0079, + "step": 10576 + }, + { + "epoch": 0.7949045543363896, + "grad_norm": 2.288837929603834, + "learning_rate": 4.252982084209225e-07, + "loss": 1.017, + "step": 10577 + }, + { + "epoch": 0.7949797084022245, + "grad_norm": 1.3768876729789779, + "learning_rate": 4.24998114372985e-07, + "loss": 0.9397, + "step": 10578 + }, + { + "epoch": 0.7950548624680596, + "grad_norm": 1.968202302831344, + "learning_rate": 4.2469811364994037e-07, + "loss": 0.9294, + "step": 10579 + }, + { + "epoch": 0.7951300165338945, + "grad_norm": 2.0049580867886125, + "learning_rate": 4.2439820626956455e-07, + "loss": 0.9929, + "step": 10580 + }, + { + "epoch": 0.7952051705997294, + "grad_norm": 1.762789381903863, + "learning_rate": 4.2409839224962795e-07, + "loss": 0.9666, + "step": 10581 + }, + { + "epoch": 0.7952803246655644, + "grad_norm": 1.3535218782266716, + "learning_rate": 4.237986716078965e-07, + "loss": 0.9429, + "step": 10582 + }, + { + "epoch": 0.7953554787313993, + "grad_norm": 1.6495849809913792, + "learning_rate": 4.234990443621298e-07, + "loss": 0.9789, + "step": 10583 + }, + { + "epoch": 0.7954306327972344, + "grad_norm": 1.4238948518582963, + "learning_rate": 4.2319951053008116e-07, + "loss": 0.9538, + "step": 10584 + }, + { + "epoch": 0.7955057868630693, + "grad_norm": 1.7584135257218099, + "learning_rate": 4.229000701294998e-07, + "loss": 0.9125, + "step": 10585 + }, + { + "epoch": 0.7955809409289043, + "grad_norm": 1.7571449877341685, + "learning_rate": 4.2260072317812766e-07, + "loss": 0.9101, + "step": 10586 + }, + { + "epoch": 0.7956560949947392, + "grad_norm": 1.9154071677786124, + "learning_rate": 4.223014696937035e-07, + "loss": 0.8663, + "step": 10587 + }, + { + "epoch": 0.7957312490605741, + "grad_norm": 1.9275369127551278, + "learning_rate": 4.220023096939582e-07, + "loss": 0.8876, + "step": 10588 + }, + { + "epoch": 0.7958064031264092, + "grad_norm": 3.7153359777976314, + "learning_rate": 4.217032431966192e-07, + "loss": 0.9182, + "step": 10589 + }, + { + "epoch": 0.7958815571922441, + "grad_norm": 1.6922355187137839, + "learning_rate": 4.214042702194067e-07, + "loss": 0.9539, + "step": 10590 + }, + { + "epoch": 0.7959567112580791, + "grad_norm": 1.4773603433355955, + "learning_rate": 4.211053907800359e-07, + "loss": 0.9522, + "step": 10591 + }, + { + "epoch": 0.796031865323914, + "grad_norm": 1.9351474877268067, + "learning_rate": 4.208066048962169e-07, + "loss": 0.9764, + "step": 10592 + }, + { + "epoch": 0.796107019389749, + "grad_norm": 1.931683083772494, + "learning_rate": 4.20507912585653e-07, + "loss": 1.0185, + "step": 10593 + }, + { + "epoch": 0.796182173455584, + "grad_norm": 2.8769131804656554, + "learning_rate": 4.202093138660443e-07, + "loss": 0.9498, + "step": 10594 + }, + { + "epoch": 0.7962573275214189, + "grad_norm": 1.7772668172094568, + "learning_rate": 4.199108087550829e-07, + "loss": 0.9769, + "step": 10595 + }, + { + "epoch": 0.7963324815872539, + "grad_norm": 2.023856336923069, + "learning_rate": 4.196123972704568e-07, + "loss": 0.8876, + "step": 10596 + }, + { + "epoch": 0.7964076356530888, + "grad_norm": 2.019524284660679, + "learning_rate": 4.1931407942984777e-07, + "loss": 0.8715, + "step": 10597 + }, + { + "epoch": 0.7964827897189238, + "grad_norm": 1.462089777740484, + "learning_rate": 4.19015855250932e-07, + "loss": 0.9438, + "step": 10598 + }, + { + "epoch": 0.7965579437847587, + "grad_norm": 1.7146080741472403, + "learning_rate": 4.1871772475138136e-07, + "loss": 0.9607, + "step": 10599 + }, + { + "epoch": 0.7966330978505937, + "grad_norm": 2.55379402798619, + "learning_rate": 4.184196879488604e-07, + "loss": 0.8686, + "step": 10600 + }, + { + "epoch": 0.7967082519164287, + "grad_norm": 0.6890653432843634, + "learning_rate": 4.181217448610295e-07, + "loss": 0.7849, + "step": 10601 + }, + { + "epoch": 0.7967834059822636, + "grad_norm": 1.550773793546109, + "learning_rate": 4.178238955055424e-07, + "loss": 1.0143, + "step": 10602 + }, + { + "epoch": 0.7968585600480986, + "grad_norm": 1.6209838829957208, + "learning_rate": 4.175261399000476e-07, + "loss": 0.9456, + "step": 10603 + }, + { + "epoch": 0.7969337141139335, + "grad_norm": 2.333235999756626, + "learning_rate": 4.172284780621893e-07, + "loss": 0.9285, + "step": 10604 + }, + { + "epoch": 0.7970088681797686, + "grad_norm": 2.067124701574714, + "learning_rate": 4.1693091000960454e-07, + "loss": 0.9538, + "step": 10605 + }, + { + "epoch": 0.7970840222456035, + "grad_norm": 2.1883758971158582, + "learning_rate": 4.1663343575992526e-07, + "loss": 0.9967, + "step": 10606 + }, + { + "epoch": 0.7971591763114384, + "grad_norm": 1.4956928243852754, + "learning_rate": 4.16336055330778e-07, + "loss": 1.0089, + "step": 10607 + }, + { + "epoch": 0.7972343303772734, + "grad_norm": 1.517367668729337, + "learning_rate": 4.1603876873978327e-07, + "loss": 1.0014, + "step": 10608 + }, + { + "epoch": 0.7973094844431083, + "grad_norm": 3.332699779065437, + "learning_rate": 4.157415760045573e-07, + "loss": 1.0489, + "step": 10609 + }, + { + "epoch": 0.7973846385089434, + "grad_norm": 1.7283059547975503, + "learning_rate": 4.15444477142709e-07, + "loss": 0.9026, + "step": 10610 + }, + { + "epoch": 0.7974597925747783, + "grad_norm": 1.6628989149787121, + "learning_rate": 4.1514747217184355e-07, + "loss": 1.0361, + "step": 10611 + }, + { + "epoch": 0.7975349466406133, + "grad_norm": 3.3250768609592503, + "learning_rate": 4.148505611095594e-07, + "loss": 0.8862, + "step": 10612 + }, + { + "epoch": 0.7976101007064482, + "grad_norm": 1.4902482938303954, + "learning_rate": 4.145537439734492e-07, + "loss": 0.9476, + "step": 10613 + }, + { + "epoch": 0.7976852547722831, + "grad_norm": 1.9031505088161456, + "learning_rate": 4.142570207811009e-07, + "loss": 0.959, + "step": 10614 + }, + { + "epoch": 0.7977604088381182, + "grad_norm": 2.0935841964434236, + "learning_rate": 4.139603915500958e-07, + "loss": 1.0459, + "step": 10615 + }, + { + "epoch": 0.7978355629039531, + "grad_norm": 1.6601761401611799, + "learning_rate": 4.1366385629801126e-07, + "loss": 0.9683, + "step": 10616 + }, + { + "epoch": 0.7979107169697881, + "grad_norm": 2.4446515539843197, + "learning_rate": 4.1336741504241803e-07, + "loss": 0.9254, + "step": 10617 + }, + { + "epoch": 0.797985871035623, + "grad_norm": 1.5665617982112119, + "learning_rate": 4.1307106780088065e-07, + "loss": 0.9404, + "step": 10618 + }, + { + "epoch": 0.798061025101458, + "grad_norm": 1.8612318624479933, + "learning_rate": 4.1277481459095954e-07, + "loss": 0.9441, + "step": 10619 + }, + { + "epoch": 0.798136179167293, + "grad_norm": 4.9900998330688875, + "learning_rate": 4.1247865543020797e-07, + "loss": 1.0287, + "step": 10620 + }, + { + "epoch": 0.7982113332331279, + "grad_norm": 1.5790771029529598, + "learning_rate": 4.121825903361755e-07, + "loss": 0.9197, + "step": 10621 + }, + { + "epoch": 0.7982864872989629, + "grad_norm": 2.4096510989942668, + "learning_rate": 4.1188661932640503e-07, + "loss": 1.0054, + "step": 10622 + }, + { + "epoch": 0.7983616413647978, + "grad_norm": 1.964993783974947, + "learning_rate": 4.1159074241843326e-07, + "loss": 0.9417, + "step": 10623 + }, + { + "epoch": 0.7984367954306328, + "grad_norm": 1.5476261973652554, + "learning_rate": 4.112949596297928e-07, + "loss": 0.8797, + "step": 10624 + }, + { + "epoch": 0.7985119494964678, + "grad_norm": 4.735534123201609, + "learning_rate": 4.109992709780088e-07, + "loss": 0.9844, + "step": 10625 + }, + { + "epoch": 0.7985871035623027, + "grad_norm": 1.6348606387365896, + "learning_rate": 4.107036764806031e-07, + "loss": 0.9198, + "step": 10626 + }, + { + "epoch": 0.7986622576281377, + "grad_norm": 2.1945139019539477, + "learning_rate": 4.104081761550902e-07, + "loss": 0.8948, + "step": 10627 + }, + { + "epoch": 0.7987374116939726, + "grad_norm": 2.012753829586625, + "learning_rate": 4.101127700189806e-07, + "loss": 1.0002, + "step": 10628 + }, + { + "epoch": 0.7988125657598076, + "grad_norm": 1.9428208414861718, + "learning_rate": 4.0981745808977707e-07, + "loss": 0.9867, + "step": 10629 + }, + { + "epoch": 0.7988877198256426, + "grad_norm": 1.4466019111121755, + "learning_rate": 4.0952224038497764e-07, + "loss": 0.9506, + "step": 10630 + }, + { + "epoch": 0.7989628738914776, + "grad_norm": 1.563416936763178, + "learning_rate": 4.0922711692207645e-07, + "loss": 0.9576, + "step": 10631 + }, + { + "epoch": 0.7990380279573125, + "grad_norm": 1.5925356408507034, + "learning_rate": 4.089320877185596e-07, + "loss": 0.9158, + "step": 10632 + }, + { + "epoch": 0.7991131820231474, + "grad_norm": 2.5325316963584688, + "learning_rate": 4.086371527919097e-07, + "loss": 0.9876, + "step": 10633 + }, + { + "epoch": 0.7991883360889824, + "grad_norm": 2.1416516591695376, + "learning_rate": 4.083423121596021e-07, + "loss": 0.9777, + "step": 10634 + }, + { + "epoch": 0.7992634901548173, + "grad_norm": 0.7104974160117208, + "learning_rate": 4.080475658391076e-07, + "loss": 0.821, + "step": 10635 + }, + { + "epoch": 0.7993386442206524, + "grad_norm": 1.8176435296895646, + "learning_rate": 4.077529138478906e-07, + "loss": 1.0106, + "step": 10636 + }, + { + "epoch": 0.7994137982864873, + "grad_norm": 3.3076716807955733, + "learning_rate": 4.074583562034102e-07, + "loss": 0.9537, + "step": 10637 + }, + { + "epoch": 0.7994889523523223, + "grad_norm": 1.4814358976865882, + "learning_rate": 4.071638929231207e-07, + "loss": 0.8722, + "step": 10638 + }, + { + "epoch": 0.7995641064181572, + "grad_norm": 2.3040163258994033, + "learning_rate": 4.0686952402447016e-07, + "loss": 1.0085, + "step": 10639 + }, + { + "epoch": 0.7996392604839921, + "grad_norm": 1.4768214730318652, + "learning_rate": 4.0657524952490087e-07, + "loss": 0.9364, + "step": 10640 + }, + { + "epoch": 0.7997144145498272, + "grad_norm": 1.626775212079778, + "learning_rate": 4.0628106944184947e-07, + "loss": 0.8008, + "step": 10641 + }, + { + "epoch": 0.7997895686156621, + "grad_norm": 1.5800078085211455, + "learning_rate": 4.05986983792747e-07, + "loss": 0.9505, + "step": 10642 + }, + { + "epoch": 0.7998647226814971, + "grad_norm": 2.056111469646126, + "learning_rate": 4.0569299259502035e-07, + "loss": 1.0467, + "step": 10643 + }, + { + "epoch": 0.799939876747332, + "grad_norm": 2.0530308764291716, + "learning_rate": 4.0539909586608866e-07, + "loss": 0.9109, + "step": 10644 + }, + { + "epoch": 0.8000150308131669, + "grad_norm": 1.4494131939762929, + "learning_rate": 4.051052936233668e-07, + "loss": 0.9774, + "step": 10645 + }, + { + "epoch": 0.800090184879002, + "grad_norm": 2.070119481715754, + "learning_rate": 4.0481158588426334e-07, + "loss": 1.0383, + "step": 10646 + }, + { + "epoch": 0.8001653389448369, + "grad_norm": 2.287982047730822, + "learning_rate": 4.045179726661816e-07, + "loss": 0.8795, + "step": 10647 + }, + { + "epoch": 0.8002404930106719, + "grad_norm": 1.7196621641787904, + "learning_rate": 4.0422445398651985e-07, + "loss": 0.9508, + "step": 10648 + }, + { + "epoch": 0.8003156470765068, + "grad_norm": 2.2309554816591524, + "learning_rate": 4.0393102986266925e-07, + "loss": 1.0386, + "step": 10649 + }, + { + "epoch": 0.8003908011423418, + "grad_norm": 2.802221122294099, + "learning_rate": 4.0363770031201793e-07, + "loss": 1.0377, + "step": 10650 + }, + { + "epoch": 0.8004659552081768, + "grad_norm": 1.5741986824820113, + "learning_rate": 4.033444653519449e-07, + "loss": 0.9785, + "step": 10651 + }, + { + "epoch": 0.8005411092740117, + "grad_norm": 1.7485203618518563, + "learning_rate": 4.030513249998266e-07, + "loss": 0.9853, + "step": 10652 + }, + { + "epoch": 0.8006162633398467, + "grad_norm": 1.4680337173841025, + "learning_rate": 4.0275827927303265e-07, + "loss": 0.9873, + "step": 10653 + }, + { + "epoch": 0.8006914174056816, + "grad_norm": 1.556716515044496, + "learning_rate": 4.024653281889261e-07, + "loss": 0.998, + "step": 10654 + }, + { + "epoch": 0.8007665714715166, + "grad_norm": 4.936253837068119, + "learning_rate": 4.021724717648669e-07, + "loss": 0.947, + "step": 10655 + }, + { + "epoch": 0.8008417255373516, + "grad_norm": 3.954164029205268, + "learning_rate": 4.018797100182072e-07, + "loss": 0.9578, + "step": 10656 + }, + { + "epoch": 0.8009168796031866, + "grad_norm": 1.4132534684402334, + "learning_rate": 4.0158704296629445e-07, + "loss": 0.9376, + "step": 10657 + }, + { + "epoch": 0.8009920336690215, + "grad_norm": 1.9467341731833956, + "learning_rate": 4.0129447062646983e-07, + "loss": 1.0123, + "step": 10658 + }, + { + "epoch": 0.8010671877348564, + "grad_norm": 2.405357988167045, + "learning_rate": 4.010019930160695e-07, + "loss": 0.9812, + "step": 10659 + }, + { + "epoch": 0.8011423418006914, + "grad_norm": 1.503087759845378, + "learning_rate": 4.0070961015242433e-07, + "loss": 0.9663, + "step": 10660 + }, + { + "epoch": 0.8012174958665264, + "grad_norm": 1.6804840328385284, + "learning_rate": 4.0041732205285883e-07, + "loss": 0.8992, + "step": 10661 + }, + { + "epoch": 0.8012926499323614, + "grad_norm": 2.0676826328118487, + "learning_rate": 4.001251287346925e-07, + "loss": 0.947, + "step": 10662 + }, + { + "epoch": 0.8013678039981963, + "grad_norm": 1.3641775950042832, + "learning_rate": 3.998330302152384e-07, + "loss": 0.9411, + "step": 10663 + }, + { + "epoch": 0.8014429580640313, + "grad_norm": 2.0439541775966372, + "learning_rate": 3.995410265118042e-07, + "loss": 0.9453, + "step": 10664 + }, + { + "epoch": 0.8015181121298662, + "grad_norm": 1.8865199013880374, + "learning_rate": 3.992491176416932e-07, + "loss": 0.9767, + "step": 10665 + }, + { + "epoch": 0.8015932661957011, + "grad_norm": 1.6300200901080586, + "learning_rate": 3.989573036222018e-07, + "loss": 1.0509, + "step": 10666 + }, + { + "epoch": 0.8016684202615362, + "grad_norm": 2.2292211888380766, + "learning_rate": 3.986655844706208e-07, + "loss": 0.9742, + "step": 10667 + }, + { + "epoch": 0.8017435743273711, + "grad_norm": 1.6539232717150767, + "learning_rate": 3.9837396020423595e-07, + "loss": 0.9504, + "step": 10668 + }, + { + "epoch": 0.8018187283932061, + "grad_norm": 3.6098158265788087, + "learning_rate": 3.9808243084032657e-07, + "loss": 0.9349, + "step": 10669 + }, + { + "epoch": 0.801893882459041, + "grad_norm": 1.949029152618387, + "learning_rate": 3.9779099639616766e-07, + "loss": 0.9097, + "step": 10670 + }, + { + "epoch": 0.801969036524876, + "grad_norm": 1.8311152333739915, + "learning_rate": 3.9749965688902696e-07, + "loss": 0.959, + "step": 10671 + }, + { + "epoch": 0.802044190590711, + "grad_norm": 2.3223786579211887, + "learning_rate": 3.9720841233616875e-07, + "loss": 0.9117, + "step": 10672 + }, + { + "epoch": 0.8021193446565459, + "grad_norm": 1.4641240093781833, + "learning_rate": 3.969172627548494e-07, + "loss": 0.9299, + "step": 10673 + }, + { + "epoch": 0.8021944987223809, + "grad_norm": 2.700810847562999, + "learning_rate": 3.966262081623208e-07, + "loss": 1.0417, + "step": 10674 + }, + { + "epoch": 0.8022696527882158, + "grad_norm": 1.6177426919360884, + "learning_rate": 3.963352485758291e-07, + "loss": 0.9615, + "step": 10675 + }, + { + "epoch": 0.8023448068540509, + "grad_norm": 1.9399675223001405, + "learning_rate": 3.960443840126144e-07, + "loss": 0.828, + "step": 10676 + }, + { + "epoch": 0.8024199609198858, + "grad_norm": 1.4333132435625344, + "learning_rate": 3.957536144899123e-07, + "loss": 1.054, + "step": 10677 + }, + { + "epoch": 0.8024951149857207, + "grad_norm": 2.159675530858765, + "learning_rate": 3.954629400249516e-07, + "loss": 0.9502, + "step": 10678 + }, + { + "epoch": 0.8025702690515557, + "grad_norm": 2.0653588104712486, + "learning_rate": 3.9517236063495596e-07, + "loss": 0.955, + "step": 10679 + }, + { + "epoch": 0.8026454231173906, + "grad_norm": 0.6811566533871841, + "learning_rate": 3.9488187633714333e-07, + "loss": 0.8289, + "step": 10680 + }, + { + "epoch": 0.8027205771832256, + "grad_norm": 1.8487955012728512, + "learning_rate": 3.9459148714872526e-07, + "loss": 0.9855, + "step": 10681 + }, + { + "epoch": 0.8027957312490606, + "grad_norm": 2.110036110591591, + "learning_rate": 3.943011930869098e-07, + "loss": 0.9759, + "step": 10682 + }, + { + "epoch": 0.8028708853148956, + "grad_norm": 1.7283714794681202, + "learning_rate": 3.940109941688969e-07, + "loss": 1.0512, + "step": 10683 + }, + { + "epoch": 0.8029460393807305, + "grad_norm": 1.3054553302323975, + "learning_rate": 3.9372089041188275e-07, + "loss": 0.8706, + "step": 10684 + }, + { + "epoch": 0.8030211934465654, + "grad_norm": 6.896086105216251, + "learning_rate": 3.934308818330565e-07, + "loss": 0.962, + "step": 10685 + }, + { + "epoch": 0.8030963475124004, + "grad_norm": 1.4047237166678248, + "learning_rate": 3.9314096844960186e-07, + "loss": 0.988, + "step": 10686 + }, + { + "epoch": 0.8031715015782354, + "grad_norm": 0.7473783767912293, + "learning_rate": 3.9285115027869863e-07, + "loss": 0.8297, + "step": 10687 + }, + { + "epoch": 0.8032466556440704, + "grad_norm": 1.6208969548282168, + "learning_rate": 3.9256142733751886e-07, + "loss": 0.9907, + "step": 10688 + }, + { + "epoch": 0.8033218097099053, + "grad_norm": 1.6815827079484567, + "learning_rate": 3.9227179964322985e-07, + "loss": 0.9529, + "step": 10689 + }, + { + "epoch": 0.8033969637757402, + "grad_norm": 19.067490221568473, + "learning_rate": 3.919822672129931e-07, + "loss": 0.9907, + "step": 10690 + }, + { + "epoch": 0.8034721178415752, + "grad_norm": 1.715239876634275, + "learning_rate": 3.9169283006396394e-07, + "loss": 0.965, + "step": 10691 + }, + { + "epoch": 0.8035472719074102, + "grad_norm": 1.8622547577044268, + "learning_rate": 3.914034882132937e-07, + "loss": 0.9431, + "step": 10692 + }, + { + "epoch": 0.8036224259732452, + "grad_norm": 1.891385201451673, + "learning_rate": 3.911142416781261e-07, + "loss": 0.9008, + "step": 10693 + }, + { + "epoch": 0.8036975800390801, + "grad_norm": 1.8789350543124088, + "learning_rate": 3.90825090475601e-07, + "loss": 0.8942, + "step": 10694 + }, + { + "epoch": 0.8037727341049151, + "grad_norm": 1.8121023070651259, + "learning_rate": 3.9053603462285124e-07, + "loss": 1.0025, + "step": 10695 + }, + { + "epoch": 0.80384788817075, + "grad_norm": 1.8471791334871086, + "learning_rate": 3.902470741370045e-07, + "loss": 0.9185, + "step": 10696 + }, + { + "epoch": 0.803923042236585, + "grad_norm": 1.5601371446871406, + "learning_rate": 3.899582090351827e-07, + "loss": 1.0041, + "step": 10697 + }, + { + "epoch": 0.80399819630242, + "grad_norm": 1.8957921187842968, + "learning_rate": 3.8966943933450167e-07, + "loss": 0.94, + "step": 10698 + }, + { + "epoch": 0.8040733503682549, + "grad_norm": 4.8944042094945495, + "learning_rate": 3.893807650520735e-07, + "loss": 0.9849, + "step": 10699 + }, + { + "epoch": 0.8041485044340899, + "grad_norm": 1.8962009897857746, + "learning_rate": 3.890921862050023e-07, + "loss": 0.9657, + "step": 10700 + }, + { + "epoch": 0.8042236584999248, + "grad_norm": 2.784700460710952, + "learning_rate": 3.888037028103877e-07, + "loss": 1.036, + "step": 10701 + }, + { + "epoch": 0.8042988125657599, + "grad_norm": 2.295185220096531, + "learning_rate": 3.8851531488532353e-07, + "loss": 0.891, + "step": 10702 + }, + { + "epoch": 0.8043739666315948, + "grad_norm": 1.8630343419552977, + "learning_rate": 3.882270224468969e-07, + "loss": 1.0284, + "step": 10703 + }, + { + "epoch": 0.8044491206974297, + "grad_norm": 1.5114689110408355, + "learning_rate": 3.879388255121918e-07, + "loss": 1.0314, + "step": 10704 + }, + { + "epoch": 0.8045242747632647, + "grad_norm": 2.3338245033831386, + "learning_rate": 3.8765072409828424e-07, + "loss": 0.8478, + "step": 10705 + }, + { + "epoch": 0.8045994288290996, + "grad_norm": 2.3966544914024777, + "learning_rate": 3.873627182222454e-07, + "loss": 0.9476, + "step": 10706 + }, + { + "epoch": 0.8046745828949347, + "grad_norm": 1.677343605016623, + "learning_rate": 3.870748079011408e-07, + "loss": 0.9639, + "step": 10707 + }, + { + "epoch": 0.8047497369607696, + "grad_norm": 2.060942900146628, + "learning_rate": 3.867869931520296e-07, + "loss": 0.9603, + "step": 10708 + }, + { + "epoch": 0.8048248910266046, + "grad_norm": 0.7864630911643171, + "learning_rate": 3.864992739919668e-07, + "loss": 0.8907, + "step": 10709 + }, + { + "epoch": 0.8049000450924395, + "grad_norm": 1.5344496792335602, + "learning_rate": 3.8621165043800065e-07, + "loss": 1.0789, + "step": 10710 + }, + { + "epoch": 0.8049751991582744, + "grad_norm": 1.5349381075878488, + "learning_rate": 3.8592412250717366e-07, + "loss": 0.9813, + "step": 10711 + }, + { + "epoch": 0.8050503532241094, + "grad_norm": 1.4289306972241453, + "learning_rate": 3.8563669021652334e-07, + "loss": 1.0039, + "step": 10712 + }, + { + "epoch": 0.8051255072899444, + "grad_norm": 4.288422305719481, + "learning_rate": 3.853493535830803e-07, + "loss": 0.9651, + "step": 10713 + }, + { + "epoch": 0.8052006613557794, + "grad_norm": 1.8372523367962035, + "learning_rate": 3.8506211262387155e-07, + "loss": 0.9991, + "step": 10714 + }, + { + "epoch": 0.8052758154216143, + "grad_norm": 1.9003507917724456, + "learning_rate": 3.84774967355916e-07, + "loss": 0.9406, + "step": 10715 + }, + { + "epoch": 0.8053509694874492, + "grad_norm": 1.6377773457717384, + "learning_rate": 3.844879177962295e-07, + "loss": 0.9384, + "step": 10716 + }, + { + "epoch": 0.8054261235532842, + "grad_norm": 1.8251539687958935, + "learning_rate": 3.842009639618198e-07, + "loss": 0.9735, + "step": 10717 + }, + { + "epoch": 0.8055012776191192, + "grad_norm": 1.79815714330388, + "learning_rate": 3.839141058696904e-07, + "loss": 0.976, + "step": 10718 + }, + { + "epoch": 0.8055764316849542, + "grad_norm": 2.333225270448332, + "learning_rate": 3.836273435368387e-07, + "loss": 0.9265, + "step": 10719 + }, + { + "epoch": 0.8056515857507891, + "grad_norm": 4.349506104591858, + "learning_rate": 3.8334067698025583e-07, + "loss": 0.8118, + "step": 10720 + }, + { + "epoch": 0.8057267398166241, + "grad_norm": 2.1692545059401738, + "learning_rate": 3.83054106216929e-07, + "loss": 1.0566, + "step": 10721 + }, + { + "epoch": 0.805801893882459, + "grad_norm": 1.443665344266448, + "learning_rate": 3.827676312638379e-07, + "loss": 0.9929, + "step": 10722 + }, + { + "epoch": 0.805877047948294, + "grad_norm": 1.8468253785775552, + "learning_rate": 3.824812521379577e-07, + "loss": 0.9149, + "step": 10723 + }, + { + "epoch": 0.805952202014129, + "grad_norm": 2.598659786183288, + "learning_rate": 3.821949688562571e-07, + "loss": 0.9439, + "step": 10724 + }, + { + "epoch": 0.8060273560799639, + "grad_norm": 3.485605202314773, + "learning_rate": 3.8190878143569896e-07, + "loss": 0.9766, + "step": 10725 + }, + { + "epoch": 0.8061025101457989, + "grad_norm": 1.6803527933900984, + "learning_rate": 3.816226898932422e-07, + "loss": 0.9873, + "step": 10726 + }, + { + "epoch": 0.8061776642116338, + "grad_norm": 1.335272193905657, + "learning_rate": 3.8133669424583847e-07, + "loss": 1.0015, + "step": 10727 + }, + { + "epoch": 0.8062528182774689, + "grad_norm": 1.583690310273947, + "learning_rate": 3.8105079451043355e-07, + "loss": 1.0107, + "step": 10728 + }, + { + "epoch": 0.8063279723433038, + "grad_norm": 1.555444964193961, + "learning_rate": 3.807649907039685e-07, + "loss": 0.8882, + "step": 10729 + }, + { + "epoch": 0.8064031264091387, + "grad_norm": 1.4694033853571067, + "learning_rate": 3.804792828433778e-07, + "loss": 1.0127, + "step": 10730 + }, + { + "epoch": 0.8064782804749737, + "grad_norm": 1.957389935727905, + "learning_rate": 3.8019367094559173e-07, + "loss": 0.9246, + "step": 10731 + }, + { + "epoch": 0.8065534345408086, + "grad_norm": 2.202753455033243, + "learning_rate": 3.7990815502753317e-07, + "loss": 0.9475, + "step": 10732 + }, + { + "epoch": 0.8066285886066437, + "grad_norm": 1.6950215032783293, + "learning_rate": 3.796227351061201e-07, + "loss": 0.868, + "step": 10733 + }, + { + "epoch": 0.8067037426724786, + "grad_norm": 2.3738918730898573, + "learning_rate": 3.79337411198265e-07, + "loss": 0.8687, + "step": 10734 + }, + { + "epoch": 0.8067788967383135, + "grad_norm": 1.9881155848107788, + "learning_rate": 3.790521833208735e-07, + "loss": 0.9945, + "step": 10735 + }, + { + "epoch": 0.8068540508041485, + "grad_norm": 0.6409543749123641, + "learning_rate": 3.7876705149084786e-07, + "loss": 0.8391, + "step": 10736 + }, + { + "epoch": 0.8069292048699834, + "grad_norm": 1.7277281955761143, + "learning_rate": 3.784820157250819e-07, + "loss": 0.9129, + "step": 10737 + }, + { + "epoch": 0.8070043589358185, + "grad_norm": 0.9866109914525824, + "learning_rate": 3.781970760404665e-07, + "loss": 0.9133, + "step": 10738 + }, + { + "epoch": 0.8070795130016534, + "grad_norm": 1.9946280695377683, + "learning_rate": 3.779122324538844e-07, + "loss": 1.0088, + "step": 10739 + }, + { + "epoch": 0.8071546670674884, + "grad_norm": 2.527846603248394, + "learning_rate": 3.7762748498221385e-07, + "loss": 0.8321, + "step": 10740 + }, + { + "epoch": 0.8072298211333233, + "grad_norm": 1.926944280882869, + "learning_rate": 3.7734283364232745e-07, + "loss": 0.8606, + "step": 10741 + }, + { + "epoch": 0.8073049751991582, + "grad_norm": 1.4556871202250625, + "learning_rate": 3.7705827845109117e-07, + "loss": 0.9616, + "step": 10742 + }, + { + "epoch": 0.8073801292649933, + "grad_norm": 1.7850176801260056, + "learning_rate": 3.767738194253669e-07, + "loss": 0.9497, + "step": 10743 + }, + { + "epoch": 0.8074552833308282, + "grad_norm": 2.2359893942088376, + "learning_rate": 3.7648945658200983e-07, + "loss": 1.0046, + "step": 10744 + }, + { + "epoch": 0.8075304373966632, + "grad_norm": 2.1146957542965246, + "learning_rate": 3.762051899378691e-07, + "loss": 0.9224, + "step": 10745 + }, + { + "epoch": 0.8076055914624981, + "grad_norm": 1.6309586363411583, + "learning_rate": 3.7592101950978883e-07, + "loss": 0.9211, + "step": 10746 + }, + { + "epoch": 0.8076807455283331, + "grad_norm": 3.947556142860744, + "learning_rate": 3.7563694531460686e-07, + "loss": 0.9782, + "step": 10747 + }, + { + "epoch": 0.807755899594168, + "grad_norm": 2.1155952575486188, + "learning_rate": 3.7535296736915623e-07, + "loss": 0.8742, + "step": 10748 + }, + { + "epoch": 0.807831053660003, + "grad_norm": 2.1910319561144798, + "learning_rate": 3.750690856902636e-07, + "loss": 1.0116, + "step": 10749 + }, + { + "epoch": 0.807906207725838, + "grad_norm": 7.075069888143801, + "learning_rate": 3.7478530029474987e-07, + "loss": 0.9695, + "step": 10750 + }, + { + "epoch": 0.8079813617916729, + "grad_norm": 1.8030191611923683, + "learning_rate": 3.7450161119943056e-07, + "loss": 0.9537, + "step": 10751 + }, + { + "epoch": 0.8080565158575079, + "grad_norm": 1.7536517597940893, + "learning_rate": 3.7421801842111454e-07, + "loss": 0.9439, + "step": 10752 + }, + { + "epoch": 0.8081316699233428, + "grad_norm": 3.78957969045997, + "learning_rate": 3.7393452197660723e-07, + "loss": 0.9758, + "step": 10753 + }, + { + "epoch": 0.8082068239891779, + "grad_norm": 1.6650981993255582, + "learning_rate": 3.7365112188270585e-07, + "loss": 0.9448, + "step": 10754 + }, + { + "epoch": 0.8082819780550128, + "grad_norm": 1.8016706740234487, + "learning_rate": 3.7336781815620345e-07, + "loss": 1.0483, + "step": 10755 + }, + { + "epoch": 0.8083571321208477, + "grad_norm": 2.0818942885591842, + "learning_rate": 3.730846108138863e-07, + "loss": 1.1174, + "step": 10756 + }, + { + "epoch": 0.8084322861866827, + "grad_norm": 1.2681618215029098, + "learning_rate": 3.728014998725357e-07, + "loss": 1.0112, + "step": 10757 + }, + { + "epoch": 0.8085074402525176, + "grad_norm": 1.5299944498547384, + "learning_rate": 3.725184853489274e-07, + "loss": 0.9927, + "step": 10758 + }, + { + "epoch": 0.8085825943183527, + "grad_norm": 1.6085871277163801, + "learning_rate": 3.722355672598305e-07, + "loss": 0.9942, + "step": 10759 + }, + { + "epoch": 0.8086577483841876, + "grad_norm": 2.5537484685937115, + "learning_rate": 3.7195274562200996e-07, + "loss": 1.0242, + "step": 10760 + }, + { + "epoch": 0.8087329024500225, + "grad_norm": 1.8261144292435254, + "learning_rate": 3.716700204522234e-07, + "loss": 1.0287, + "step": 10761 + }, + { + "epoch": 0.8088080565158575, + "grad_norm": 2.9384189548087747, + "learning_rate": 3.7138739176722323e-07, + "loss": 0.9076, + "step": 10762 + }, + { + "epoch": 0.8088832105816924, + "grad_norm": 2.3542774028108893, + "learning_rate": 3.711048595837567e-07, + "loss": 1.0285, + "step": 10763 + }, + { + "epoch": 0.8089583646475275, + "grad_norm": 2.188263024360129, + "learning_rate": 3.70822423918564e-07, + "loss": 1.0254, + "step": 10764 + }, + { + "epoch": 0.8090335187133624, + "grad_norm": 1.7678288571735714, + "learning_rate": 3.7054008478838197e-07, + "loss": 0.9437, + "step": 10765 + }, + { + "epoch": 0.8091086727791974, + "grad_norm": 1.7659838826533791, + "learning_rate": 3.702578422099394e-07, + "loss": 0.9175, + "step": 10766 + }, + { + "epoch": 0.8091838268450323, + "grad_norm": 2.07316582746213, + "learning_rate": 3.6997569619996027e-07, + "loss": 0.9235, + "step": 10767 + }, + { + "epoch": 0.8092589809108672, + "grad_norm": 3.665577004094883, + "learning_rate": 3.69693646775163e-07, + "loss": 0.9975, + "step": 10768 + }, + { + "epoch": 0.8093341349767023, + "grad_norm": 2.0352793709383814, + "learning_rate": 3.6941169395225956e-07, + "loss": 1.0082, + "step": 10769 + }, + { + "epoch": 0.8094092890425372, + "grad_norm": 2.306375486686762, + "learning_rate": 3.691298377479577e-07, + "loss": 0.9706, + "step": 10770 + }, + { + "epoch": 0.8094844431083722, + "grad_norm": 1.474378959388478, + "learning_rate": 3.6884807817895804e-07, + "loss": 0.992, + "step": 10771 + }, + { + "epoch": 0.8095595971742071, + "grad_norm": 1.8227407606720256, + "learning_rate": 3.685664152619556e-07, + "loss": 0.9296, + "step": 10772 + }, + { + "epoch": 0.8096347512400421, + "grad_norm": 1.5391167955454843, + "learning_rate": 3.6828484901364054e-07, + "loss": 0.9163, + "step": 10773 + }, + { + "epoch": 0.809709905305877, + "grad_norm": 2.46429853258294, + "learning_rate": 3.680033794506958e-07, + "loss": 0.9215, + "step": 10774 + }, + { + "epoch": 0.809785059371712, + "grad_norm": 1.616599081861808, + "learning_rate": 3.6772200658980057e-07, + "loss": 0.9295, + "step": 10775 + }, + { + "epoch": 0.809860213437547, + "grad_norm": 1.6208620939996383, + "learning_rate": 3.67440730447627e-07, + "loss": 0.9794, + "step": 10776 + }, + { + "epoch": 0.8099353675033819, + "grad_norm": 2.4861795841024756, + "learning_rate": 3.671595510408416e-07, + "loss": 1.1114, + "step": 10777 + }, + { + "epoch": 0.8100105215692169, + "grad_norm": 2.0939757951286633, + "learning_rate": 3.6687846838610527e-07, + "loss": 0.9956, + "step": 10778 + }, + { + "epoch": 0.8100856756350518, + "grad_norm": 1.490003175412065, + "learning_rate": 3.6659748250007283e-07, + "loss": 0.9206, + "step": 10779 + }, + { + "epoch": 0.8101608297008868, + "grad_norm": 1.7312948007674218, + "learning_rate": 3.663165933993948e-07, + "loss": 0.9721, + "step": 10780 + }, + { + "epoch": 0.8102359837667218, + "grad_norm": 2.2375816330323333, + "learning_rate": 3.660358011007141e-07, + "loss": 1.041, + "step": 10781 + }, + { + "epoch": 0.8103111378325567, + "grad_norm": 1.4695199612108352, + "learning_rate": 3.6575510562066937e-07, + "loss": 0.9228, + "step": 10782 + }, + { + "epoch": 0.8103862918983917, + "grad_norm": 2.0168238187146796, + "learning_rate": 3.6547450697589243e-07, + "loss": 0.9495, + "step": 10783 + }, + { + "epoch": 0.8104614459642266, + "grad_norm": 3.520608311672332, + "learning_rate": 3.6519400518301023e-07, + "loss": 0.9436, + "step": 10784 + }, + { + "epoch": 0.8105366000300617, + "grad_norm": 1.507304684209951, + "learning_rate": 3.6491360025864324e-07, + "loss": 0.8566, + "step": 10785 + }, + { + "epoch": 0.8106117540958966, + "grad_norm": 1.490020776642205, + "learning_rate": 3.6463329221940597e-07, + "loss": 1.0333, + "step": 10786 + }, + { + "epoch": 0.8106869081617315, + "grad_norm": 1.6034499049222632, + "learning_rate": 3.643530810819091e-07, + "loss": 0.9546, + "step": 10787 + }, + { + "epoch": 0.8107620622275665, + "grad_norm": 1.5557253646506137, + "learning_rate": 3.640729668627553e-07, + "loss": 0.9242, + "step": 10788 + }, + { + "epoch": 0.8108372162934014, + "grad_norm": 1.7301080814402798, + "learning_rate": 3.6379294957854257e-07, + "loss": 0.8254, + "step": 10789 + }, + { + "epoch": 0.8109123703592365, + "grad_norm": 2.3979833713492633, + "learning_rate": 3.6351302924586326e-07, + "loss": 1.0442, + "step": 10790 + }, + { + "epoch": 0.8109875244250714, + "grad_norm": 1.418103088236524, + "learning_rate": 3.6323320588130277e-07, + "loss": 1.0132, + "step": 10791 + }, + { + "epoch": 0.8110626784909064, + "grad_norm": 2.229321186259972, + "learning_rate": 3.6295347950144305e-07, + "loss": 1.0123, + "step": 10792 + }, + { + "epoch": 0.8111378325567413, + "grad_norm": 1.8584027994398362, + "learning_rate": 3.6267385012285836e-07, + "loss": 0.9229, + "step": 10793 + }, + { + "epoch": 0.8112129866225762, + "grad_norm": 1.820641811200257, + "learning_rate": 3.6239431776211757e-07, + "loss": 0.9363, + "step": 10794 + }, + { + "epoch": 0.8112881406884113, + "grad_norm": 0.7507324615726616, + "learning_rate": 3.6211488243578445e-07, + "loss": 0.8978, + "step": 10795 + }, + { + "epoch": 0.8113632947542462, + "grad_norm": 1.9444923357136927, + "learning_rate": 3.6183554416041597e-07, + "loss": 0.9573, + "step": 10796 + }, + { + "epoch": 0.8114384488200812, + "grad_norm": 1.7085475864810982, + "learning_rate": 3.615563029525648e-07, + "loss": 1.0784, + "step": 10797 + }, + { + "epoch": 0.8115136028859161, + "grad_norm": 1.6175751889424874, + "learning_rate": 3.612771588287764e-07, + "loss": 0.9917, + "step": 10798 + }, + { + "epoch": 0.8115887569517511, + "grad_norm": 1.930242111259104, + "learning_rate": 3.609981118055923e-07, + "loss": 0.9966, + "step": 10799 + }, + { + "epoch": 0.8116639110175861, + "grad_norm": 1.7986788775795637, + "learning_rate": 3.6071916189954575e-07, + "loss": 0.9168, + "step": 10800 + }, + { + "epoch": 0.811739065083421, + "grad_norm": 1.9944762602239832, + "learning_rate": 3.604403091271655e-07, + "loss": 0.9377, + "step": 10801 + }, + { + "epoch": 0.811814219149256, + "grad_norm": 1.5704458758899433, + "learning_rate": 3.601615535049758e-07, + "loss": 0.9778, + "step": 10802 + }, + { + "epoch": 0.8118893732150909, + "grad_norm": 1.4552751446157028, + "learning_rate": 3.5988289504949297e-07, + "loss": 0.9915, + "step": 10803 + }, + { + "epoch": 0.8119645272809259, + "grad_norm": 1.539615280102103, + "learning_rate": 3.5960433377722945e-07, + "loss": 0.9667, + "step": 10804 + }, + { + "epoch": 0.8120396813467609, + "grad_norm": 1.6062306176585857, + "learning_rate": 3.5932586970469057e-07, + "loss": 0.8594, + "step": 10805 + }, + { + "epoch": 0.8121148354125958, + "grad_norm": 2.168606501072553, + "learning_rate": 3.5904750284837657e-07, + "loss": 0.9549, + "step": 10806 + }, + { + "epoch": 0.8121899894784308, + "grad_norm": 0.6369045431156233, + "learning_rate": 3.587692332247818e-07, + "loss": 0.8417, + "step": 10807 + }, + { + "epoch": 0.8122651435442657, + "grad_norm": 2.305652170825168, + "learning_rate": 3.5849106085039393e-07, + "loss": 1.0044, + "step": 10808 + }, + { + "epoch": 0.8123402976101007, + "grad_norm": 1.49525184946818, + "learning_rate": 3.582129857416971e-07, + "loss": 0.8756, + "step": 10809 + }, + { + "epoch": 0.8124154516759357, + "grad_norm": 1.6920107846964731, + "learning_rate": 3.5793500791516773e-07, + "loss": 0.9688, + "step": 10810 + }, + { + "epoch": 0.8124906057417707, + "grad_norm": 1.609409368944304, + "learning_rate": 3.576571273872768e-07, + "loss": 1.0158, + "step": 10811 + }, + { + "epoch": 0.8125657598076056, + "grad_norm": 7.057048212523204, + "learning_rate": 3.573793441744901e-07, + "loss": 0.8865, + "step": 10812 + }, + { + "epoch": 0.8126409138734405, + "grad_norm": 1.750010422266851, + "learning_rate": 3.5710165829326686e-07, + "loss": 0.9333, + "step": 10813 + }, + { + "epoch": 0.8127160679392755, + "grad_norm": 1.6488135754038893, + "learning_rate": 3.5682406976006196e-07, + "loss": 0.9557, + "step": 10814 + }, + { + "epoch": 0.8127912220051104, + "grad_norm": 1.4170983535671973, + "learning_rate": 3.565465785913231e-07, + "loss": 0.9778, + "step": 10815 + }, + { + "epoch": 0.8128663760709455, + "grad_norm": 1.5617144326498134, + "learning_rate": 3.5626918480349244e-07, + "loss": 0.9112, + "step": 10816 + }, + { + "epoch": 0.8129415301367804, + "grad_norm": 0.8751555032153111, + "learning_rate": 3.559918884130071e-07, + "loss": 0.8691, + "step": 10817 + }, + { + "epoch": 0.8130166842026154, + "grad_norm": 1.7597382571921552, + "learning_rate": 3.55714689436297e-07, + "loss": 0.8249, + "step": 10818 + }, + { + "epoch": 0.8130918382684503, + "grad_norm": 1.7911977893703257, + "learning_rate": 3.554375878897886e-07, + "loss": 0.9557, + "step": 10819 + }, + { + "epoch": 0.8131669923342852, + "grad_norm": 1.5211770527134256, + "learning_rate": 3.551605837898999e-07, + "loss": 0.9506, + "step": 10820 + }, + { + "epoch": 0.8132421464001203, + "grad_norm": 1.819752619352996, + "learning_rate": 3.5488367715304637e-07, + "loss": 1.0197, + "step": 10821 + }, + { + "epoch": 0.8133173004659552, + "grad_norm": 1.5617580941278886, + "learning_rate": 3.5460686799563375e-07, + "loss": 0.9347, + "step": 10822 + }, + { + "epoch": 0.8133924545317902, + "grad_norm": 1.5485871212882563, + "learning_rate": 3.543301563340646e-07, + "loss": 0.9891, + "step": 10823 + }, + { + "epoch": 0.8134676085976251, + "grad_norm": 1.8781711464796678, + "learning_rate": 3.540535421847357e-07, + "loss": 1.0294, + "step": 10824 + }, + { + "epoch": 0.81354276266346, + "grad_norm": 1.7467844576223452, + "learning_rate": 3.5377702556403664e-07, + "loss": 1.0219, + "step": 10825 + }, + { + "epoch": 0.8136179167292951, + "grad_norm": 2.305793729478071, + "learning_rate": 3.535006064883532e-07, + "loss": 0.936, + "step": 10826 + }, + { + "epoch": 0.81369307079513, + "grad_norm": 1.6964975601165253, + "learning_rate": 3.5322428497406387e-07, + "loss": 1.0134, + "step": 10827 + }, + { + "epoch": 0.813768224860965, + "grad_norm": 2.13806790398014, + "learning_rate": 3.5294806103754124e-07, + "loss": 0.8911, + "step": 10828 + }, + { + "epoch": 0.8138433789267999, + "grad_norm": 3.6463653757531227, + "learning_rate": 3.5267193469515324e-07, + "loss": 1.1312, + "step": 10829 + }, + { + "epoch": 0.813918532992635, + "grad_norm": 1.9082672202084077, + "learning_rate": 3.523959059632606e-07, + "loss": 1.0375, + "step": 10830 + }, + { + "epoch": 0.8139936870584699, + "grad_norm": 1.275926556228468, + "learning_rate": 3.5211997485822e-07, + "loss": 0.9872, + "step": 10831 + }, + { + "epoch": 0.8140688411243048, + "grad_norm": 2.094211925964235, + "learning_rate": 3.518441413963811e-07, + "loss": 0.9635, + "step": 10832 + }, + { + "epoch": 0.8141439951901398, + "grad_norm": 1.5612378172326589, + "learning_rate": 3.5156840559408816e-07, + "loss": 0.8933, + "step": 10833 + }, + { + "epoch": 0.8142191492559747, + "grad_norm": 1.2776288755473975, + "learning_rate": 3.5129276746767886e-07, + "loss": 1.0377, + "step": 10834 + }, + { + "epoch": 0.8142943033218097, + "grad_norm": 2.953376880379237, + "learning_rate": 3.510172270334875e-07, + "loss": 0.8211, + "step": 10835 + }, + { + "epoch": 0.8143694573876447, + "grad_norm": 2.0033231782999636, + "learning_rate": 3.507417843078386e-07, + "loss": 1.0425, + "step": 10836 + }, + { + "epoch": 0.8144446114534797, + "grad_norm": 2.056453859375047, + "learning_rate": 3.504664393070551e-07, + "loss": 0.915, + "step": 10837 + }, + { + "epoch": 0.8145197655193146, + "grad_norm": 2.557645805580393, + "learning_rate": 3.5019119204745097e-07, + "loss": 0.9386, + "step": 10838 + }, + { + "epoch": 0.8145949195851495, + "grad_norm": 1.286763771441465, + "learning_rate": 3.499160425453371e-07, + "loss": 0.9227, + "step": 10839 + }, + { + "epoch": 0.8146700736509845, + "grad_norm": 1.4814940754478765, + "learning_rate": 3.496409908170157e-07, + "loss": 0.9925, + "step": 10840 + }, + { + "epoch": 0.8147452277168195, + "grad_norm": 1.7025942894232997, + "learning_rate": 3.4936603687878496e-07, + "loss": 0.9374, + "step": 10841 + }, + { + "epoch": 0.8148203817826545, + "grad_norm": 1.9622907490547072, + "learning_rate": 3.490911807469383e-07, + "loss": 0.9441, + "step": 10842 + }, + { + "epoch": 0.8148955358484894, + "grad_norm": 1.627362954131241, + "learning_rate": 3.488164224377599e-07, + "loss": 0.9746, + "step": 10843 + }, + { + "epoch": 0.8149706899143244, + "grad_norm": 2.0792178385316817, + "learning_rate": 3.485417619675317e-07, + "loss": 0.9581, + "step": 10844 + }, + { + "epoch": 0.8150458439801593, + "grad_norm": 1.3609482879736658, + "learning_rate": 3.482671993525286e-07, + "loss": 0.9865, + "step": 10845 + }, + { + "epoch": 0.8151209980459942, + "grad_norm": 9.473046464256619, + "learning_rate": 3.479927346090179e-07, + "loss": 0.8752, + "step": 10846 + }, + { + "epoch": 0.8151961521118293, + "grad_norm": 1.4725336079413345, + "learning_rate": 3.4771836775326333e-07, + "loss": 0.91, + "step": 10847 + }, + { + "epoch": 0.8152713061776642, + "grad_norm": 1.5056635114495847, + "learning_rate": 3.474440988015233e-07, + "loss": 0.9874, + "step": 10848 + }, + { + "epoch": 0.8153464602434992, + "grad_norm": 1.7057799897132908, + "learning_rate": 3.471699277700484e-07, + "loss": 0.9955, + "step": 10849 + }, + { + "epoch": 0.8154216143093341, + "grad_norm": 1.8999893062692164, + "learning_rate": 3.468958546750844e-07, + "loss": 0.9715, + "step": 10850 + }, + { + "epoch": 0.815496768375169, + "grad_norm": 2.291623918539208, + "learning_rate": 3.466218795328706e-07, + "loss": 0.9716, + "step": 10851 + }, + { + "epoch": 0.8155719224410041, + "grad_norm": 1.805408944592431, + "learning_rate": 3.4634800235964255e-07, + "loss": 0.8739, + "step": 10852 + }, + { + "epoch": 0.815647076506839, + "grad_norm": 1.5913891229096473, + "learning_rate": 3.460742231716267e-07, + "loss": 0.9904, + "step": 10853 + }, + { + "epoch": 0.815722230572674, + "grad_norm": 1.4759738293630411, + "learning_rate": 3.4580054198504716e-07, + "loss": 0.9339, + "step": 10854 + }, + { + "epoch": 0.8157973846385089, + "grad_norm": 1.8299443899366505, + "learning_rate": 3.455269588161196e-07, + "loss": 0.972, + "step": 10855 + }, + { + "epoch": 0.815872538704344, + "grad_norm": 1.292202817740746, + "learning_rate": 3.4525347368105504e-07, + "loss": 1.0007, + "step": 10856 + }, + { + "epoch": 0.8159476927701789, + "grad_norm": 1.8946636350331403, + "learning_rate": 3.4498008659605836e-07, + "loss": 0.949, + "step": 10857 + }, + { + "epoch": 0.8160228468360138, + "grad_norm": 1.7493095398240373, + "learning_rate": 3.4470679757732945e-07, + "loss": 1.0295, + "step": 10858 + }, + { + "epoch": 0.8160980009018488, + "grad_norm": 1.966543444582535, + "learning_rate": 3.4443360664106135e-07, + "loss": 0.8542, + "step": 10859 + }, + { + "epoch": 0.8161731549676837, + "grad_norm": 1.8254589627202813, + "learning_rate": 3.441605138034416e-07, + "loss": 1.0274, + "step": 10860 + }, + { + "epoch": 0.8162483090335187, + "grad_norm": 1.96596445084852, + "learning_rate": 3.438875190806516e-07, + "loss": 0.9669, + "step": 10861 + }, + { + "epoch": 0.8163234630993537, + "grad_norm": 2.0360172376146557, + "learning_rate": 3.4361462248886875e-07, + "loss": 0.7731, + "step": 10862 + }, + { + "epoch": 0.8163986171651887, + "grad_norm": 1.763509309865779, + "learning_rate": 3.433418240442611e-07, + "loss": 1.0137, + "step": 10863 + }, + { + "epoch": 0.8164737712310236, + "grad_norm": 2.4415156225501096, + "learning_rate": 3.4306912376299437e-07, + "loss": 1.0172, + "step": 10864 + }, + { + "epoch": 0.8165489252968585, + "grad_norm": 2.898636063100394, + "learning_rate": 3.4279652166122717e-07, + "loss": 0.9368, + "step": 10865 + }, + { + "epoch": 0.8166240793626935, + "grad_norm": 1.7323220965048094, + "learning_rate": 3.4252401775511255e-07, + "loss": 0.9688, + "step": 10866 + }, + { + "epoch": 0.8166992334285285, + "grad_norm": 1.5680236076942524, + "learning_rate": 3.422516120607957e-07, + "loss": 0.9207, + "step": 10867 + }, + { + "epoch": 0.8167743874943635, + "grad_norm": 2.0387839642285326, + "learning_rate": 3.4197930459441883e-07, + "loss": 0.8981, + "step": 10868 + }, + { + "epoch": 0.8168495415601984, + "grad_norm": 1.6789350376705388, + "learning_rate": 3.4170709537211815e-07, + "loss": 0.9836, + "step": 10869 + }, + { + "epoch": 0.8169246956260333, + "grad_norm": 1.7832686379839928, + "learning_rate": 3.4143498441002105e-07, + "loss": 1.0005, + "step": 10870 + }, + { + "epoch": 0.8169998496918683, + "grad_norm": 1.696589819336331, + "learning_rate": 3.4116297172425277e-07, + "loss": 0.9908, + "step": 10871 + }, + { + "epoch": 0.8170750037577033, + "grad_norm": 1.7843967549990385, + "learning_rate": 3.408910573309305e-07, + "loss": 0.9222, + "step": 10872 + }, + { + "epoch": 0.8171501578235383, + "grad_norm": 1.4189259054559369, + "learning_rate": 3.4061924124616613e-07, + "loss": 0.9599, + "step": 10873 + }, + { + "epoch": 0.8172253118893732, + "grad_norm": 1.6071206015605068, + "learning_rate": 3.4034752348606553e-07, + "loss": 0.8757, + "step": 10874 + }, + { + "epoch": 0.8173004659552082, + "grad_norm": 2.110390877699495, + "learning_rate": 3.400759040667298e-07, + "loss": 0.9808, + "step": 10875 + }, + { + "epoch": 0.8173756200210431, + "grad_norm": 1.4154374829995244, + "learning_rate": 3.398043830042532e-07, + "loss": 0.9592, + "step": 10876 + }, + { + "epoch": 0.817450774086878, + "grad_norm": 1.6781026060858508, + "learning_rate": 3.395329603147241e-07, + "loss": 1.047, + "step": 10877 + }, + { + "epoch": 0.8175259281527131, + "grad_norm": 1.9023205640190601, + "learning_rate": 3.3926163601422485e-07, + "loss": 0.8517, + "step": 10878 + }, + { + "epoch": 0.817601082218548, + "grad_norm": 1.6141026509434253, + "learning_rate": 3.3899041011883433e-07, + "loss": 1.0125, + "step": 10879 + }, + { + "epoch": 0.817676236284383, + "grad_norm": 1.4226646326730399, + "learning_rate": 3.3871928264462124e-07, + "loss": 0.9738, + "step": 10880 + }, + { + "epoch": 0.8177513903502179, + "grad_norm": 1.8289038148494707, + "learning_rate": 3.3844825360765273e-07, + "loss": 0.8845, + "step": 10881 + }, + { + "epoch": 0.817826544416053, + "grad_norm": 1.9585156930738339, + "learning_rate": 3.38177323023987e-07, + "loss": 0.9086, + "step": 10882 + }, + { + "epoch": 0.8179016984818879, + "grad_norm": 1.6048490715959987, + "learning_rate": 3.379064909096796e-07, + "loss": 0.9304, + "step": 10883 + }, + { + "epoch": 0.8179768525477228, + "grad_norm": 2.298497775057035, + "learning_rate": 3.376357572807762e-07, + "loss": 0.9385, + "step": 10884 + }, + { + "epoch": 0.8180520066135578, + "grad_norm": 2.7898422561679963, + "learning_rate": 3.373651221533198e-07, + "loss": 0.9608, + "step": 10885 + }, + { + "epoch": 0.8181271606793927, + "grad_norm": 1.5025090055782935, + "learning_rate": 3.3709458554334735e-07, + "loss": 0.9169, + "step": 10886 + }, + { + "epoch": 0.8182023147452278, + "grad_norm": 1.474593691710888, + "learning_rate": 3.368241474668876e-07, + "loss": 0.9985, + "step": 10887 + }, + { + "epoch": 0.8182774688110627, + "grad_norm": 3.035268106449139, + "learning_rate": 3.3655380793996636e-07, + "loss": 0.8906, + "step": 10888 + }, + { + "epoch": 0.8183526228768977, + "grad_norm": 1.5045342261344083, + "learning_rate": 3.3628356697860216e-07, + "loss": 0.9106, + "step": 10889 + }, + { + "epoch": 0.8184277769427326, + "grad_norm": 2.018677049387574, + "learning_rate": 3.3601342459880643e-07, + "loss": 0.8983, + "step": 10890 + }, + { + "epoch": 0.8185029310085675, + "grad_norm": 1.986895124892088, + "learning_rate": 3.3574338081658724e-07, + "loss": 0.9285, + "step": 10891 + }, + { + "epoch": 0.8185780850744026, + "grad_norm": 1.5784315812342242, + "learning_rate": 3.3547343564794605e-07, + "loss": 0.9797, + "step": 10892 + }, + { + "epoch": 0.8186532391402375, + "grad_norm": 1.4993316432954098, + "learning_rate": 3.352035891088776e-07, + "loss": 0.9166, + "step": 10893 + }, + { + "epoch": 0.8187283932060725, + "grad_norm": 2.5833786581021556, + "learning_rate": 3.3493384121537147e-07, + "loss": 0.9304, + "step": 10894 + }, + { + "epoch": 0.8188035472719074, + "grad_norm": 2.133427457918521, + "learning_rate": 3.346641919834108e-07, + "loss": 0.9628, + "step": 10895 + }, + { + "epoch": 0.8188787013377423, + "grad_norm": 1.8234585928330294, + "learning_rate": 3.3439464142897467e-07, + "loss": 0.9684, + "step": 10896 + }, + { + "epoch": 0.8189538554035773, + "grad_norm": 2.282944768250071, + "learning_rate": 3.3412518956803306e-07, + "loss": 1.0581, + "step": 10897 + }, + { + "epoch": 0.8190290094694123, + "grad_norm": 1.5702578406998313, + "learning_rate": 3.338558364165536e-07, + "loss": 0.9902, + "step": 10898 + }, + { + "epoch": 0.8191041635352473, + "grad_norm": 1.8184295214635153, + "learning_rate": 3.335865819904957e-07, + "loss": 1.0032, + "step": 10899 + }, + { + "epoch": 0.8191793176010822, + "grad_norm": 1.6779097974946218, + "learning_rate": 3.3331742630581405e-07, + "loss": 0.9983, + "step": 10900 + }, + { + "epoch": 0.8192544716669172, + "grad_norm": 3.0682210617895165, + "learning_rate": 3.330483693784567e-07, + "loss": 1.0711, + "step": 10901 + }, + { + "epoch": 0.8193296257327521, + "grad_norm": 1.8064066307503681, + "learning_rate": 3.3277941122436714e-07, + "loss": 1.0288, + "step": 10902 + }, + { + "epoch": 0.8194047797985871, + "grad_norm": 1.787452939887885, + "learning_rate": 3.325105518594815e-07, + "loss": 1.0143, + "step": 10903 + }, + { + "epoch": 0.8194799338644221, + "grad_norm": 1.4698125364181358, + "learning_rate": 3.322417912997311e-07, + "loss": 0.9568, + "step": 10904 + }, + { + "epoch": 0.819555087930257, + "grad_norm": 0.8250717883231985, + "learning_rate": 3.3197312956104016e-07, + "loss": 0.8929, + "step": 10905 + }, + { + "epoch": 0.819630241996092, + "grad_norm": 2.1827078508155227, + "learning_rate": 3.317045666593297e-07, + "loss": 0.9846, + "step": 10906 + }, + { + "epoch": 0.8197053960619269, + "grad_norm": 5.566482489883595, + "learning_rate": 3.314361026105108e-07, + "loss": 0.9236, + "step": 10907 + }, + { + "epoch": 0.819780550127762, + "grad_norm": 1.582943801479951, + "learning_rate": 3.3116773743049244e-07, + "loss": 0.8824, + "step": 10908 + }, + { + "epoch": 0.8198557041935969, + "grad_norm": 1.8043700099938171, + "learning_rate": 3.3089947113517647e-07, + "loss": 0.9881, + "step": 10909 + }, + { + "epoch": 0.8199308582594318, + "grad_norm": 1.842291044604458, + "learning_rate": 3.306313037404582e-07, + "loss": 1.02, + "step": 10910 + }, + { + "epoch": 0.8200060123252668, + "grad_norm": 5.416089579823244, + "learning_rate": 3.303632352622276e-07, + "loss": 1.0543, + "step": 10911 + }, + { + "epoch": 0.8200811663911017, + "grad_norm": 1.8963690269898654, + "learning_rate": 3.3009526571636827e-07, + "loss": 1.026, + "step": 10912 + }, + { + "epoch": 0.8201563204569368, + "grad_norm": 1.978287076411967, + "learning_rate": 3.2982739511876e-07, + "loss": 0.9972, + "step": 10913 + }, + { + "epoch": 0.8202314745227717, + "grad_norm": 0.6792123383297731, + "learning_rate": 3.295596234852731e-07, + "loss": 0.8254, + "step": 10914 + }, + { + "epoch": 0.8203066285886066, + "grad_norm": 1.650468271449174, + "learning_rate": 3.2929195083177554e-07, + "loss": 0.9038, + "step": 10915 + }, + { + "epoch": 0.8203817826544416, + "grad_norm": 1.829942109906855, + "learning_rate": 3.2902437717412743e-07, + "loss": 0.9997, + "step": 10916 + }, + { + "epoch": 0.8204569367202765, + "grad_norm": 1.796641259954481, + "learning_rate": 3.2875690252818357e-07, + "loss": 1.0082, + "step": 10917 + }, + { + "epoch": 0.8205320907861116, + "grad_norm": 1.6124830141577509, + "learning_rate": 3.2848952690979224e-07, + "loss": 0.8767, + "step": 10918 + }, + { + "epoch": 0.8206072448519465, + "grad_norm": 1.4403420424871585, + "learning_rate": 3.282222503347978e-07, + "loss": 1.0355, + "step": 10919 + }, + { + "epoch": 0.8206823989177815, + "grad_norm": 0.9275160361906133, + "learning_rate": 3.2795507281903655e-07, + "loss": 0.8414, + "step": 10920 + }, + { + "epoch": 0.8207575529836164, + "grad_norm": 1.7304963051558946, + "learning_rate": 3.2768799437833994e-07, + "loss": 0.9743, + "step": 10921 + }, + { + "epoch": 0.8208327070494513, + "grad_norm": 1.9424968396410278, + "learning_rate": 3.274210150285328e-07, + "loss": 0.9574, + "step": 10922 + }, + { + "epoch": 0.8209078611152864, + "grad_norm": 1.677371464337333, + "learning_rate": 3.271541347854363e-07, + "loss": 0.9568, + "step": 10923 + }, + { + "epoch": 0.8209830151811213, + "grad_norm": 1.5032760290458251, + "learning_rate": 3.268873536648622e-07, + "loss": 0.9815, + "step": 10924 + }, + { + "epoch": 0.8210581692469563, + "grad_norm": 0.8447979670868603, + "learning_rate": 3.2662067168261966e-07, + "loss": 0.8141, + "step": 10925 + }, + { + "epoch": 0.8211333233127912, + "grad_norm": 1.8276243257828555, + "learning_rate": 3.2635408885450956e-07, + "loss": 0.9056, + "step": 10926 + }, + { + "epoch": 0.8212084773786262, + "grad_norm": 1.646541209840718, + "learning_rate": 3.260876051963295e-07, + "loss": 0.9423, + "step": 10927 + }, + { + "epoch": 0.8212836314444611, + "grad_norm": 2.132111039281417, + "learning_rate": 3.2582122072386755e-07, + "loss": 0.9131, + "step": 10928 + }, + { + "epoch": 0.8213587855102961, + "grad_norm": 1.574240710386916, + "learning_rate": 3.2555493545290927e-07, + "loss": 0.9014, + "step": 10929 + }, + { + "epoch": 0.8214339395761311, + "grad_norm": 2.0450269203932394, + "learning_rate": 3.25288749399234e-07, + "loss": 0.9051, + "step": 10930 + }, + { + "epoch": 0.821509093641966, + "grad_norm": 2.584901077620806, + "learning_rate": 3.2502266257861213e-07, + "loss": 1.0245, + "step": 10931 + }, + { + "epoch": 0.821584247707801, + "grad_norm": 1.8882482422424403, + "learning_rate": 3.247566750068118e-07, + "loss": 1.0475, + "step": 10932 + }, + { + "epoch": 0.821659401773636, + "grad_norm": 2.7295822428960888, + "learning_rate": 3.2449078669959406e-07, + "loss": 0.9824, + "step": 10933 + }, + { + "epoch": 0.821734555839471, + "grad_norm": 1.5553077725581392, + "learning_rate": 3.242249976727123e-07, + "loss": 0.9977, + "step": 10934 + }, + { + "epoch": 0.8218097099053059, + "grad_norm": 1.5878316908218881, + "learning_rate": 3.2395930794191607e-07, + "loss": 1.0578, + "step": 10935 + }, + { + "epoch": 0.8218848639711408, + "grad_norm": 2.632643878189315, + "learning_rate": 3.236937175229495e-07, + "loss": 0.9386, + "step": 10936 + }, + { + "epoch": 0.8219600180369758, + "grad_norm": 4.985209906318749, + "learning_rate": 3.234282264315493e-07, + "loss": 0.919, + "step": 10937 + }, + { + "epoch": 0.8220351721028107, + "grad_norm": 1.441733718518196, + "learning_rate": 3.2316283468344653e-07, + "loss": 0.9863, + "step": 10938 + }, + { + "epoch": 0.8221103261686458, + "grad_norm": 2.494375577776982, + "learning_rate": 3.228975422943665e-07, + "loss": 0.9586, + "step": 10939 + }, + { + "epoch": 0.8221854802344807, + "grad_norm": 1.7818397081212503, + "learning_rate": 3.2263234928003e-07, + "loss": 1.0387, + "step": 10940 + }, + { + "epoch": 0.8222606343003156, + "grad_norm": 1.5577973360262931, + "learning_rate": 3.22367255656149e-07, + "loss": 0.9909, + "step": 10941 + }, + { + "epoch": 0.8223357883661506, + "grad_norm": 1.9717844632458388, + "learning_rate": 3.2210226143843257e-07, + "loss": 1.0058, + "step": 10942 + }, + { + "epoch": 0.8224109424319855, + "grad_norm": 1.895877446370544, + "learning_rate": 3.218373666425822e-07, + "loss": 0.915, + "step": 10943 + }, + { + "epoch": 0.8224860964978206, + "grad_norm": 1.4488863910921352, + "learning_rate": 3.2157257128429406e-07, + "loss": 0.9553, + "step": 10944 + }, + { + "epoch": 0.8225612505636555, + "grad_norm": 1.9088540972769925, + "learning_rate": 3.2130787537925776e-07, + "loss": 0.908, + "step": 10945 + }, + { + "epoch": 0.8226364046294905, + "grad_norm": 1.9109801093503092, + "learning_rate": 3.2104327894315785e-07, + "loss": 0.9181, + "step": 10946 + }, + { + "epoch": 0.8227115586953254, + "grad_norm": 1.6690260955205842, + "learning_rate": 3.2077878199167384e-07, + "loss": 0.9737, + "step": 10947 + }, + { + "epoch": 0.8227867127611603, + "grad_norm": 2.138433964007402, + "learning_rate": 3.2051438454047677e-07, + "loss": 1.0563, + "step": 10948 + }, + { + "epoch": 0.8228618668269954, + "grad_norm": 1.6486662213352794, + "learning_rate": 3.202500866052331e-07, + "loss": 1.064, + "step": 10949 + }, + { + "epoch": 0.8229370208928303, + "grad_norm": 3.06286945839238, + "learning_rate": 3.1998588820160486e-07, + "loss": 0.9313, + "step": 10950 + }, + { + "epoch": 0.8230121749586653, + "grad_norm": 3.3687407767712765, + "learning_rate": 3.1972178934524506e-07, + "loss": 1.0426, + "step": 10951 + }, + { + "epoch": 0.8230873290245002, + "grad_norm": 1.5099551142406467, + "learning_rate": 3.194577900518034e-07, + "loss": 1.0774, + "step": 10952 + }, + { + "epoch": 0.8231624830903352, + "grad_norm": 1.61605522581196, + "learning_rate": 3.1919389033692336e-07, + "loss": 0.9868, + "step": 10953 + }, + { + "epoch": 0.8232376371561702, + "grad_norm": 2.1594243663659944, + "learning_rate": 3.189300902162417e-07, + "loss": 0.8225, + "step": 10954 + }, + { + "epoch": 0.8233127912220051, + "grad_norm": 2.252106951631768, + "learning_rate": 3.186663897053892e-07, + "loss": 0.9111, + "step": 10955 + }, + { + "epoch": 0.8233879452878401, + "grad_norm": 1.667438074246031, + "learning_rate": 3.1840278881999115e-07, + "loss": 0.9693, + "step": 10956 + }, + { + "epoch": 0.823463099353675, + "grad_norm": 1.3388764028419864, + "learning_rate": 3.1813928757566786e-07, + "loss": 0.9179, + "step": 10957 + }, + { + "epoch": 0.82353825341951, + "grad_norm": 2.799911810984766, + "learning_rate": 3.1787588598803126e-07, + "loss": 0.9488, + "step": 10958 + }, + { + "epoch": 0.823613407485345, + "grad_norm": 2.391628877383318, + "learning_rate": 3.176125840726902e-07, + "loss": 0.7809, + "step": 10959 + }, + { + "epoch": 0.8236885615511799, + "grad_norm": 1.8054111235470622, + "learning_rate": 3.1734938184524576e-07, + "loss": 1.0003, + "step": 10960 + }, + { + "epoch": 0.8237637156170149, + "grad_norm": 1.5218785844012588, + "learning_rate": 3.170862793212936e-07, + "loss": 0.9717, + "step": 10961 + }, + { + "epoch": 0.8238388696828498, + "grad_norm": 2.80753209789539, + "learning_rate": 3.1682327651642336e-07, + "loss": 0.9041, + "step": 10962 + }, + { + "epoch": 0.8239140237486848, + "grad_norm": 2.125902208772414, + "learning_rate": 3.1656037344621987e-07, + "loss": 0.9986, + "step": 10963 + }, + { + "epoch": 0.8239891778145197, + "grad_norm": 1.9911156856547232, + "learning_rate": 3.1629757012626044e-07, + "loss": 0.9734, + "step": 10964 + }, + { + "epoch": 0.8240643318803548, + "grad_norm": 1.6325248209306313, + "learning_rate": 3.160348665721173e-07, + "loss": 0.9477, + "step": 10965 + }, + { + "epoch": 0.8241394859461897, + "grad_norm": 1.7748670823741182, + "learning_rate": 3.157722627993562e-07, + "loss": 0.939, + "step": 10966 + }, + { + "epoch": 0.8242146400120246, + "grad_norm": 2.151379821158434, + "learning_rate": 3.155097588235389e-07, + "loss": 0.796, + "step": 10967 + }, + { + "epoch": 0.8242897940778596, + "grad_norm": 1.5829718913404003, + "learning_rate": 3.1524735466021766e-07, + "loss": 1.0368, + "step": 10968 + }, + { + "epoch": 0.8243649481436945, + "grad_norm": 1.6763837393732772, + "learning_rate": 3.1498505032494204e-07, + "loss": 1.0261, + "step": 10969 + }, + { + "epoch": 0.8244401022095296, + "grad_norm": 1.5148819350515028, + "learning_rate": 3.1472284583325516e-07, + "loss": 0.932, + "step": 10970 + }, + { + "epoch": 0.8245152562753645, + "grad_norm": 2.660395898541249, + "learning_rate": 3.1446074120069346e-07, + "loss": 0.9415, + "step": 10971 + }, + { + "epoch": 0.8245904103411995, + "grad_norm": 1.835589891357086, + "learning_rate": 3.1419873644278606e-07, + "loss": 0.9571, + "step": 10972 + }, + { + "epoch": 0.8246655644070344, + "grad_norm": 1.4994055841784366, + "learning_rate": 3.13936831575059e-07, + "loss": 0.8688, + "step": 10973 + }, + { + "epoch": 0.8247407184728693, + "grad_norm": 0.7520883810097614, + "learning_rate": 3.1367502661303215e-07, + "loss": 0.8793, + "step": 10974 + }, + { + "epoch": 0.8248158725387044, + "grad_norm": 2.396209079192046, + "learning_rate": 3.134133215722161e-07, + "loss": 0.9236, + "step": 10975 + }, + { + "epoch": 0.8248910266045393, + "grad_norm": 1.519604835591727, + "learning_rate": 3.1315171646811964e-07, + "loss": 0.9222, + "step": 10976 + }, + { + "epoch": 0.8249661806703743, + "grad_norm": 1.3870101407649178, + "learning_rate": 3.1289021131624347e-07, + "loss": 0.9962, + "step": 10977 + }, + { + "epoch": 0.8250413347362092, + "grad_norm": 2.061147535378692, + "learning_rate": 3.1262880613208274e-07, + "loss": 0.924, + "step": 10978 + }, + { + "epoch": 0.8251164888020442, + "grad_norm": 0.7370057567387133, + "learning_rate": 3.12367500931126e-07, + "loss": 0.8345, + "step": 10979 + }, + { + "epoch": 0.8251916428678792, + "grad_norm": 3.3367245748438243, + "learning_rate": 3.121062957288576e-07, + "loss": 1.004, + "step": 10980 + }, + { + "epoch": 0.8252667969337141, + "grad_norm": 1.4830482576869493, + "learning_rate": 3.118451905407549e-07, + "loss": 0.8533, + "step": 10981 + }, + { + "epoch": 0.8253419509995491, + "grad_norm": 1.9742520666537917, + "learning_rate": 3.115841853822887e-07, + "loss": 1.0213, + "step": 10982 + }, + { + "epoch": 0.825417105065384, + "grad_norm": 1.6426440213260876, + "learning_rate": 3.1132328026892454e-07, + "loss": 0.9057, + "step": 10983 + }, + { + "epoch": 0.825492259131219, + "grad_norm": 2.2209507947518197, + "learning_rate": 3.110624752161233e-07, + "loss": 1.0076, + "step": 10984 + }, + { + "epoch": 0.825567413197054, + "grad_norm": 3.7060459383131246, + "learning_rate": 3.1080177023933685e-07, + "loss": 1.0393, + "step": 10985 + }, + { + "epoch": 0.8256425672628889, + "grad_norm": 0.735844825368319, + "learning_rate": 3.105411653540144e-07, + "loss": 0.8682, + "step": 10986 + }, + { + "epoch": 0.8257177213287239, + "grad_norm": 1.4487920168822586, + "learning_rate": 3.102806605755972e-07, + "loss": 1.0509, + "step": 10987 + }, + { + "epoch": 0.8257928753945588, + "grad_norm": 2.106505299838353, + "learning_rate": 3.100202559195213e-07, + "loss": 0.944, + "step": 10988 + }, + { + "epoch": 0.8258680294603938, + "grad_norm": 2.049911576201608, + "learning_rate": 3.0975995140121613e-07, + "loss": 0.9602, + "step": 10989 + }, + { + "epoch": 0.8259431835262288, + "grad_norm": 2.109974980680553, + "learning_rate": 3.0949974703610604e-07, + "loss": 0.9991, + "step": 10990 + }, + { + "epoch": 0.8260183375920638, + "grad_norm": 1.6206637260261783, + "learning_rate": 3.0923964283961046e-07, + "loss": 0.996, + "step": 10991 + }, + { + "epoch": 0.8260934916578987, + "grad_norm": 2.348121533486491, + "learning_rate": 3.0897963882713976e-07, + "loss": 0.964, + "step": 10992 + }, + { + "epoch": 0.8261686457237336, + "grad_norm": 1.4484318244436067, + "learning_rate": 3.087197350141004e-07, + "loss": 0.9726, + "step": 10993 + }, + { + "epoch": 0.8262437997895686, + "grad_norm": 1.88605824157495, + "learning_rate": 3.08459931415894e-07, + "loss": 0.8868, + "step": 10994 + }, + { + "epoch": 0.8263189538554035, + "grad_norm": 1.7036696096977162, + "learning_rate": 3.082002280479132e-07, + "loss": 0.9926, + "step": 10995 + }, + { + "epoch": 0.8263941079212386, + "grad_norm": 1.4345888431357214, + "learning_rate": 3.0794062492554716e-07, + "loss": 0.9877, + "step": 10996 + }, + { + "epoch": 0.8264692619870735, + "grad_norm": 2.022420738185413, + "learning_rate": 3.0768112206417885e-07, + "loss": 0.9559, + "step": 10997 + }, + { + "epoch": 0.8265444160529085, + "grad_norm": 1.9620005245344139, + "learning_rate": 3.074217194791844e-07, + "loss": 0.9814, + "step": 10998 + }, + { + "epoch": 0.8266195701187434, + "grad_norm": 1.8840133193369322, + "learning_rate": 3.071624171859344e-07, + "loss": 1.0354, + "step": 10999 + }, + { + "epoch": 0.8266947241845783, + "grad_norm": 2.097269658386592, + "learning_rate": 3.069032151997928e-07, + "loss": 0.887, + "step": 11000 + }, + { + "epoch": 0.8267698782504134, + "grad_norm": 1.625707325682226, + "learning_rate": 3.066441135361202e-07, + "loss": 0.9753, + "step": 11001 + }, + { + "epoch": 0.8268450323162483, + "grad_norm": 1.9932817034695816, + "learning_rate": 3.063851122102672e-07, + "loss": 1.0868, + "step": 11002 + }, + { + "epoch": 0.8269201863820833, + "grad_norm": 2.509322807937882, + "learning_rate": 3.0612621123758196e-07, + "loss": 0.9241, + "step": 11003 + }, + { + "epoch": 0.8269953404479182, + "grad_norm": 6.554155548661832, + "learning_rate": 3.0586741063340494e-07, + "loss": 0.8901, + "step": 11004 + }, + { + "epoch": 0.8270704945137531, + "grad_norm": 3.7736251500071005, + "learning_rate": 3.0560871041307137e-07, + "loss": 1.0436, + "step": 11005 + }, + { + "epoch": 0.8271456485795882, + "grad_norm": 1.3643909299290091, + "learning_rate": 3.0535011059190916e-07, + "loss": 0.9424, + "step": 11006 + }, + { + "epoch": 0.8272208026454231, + "grad_norm": 0.6478545544485951, + "learning_rate": 3.0509161118524283e-07, + "loss": 0.8171, + "step": 11007 + }, + { + "epoch": 0.8272959567112581, + "grad_norm": 1.6972760901441246, + "learning_rate": 3.0483321220838876e-07, + "loss": 0.993, + "step": 11008 + }, + { + "epoch": 0.827371110777093, + "grad_norm": 1.4882350323727167, + "learning_rate": 3.04574913676658e-07, + "loss": 0.9229, + "step": 11009 + }, + { + "epoch": 0.827446264842928, + "grad_norm": 1.6994978752201508, + "learning_rate": 3.0431671560535545e-07, + "loss": 0.9697, + "step": 11010 + }, + { + "epoch": 0.827521418908763, + "grad_norm": 4.858580104172629, + "learning_rate": 3.040586180097815e-07, + "loss": 0.9856, + "step": 11011 + }, + { + "epoch": 0.8275965729745979, + "grad_norm": 1.6140528720151266, + "learning_rate": 3.0380062090522796e-07, + "loss": 0.9675, + "step": 11012 + }, + { + "epoch": 0.8276717270404329, + "grad_norm": 1.701790873297121, + "learning_rate": 3.035427243069826e-07, + "loss": 0.9447, + "step": 11013 + }, + { + "epoch": 0.8277468811062678, + "grad_norm": 1.5867535976563258, + "learning_rate": 3.0328492823032756e-07, + "loss": 0.9452, + "step": 11014 + }, + { + "epoch": 0.8278220351721028, + "grad_norm": 1.527207078209126, + "learning_rate": 3.030272326905381e-07, + "loss": 0.9175, + "step": 11015 + }, + { + "epoch": 0.8278971892379378, + "grad_norm": 1.8004619455974502, + "learning_rate": 3.027696377028821e-07, + "loss": 0.9037, + "step": 11016 + }, + { + "epoch": 0.8279723433037728, + "grad_norm": 1.472579751696597, + "learning_rate": 3.025121432826245e-07, + "loss": 0.9526, + "step": 11017 + }, + { + "epoch": 0.8280474973696077, + "grad_norm": 1.672166674118374, + "learning_rate": 3.022547494450234e-07, + "loss": 0.8973, + "step": 11018 + }, + { + "epoch": 0.8281226514354426, + "grad_norm": 1.614471218359401, + "learning_rate": 3.019974562053285e-07, + "loss": 1.0139, + "step": 11019 + }, + { + "epoch": 0.8281978055012776, + "grad_norm": 1.4783331532928001, + "learning_rate": 3.017402635787869e-07, + "loss": 0.9629, + "step": 11020 + }, + { + "epoch": 0.8282729595671126, + "grad_norm": 1.8006585135182929, + "learning_rate": 3.0148317158063763e-07, + "loss": 0.9649, + "step": 11021 + }, + { + "epoch": 0.8283481136329476, + "grad_norm": 1.9660364251085267, + "learning_rate": 3.0122618022611467e-07, + "loss": 1.041, + "step": 11022 + }, + { + "epoch": 0.8284232676987825, + "grad_norm": 2.159858227774896, + "learning_rate": 3.00969289530445e-07, + "loss": 0.9661, + "step": 11023 + }, + { + "epoch": 0.8284984217646175, + "grad_norm": 3.621578575030793, + "learning_rate": 3.0071249950885145e-07, + "loss": 0.938, + "step": 11024 + }, + { + "epoch": 0.8285735758304524, + "grad_norm": 2.3198227285204904, + "learning_rate": 3.0045581017654933e-07, + "loss": 0.8317, + "step": 11025 + }, + { + "epoch": 0.8286487298962874, + "grad_norm": 7.965514719316784, + "learning_rate": 3.0019922154874853e-07, + "loss": 0.944, + "step": 11026 + }, + { + "epoch": 0.8287238839621224, + "grad_norm": 1.8130579122645525, + "learning_rate": 2.9994273364065235e-07, + "loss": 1.0225, + "step": 11027 + }, + { + "epoch": 0.8287990380279573, + "grad_norm": 1.81601459833823, + "learning_rate": 2.9968634646745995e-07, + "loss": 0.9702, + "step": 11028 + }, + { + "epoch": 0.8288741920937923, + "grad_norm": 1.4128457414259887, + "learning_rate": 2.9943006004436153e-07, + "loss": 0.8902, + "step": 11029 + }, + { + "epoch": 0.8289493461596272, + "grad_norm": 1.7400127573751427, + "learning_rate": 2.991738743865444e-07, + "loss": 0.968, + "step": 11030 + }, + { + "epoch": 0.8290245002254621, + "grad_norm": 1.6963810519683284, + "learning_rate": 2.9891778950918836e-07, + "loss": 1.0454, + "step": 11031 + }, + { + "epoch": 0.8290996542912972, + "grad_norm": 7.079478422877651, + "learning_rate": 2.98661805427467e-07, + "loss": 1.0287, + "step": 11032 + }, + { + "epoch": 0.8291748083571321, + "grad_norm": 1.8818297927985759, + "learning_rate": 2.98405922156548e-07, + "loss": 0.8813, + "step": 11033 + }, + { + "epoch": 0.8292499624229671, + "grad_norm": 1.45995954940883, + "learning_rate": 2.9815013971159395e-07, + "loss": 0.9969, + "step": 11034 + }, + { + "epoch": 0.829325116488802, + "grad_norm": 1.4508379915364593, + "learning_rate": 2.97894458107762e-07, + "loss": 0.839, + "step": 11035 + }, + { + "epoch": 0.829400270554637, + "grad_norm": 2.263484812158687, + "learning_rate": 2.9763887736020035e-07, + "loss": 0.8654, + "step": 11036 + }, + { + "epoch": 0.829475424620472, + "grad_norm": 1.712138412058768, + "learning_rate": 2.9738339748405426e-07, + "loss": 1.0618, + "step": 11037 + }, + { + "epoch": 0.8295505786863069, + "grad_norm": 3.5232503295716384, + "learning_rate": 2.9712801849446223e-07, + "loss": 0.8796, + "step": 11038 + }, + { + "epoch": 0.8296257327521419, + "grad_norm": 1.5281047827508558, + "learning_rate": 2.9687274040655477e-07, + "loss": 0.9592, + "step": 11039 + }, + { + "epoch": 0.8297008868179768, + "grad_norm": 1.8247815719948357, + "learning_rate": 2.966175632354593e-07, + "loss": 0.9409, + "step": 11040 + }, + { + "epoch": 0.8297760408838118, + "grad_norm": 1.7627983084305034, + "learning_rate": 2.963624869962962e-07, + "loss": 0.9639, + "step": 11041 + }, + { + "epoch": 0.8298511949496468, + "grad_norm": 2.0100008546058774, + "learning_rate": 2.9610751170417935e-07, + "loss": 0.9703, + "step": 11042 + }, + { + "epoch": 0.8299263490154818, + "grad_norm": 0.6701982334084652, + "learning_rate": 2.9585263737421717e-07, + "loss": 0.7368, + "step": 11043 + }, + { + "epoch": 0.8300015030813167, + "grad_norm": 1.4254965301921017, + "learning_rate": 2.955978640215115e-07, + "loss": 1.0214, + "step": 11044 + }, + { + "epoch": 0.8300766571471516, + "grad_norm": 1.5716119959529973, + "learning_rate": 2.9534319166115975e-07, + "loss": 1.0053, + "step": 11045 + }, + { + "epoch": 0.8301518112129866, + "grad_norm": 2.3435424712808737, + "learning_rate": 2.9508862030825075e-07, + "loss": 1.0861, + "step": 11046 + }, + { + "epoch": 0.8302269652788216, + "grad_norm": 6.263555855742632, + "learning_rate": 2.948341499778697e-07, + "loss": 0.9235, + "step": 11047 + }, + { + "epoch": 0.8303021193446566, + "grad_norm": 2.235858131562165, + "learning_rate": 2.9457978068509494e-07, + "loss": 0.9369, + "step": 11048 + }, + { + "epoch": 0.8303772734104915, + "grad_norm": 1.2996855006898516, + "learning_rate": 2.943255124449988e-07, + "loss": 1.0296, + "step": 11049 + }, + { + "epoch": 0.8304524274763264, + "grad_norm": 4.277331710171412, + "learning_rate": 2.940713452726469e-07, + "loss": 0.957, + "step": 11050 + }, + { + "epoch": 0.8305275815421614, + "grad_norm": 2.1795663013739595, + "learning_rate": 2.938172791831006e-07, + "loss": 1.0628, + "step": 11051 + }, + { + "epoch": 0.8306027356079964, + "grad_norm": 1.2837727529260758, + "learning_rate": 2.93563314191414e-07, + "loss": 0.9328, + "step": 11052 + }, + { + "epoch": 0.8306778896738314, + "grad_norm": 2.5210796477564985, + "learning_rate": 2.9330945031263545e-07, + "loss": 0.8992, + "step": 11053 + }, + { + "epoch": 0.8307530437396663, + "grad_norm": 1.7785630288311474, + "learning_rate": 2.9305568756180686e-07, + "loss": 0.8715, + "step": 11054 + }, + { + "epoch": 0.8308281978055013, + "grad_norm": 1.9029244353245285, + "learning_rate": 2.928020259539661e-07, + "loss": 0.9444, + "step": 11055 + }, + { + "epoch": 0.8309033518713362, + "grad_norm": 1.5426568210955829, + "learning_rate": 2.9254846550414146e-07, + "loss": 0.9537, + "step": 11056 + }, + { + "epoch": 0.8309785059371712, + "grad_norm": 1.9261647552500996, + "learning_rate": 2.922950062273586e-07, + "loss": 0.9353, + "step": 11057 + }, + { + "epoch": 0.8310536600030062, + "grad_norm": 2.0868547687309595, + "learning_rate": 2.9204164813863654e-07, + "loss": 0.9401, + "step": 11058 + }, + { + "epoch": 0.8311288140688411, + "grad_norm": 2.820797611688435, + "learning_rate": 2.917883912529873e-07, + "loss": 0.986, + "step": 11059 + }, + { + "epoch": 0.8312039681346761, + "grad_norm": 2.522781617518877, + "learning_rate": 2.9153523558541613e-07, + "loss": 0.9746, + "step": 11060 + }, + { + "epoch": 0.831279122200511, + "grad_norm": 1.385047485763181, + "learning_rate": 2.912821811509243e-07, + "loss": 0.8892, + "step": 11061 + }, + { + "epoch": 0.8313542762663461, + "grad_norm": 1.5362298138179187, + "learning_rate": 2.9102922796450745e-07, + "loss": 1.074, + "step": 11062 + }, + { + "epoch": 0.831429430332181, + "grad_norm": 1.5102331466555354, + "learning_rate": 2.9077637604115193e-07, + "loss": 1.0719, + "step": 11063 + }, + { + "epoch": 0.8315045843980159, + "grad_norm": 0.8105873925066983, + "learning_rate": 2.9052362539584164e-07, + "loss": 0.8756, + "step": 11064 + }, + { + "epoch": 0.8315797384638509, + "grad_norm": 1.6432943517589376, + "learning_rate": 2.9027097604355265e-07, + "loss": 0.9517, + "step": 11065 + }, + { + "epoch": 0.8316548925296858, + "grad_norm": 1.7467701261154485, + "learning_rate": 2.9001842799925526e-07, + "loss": 0.9384, + "step": 11066 + }, + { + "epoch": 0.8317300465955209, + "grad_norm": 1.668867454252884, + "learning_rate": 2.8976598127791364e-07, + "loss": 0.9663, + "step": 11067 + }, + { + "epoch": 0.8318052006613558, + "grad_norm": 1.5466309942593173, + "learning_rate": 2.8951363589448676e-07, + "loss": 1.0604, + "step": 11068 + }, + { + "epoch": 0.8318803547271908, + "grad_norm": 1.890417986168617, + "learning_rate": 2.8926139186392707e-07, + "loss": 0.8387, + "step": 11069 + }, + { + "epoch": 0.8319555087930257, + "grad_norm": 1.672146356255056, + "learning_rate": 2.8900924920118064e-07, + "loss": 0.9784, + "step": 11070 + }, + { + "epoch": 0.8320306628588606, + "grad_norm": 1.4958530162624926, + "learning_rate": 2.8875720792118754e-07, + "loss": 0.9974, + "step": 11071 + }, + { + "epoch": 0.8321058169246957, + "grad_norm": 1.7453031179368885, + "learning_rate": 2.8850526803888376e-07, + "loss": 0.9667, + "step": 11072 + }, + { + "epoch": 0.8321809709905306, + "grad_norm": 1.738056502653998, + "learning_rate": 2.882534295691954e-07, + "loss": 0.998, + "step": 11073 + }, + { + "epoch": 0.8322561250563656, + "grad_norm": 1.8503446799161551, + "learning_rate": 2.8800169252704675e-07, + "loss": 0.9379, + "step": 11074 + }, + { + "epoch": 0.8323312791222005, + "grad_norm": 2.194131227934635, + "learning_rate": 2.8775005692735344e-07, + "loss": 0.9377, + "step": 11075 + }, + { + "epoch": 0.8324064331880354, + "grad_norm": 1.358098625702946, + "learning_rate": 2.8749852278502573e-07, + "loss": 0.9739, + "step": 11076 + }, + { + "epoch": 0.8324815872538704, + "grad_norm": 2.05699556046648, + "learning_rate": 2.8724709011496795e-07, + "loss": 0.9923, + "step": 11077 + }, + { + "epoch": 0.8325567413197054, + "grad_norm": 2.396329170554902, + "learning_rate": 2.8699575893207837e-07, + "loss": 0.9482, + "step": 11078 + }, + { + "epoch": 0.8326318953855404, + "grad_norm": 1.8924452791151427, + "learning_rate": 2.867445292512507e-07, + "loss": 0.9305, + "step": 11079 + }, + { + "epoch": 0.8327070494513753, + "grad_norm": 1.674656203558201, + "learning_rate": 2.864934010873692e-07, + "loss": 0.9879, + "step": 11080 + }, + { + "epoch": 0.8327822035172103, + "grad_norm": 1.9636311059577953, + "learning_rate": 2.862423744553157e-07, + "loss": 0.9311, + "step": 11081 + }, + { + "epoch": 0.8328573575830452, + "grad_norm": 1.759234450910817, + "learning_rate": 2.8599144936996424e-07, + "loss": 0.95, + "step": 11082 + }, + { + "epoch": 0.8329325116488802, + "grad_norm": 1.5978798558232186, + "learning_rate": 2.8574062584618206e-07, + "loss": 1.041, + "step": 11083 + }, + { + "epoch": 0.8330076657147152, + "grad_norm": 1.35331700396591, + "learning_rate": 2.854899038988319e-07, + "loss": 0.9743, + "step": 11084 + }, + { + "epoch": 0.8330828197805501, + "grad_norm": 2.517926413250869, + "learning_rate": 2.8523928354277085e-07, + "loss": 0.9238, + "step": 11085 + }, + { + "epoch": 0.8331579738463851, + "grad_norm": 2.4085255551652414, + "learning_rate": 2.849887647928484e-07, + "loss": 0.9862, + "step": 11086 + }, + { + "epoch": 0.83323312791222, + "grad_norm": 1.844661745838035, + "learning_rate": 2.847383476639089e-07, + "loss": 0.9074, + "step": 11087 + }, + { + "epoch": 0.8333082819780551, + "grad_norm": 1.6806542741489363, + "learning_rate": 2.844880321707901e-07, + "loss": 0.9981, + "step": 11088 + }, + { + "epoch": 0.83338343604389, + "grad_norm": 0.7462786300636832, + "learning_rate": 2.842378183283254e-07, + "loss": 0.8289, + "step": 11089 + }, + { + "epoch": 0.8334585901097249, + "grad_norm": 1.2833347107933546, + "learning_rate": 2.8398770615133915e-07, + "loss": 0.933, + "step": 11090 + }, + { + "epoch": 0.8335337441755599, + "grad_norm": 1.7194661815764103, + "learning_rate": 2.837376956546527e-07, + "loss": 0.9836, + "step": 11091 + }, + { + "epoch": 0.8336088982413948, + "grad_norm": 1.4550442890099073, + "learning_rate": 2.8348778685307983e-07, + "loss": 0.9272, + "step": 11092 + }, + { + "epoch": 0.8336840523072299, + "grad_norm": 2.1374398920185307, + "learning_rate": 2.832379797614286e-07, + "loss": 0.9029, + "step": 11093 + }, + { + "epoch": 0.8337592063730648, + "grad_norm": 2.034551197396886, + "learning_rate": 2.829882743945007e-07, + "loss": 0.9507, + "step": 11094 + }, + { + "epoch": 0.8338343604388997, + "grad_norm": 2.811707109372611, + "learning_rate": 2.8273867076709225e-07, + "loss": 1.0155, + "step": 11095 + }, + { + "epoch": 0.8339095145047347, + "grad_norm": 1.4799096654696324, + "learning_rate": 2.8248916889399434e-07, + "loss": 0.9785, + "step": 11096 + }, + { + "epoch": 0.8339846685705696, + "grad_norm": 2.3642488374661847, + "learning_rate": 2.822397687899893e-07, + "loss": 0.9269, + "step": 11097 + }, + { + "epoch": 0.8340598226364047, + "grad_norm": 1.6910665051936287, + "learning_rate": 2.819904704698555e-07, + "loss": 0.9996, + "step": 11098 + }, + { + "epoch": 0.8341349767022396, + "grad_norm": 2.3568958115853786, + "learning_rate": 2.8174127394836577e-07, + "loss": 1.0625, + "step": 11099 + }, + { + "epoch": 0.8342101307680746, + "grad_norm": 1.893410136630626, + "learning_rate": 2.8149217924028443e-07, + "loss": 0.9629, + "step": 11100 + }, + { + "epoch": 0.8342852848339095, + "grad_norm": 2.2376391703197083, + "learning_rate": 2.8124318636037193e-07, + "loss": 0.9416, + "step": 11101 + }, + { + "epoch": 0.8343604388997444, + "grad_norm": 0.7866470076911706, + "learning_rate": 2.809942953233828e-07, + "loss": 0.87, + "step": 11102 + }, + { + "epoch": 0.8344355929655795, + "grad_norm": 2.1321808155530344, + "learning_rate": 2.8074550614406424e-07, + "loss": 0.9783, + "step": 11103 + }, + { + "epoch": 0.8345107470314144, + "grad_norm": 1.934240799348388, + "learning_rate": 2.804968188371577e-07, + "loss": 0.9381, + "step": 11104 + }, + { + "epoch": 0.8345859010972494, + "grad_norm": 2.4416134677685513, + "learning_rate": 2.8024823341739876e-07, + "loss": 1.0401, + "step": 11105 + }, + { + "epoch": 0.8346610551630843, + "grad_norm": 2.5213687317654787, + "learning_rate": 2.7999974989951813e-07, + "loss": 0.9886, + "step": 11106 + }, + { + "epoch": 0.8347362092289193, + "grad_norm": 2.1686854372116144, + "learning_rate": 2.7975136829823775e-07, + "loss": 0.8682, + "step": 11107 + }, + { + "epoch": 0.8348113632947542, + "grad_norm": 2.3335185204541946, + "learning_rate": 2.7950308862827675e-07, + "loss": 0.942, + "step": 11108 + }, + { + "epoch": 0.8348865173605892, + "grad_norm": 1.8896331076776438, + "learning_rate": 2.7925491090434583e-07, + "loss": 1.0756, + "step": 11109 + }, + { + "epoch": 0.8349616714264242, + "grad_norm": 2.214392560587222, + "learning_rate": 2.7900683514115054e-07, + "loss": 0.9056, + "step": 11110 + }, + { + "epoch": 0.8350368254922591, + "grad_norm": 1.685908591488391, + "learning_rate": 2.7875886135339e-07, + "loss": 0.9595, + "step": 11111 + }, + { + "epoch": 0.8351119795580941, + "grad_norm": 1.8005829079185924, + "learning_rate": 2.7851098955575845e-07, + "loss": 1.0549, + "step": 11112 + }, + { + "epoch": 0.835187133623929, + "grad_norm": 1.9652474142304333, + "learning_rate": 2.782632197629426e-07, + "loss": 0.9181, + "step": 11113 + }, + { + "epoch": 0.8352622876897641, + "grad_norm": 1.7810216389219153, + "learning_rate": 2.7801555198962433e-07, + "loss": 0.9694, + "step": 11114 + }, + { + "epoch": 0.835337441755599, + "grad_norm": 1.9361320711871564, + "learning_rate": 2.7776798625047784e-07, + "loss": 0.9883, + "step": 11115 + }, + { + "epoch": 0.8354125958214339, + "grad_norm": 1.631524412963292, + "learning_rate": 2.77520522560174e-07, + "loss": 1.0349, + "step": 11116 + }, + { + "epoch": 0.8354877498872689, + "grad_norm": 2.396873436321758, + "learning_rate": 2.7727316093337406e-07, + "loss": 1.0142, + "step": 11117 + }, + { + "epoch": 0.8355629039531038, + "grad_norm": 1.9874611832768831, + "learning_rate": 2.770259013847365e-07, + "loss": 0.9939, + "step": 11118 + }, + { + "epoch": 0.8356380580189389, + "grad_norm": 0.8212246863936187, + "learning_rate": 2.76778743928912e-07, + "loss": 0.8467, + "step": 11119 + }, + { + "epoch": 0.8357132120847738, + "grad_norm": 1.6252125821080257, + "learning_rate": 2.765316885805458e-07, + "loss": 0.989, + "step": 11120 + }, + { + "epoch": 0.8357883661506087, + "grad_norm": 1.9018836120952036, + "learning_rate": 2.762847353542759e-07, + "loss": 0.973, + "step": 11121 + }, + { + "epoch": 0.8358635202164437, + "grad_norm": 1.6836491197300938, + "learning_rate": 2.7603788426473615e-07, + "loss": 0.9642, + "step": 11122 + }, + { + "epoch": 0.8359386742822786, + "grad_norm": 1.596235020865207, + "learning_rate": 2.7579113532655407e-07, + "loss": 0.9609, + "step": 11123 + }, + { + "epoch": 0.8360138283481137, + "grad_norm": 2.045417324844952, + "learning_rate": 2.755444885543488e-07, + "loss": 0.9364, + "step": 11124 + }, + { + "epoch": 0.8360889824139486, + "grad_norm": 1.8149218328578658, + "learning_rate": 2.752979439627363e-07, + "loss": 0.9834, + "step": 11125 + }, + { + "epoch": 0.8361641364797836, + "grad_norm": 1.8263319302352934, + "learning_rate": 2.750515015663251e-07, + "loss": 0.9816, + "step": 11126 + }, + { + "epoch": 0.8362392905456185, + "grad_norm": 6.352904981414476, + "learning_rate": 2.7480516137971776e-07, + "loss": 0.9977, + "step": 11127 + }, + { + "epoch": 0.8363144446114534, + "grad_norm": 1.8846046497602162, + "learning_rate": 2.745589234175103e-07, + "loss": 0.9876, + "step": 11128 + }, + { + "epoch": 0.8363895986772885, + "grad_norm": 1.8478038640161587, + "learning_rate": 2.7431278769429414e-07, + "loss": 0.977, + "step": 11129 + }, + { + "epoch": 0.8364647527431234, + "grad_norm": 2.0312294592185576, + "learning_rate": 2.740667542246535e-07, + "loss": 1.0201, + "step": 11130 + }, + { + "epoch": 0.8365399068089584, + "grad_norm": 2.734190754813884, + "learning_rate": 2.738208230231667e-07, + "loss": 1.0916, + "step": 11131 + }, + { + "epoch": 0.8366150608747933, + "grad_norm": 1.381495435839033, + "learning_rate": 2.7357499410440564e-07, + "loss": 0.9814, + "step": 11132 + }, + { + "epoch": 0.8366902149406283, + "grad_norm": 3.5173128272647727, + "learning_rate": 2.7332926748293797e-07, + "loss": 0.9223, + "step": 11133 + }, + { + "epoch": 0.8367653690064633, + "grad_norm": 2.440805394811679, + "learning_rate": 2.730836431733221e-07, + "loss": 0.8741, + "step": 11134 + }, + { + "epoch": 0.8368405230722982, + "grad_norm": 1.708493023673649, + "learning_rate": 2.7283812119011386e-07, + "loss": 0.8994, + "step": 11135 + }, + { + "epoch": 0.8369156771381332, + "grad_norm": 1.7336279144977746, + "learning_rate": 2.7259270154786063e-07, + "loss": 0.8689, + "step": 11136 + }, + { + "epoch": 0.8369908312039681, + "grad_norm": 2.070471300476636, + "learning_rate": 2.723473842611044e-07, + "loss": 0.9551, + "step": 11137 + }, + { + "epoch": 0.8370659852698031, + "grad_norm": 2.8546702550942418, + "learning_rate": 2.721021693443808e-07, + "loss": 0.8899, + "step": 11138 + }, + { + "epoch": 0.837141139335638, + "grad_norm": 7.204279411967801, + "learning_rate": 2.718570568122203e-07, + "loss": 0.9502, + "step": 11139 + }, + { + "epoch": 0.837216293401473, + "grad_norm": 1.7257417825422803, + "learning_rate": 2.716120466791476e-07, + "loss": 1.0079, + "step": 11140 + }, + { + "epoch": 0.837291447467308, + "grad_norm": 1.52419915355323, + "learning_rate": 2.71367138959679e-07, + "loss": 1.0228, + "step": 11141 + }, + { + "epoch": 0.8373666015331429, + "grad_norm": 1.9277536068529975, + "learning_rate": 2.7112233366832657e-07, + "loss": 0.8489, + "step": 11142 + }, + { + "epoch": 0.8374417555989779, + "grad_norm": 1.6851236070522904, + "learning_rate": 2.70877630819597e-07, + "loss": 1.0345, + "step": 11143 + }, + { + "epoch": 0.8375169096648128, + "grad_norm": 2.085942414810163, + "learning_rate": 2.7063303042798803e-07, + "loss": 0.9906, + "step": 11144 + }, + { + "epoch": 0.8375920637306479, + "grad_norm": 1.296684434976576, + "learning_rate": 2.703885325079944e-07, + "loss": 1.0046, + "step": 11145 + }, + { + "epoch": 0.8376672177964828, + "grad_norm": 1.8849384129832323, + "learning_rate": 2.7014413707410356e-07, + "loss": 0.9115, + "step": 11146 + }, + { + "epoch": 0.8377423718623177, + "grad_norm": 1.4478159075836254, + "learning_rate": 2.698998441407969e-07, + "loss": 0.9644, + "step": 11147 + }, + { + "epoch": 0.8378175259281527, + "grad_norm": 1.6473440729195372, + "learning_rate": 2.696556537225492e-07, + "loss": 1.0226, + "step": 11148 + }, + { + "epoch": 0.8378926799939876, + "grad_norm": 1.7593346680918829, + "learning_rate": 2.6941156583382965e-07, + "loss": 1.0384, + "step": 11149 + }, + { + "epoch": 0.8379678340598227, + "grad_norm": 2.574872178543984, + "learning_rate": 2.691675804891027e-07, + "loss": 0.999, + "step": 11150 + }, + { + "epoch": 0.8380429881256576, + "grad_norm": 1.9281255292350747, + "learning_rate": 2.6892369770282333e-07, + "loss": 1.0695, + "step": 11151 + }, + { + "epoch": 0.8381181421914926, + "grad_norm": 1.9176788421744744, + "learning_rate": 2.686799174894441e-07, + "loss": 1.0426, + "step": 11152 + }, + { + "epoch": 0.8381932962573275, + "grad_norm": 1.8107541324705785, + "learning_rate": 2.684362398634095e-07, + "loss": 0.9875, + "step": 11153 + }, + { + "epoch": 0.8382684503231624, + "grad_norm": 1.9653062522571725, + "learning_rate": 2.6819266483915813e-07, + "loss": 0.9368, + "step": 11154 + }, + { + "epoch": 0.8383436043889975, + "grad_norm": 2.434003816838277, + "learning_rate": 2.6794919243112256e-07, + "loss": 0.9067, + "step": 11155 + }, + { + "epoch": 0.8384187584548324, + "grad_norm": 5.361286322981389, + "learning_rate": 2.6770582265373033e-07, + "loss": 0.8324, + "step": 11156 + }, + { + "epoch": 0.8384939125206674, + "grad_norm": 1.505635800320917, + "learning_rate": 2.674625555214014e-07, + "loss": 1.0459, + "step": 11157 + }, + { + "epoch": 0.8385690665865023, + "grad_norm": 2.989797568980048, + "learning_rate": 2.672193910485505e-07, + "loss": 0.9141, + "step": 11158 + }, + { + "epoch": 0.8386442206523373, + "grad_norm": 1.688136298697213, + "learning_rate": 2.6697632924958524e-07, + "loss": 0.8915, + "step": 11159 + }, + { + "epoch": 0.8387193747181723, + "grad_norm": 1.3972151118216185, + "learning_rate": 2.6673337013890986e-07, + "loss": 0.9818, + "step": 11160 + }, + { + "epoch": 0.8387945287840072, + "grad_norm": 1.560366891118849, + "learning_rate": 2.6649051373091834e-07, + "loss": 0.9787, + "step": 11161 + }, + { + "epoch": 0.8388696828498422, + "grad_norm": 0.8154070274265405, + "learning_rate": 2.6624776004000194e-07, + "loss": 0.8628, + "step": 11162 + }, + { + "epoch": 0.8389448369156771, + "grad_norm": 1.8169887470113146, + "learning_rate": 2.660051090805453e-07, + "loss": 1.0463, + "step": 11163 + }, + { + "epoch": 0.8390199909815121, + "grad_norm": 3.375860740257196, + "learning_rate": 2.657625608669263e-07, + "loss": 0.9635, + "step": 11164 + }, + { + "epoch": 0.8390951450473471, + "grad_norm": 1.898881271676389, + "learning_rate": 2.655201154135154e-07, + "loss": 0.9248, + "step": 11165 + }, + { + "epoch": 0.839170299113182, + "grad_norm": 1.8054437415232694, + "learning_rate": 2.6527777273467934e-07, + "loss": 0.9158, + "step": 11166 + }, + { + "epoch": 0.839245453179017, + "grad_norm": 2.1715220432133258, + "learning_rate": 2.650355328447791e-07, + "loss": 0.8866, + "step": 11167 + }, + { + "epoch": 0.8393206072448519, + "grad_norm": 1.5830390639988523, + "learning_rate": 2.6479339575816607e-07, + "loss": 0.908, + "step": 11168 + }, + { + "epoch": 0.8393957613106869, + "grad_norm": 1.9162007124821157, + "learning_rate": 2.6455136148918946e-07, + "loss": 1.0343, + "step": 11169 + }, + { + "epoch": 0.8394709153765219, + "grad_norm": 2.191018462931617, + "learning_rate": 2.6430943005219e-07, + "loss": 0.9862, + "step": 11170 + }, + { + "epoch": 0.8395460694423569, + "grad_norm": 1.5384559016858055, + "learning_rate": 2.640676014615033e-07, + "loss": 0.9987, + "step": 11171 + }, + { + "epoch": 0.8396212235081918, + "grad_norm": 1.789377642865083, + "learning_rate": 2.638258757314582e-07, + "loss": 1.0397, + "step": 11172 + }, + { + "epoch": 0.8396963775740267, + "grad_norm": 1.8933962223976322, + "learning_rate": 2.635842528763785e-07, + "loss": 1.0329, + "step": 11173 + }, + { + "epoch": 0.8397715316398617, + "grad_norm": 1.9445117696580978, + "learning_rate": 2.6334273291058105e-07, + "loss": 0.8907, + "step": 11174 + }, + { + "epoch": 0.8398466857056966, + "grad_norm": 2.7903177978656726, + "learning_rate": 2.631013158483768e-07, + "loss": 0.9623, + "step": 11175 + }, + { + "epoch": 0.8399218397715317, + "grad_norm": 2.400227925286004, + "learning_rate": 2.628600017040703e-07, + "loss": 1.0382, + "step": 11176 + }, + { + "epoch": 0.8399969938373666, + "grad_norm": 2.4766887067051147, + "learning_rate": 2.6261879049196125e-07, + "loss": 0.963, + "step": 11177 + }, + { + "epoch": 0.8400721479032016, + "grad_norm": 1.6194359817123818, + "learning_rate": 2.6237768222634103e-07, + "loss": 0.9111, + "step": 11178 + }, + { + "epoch": 0.8401473019690365, + "grad_norm": 1.8533279436476757, + "learning_rate": 2.621366769214974e-07, + "loss": 0.9362, + "step": 11179 + }, + { + "epoch": 0.8402224560348714, + "grad_norm": 1.7534544093000464, + "learning_rate": 2.6189577459171033e-07, + "loss": 0.9502, + "step": 11180 + }, + { + "epoch": 0.8402976101007065, + "grad_norm": 2.056360760056583, + "learning_rate": 2.6165497525125423e-07, + "loss": 1.0086, + "step": 11181 + }, + { + "epoch": 0.8403727641665414, + "grad_norm": 4.13762374646777, + "learning_rate": 2.614142789143972e-07, + "loss": 0.8474, + "step": 11182 + }, + { + "epoch": 0.8404479182323764, + "grad_norm": 1.3519210339881496, + "learning_rate": 2.611736855954014e-07, + "loss": 0.9695, + "step": 11183 + }, + { + "epoch": 0.8405230722982113, + "grad_norm": 2.301677950095497, + "learning_rate": 2.6093319530852407e-07, + "loss": 1.01, + "step": 11184 + }, + { + "epoch": 0.8405982263640462, + "grad_norm": 1.702749228363467, + "learning_rate": 2.606928080680133e-07, + "loss": 0.9609, + "step": 11185 + }, + { + "epoch": 0.8406733804298813, + "grad_norm": 3.047445547041426, + "learning_rate": 2.6045252388811434e-07, + "loss": 0.9788, + "step": 11186 + }, + { + "epoch": 0.8407485344957162, + "grad_norm": 2.4793981449401814, + "learning_rate": 2.602123427830651e-07, + "loss": 0.9898, + "step": 11187 + }, + { + "epoch": 0.8408236885615512, + "grad_norm": 1.532063326580077, + "learning_rate": 2.5997226476709567e-07, + "loss": 0.9316, + "step": 11188 + }, + { + "epoch": 0.8408988426273861, + "grad_norm": 3.576659856281984, + "learning_rate": 2.5973228985443274e-07, + "loss": 0.9919, + "step": 11189 + }, + { + "epoch": 0.8409739966932211, + "grad_norm": 1.6574770862267076, + "learning_rate": 2.594924180592957e-07, + "loss": 1.031, + "step": 11190 + }, + { + "epoch": 0.8410491507590561, + "grad_norm": 2.018977961940948, + "learning_rate": 2.59252649395898e-07, + "loss": 0.8919, + "step": 11191 + }, + { + "epoch": 0.841124304824891, + "grad_norm": 1.6343205984481635, + "learning_rate": 2.590129838784465e-07, + "loss": 1.059, + "step": 11192 + }, + { + "epoch": 0.841199458890726, + "grad_norm": 7.068554738768789, + "learning_rate": 2.587734215211419e-07, + "loss": 0.979, + "step": 11193 + }, + { + "epoch": 0.8412746129565609, + "grad_norm": 1.9050225230009405, + "learning_rate": 2.585339623381806e-07, + "loss": 1.0811, + "step": 11194 + }, + { + "epoch": 0.841349767022396, + "grad_norm": 2.176229050747448, + "learning_rate": 2.582946063437497e-07, + "loss": 0.9823, + "step": 11195 + }, + { + "epoch": 0.8414249210882309, + "grad_norm": 1.95359058171485, + "learning_rate": 2.5805535355203313e-07, + "loss": 0.9892, + "step": 11196 + }, + { + "epoch": 0.8415000751540659, + "grad_norm": 1.7064756977755027, + "learning_rate": 2.5781620397720715e-07, + "loss": 0.898, + "step": 11197 + }, + { + "epoch": 0.8415752292199008, + "grad_norm": 1.370246081770661, + "learning_rate": 2.575771576334427e-07, + "loss": 0.9672, + "step": 11198 + }, + { + "epoch": 0.8416503832857357, + "grad_norm": 1.4608164038423388, + "learning_rate": 2.5733821453490303e-07, + "loss": 0.9664, + "step": 11199 + }, + { + "epoch": 0.8417255373515707, + "grad_norm": 1.5550345028793617, + "learning_rate": 2.5709937469574794e-07, + "loss": 0.9768, + "step": 11200 + }, + { + "epoch": 0.8418006914174057, + "grad_norm": 0.6363577023839478, + "learning_rate": 2.568606381301288e-07, + "loss": 0.8018, + "step": 11201 + }, + { + "epoch": 0.8418758454832407, + "grad_norm": 2.278800171484463, + "learning_rate": 2.566220048521919e-07, + "loss": 0.9491, + "step": 11202 + }, + { + "epoch": 0.8419509995490756, + "grad_norm": 2.0785637872711837, + "learning_rate": 2.5638347487607646e-07, + "loss": 1.0344, + "step": 11203 + }, + { + "epoch": 0.8420261536149106, + "grad_norm": 0.7378703189800246, + "learning_rate": 2.5614504821591776e-07, + "loss": 0.8492, + "step": 11204 + }, + { + "epoch": 0.8421013076807455, + "grad_norm": 1.6515407910900655, + "learning_rate": 2.5590672488584177e-07, + "loss": 0.895, + "step": 11205 + }, + { + "epoch": 0.8421764617465805, + "grad_norm": 1.7956228786393569, + "learning_rate": 2.5566850489997096e-07, + "loss": 1.0174, + "step": 11206 + }, + { + "epoch": 0.8422516158124155, + "grad_norm": 1.6509689491535982, + "learning_rate": 2.5543038827242113e-07, + "loss": 0.935, + "step": 11207 + }, + { + "epoch": 0.8423267698782504, + "grad_norm": 0.7716856488707465, + "learning_rate": 2.5519237501730174e-07, + "loss": 0.7887, + "step": 11208 + }, + { + "epoch": 0.8424019239440854, + "grad_norm": 1.5608612622343625, + "learning_rate": 2.5495446514871457e-07, + "loss": 0.8889, + "step": 11209 + }, + { + "epoch": 0.8424770780099203, + "grad_norm": 1.3045844762637309, + "learning_rate": 2.547166586807574e-07, + "loss": 0.9055, + "step": 11210 + }, + { + "epoch": 0.8425522320757552, + "grad_norm": 1.7944036322966372, + "learning_rate": 2.544789556275222e-07, + "loss": 0.9879, + "step": 11211 + }, + { + "epoch": 0.8426273861415903, + "grad_norm": 1.6794246734098666, + "learning_rate": 2.542413560030923e-07, + "loss": 0.98, + "step": 11212 + }, + { + "epoch": 0.8427025402074252, + "grad_norm": 1.607013043201737, + "learning_rate": 2.5400385982154726e-07, + "loss": 1.035, + "step": 11213 + }, + { + "epoch": 0.8427776942732602, + "grad_norm": 2.0728274634511443, + "learning_rate": 2.5376646709695925e-07, + "loss": 0.8651, + "step": 11214 + }, + { + "epoch": 0.8428528483390951, + "grad_norm": 2.1528715120096926, + "learning_rate": 2.53529177843395e-07, + "loss": 0.9103, + "step": 11215 + }, + { + "epoch": 0.8429280024049302, + "grad_norm": 2.2460481058541544, + "learning_rate": 2.532919920749144e-07, + "loss": 0.8936, + "step": 11216 + }, + { + "epoch": 0.8430031564707651, + "grad_norm": 1.8496535879004896, + "learning_rate": 2.530549098055721e-07, + "loss": 0.8862, + "step": 11217 + }, + { + "epoch": 0.8430783105366, + "grad_norm": 1.8648786115643992, + "learning_rate": 2.528179310494158e-07, + "loss": 0.9511, + "step": 11218 + }, + { + "epoch": 0.843153464602435, + "grad_norm": 1.5261569277966531, + "learning_rate": 2.5258105582048775e-07, + "loss": 0.9037, + "step": 11219 + }, + { + "epoch": 0.8432286186682699, + "grad_norm": 2.3495932409631535, + "learning_rate": 2.5234428413282273e-07, + "loss": 1.0452, + "step": 11220 + }, + { + "epoch": 0.843303772734105, + "grad_norm": 1.8034378943858478, + "learning_rate": 2.521076160004523e-07, + "loss": 0.9761, + "step": 11221 + }, + { + "epoch": 0.8433789267999399, + "grad_norm": 1.8309265580170195, + "learning_rate": 2.5187105143739764e-07, + "loss": 0.954, + "step": 11222 + }, + { + "epoch": 0.8434540808657749, + "grad_norm": 2.2434347584749257, + "learning_rate": 2.5163459045767754e-07, + "loss": 0.8668, + "step": 11223 + }, + { + "epoch": 0.8435292349316098, + "grad_norm": 2.550085933957687, + "learning_rate": 2.5139823307530307e-07, + "loss": 0.9291, + "step": 11224 + }, + { + "epoch": 0.8436043889974447, + "grad_norm": 1.7140859958703978, + "learning_rate": 2.511619793042792e-07, + "loss": 0.9966, + "step": 11225 + }, + { + "epoch": 0.8436795430632797, + "grad_norm": 2.1604706740610444, + "learning_rate": 2.5092582915860427e-07, + "loss": 0.9395, + "step": 11226 + }, + { + "epoch": 0.8437546971291147, + "grad_norm": 1.7802817406724911, + "learning_rate": 2.5068978265227157e-07, + "loss": 1.0501, + "step": 11227 + }, + { + "epoch": 0.8438298511949497, + "grad_norm": 0.6024202696489809, + "learning_rate": 2.5045383979926884e-07, + "loss": 0.8204, + "step": 11228 + }, + { + "epoch": 0.8439050052607846, + "grad_norm": 3.3481239873426345, + "learning_rate": 2.5021800061357454e-07, + "loss": 0.9825, + "step": 11229 + }, + { + "epoch": 0.8439801593266195, + "grad_norm": 1.9426349152165363, + "learning_rate": 2.499822651091645e-07, + "loss": 0.8584, + "step": 11230 + }, + { + "epoch": 0.8440553133924545, + "grad_norm": 1.6687987359748946, + "learning_rate": 2.497466333000071e-07, + "loss": 0.8982, + "step": 11231 + }, + { + "epoch": 0.8441304674582895, + "grad_norm": 1.9661398039129576, + "learning_rate": 2.4951110520006294e-07, + "loss": 1.0378, + "step": 11232 + }, + { + "epoch": 0.8442056215241245, + "grad_norm": 2.4441001738763957, + "learning_rate": 2.492756808232888e-07, + "loss": 0.9057, + "step": 11233 + }, + { + "epoch": 0.8442807755899594, + "grad_norm": 1.4050868840351554, + "learning_rate": 2.4904036018363483e-07, + "loss": 0.8629, + "step": 11234 + }, + { + "epoch": 0.8443559296557944, + "grad_norm": 1.4417282613190614, + "learning_rate": 2.4880514329504463e-07, + "loss": 0.9545, + "step": 11235 + }, + { + "epoch": 0.8444310837216293, + "grad_norm": 2.275411685960738, + "learning_rate": 2.4857003017145526e-07, + "loss": 0.9841, + "step": 11236 + }, + { + "epoch": 0.8445062377874643, + "grad_norm": 1.6959663220246723, + "learning_rate": 2.48335020826798e-07, + "loss": 1.0414, + "step": 11237 + }, + { + "epoch": 0.8445813918532993, + "grad_norm": 2.0703872379272252, + "learning_rate": 2.481001152749993e-07, + "loss": 0.9051, + "step": 11238 + }, + { + "epoch": 0.8446565459191342, + "grad_norm": 1.5120616592923966, + "learning_rate": 2.478653135299762e-07, + "loss": 0.9399, + "step": 11239 + }, + { + "epoch": 0.8447316999849692, + "grad_norm": 2.056794105591965, + "learning_rate": 2.476306156056431e-07, + "loss": 1.0135, + "step": 11240 + }, + { + "epoch": 0.8448068540508041, + "grad_norm": 0.7355845210266494, + "learning_rate": 2.4739602151590635e-07, + "loss": 0.8569, + "step": 11241 + }, + { + "epoch": 0.8448820081166392, + "grad_norm": 1.6615631963309765, + "learning_rate": 2.471615312746664e-07, + "loss": 1.0479, + "step": 11242 + }, + { + "epoch": 0.8449571621824741, + "grad_norm": 2.3355109294161136, + "learning_rate": 2.4692714489581746e-07, + "loss": 0.9586, + "step": 11243 + }, + { + "epoch": 0.845032316248309, + "grad_norm": 0.7368862556329112, + "learning_rate": 2.466928623932485e-07, + "loss": 0.84, + "step": 11244 + }, + { + "epoch": 0.845107470314144, + "grad_norm": 2.7137267964204415, + "learning_rate": 2.4645868378084133e-07, + "loss": 0.9087, + "step": 11245 + }, + { + "epoch": 0.8451826243799789, + "grad_norm": 1.7718042591647052, + "learning_rate": 2.462246090724718e-07, + "loss": 0.9204, + "step": 11246 + }, + { + "epoch": 0.845257778445814, + "grad_norm": 1.781651769468132, + "learning_rate": 2.459906382820096e-07, + "loss": 1.0012, + "step": 11247 + }, + { + "epoch": 0.8453329325116489, + "grad_norm": 1.831799848375711, + "learning_rate": 2.457567714233193e-07, + "loss": 1.0213, + "step": 11248 + }, + { + "epoch": 0.8454080865774839, + "grad_norm": 1.5328255935716326, + "learning_rate": 2.455230085102571e-07, + "loss": 0.9422, + "step": 11249 + }, + { + "epoch": 0.8454832406433188, + "grad_norm": 1.4608735259221568, + "learning_rate": 2.452893495566748e-07, + "loss": 1.08, + "step": 11250 + }, + { + "epoch": 0.8455583947091537, + "grad_norm": 1.5287391161901605, + "learning_rate": 2.4505579457641824e-07, + "loss": 0.9702, + "step": 11251 + }, + { + "epoch": 0.8456335487749888, + "grad_norm": 2.1355058480845632, + "learning_rate": 2.448223435833261e-07, + "loss": 1.0994, + "step": 11252 + }, + { + "epoch": 0.8457087028408237, + "grad_norm": 1.7734346767856795, + "learning_rate": 2.445889965912311e-07, + "loss": 0.938, + "step": 11253 + }, + { + "epoch": 0.8457838569066587, + "grad_norm": 1.7124543455400736, + "learning_rate": 2.4435575361395934e-07, + "loss": 1.0756, + "step": 11254 + }, + { + "epoch": 0.8458590109724936, + "grad_norm": 1.8345834486897288, + "learning_rate": 2.44122614665333e-07, + "loss": 0.9967, + "step": 11255 + }, + { + "epoch": 0.8459341650383285, + "grad_norm": 1.8546156678915957, + "learning_rate": 2.4388957975916466e-07, + "loss": 0.8858, + "step": 11256 + }, + { + "epoch": 0.8460093191041635, + "grad_norm": 2.042544377380657, + "learning_rate": 2.436566489092635e-07, + "loss": 0.9697, + "step": 11257 + }, + { + "epoch": 0.8460844731699985, + "grad_norm": 2.707289822381979, + "learning_rate": 2.434238221294316e-07, + "loss": 0.7759, + "step": 11258 + }, + { + "epoch": 0.8461596272358335, + "grad_norm": 1.4192041317259243, + "learning_rate": 2.4319109943346473e-07, + "loss": 1.0206, + "step": 11259 + }, + { + "epoch": 0.8462347813016684, + "grad_norm": 1.362169528395438, + "learning_rate": 2.429584808351517e-07, + "loss": 0.926, + "step": 11260 + }, + { + "epoch": 0.8463099353675034, + "grad_norm": 3.761689341460849, + "learning_rate": 2.427259663482775e-07, + "loss": 0.9408, + "step": 11261 + }, + { + "epoch": 0.8463850894333383, + "grad_norm": 1.8438520241651233, + "learning_rate": 2.424935559866188e-07, + "loss": 1.0148, + "step": 11262 + }, + { + "epoch": 0.8464602434991733, + "grad_norm": 2.374083091020887, + "learning_rate": 2.422612497639469e-07, + "loss": 0.958, + "step": 11263 + }, + { + "epoch": 0.8465353975650083, + "grad_norm": 2.549031380335635, + "learning_rate": 2.4202904769402633e-07, + "loss": 0.9747, + "step": 11264 + }, + { + "epoch": 0.8466105516308432, + "grad_norm": 1.5885546657676601, + "learning_rate": 2.4179694979061717e-07, + "loss": 0.7789, + "step": 11265 + }, + { + "epoch": 0.8466857056966782, + "grad_norm": 1.643248649114915, + "learning_rate": 2.4156495606747065e-07, + "loss": 0.9484, + "step": 11266 + }, + { + "epoch": 0.8467608597625131, + "grad_norm": 0.6995852271874766, + "learning_rate": 2.413330665383342e-07, + "loss": 0.8049, + "step": 11267 + }, + { + "epoch": 0.8468360138283482, + "grad_norm": 1.8735437460099218, + "learning_rate": 2.4110128121694816e-07, + "loss": 0.914, + "step": 11268 + }, + { + "epoch": 0.8469111678941831, + "grad_norm": 1.687907558251167, + "learning_rate": 2.408696001170463e-07, + "loss": 0.9849, + "step": 11269 + }, + { + "epoch": 0.846986321960018, + "grad_norm": 1.6206982233502492, + "learning_rate": 2.406380232523566e-07, + "loss": 0.9782, + "step": 11270 + }, + { + "epoch": 0.847061476025853, + "grad_norm": 1.747954944771929, + "learning_rate": 2.4040655063660085e-07, + "loss": 0.9503, + "step": 11271 + }, + { + "epoch": 0.8471366300916879, + "grad_norm": 1.5329801166436903, + "learning_rate": 2.4017518228349586e-07, + "loss": 0.9554, + "step": 11272 + }, + { + "epoch": 0.847211784157523, + "grad_norm": 1.8058555093039417, + "learning_rate": 2.399439182067491e-07, + "loss": 0.9316, + "step": 11273 + }, + { + "epoch": 0.8472869382233579, + "grad_norm": 1.5360971144731617, + "learning_rate": 2.397127584200656e-07, + "loss": 1.0363, + "step": 11274 + }, + { + "epoch": 0.8473620922891928, + "grad_norm": 1.722061573589174, + "learning_rate": 2.394817029371421e-07, + "loss": 0.991, + "step": 11275 + }, + { + "epoch": 0.8474372463550278, + "grad_norm": 1.7490815068354335, + "learning_rate": 2.392507517716682e-07, + "loss": 0.9303, + "step": 11276 + }, + { + "epoch": 0.8475124004208627, + "grad_norm": 1.800829640563728, + "learning_rate": 2.3901990493732957e-07, + "loss": 0.9606, + "step": 11277 + }, + { + "epoch": 0.8475875544866978, + "grad_norm": 1.7155389788520736, + "learning_rate": 2.387891624478056e-07, + "loss": 0.9848, + "step": 11278 + }, + { + "epoch": 0.8476627085525327, + "grad_norm": 3.029866797663632, + "learning_rate": 2.385585243167676e-07, + "loss": 1.0025, + "step": 11279 + }, + { + "epoch": 0.8477378626183677, + "grad_norm": 1.722565386877869, + "learning_rate": 2.383279905578821e-07, + "loss": 0.9449, + "step": 11280 + }, + { + "epoch": 0.8478130166842026, + "grad_norm": 1.935966130646089, + "learning_rate": 2.3809756118480863e-07, + "loss": 1.0012, + "step": 11281 + }, + { + "epoch": 0.8478881707500375, + "grad_norm": 1.6043913630210973, + "learning_rate": 2.378672362112022e-07, + "loss": 0.991, + "step": 11282 + }, + { + "epoch": 0.8479633248158726, + "grad_norm": 1.8957572195123646, + "learning_rate": 2.3763701565070882e-07, + "loss": 1.0007, + "step": 11283 + }, + { + "epoch": 0.8480384788817075, + "grad_norm": 1.8780452793866265, + "learning_rate": 2.3740689951697135e-07, + "loss": 1.0046, + "step": 11284 + }, + { + "epoch": 0.8481136329475425, + "grad_norm": 1.717122573249892, + "learning_rate": 2.3717688782362444e-07, + "loss": 0.9019, + "step": 11285 + }, + { + "epoch": 0.8481887870133774, + "grad_norm": 1.533353100870394, + "learning_rate": 2.369469805842972e-07, + "loss": 0.9425, + "step": 11286 + }, + { + "epoch": 0.8482639410792124, + "grad_norm": 1.2357419328250685, + "learning_rate": 2.3671717781261225e-07, + "loss": 1.0133, + "step": 11287 + }, + { + "epoch": 0.8483390951450474, + "grad_norm": 1.3576673546441316, + "learning_rate": 2.364874795221865e-07, + "loss": 1.0402, + "step": 11288 + }, + { + "epoch": 0.8484142492108823, + "grad_norm": 3.7660879270040657, + "learning_rate": 2.362578857266313e-07, + "loss": 1.0278, + "step": 11289 + }, + { + "epoch": 0.8484894032767173, + "grad_norm": 1.9496904616555808, + "learning_rate": 2.3602839643954997e-07, + "loss": 0.9034, + "step": 11290 + }, + { + "epoch": 0.8485645573425522, + "grad_norm": 1.5248873340699816, + "learning_rate": 2.3579901167454008e-07, + "loss": 0.984, + "step": 11291 + }, + { + "epoch": 0.8486397114083872, + "grad_norm": 1.4056545692410563, + "learning_rate": 2.3556973144519542e-07, + "loss": 0.9218, + "step": 11292 + }, + { + "epoch": 0.8487148654742221, + "grad_norm": 1.8501526821125591, + "learning_rate": 2.353405557650998e-07, + "loss": 1.0273, + "step": 11293 + }, + { + "epoch": 0.8487900195400572, + "grad_norm": 1.6640519692531006, + "learning_rate": 2.3511148464783348e-07, + "loss": 0.8418, + "step": 11294 + }, + { + "epoch": 0.8488651736058921, + "grad_norm": 1.959755773217308, + "learning_rate": 2.3488251810697047e-07, + "loss": 0.9602, + "step": 11295 + }, + { + "epoch": 0.848940327671727, + "grad_norm": 1.6605555155225755, + "learning_rate": 2.3465365615607723e-07, + "loss": 1.0745, + "step": 11296 + }, + { + "epoch": 0.849015481737562, + "grad_norm": 1.395044953432058, + "learning_rate": 2.3442489880871475e-07, + "loss": 0.9648, + "step": 11297 + }, + { + "epoch": 0.8490906358033969, + "grad_norm": 0.7561072206883713, + "learning_rate": 2.341962460784377e-07, + "loss": 0.8488, + "step": 11298 + }, + { + "epoch": 0.849165789869232, + "grad_norm": 2.3341018455663143, + "learning_rate": 2.3396769797879544e-07, + "loss": 0.9785, + "step": 11299 + }, + { + "epoch": 0.8492409439350669, + "grad_norm": 3.0288442916771237, + "learning_rate": 2.3373925452332877e-07, + "loss": 1.0233, + "step": 11300 + }, + { + "epoch": 0.8493160980009018, + "grad_norm": 1.8947444835098841, + "learning_rate": 2.3351091572557524e-07, + "loss": 0.897, + "step": 11301 + }, + { + "epoch": 0.8493912520667368, + "grad_norm": 1.6449371344429948, + "learning_rate": 2.3328268159906428e-07, + "loss": 0.8903, + "step": 11302 + }, + { + "epoch": 0.8494664061325717, + "grad_norm": 0.6651472768252242, + "learning_rate": 2.3305455215731952e-07, + "loss": 0.8308, + "step": 11303 + }, + { + "epoch": 0.8495415601984068, + "grad_norm": 1.7530870457312975, + "learning_rate": 2.3282652741385834e-07, + "loss": 1.0869, + "step": 11304 + }, + { + "epoch": 0.8496167142642417, + "grad_norm": 1.7375462038771545, + "learning_rate": 2.3259860738219284e-07, + "loss": 0.9193, + "step": 11305 + }, + { + "epoch": 0.8496918683300767, + "grad_norm": 2.4470220567888497, + "learning_rate": 2.3237079207582755e-07, + "loss": 0.9898, + "step": 11306 + }, + { + "epoch": 0.8497670223959116, + "grad_norm": 1.460872220299561, + "learning_rate": 2.3214308150826166e-07, + "loss": 0.9437, + "step": 11307 + }, + { + "epoch": 0.8498421764617465, + "grad_norm": 1.4211084164811045, + "learning_rate": 2.319154756929873e-07, + "loss": 1.0158, + "step": 11308 + }, + { + "epoch": 0.8499173305275816, + "grad_norm": 2.0421959197520936, + "learning_rate": 2.3168797464349232e-07, + "loss": 0.9145, + "step": 11309 + }, + { + "epoch": 0.8499924845934165, + "grad_norm": 1.6489598319006076, + "learning_rate": 2.3146057837325527e-07, + "loss": 0.996, + "step": 11310 + }, + { + "epoch": 0.8500676386592515, + "grad_norm": 2.127464660279297, + "learning_rate": 2.3123328689575115e-07, + "loss": 1.0633, + "step": 11311 + }, + { + "epoch": 0.8501427927250864, + "grad_norm": 1.838031629611922, + "learning_rate": 2.3100610022444877e-07, + "loss": 0.9303, + "step": 11312 + }, + { + "epoch": 0.8502179467909214, + "grad_norm": 1.7272623982417266, + "learning_rate": 2.3077901837280844e-07, + "loss": 0.9946, + "step": 11313 + }, + { + "epoch": 0.8502931008567564, + "grad_norm": 2.8794776214932916, + "learning_rate": 2.305520413542854e-07, + "loss": 0.9293, + "step": 11314 + }, + { + "epoch": 0.8503682549225913, + "grad_norm": 2.9090882255260055, + "learning_rate": 2.303251691823298e-07, + "loss": 0.9652, + "step": 11315 + }, + { + "epoch": 0.8504434089884263, + "grad_norm": 1.6042397800314412, + "learning_rate": 2.3009840187038533e-07, + "loss": 0.9982, + "step": 11316 + }, + { + "epoch": 0.8505185630542612, + "grad_norm": 1.6523990847293015, + "learning_rate": 2.2987173943188697e-07, + "loss": 0.9565, + "step": 11317 + }, + { + "epoch": 0.8505937171200962, + "grad_norm": 1.5982761812516624, + "learning_rate": 2.2964518188026693e-07, + "loss": 1.0717, + "step": 11318 + }, + { + "epoch": 0.8506688711859312, + "grad_norm": 1.5051095879536074, + "learning_rate": 2.294187292289489e-07, + "loss": 1.0063, + "step": 11319 + }, + { + "epoch": 0.8507440252517661, + "grad_norm": 1.9552561615053135, + "learning_rate": 2.2919238149135122e-07, + "loss": 0.9491, + "step": 11320 + }, + { + "epoch": 0.8508191793176011, + "grad_norm": 1.9790305308318823, + "learning_rate": 2.2896613868088543e-07, + "loss": 0.9281, + "step": 11321 + }, + { + "epoch": 0.850894333383436, + "grad_norm": 1.4136862491362818, + "learning_rate": 2.287400008109579e-07, + "loss": 0.8416, + "step": 11322 + }, + { + "epoch": 0.850969487449271, + "grad_norm": 1.8158697176762122, + "learning_rate": 2.2851396789496812e-07, + "loss": 0.9679, + "step": 11323 + }, + { + "epoch": 0.851044641515106, + "grad_norm": 1.8965327750727243, + "learning_rate": 2.2828803994630917e-07, + "loss": 0.9371, + "step": 11324 + }, + { + "epoch": 0.851119795580941, + "grad_norm": 1.4690544949077924, + "learning_rate": 2.280622169783677e-07, + "loss": 0.9653, + "step": 11325 + }, + { + "epoch": 0.8511949496467759, + "grad_norm": 1.5892822623789422, + "learning_rate": 2.2783649900452584e-07, + "loss": 0.8901, + "step": 11326 + }, + { + "epoch": 0.8512701037126108, + "grad_norm": 1.5370524491623017, + "learning_rate": 2.276108860381567e-07, + "loss": 0.9629, + "step": 11327 + }, + { + "epoch": 0.8513452577784458, + "grad_norm": 1.6471119111284978, + "learning_rate": 2.2738537809263003e-07, + "loss": 0.9988, + "step": 11328 + }, + { + "epoch": 0.8514204118442807, + "grad_norm": 1.5188764323288317, + "learning_rate": 2.2715997518130758e-07, + "loss": 0.9205, + "step": 11329 + }, + { + "epoch": 0.8514955659101158, + "grad_norm": 2.7077307299700557, + "learning_rate": 2.2693467731754511e-07, + "loss": 1.0418, + "step": 11330 + }, + { + "epoch": 0.8515707199759507, + "grad_norm": 1.6315062193779766, + "learning_rate": 2.2670948451469195e-07, + "loss": 1.035, + "step": 11331 + }, + { + "epoch": 0.8516458740417857, + "grad_norm": 1.9406238107477989, + "learning_rate": 2.2648439678609254e-07, + "loss": 0.9994, + "step": 11332 + }, + { + "epoch": 0.8517210281076206, + "grad_norm": 1.5893565188168575, + "learning_rate": 2.2625941414508442e-07, + "loss": 0.8894, + "step": 11333 + }, + { + "epoch": 0.8517961821734555, + "grad_norm": 1.7297127409129702, + "learning_rate": 2.2603453660499782e-07, + "loss": 1.0089, + "step": 11334 + }, + { + "epoch": 0.8518713362392906, + "grad_norm": 1.7987855125457992, + "learning_rate": 2.2580976417915766e-07, + "loss": 0.9625, + "step": 11335 + }, + { + "epoch": 0.8519464903051255, + "grad_norm": 1.5820745014824575, + "learning_rate": 2.2558509688088324e-07, + "loss": 0.9109, + "step": 11336 + }, + { + "epoch": 0.8520216443709605, + "grad_norm": 1.5412651604543335, + "learning_rate": 2.2536053472348592e-07, + "loss": 0.8497, + "step": 11337 + }, + { + "epoch": 0.8520967984367954, + "grad_norm": 2.305803449037507, + "learning_rate": 2.2513607772027243e-07, + "loss": 0.8976, + "step": 11338 + }, + { + "epoch": 0.8521719525026304, + "grad_norm": 1.5724590301028287, + "learning_rate": 2.2491172588454322e-07, + "loss": 0.942, + "step": 11339 + }, + { + "epoch": 0.8522471065684654, + "grad_norm": 1.7711175877431282, + "learning_rate": 2.2468747922959143e-07, + "loss": 0.9164, + "step": 11340 + }, + { + "epoch": 0.8523222606343003, + "grad_norm": 2.19172282660683, + "learning_rate": 2.2446333776870484e-07, + "loss": 1.0214, + "step": 11341 + }, + { + "epoch": 0.8523974147001353, + "grad_norm": 1.2754323581568172, + "learning_rate": 2.242393015151638e-07, + "loss": 1.0101, + "step": 11342 + }, + { + "epoch": 0.8524725687659702, + "grad_norm": 2.6077405184804823, + "learning_rate": 2.2401537048224495e-07, + "loss": 1.0726, + "step": 11343 + }, + { + "epoch": 0.8525477228318052, + "grad_norm": 1.8347087887851266, + "learning_rate": 2.2379154468321525e-07, + "loss": 1.0265, + "step": 11344 + }, + { + "epoch": 0.8526228768976402, + "grad_norm": 1.954212953823196, + "learning_rate": 2.2356782413133834e-07, + "loss": 0.8637, + "step": 11345 + }, + { + "epoch": 0.8526980309634751, + "grad_norm": 1.701257994757036, + "learning_rate": 2.233442088398705e-07, + "loss": 1.011, + "step": 11346 + }, + { + "epoch": 0.8527731850293101, + "grad_norm": 2.043789350498286, + "learning_rate": 2.231206988220613e-07, + "loss": 0.9895, + "step": 11347 + }, + { + "epoch": 0.852848339095145, + "grad_norm": 1.7732352927122539, + "learning_rate": 2.2289729409115466e-07, + "loss": 0.9701, + "step": 11348 + }, + { + "epoch": 0.85292349316098, + "grad_norm": 2.3303666555693865, + "learning_rate": 2.226739946603886e-07, + "loss": 0.9121, + "step": 11349 + }, + { + "epoch": 0.852998647226815, + "grad_norm": 2.569488950005989, + "learning_rate": 2.2245080054299415e-07, + "loss": 0.886, + "step": 11350 + }, + { + "epoch": 0.85307380129265, + "grad_norm": 1.8724080926726367, + "learning_rate": 2.2222771175219668e-07, + "loss": 1.015, + "step": 11351 + }, + { + "epoch": 0.8531489553584849, + "grad_norm": 8.034346759023927, + "learning_rate": 2.2200472830121431e-07, + "loss": 0.8987, + "step": 11352 + }, + { + "epoch": 0.8532241094243198, + "grad_norm": 1.6944713703302203, + "learning_rate": 2.2178185020326091e-07, + "loss": 0.9541, + "step": 11353 + }, + { + "epoch": 0.8532992634901548, + "grad_norm": 1.3865392434433288, + "learning_rate": 2.2155907747154122e-07, + "loss": 0.8985, + "step": 11354 + }, + { + "epoch": 0.8533744175559898, + "grad_norm": 1.6044113501014858, + "learning_rate": 2.2133641011925652e-07, + "loss": 1.0207, + "step": 11355 + }, + { + "epoch": 0.8534495716218248, + "grad_norm": 2.0321378674782293, + "learning_rate": 2.2111384815960132e-07, + "loss": 0.9352, + "step": 11356 + }, + { + "epoch": 0.8535247256876597, + "grad_norm": 1.8721371093040502, + "learning_rate": 2.2089139160576197e-07, + "loss": 0.9702, + "step": 11357 + }, + { + "epoch": 0.8535998797534947, + "grad_norm": 6.211447256989887, + "learning_rate": 2.206690404709197e-07, + "loss": 1.0296, + "step": 11358 + }, + { + "epoch": 0.8536750338193296, + "grad_norm": 2.308417040779542, + "learning_rate": 2.2044679476825045e-07, + "loss": 0.971, + "step": 11359 + }, + { + "epoch": 0.8537501878851645, + "grad_norm": 1.8155514254082703, + "learning_rate": 2.2022465451092386e-07, + "loss": 0.933, + "step": 11360 + }, + { + "epoch": 0.8538253419509996, + "grad_norm": 1.459332568840383, + "learning_rate": 2.2000261971210098e-07, + "loss": 0.9425, + "step": 11361 + }, + { + "epoch": 0.8539004960168345, + "grad_norm": 3.087725852705723, + "learning_rate": 2.1978069038493906e-07, + "loss": 0.9987, + "step": 11362 + }, + { + "epoch": 0.8539756500826695, + "grad_norm": 1.7364854773505354, + "learning_rate": 2.1955886654258848e-07, + "loss": 1.0029, + "step": 11363 + }, + { + "epoch": 0.8540508041485044, + "grad_norm": 1.6018544745051173, + "learning_rate": 2.1933714819819248e-07, + "loss": 1.1111, + "step": 11364 + }, + { + "epoch": 0.8541259582143393, + "grad_norm": 2.0175525999716446, + "learning_rate": 2.191155353648888e-07, + "loss": 0.9948, + "step": 11365 + }, + { + "epoch": 0.8542011122801744, + "grad_norm": 0.7488159130933402, + "learning_rate": 2.188940280558096e-07, + "loss": 0.7919, + "step": 11366 + }, + { + "epoch": 0.8542762663460093, + "grad_norm": 2.3781517598755695, + "learning_rate": 2.1867262628407945e-07, + "loss": 1.0571, + "step": 11367 + }, + { + "epoch": 0.8543514204118443, + "grad_norm": 1.5253061425018526, + "learning_rate": 2.1845133006281745e-07, + "loss": 0.9386, + "step": 11368 + }, + { + "epoch": 0.8544265744776792, + "grad_norm": 1.5802357975738424, + "learning_rate": 2.182301394051358e-07, + "loss": 0.9655, + "step": 11369 + }, + { + "epoch": 0.8545017285435142, + "grad_norm": 2.1159048101074376, + "learning_rate": 2.1800905432414197e-07, + "loss": 0.913, + "step": 11370 + }, + { + "epoch": 0.8545768826093492, + "grad_norm": 1.8862707265596135, + "learning_rate": 2.1778807483293437e-07, + "loss": 0.9342, + "step": 11371 + }, + { + "epoch": 0.8546520366751841, + "grad_norm": 1.559986381471318, + "learning_rate": 2.1756720094460856e-07, + "loss": 0.9254, + "step": 11372 + }, + { + "epoch": 0.8547271907410191, + "grad_norm": 1.3321117477480875, + "learning_rate": 2.173464326722514e-07, + "loss": 0.9285, + "step": 11373 + }, + { + "epoch": 0.854802344806854, + "grad_norm": 1.664042584653788, + "learning_rate": 2.1712577002894417e-07, + "loss": 1.0517, + "step": 11374 + }, + { + "epoch": 0.854877498872689, + "grad_norm": 2.1876264263131207, + "learning_rate": 2.1690521302776198e-07, + "loss": 0.9392, + "step": 11375 + }, + { + "epoch": 0.854952652938524, + "grad_norm": 3.766471799237733, + "learning_rate": 2.1668476168177374e-07, + "loss": 1.0411, + "step": 11376 + }, + { + "epoch": 0.855027807004359, + "grad_norm": 1.4007013845999405, + "learning_rate": 2.1646441600404297e-07, + "loss": 0.9583, + "step": 11377 + }, + { + "epoch": 0.8551029610701939, + "grad_norm": 2.0589026087311635, + "learning_rate": 2.1624417600762435e-07, + "loss": 1.0771, + "step": 11378 + }, + { + "epoch": 0.8551781151360288, + "grad_norm": 3.3609281985811696, + "learning_rate": 2.160240417055692e-07, + "loss": 0.9112, + "step": 11379 + }, + { + "epoch": 0.8552532692018638, + "grad_norm": 1.6162709758014024, + "learning_rate": 2.1580401311092112e-07, + "loss": 0.9584, + "step": 11380 + }, + { + "epoch": 0.8553284232676988, + "grad_norm": 1.5914732434381258, + "learning_rate": 2.1558409023671677e-07, + "loss": 1.0115, + "step": 11381 + }, + { + "epoch": 0.8554035773335338, + "grad_norm": 1.8250456085776794, + "learning_rate": 2.1536427309598792e-07, + "loss": 1.0077, + "step": 11382 + }, + { + "epoch": 0.8554787313993687, + "grad_norm": 1.4627846041510963, + "learning_rate": 2.1514456170176021e-07, + "loss": 0.9194, + "step": 11383 + }, + { + "epoch": 0.8555538854652037, + "grad_norm": 3.277227642926193, + "learning_rate": 2.1492495606705184e-07, + "loss": 1.0218, + "step": 11384 + }, + { + "epoch": 0.8556290395310386, + "grad_norm": 2.8195525014514558, + "learning_rate": 2.1470545620487557e-07, + "loss": 0.9602, + "step": 11385 + }, + { + "epoch": 0.8557041935968736, + "grad_norm": 1.3605086324236355, + "learning_rate": 2.1448606212823673e-07, + "loss": 0.914, + "step": 11386 + }, + { + "epoch": 0.8557793476627086, + "grad_norm": 0.7217484829830219, + "learning_rate": 2.142667738501367e-07, + "loss": 0.7689, + "step": 11387 + }, + { + "epoch": 0.8558545017285435, + "grad_norm": 2.7639234232914878, + "learning_rate": 2.1404759138356753e-07, + "loss": 0.8837, + "step": 11388 + }, + { + "epoch": 0.8559296557943785, + "grad_norm": 1.6235195899159771, + "learning_rate": 2.1382851474151799e-07, + "loss": 0.9894, + "step": 11389 + }, + { + "epoch": 0.8560048098602134, + "grad_norm": 1.6914828593676454, + "learning_rate": 2.136095439369685e-07, + "loss": 1.0108, + "step": 11390 + }, + { + "epoch": 0.8560799639260483, + "grad_norm": 0.7821918151224361, + "learning_rate": 2.133906789828941e-07, + "loss": 0.8616, + "step": 11391 + }, + { + "epoch": 0.8561551179918834, + "grad_norm": 2.2443434289142115, + "learning_rate": 2.1317191989226302e-07, + "loss": 1.0454, + "step": 11392 + }, + { + "epoch": 0.8562302720577183, + "grad_norm": 1.798170203422038, + "learning_rate": 2.129532666780385e-07, + "loss": 1.0398, + "step": 11393 + }, + { + "epoch": 0.8563054261235533, + "grad_norm": 6.142564772938589, + "learning_rate": 2.1273471935317567e-07, + "loss": 1.0291, + "step": 11394 + }, + { + "epoch": 0.8563805801893882, + "grad_norm": 2.310060090278394, + "learning_rate": 2.1251627793062466e-07, + "loss": 0.9907, + "step": 11395 + }, + { + "epoch": 0.8564557342552233, + "grad_norm": 2.2578680434209315, + "learning_rate": 2.1229794242332866e-07, + "loss": 0.9564, + "step": 11396 + }, + { + "epoch": 0.8565308883210582, + "grad_norm": 0.9168797953245572, + "learning_rate": 2.1207971284422577e-07, + "loss": 0.8189, + "step": 11397 + }, + { + "epoch": 0.8566060423868931, + "grad_norm": 2.1980228729862885, + "learning_rate": 2.1186158920624563e-07, + "loss": 0.9669, + "step": 11398 + }, + { + "epoch": 0.8566811964527281, + "grad_norm": 2.4705562502663203, + "learning_rate": 2.116435715223135e-07, + "loss": 0.949, + "step": 11399 + }, + { + "epoch": 0.856756350518563, + "grad_norm": 1.4859498353399998, + "learning_rate": 2.1142565980534877e-07, + "loss": 1.005, + "step": 11400 + }, + { + "epoch": 0.856831504584398, + "grad_norm": 2.5804186241831952, + "learning_rate": 2.1120785406826204e-07, + "loss": 0.9885, + "step": 11401 + }, + { + "epoch": 0.856906658650233, + "grad_norm": 1.7123279930661686, + "learning_rate": 2.1099015432395939e-07, + "loss": 0.962, + "step": 11402 + }, + { + "epoch": 0.856981812716068, + "grad_norm": 1.7283267160409752, + "learning_rate": 2.1077256058534055e-07, + "loss": 0.832, + "step": 11403 + }, + { + "epoch": 0.8570569667819029, + "grad_norm": 1.4979049038414116, + "learning_rate": 2.1055507286529984e-07, + "loss": 1.0257, + "step": 11404 + }, + { + "epoch": 0.8571321208477378, + "grad_norm": 1.955031418218624, + "learning_rate": 2.1033769117672229e-07, + "loss": 0.9764, + "step": 11405 + }, + { + "epoch": 0.8572072749135728, + "grad_norm": 1.79949209941922, + "learning_rate": 2.1012041553249028e-07, + "loss": 1.0045, + "step": 11406 + }, + { + "epoch": 0.8572824289794078, + "grad_norm": 2.0458295642121422, + "learning_rate": 2.0990324594547748e-07, + "loss": 0.8479, + "step": 11407 + }, + { + "epoch": 0.8573575830452428, + "grad_norm": 1.7141720231707676, + "learning_rate": 2.0968618242855207e-07, + "loss": 0.9741, + "step": 11408 + }, + { + "epoch": 0.8574327371110777, + "grad_norm": 1.529220637617216, + "learning_rate": 2.0946922499457552e-07, + "loss": 0.9127, + "step": 11409 + }, + { + "epoch": 0.8575078911769126, + "grad_norm": 3.2137689659481903, + "learning_rate": 2.0925237365640424e-07, + "loss": 0.9034, + "step": 11410 + }, + { + "epoch": 0.8575830452427476, + "grad_norm": 1.6529736772300267, + "learning_rate": 2.090356284268868e-07, + "loss": 1.0223, + "step": 11411 + }, + { + "epoch": 0.8576581993085826, + "grad_norm": 1.799965034251215, + "learning_rate": 2.0881898931886677e-07, + "loss": 1.0285, + "step": 11412 + }, + { + "epoch": 0.8577333533744176, + "grad_norm": 2.025832243019019, + "learning_rate": 2.0860245634518002e-07, + "loss": 1.0252, + "step": 11413 + }, + { + "epoch": 0.8578085074402525, + "grad_norm": 1.4664963822058434, + "learning_rate": 2.0838602951865812e-07, + "loss": 0.9351, + "step": 11414 + }, + { + "epoch": 0.8578836615060875, + "grad_norm": 1.7190526522416931, + "learning_rate": 2.0816970885212392e-07, + "loss": 1.0078, + "step": 11415 + }, + { + "epoch": 0.8579588155719224, + "grad_norm": 1.4921773540691148, + "learning_rate": 2.0795349435839605e-07, + "loss": 0.9785, + "step": 11416 + }, + { + "epoch": 0.8580339696377574, + "grad_norm": 1.827916125241018, + "learning_rate": 2.0773738605028602e-07, + "loss": 0.968, + "step": 11417 + }, + { + "epoch": 0.8581091237035924, + "grad_norm": 1.7217556422243465, + "learning_rate": 2.075213839405987e-07, + "loss": 0.9936, + "step": 11418 + }, + { + "epoch": 0.8581842777694273, + "grad_norm": 2.2060864612170654, + "learning_rate": 2.07305488042133e-07, + "loss": 1.0345, + "step": 11419 + }, + { + "epoch": 0.8582594318352623, + "grad_norm": 1.5012789677883105, + "learning_rate": 2.0708969836768176e-07, + "loss": 1.0316, + "step": 11420 + }, + { + "epoch": 0.8583345859010972, + "grad_norm": 1.398545074587543, + "learning_rate": 2.068740149300321e-07, + "loss": 0.976, + "step": 11421 + }, + { + "epoch": 0.8584097399669323, + "grad_norm": 1.8235478934087979, + "learning_rate": 2.0665843774196269e-07, + "loss": 0.9749, + "step": 11422 + }, + { + "epoch": 0.8584848940327672, + "grad_norm": 2.330511112064697, + "learning_rate": 2.064429668162484e-07, + "loss": 0.9234, + "step": 11423 + }, + { + "epoch": 0.8585600480986021, + "grad_norm": 1.9261808464415386, + "learning_rate": 2.0622760216565683e-07, + "loss": 0.9236, + "step": 11424 + }, + { + "epoch": 0.8586352021644371, + "grad_norm": 1.9805274367807144, + "learning_rate": 2.0601234380294775e-07, + "loss": 0.9869, + "step": 11425 + }, + { + "epoch": 0.858710356230272, + "grad_norm": 1.9487391552675353, + "learning_rate": 2.0579719174087696e-07, + "loss": 0.9623, + "step": 11426 + }, + { + "epoch": 0.8587855102961071, + "grad_norm": 2.242214721089118, + "learning_rate": 2.0558214599219337e-07, + "loss": 0.9003, + "step": 11427 + }, + { + "epoch": 0.858860664361942, + "grad_norm": 1.5930359680751263, + "learning_rate": 2.0536720656963902e-07, + "loss": 1.0518, + "step": 11428 + }, + { + "epoch": 0.858935818427777, + "grad_norm": 1.4670220619992242, + "learning_rate": 2.0515237348594972e-07, + "loss": 1.0271, + "step": 11429 + }, + { + "epoch": 0.8590109724936119, + "grad_norm": 1.8185093015821492, + "learning_rate": 2.0493764675385484e-07, + "loss": 0.9246, + "step": 11430 + }, + { + "epoch": 0.8590861265594468, + "grad_norm": 1.758288238573877, + "learning_rate": 2.0472302638607885e-07, + "loss": 0.9356, + "step": 11431 + }, + { + "epoch": 0.8591612806252819, + "grad_norm": 1.556851361989987, + "learning_rate": 2.045085123953374e-07, + "loss": 0.9532, + "step": 11432 + }, + { + "epoch": 0.8592364346911168, + "grad_norm": 1.777781673599319, + "learning_rate": 2.0429410479434228e-07, + "loss": 0.948, + "step": 11433 + }, + { + "epoch": 0.8593115887569518, + "grad_norm": 1.759577090106029, + "learning_rate": 2.040798035957978e-07, + "loss": 1.0175, + "step": 11434 + }, + { + "epoch": 0.8593867428227867, + "grad_norm": 1.8471635802799442, + "learning_rate": 2.03865608812402e-07, + "loss": 0.911, + "step": 11435 + }, + { + "epoch": 0.8594618968886216, + "grad_norm": 1.573445776393909, + "learning_rate": 2.036515204568463e-07, + "loss": 1.0606, + "step": 11436 + }, + { + "epoch": 0.8595370509544566, + "grad_norm": 1.9552515888527133, + "learning_rate": 2.0343753854181655e-07, + "loss": 0.8715, + "step": 11437 + }, + { + "epoch": 0.8596122050202916, + "grad_norm": 2.4877410733516405, + "learning_rate": 2.0322366307999307e-07, + "loss": 1.0108, + "step": 11438 + }, + { + "epoch": 0.8596873590861266, + "grad_norm": 6.832355925586465, + "learning_rate": 2.0300989408404745e-07, + "loss": 0.9364, + "step": 11439 + }, + { + "epoch": 0.8597625131519615, + "grad_norm": 1.5145421043077096, + "learning_rate": 2.027962315666465e-07, + "loss": 0.9612, + "step": 11440 + }, + { + "epoch": 0.8598376672177965, + "grad_norm": 1.8787263082202077, + "learning_rate": 2.0258267554045138e-07, + "loss": 0.9952, + "step": 11441 + }, + { + "epoch": 0.8599128212836314, + "grad_norm": 0.7838209954189139, + "learning_rate": 2.0236922601811491e-07, + "loss": 0.8317, + "step": 11442 + }, + { + "epoch": 0.8599879753494664, + "grad_norm": 1.33299269894706, + "learning_rate": 2.0215588301228515e-07, + "loss": 0.9921, + "step": 11443 + }, + { + "epoch": 0.8600631294153014, + "grad_norm": 1.7007241782275322, + "learning_rate": 2.0194264653560467e-07, + "loss": 1.034, + "step": 11444 + }, + { + "epoch": 0.8601382834811363, + "grad_norm": 2.2635203089425464, + "learning_rate": 2.017295166007067e-07, + "loss": 1.0154, + "step": 11445 + }, + { + "epoch": 0.8602134375469713, + "grad_norm": 3.182004342268613, + "learning_rate": 2.0151649322022134e-07, + "loss": 0.9874, + "step": 11446 + }, + { + "epoch": 0.8602885916128062, + "grad_norm": 1.4790905498566655, + "learning_rate": 2.0130357640677033e-07, + "loss": 1.0376, + "step": 11447 + }, + { + "epoch": 0.8603637456786413, + "grad_norm": 1.4715700064366608, + "learning_rate": 2.010907661729706e-07, + "loss": 1.0622, + "step": 11448 + }, + { + "epoch": 0.8604388997444762, + "grad_norm": 1.956020985321088, + "learning_rate": 2.0087806253143103e-07, + "loss": 0.8984, + "step": 11449 + }, + { + "epoch": 0.8605140538103111, + "grad_norm": 1.3420386395043886, + "learning_rate": 2.0066546549475573e-07, + "loss": 1.0527, + "step": 11450 + }, + { + "epoch": 0.8605892078761461, + "grad_norm": 1.7660314041576926, + "learning_rate": 2.004529750755417e-07, + "loss": 0.8518, + "step": 11451 + }, + { + "epoch": 0.860664361941981, + "grad_norm": 1.9109395611206816, + "learning_rate": 2.0024059128637983e-07, + "loss": 0.9882, + "step": 11452 + }, + { + "epoch": 0.8607395160078161, + "grad_norm": 1.413329972692009, + "learning_rate": 2.0002831413985443e-07, + "loss": 0.9135, + "step": 11453 + }, + { + "epoch": 0.860814670073651, + "grad_norm": 2.007731157725775, + "learning_rate": 1.9981614364854415e-07, + "loss": 0.9731, + "step": 11454 + }, + { + "epoch": 0.8608898241394859, + "grad_norm": 0.5852347673978293, + "learning_rate": 1.9960407982502093e-07, + "loss": 0.8335, + "step": 11455 + }, + { + "epoch": 0.8609649782053209, + "grad_norm": 1.6482391961309357, + "learning_rate": 1.9939212268185002e-07, + "loss": 1.038, + "step": 11456 + }, + { + "epoch": 0.8610401322711558, + "grad_norm": 1.991585494692374, + "learning_rate": 1.991802722315905e-07, + "loss": 0.9501, + "step": 11457 + }, + { + "epoch": 0.8611152863369909, + "grad_norm": 3.4024115817362843, + "learning_rate": 1.9896852848679634e-07, + "loss": 0.9939, + "step": 11458 + }, + { + "epoch": 0.8611904404028258, + "grad_norm": 1.7992674449537676, + "learning_rate": 1.9875689146001262e-07, + "loss": 0.9843, + "step": 11459 + }, + { + "epoch": 0.8612655944686608, + "grad_norm": 1.5165405654185642, + "learning_rate": 1.9854536116378107e-07, + "loss": 0.9597, + "step": 11460 + }, + { + "epoch": 0.8613407485344957, + "grad_norm": 5.430506099769504, + "learning_rate": 1.9833393761063523e-07, + "loss": 0.9777, + "step": 11461 + }, + { + "epoch": 0.8614159026003306, + "grad_norm": 1.719774669037186, + "learning_rate": 1.9812262081310237e-07, + "loss": 1.0672, + "step": 11462 + }, + { + "epoch": 0.8614910566661657, + "grad_norm": 1.52822834749763, + "learning_rate": 1.9791141078370366e-07, + "loss": 1.0493, + "step": 11463 + }, + { + "epoch": 0.8615662107320006, + "grad_norm": 2.3419937674604796, + "learning_rate": 1.977003075349548e-07, + "loss": 1.0271, + "step": 11464 + }, + { + "epoch": 0.8616413647978356, + "grad_norm": 1.521118825305121, + "learning_rate": 1.9748931107936473e-07, + "loss": 0.9461, + "step": 11465 + }, + { + "epoch": 0.8617165188636705, + "grad_norm": 2.148347055091964, + "learning_rate": 1.9727842142943453e-07, + "loss": 1.0555, + "step": 11466 + }, + { + "epoch": 0.8617916729295055, + "grad_norm": 1.3923078919490406, + "learning_rate": 1.970676385976613e-07, + "loss": 0.9961, + "step": 11467 + }, + { + "epoch": 0.8618668269953405, + "grad_norm": 1.9387037168151862, + "learning_rate": 1.9685696259653438e-07, + "loss": 0.8704, + "step": 11468 + }, + { + "epoch": 0.8619419810611754, + "grad_norm": 2.2490234375, + "learning_rate": 1.9664639343853717e-07, + "loss": 1.1218, + "step": 11469 + }, + { + "epoch": 0.8620171351270104, + "grad_norm": 0.7018748295444951, + "learning_rate": 1.9643593113614632e-07, + "loss": 0.7983, + "step": 11470 + }, + { + "epoch": 0.8620922891928453, + "grad_norm": 2.347007623777941, + "learning_rate": 1.9622557570183318e-07, + "loss": 0.9849, + "step": 11471 + }, + { + "epoch": 0.8621674432586803, + "grad_norm": 1.7771053615455488, + "learning_rate": 1.960153271480618e-07, + "loss": 1.0661, + "step": 11472 + }, + { + "epoch": 0.8622425973245152, + "grad_norm": 1.9528651560550117, + "learning_rate": 1.958051854872902e-07, + "loss": 0.9582, + "step": 11473 + }, + { + "epoch": 0.8623177513903503, + "grad_norm": 1.3881080223120446, + "learning_rate": 1.9559515073196952e-07, + "loss": 0.9767, + "step": 11474 + }, + { + "epoch": 0.8623929054561852, + "grad_norm": 1.6299337170001622, + "learning_rate": 1.9538522289454674e-07, + "loss": 0.9687, + "step": 11475 + }, + { + "epoch": 0.8624680595220201, + "grad_norm": 1.6193264439324897, + "learning_rate": 1.9517540198745896e-07, + "loss": 0.9372, + "step": 11476 + }, + { + "epoch": 0.8625432135878551, + "grad_norm": 2.75513516324721, + "learning_rate": 1.9496568802314007e-07, + "loss": 1.0522, + "step": 11477 + }, + { + "epoch": 0.86261836765369, + "grad_norm": 1.9731049584430298, + "learning_rate": 1.947560810140161e-07, + "loss": 0.8191, + "step": 11478 + }, + { + "epoch": 0.8626935217195251, + "grad_norm": 1.4540561440921458, + "learning_rate": 1.945465809725071e-07, + "loss": 0.9912, + "step": 11479 + }, + { + "epoch": 0.86276867578536, + "grad_norm": 1.3176849674771736, + "learning_rate": 1.943371879110265e-07, + "loss": 0.9792, + "step": 11480 + }, + { + "epoch": 0.8628438298511949, + "grad_norm": 1.6427123823420378, + "learning_rate": 1.9412790184198168e-07, + "loss": 0.9385, + "step": 11481 + }, + { + "epoch": 0.8629189839170299, + "grad_norm": 1.6364488766124017, + "learning_rate": 1.9391872277777456e-07, + "loss": 0.964, + "step": 11482 + }, + { + "epoch": 0.8629941379828648, + "grad_norm": 1.8746941634929473, + "learning_rate": 1.937096507307987e-07, + "loss": 0.8793, + "step": 11483 + }, + { + "epoch": 0.8630692920486999, + "grad_norm": 1.687706122348858, + "learning_rate": 1.9350068571344246e-07, + "loss": 0.9601, + "step": 11484 + }, + { + "epoch": 0.8631444461145348, + "grad_norm": 1.9898541717196472, + "learning_rate": 1.9329182773808904e-07, + "loss": 0.8457, + "step": 11485 + }, + { + "epoch": 0.8632196001803698, + "grad_norm": 1.9765831263977498, + "learning_rate": 1.930830768171121e-07, + "loss": 0.9718, + "step": 11486 + }, + { + "epoch": 0.8632947542462047, + "grad_norm": 2.2170863832155137, + "learning_rate": 1.928744329628822e-07, + "loss": 0.884, + "step": 11487 + }, + { + "epoch": 0.8633699083120396, + "grad_norm": 2.0819061986073737, + "learning_rate": 1.9266589618776296e-07, + "loss": 0.9049, + "step": 11488 + }, + { + "epoch": 0.8634450623778747, + "grad_norm": 1.3020972848780556, + "learning_rate": 1.924574665041092e-07, + "loss": 0.9888, + "step": 11489 + }, + { + "epoch": 0.8635202164437096, + "grad_norm": 2.0872862380904253, + "learning_rate": 1.9224914392427238e-07, + "loss": 1.0127, + "step": 11490 + }, + { + "epoch": 0.8635953705095446, + "grad_norm": 1.3931927494981926, + "learning_rate": 1.9204092846059595e-07, + "loss": 0.9878, + "step": 11491 + }, + { + "epoch": 0.8636705245753795, + "grad_norm": 1.6680600143586186, + "learning_rate": 1.9183282012541824e-07, + "loss": 0.9244, + "step": 11492 + }, + { + "epoch": 0.8637456786412145, + "grad_norm": 2.1676948993003227, + "learning_rate": 1.9162481893106918e-07, + "loss": 1.0128, + "step": 11493 + }, + { + "epoch": 0.8638208327070495, + "grad_norm": 1.7396320024028586, + "learning_rate": 1.914169248898747e-07, + "loss": 0.931, + "step": 11494 + }, + { + "epoch": 0.8638959867728844, + "grad_norm": 1.6412060389910386, + "learning_rate": 1.9120913801415294e-07, + "loss": 0.859, + "step": 11495 + }, + { + "epoch": 0.8639711408387194, + "grad_norm": 1.8116091808169845, + "learning_rate": 1.9100145831621627e-07, + "loss": 0.9761, + "step": 11496 + }, + { + "epoch": 0.8640462949045543, + "grad_norm": 1.9659966485896219, + "learning_rate": 1.9079388580836975e-07, + "loss": 0.9144, + "step": 11497 + }, + { + "epoch": 0.8641214489703893, + "grad_norm": 1.5154823944015612, + "learning_rate": 1.9058642050291374e-07, + "loss": 0.9155, + "step": 11498 + }, + { + "epoch": 0.8641966030362243, + "grad_norm": 1.5662143808563205, + "learning_rate": 1.9037906241214109e-07, + "loss": 0.9154, + "step": 11499 + }, + { + "epoch": 0.8642717571020592, + "grad_norm": 1.9185818005129205, + "learning_rate": 1.901718115483384e-07, + "loss": 0.9489, + "step": 11500 + }, + { + "epoch": 0.8643469111678942, + "grad_norm": 1.7026189349695, + "learning_rate": 1.8996466792378584e-07, + "loss": 0.9892, + "step": 11501 + }, + { + "epoch": 0.8644220652337291, + "grad_norm": 1.6625206479963002, + "learning_rate": 1.8975763155075875e-07, + "loss": 0.9596, + "step": 11502 + }, + { + "epoch": 0.8644972192995641, + "grad_norm": 1.881497442251348, + "learning_rate": 1.8955070244152284e-07, + "loss": 0.8688, + "step": 11503 + }, + { + "epoch": 0.864572373365399, + "grad_norm": 1.7182690034079384, + "learning_rate": 1.893438806083405e-07, + "loss": 1.0294, + "step": 11504 + }, + { + "epoch": 0.8646475274312341, + "grad_norm": 1.4834071969633165, + "learning_rate": 1.8913716606346776e-07, + "loss": 0.9711, + "step": 11505 + }, + { + "epoch": 0.864722681497069, + "grad_norm": 1.7271941397552937, + "learning_rate": 1.8893055881915143e-07, + "loss": 0.9703, + "step": 11506 + }, + { + "epoch": 0.8647978355629039, + "grad_norm": 1.5153443381405975, + "learning_rate": 1.8872405888763443e-07, + "loss": 0.9787, + "step": 11507 + }, + { + "epoch": 0.8648729896287389, + "grad_norm": 1.6706751145232737, + "learning_rate": 1.8851766628115273e-07, + "loss": 1.0554, + "step": 11508 + }, + { + "epoch": 0.8649481436945738, + "grad_norm": 1.6793662540277616, + "learning_rate": 1.8831138101193655e-07, + "loss": 0.9833, + "step": 11509 + }, + { + "epoch": 0.8650232977604089, + "grad_norm": 1.5445607917210313, + "learning_rate": 1.881052030922079e-07, + "loss": 0.8754, + "step": 11510 + }, + { + "epoch": 0.8650984518262438, + "grad_norm": 1.5836064538130905, + "learning_rate": 1.8789913253418433e-07, + "loss": 0.9297, + "step": 11511 + }, + { + "epoch": 0.8651736058920788, + "grad_norm": 0.6621346257889023, + "learning_rate": 1.876931693500763e-07, + "loss": 0.7814, + "step": 11512 + }, + { + "epoch": 0.8652487599579137, + "grad_norm": 1.6449975011974707, + "learning_rate": 1.874873135520878e-07, + "loss": 0.9937, + "step": 11513 + }, + { + "epoch": 0.8653239140237486, + "grad_norm": 1.4384289932413807, + "learning_rate": 1.87281565152416e-07, + "loss": 0.9321, + "step": 11514 + }, + { + "epoch": 0.8653990680895837, + "grad_norm": 2.69724467584981, + "learning_rate": 1.8707592416325336e-07, + "loss": 0.829, + "step": 11515 + }, + { + "epoch": 0.8654742221554186, + "grad_norm": 1.8390736517167001, + "learning_rate": 1.8687039059678433e-07, + "loss": 0.9835, + "step": 11516 + }, + { + "epoch": 0.8655493762212536, + "grad_norm": 2.654293461699884, + "learning_rate": 1.8666496446518765e-07, + "loss": 0.9476, + "step": 11517 + }, + { + "epoch": 0.8656245302870885, + "grad_norm": 0.6906781862242098, + "learning_rate": 1.8645964578063512e-07, + "loss": 0.8447, + "step": 11518 + }, + { + "epoch": 0.8656996843529235, + "grad_norm": 2.8309866405887356, + "learning_rate": 1.8625443455529366e-07, + "loss": 0.9474, + "step": 11519 + }, + { + "epoch": 0.8657748384187585, + "grad_norm": 0.7289410696268519, + "learning_rate": 1.860493308013218e-07, + "loss": 0.8388, + "step": 11520 + }, + { + "epoch": 0.8658499924845934, + "grad_norm": 1.5909710011264098, + "learning_rate": 1.8584433453087335e-07, + "loss": 1.022, + "step": 11521 + }, + { + "epoch": 0.8659251465504284, + "grad_norm": 2.1090678803836926, + "learning_rate": 1.8563944575609503e-07, + "loss": 1.0298, + "step": 11522 + }, + { + "epoch": 0.8660003006162633, + "grad_norm": 1.5467747357173858, + "learning_rate": 1.8543466448912713e-07, + "loss": 1.0212, + "step": 11523 + }, + { + "epoch": 0.8660754546820983, + "grad_norm": 2.2227314431824556, + "learning_rate": 1.8522999074210355e-07, + "loss": 1.0536, + "step": 11524 + }, + { + "epoch": 0.8661506087479333, + "grad_norm": 1.8060256158653296, + "learning_rate": 1.8502542452715207e-07, + "loss": 0.9342, + "step": 11525 + }, + { + "epoch": 0.8662257628137682, + "grad_norm": 1.5939863441102502, + "learning_rate": 1.8482096585639506e-07, + "loss": 0.8896, + "step": 11526 + }, + { + "epoch": 0.8663009168796032, + "grad_norm": 7.399039448436634, + "learning_rate": 1.846166147419459e-07, + "loss": 1.0107, + "step": 11527 + }, + { + "epoch": 0.8663760709454381, + "grad_norm": 1.6938655321195688, + "learning_rate": 1.8441237119591403e-07, + "loss": 0.9513, + "step": 11528 + }, + { + "epoch": 0.8664512250112731, + "grad_norm": 1.5692866376386332, + "learning_rate": 1.8420823523040197e-07, + "loss": 0.8978, + "step": 11529 + }, + { + "epoch": 0.866526379077108, + "grad_norm": 1.4877340757696917, + "learning_rate": 1.8400420685750452e-07, + "loss": 0.9042, + "step": 11530 + }, + { + "epoch": 0.8666015331429431, + "grad_norm": 3.291941200750958, + "learning_rate": 1.8380028608931152e-07, + "loss": 1.0018, + "step": 11531 + }, + { + "epoch": 0.866676687208778, + "grad_norm": 1.8448596783132358, + "learning_rate": 1.8359647293790713e-07, + "loss": 0.9343, + "step": 11532 + }, + { + "epoch": 0.8667518412746129, + "grad_norm": 2.340732615789207, + "learning_rate": 1.8339276741536657e-07, + "loss": 0.9814, + "step": 11533 + }, + { + "epoch": 0.8668269953404479, + "grad_norm": 1.5070597296397321, + "learning_rate": 1.8318916953376106e-07, + "loss": 0.9523, + "step": 11534 + }, + { + "epoch": 0.8669021494062829, + "grad_norm": 1.582340690584974, + "learning_rate": 1.8298567930515386e-07, + "loss": 0.9861, + "step": 11535 + }, + { + "epoch": 0.8669773034721179, + "grad_norm": 0.6770718549097784, + "learning_rate": 1.8278229674160373e-07, + "loss": 0.833, + "step": 11536 + }, + { + "epoch": 0.8670524575379528, + "grad_norm": 4.59065609500084, + "learning_rate": 1.825790218551604e-07, + "loss": 0.9981, + "step": 11537 + }, + { + "epoch": 0.8671276116037878, + "grad_norm": 2.5046902528629893, + "learning_rate": 1.8237585465786976e-07, + "loss": 0.9903, + "step": 11538 + }, + { + "epoch": 0.8672027656696227, + "grad_norm": 0.865125111037943, + "learning_rate": 1.8217279516176976e-07, + "loss": 0.9386, + "step": 11539 + }, + { + "epoch": 0.8672779197354576, + "grad_norm": 1.8550626611536982, + "learning_rate": 1.8196984337889276e-07, + "loss": 0.9795, + "step": 11540 + }, + { + "epoch": 0.8673530738012927, + "grad_norm": 1.5986067129531811, + "learning_rate": 1.8176699932126383e-07, + "loss": 1.0211, + "step": 11541 + }, + { + "epoch": 0.8674282278671276, + "grad_norm": 2.6681479174661136, + "learning_rate": 1.8156426300090288e-07, + "loss": 0.9187, + "step": 11542 + }, + { + "epoch": 0.8675033819329626, + "grad_norm": 1.5286309557249778, + "learning_rate": 1.8136163442982277e-07, + "loss": 1.0046, + "step": 11543 + }, + { + "epoch": 0.8675785359987975, + "grad_norm": 2.174226763727989, + "learning_rate": 1.811591136200299e-07, + "loss": 0.9879, + "step": 11544 + }, + { + "epoch": 0.8676536900646324, + "grad_norm": 1.7244212133250902, + "learning_rate": 1.8095670058352374e-07, + "loss": 0.9763, + "step": 11545 + }, + { + "epoch": 0.8677288441304675, + "grad_norm": 1.7435335852208207, + "learning_rate": 1.8075439533229964e-07, + "loss": 0.9474, + "step": 11546 + }, + { + "epoch": 0.8678039981963024, + "grad_norm": 1.6277833722958355, + "learning_rate": 1.8055219787834308e-07, + "loss": 1.0277, + "step": 11547 + }, + { + "epoch": 0.8678791522621374, + "grad_norm": 2.2819191983199434, + "learning_rate": 1.8035010823363606e-07, + "loss": 0.9321, + "step": 11548 + }, + { + "epoch": 0.8679543063279723, + "grad_norm": 1.6360719263032784, + "learning_rate": 1.8014812641015364e-07, + "loss": 0.9366, + "step": 11549 + }, + { + "epoch": 0.8680294603938074, + "grad_norm": 1.4305672362178885, + "learning_rate": 1.7994625241986293e-07, + "loss": 0.9246, + "step": 11550 + }, + { + "epoch": 0.8681046144596423, + "grad_norm": 2.1156342499163094, + "learning_rate": 1.7974448627472615e-07, + "loss": 0.9673, + "step": 11551 + }, + { + "epoch": 0.8681797685254772, + "grad_norm": 1.9241851940009251, + "learning_rate": 1.795428279866986e-07, + "loss": 1.0597, + "step": 11552 + }, + { + "epoch": 0.8682549225913122, + "grad_norm": 1.8537363417386565, + "learning_rate": 1.793412775677303e-07, + "loss": 0.9718, + "step": 11553 + }, + { + "epoch": 0.8683300766571471, + "grad_norm": 2.0971843963086223, + "learning_rate": 1.7913983502976237e-07, + "loss": 0.8755, + "step": 11554 + }, + { + "epoch": 0.8684052307229821, + "grad_norm": 2.409791000611044, + "learning_rate": 1.7893850038473192e-07, + "loss": 0.9375, + "step": 11555 + }, + { + "epoch": 0.8684803847888171, + "grad_norm": 2.1755335887717346, + "learning_rate": 1.787372736445687e-07, + "loss": 0.9686, + "step": 11556 + }, + { + "epoch": 0.8685555388546521, + "grad_norm": 0.8008672156362712, + "learning_rate": 1.7853615482119633e-07, + "loss": 0.8447, + "step": 11557 + }, + { + "epoch": 0.868630692920487, + "grad_norm": 2.725778548987448, + "learning_rate": 1.7833514392653104e-07, + "loss": 1.0006, + "step": 11558 + }, + { + "epoch": 0.8687058469863219, + "grad_norm": 1.2373776674011625, + "learning_rate": 1.7813424097248443e-07, + "loss": 0.9725, + "step": 11559 + }, + { + "epoch": 0.8687810010521569, + "grad_norm": 2.048650539327681, + "learning_rate": 1.779334459709607e-07, + "loss": 0.9656, + "step": 11560 + }, + { + "epoch": 0.8688561551179919, + "grad_norm": 1.8877912991309995, + "learning_rate": 1.777327589338571e-07, + "loss": 0.9935, + "step": 11561 + }, + { + "epoch": 0.8689313091838269, + "grad_norm": 2.0214695142116805, + "learning_rate": 1.7753217987306536e-07, + "loss": 0.8513, + "step": 11562 + }, + { + "epoch": 0.8690064632496618, + "grad_norm": 0.7400206405750943, + "learning_rate": 1.7733170880047132e-07, + "loss": 0.8274, + "step": 11563 + }, + { + "epoch": 0.8690816173154968, + "grad_norm": 2.39299621665901, + "learning_rate": 1.771313457279522e-07, + "loss": 1.0019, + "step": 11564 + }, + { + "epoch": 0.8691567713813317, + "grad_norm": 1.6974455603071041, + "learning_rate": 1.7693109066738154e-07, + "loss": 0.9483, + "step": 11565 + }, + { + "epoch": 0.8692319254471667, + "grad_norm": 2.057526690100241, + "learning_rate": 1.767309436306248e-07, + "loss": 0.9208, + "step": 11566 + }, + { + "epoch": 0.8693070795130017, + "grad_norm": 1.7540575399758533, + "learning_rate": 1.7653090462954112e-07, + "loss": 0.9666, + "step": 11567 + }, + { + "epoch": 0.8693822335788366, + "grad_norm": 1.5985790469683154, + "learning_rate": 1.763309736759837e-07, + "loss": 0.9989, + "step": 11568 + }, + { + "epoch": 0.8694573876446716, + "grad_norm": 1.3298455705833052, + "learning_rate": 1.7613115078179952e-07, + "loss": 0.9054, + "step": 11569 + }, + { + "epoch": 0.8695325417105065, + "grad_norm": 1.5494309703693474, + "learning_rate": 1.759314359588293e-07, + "loss": 0.8693, + "step": 11570 + }, + { + "epoch": 0.8696076957763414, + "grad_norm": 1.65410539944409, + "learning_rate": 1.7573182921890562e-07, + "loss": 0.9058, + "step": 11571 + }, + { + "epoch": 0.8696828498421765, + "grad_norm": 1.6509957372331954, + "learning_rate": 1.7553233057385698e-07, + "loss": 1.0378, + "step": 11572 + }, + { + "epoch": 0.8697580039080114, + "grad_norm": 3.037573599218051, + "learning_rate": 1.7533294003550436e-07, + "loss": 1.0177, + "step": 11573 + }, + { + "epoch": 0.8698331579738464, + "grad_norm": 3.9352652853665324, + "learning_rate": 1.7513365761566167e-07, + "loss": 1.0235, + "step": 11574 + }, + { + "epoch": 0.8699083120396813, + "grad_norm": 1.5692447809166299, + "learning_rate": 1.7493448332613746e-07, + "loss": 0.9239, + "step": 11575 + }, + { + "epoch": 0.8699834661055164, + "grad_norm": 1.5989977141096179, + "learning_rate": 1.7473541717873474e-07, + "loss": 0.9814, + "step": 11576 + }, + { + "epoch": 0.8700586201713513, + "grad_norm": 1.906354870022146, + "learning_rate": 1.7453645918524695e-07, + "loss": 0.9105, + "step": 11577 + }, + { + "epoch": 0.8701337742371862, + "grad_norm": 1.3082559177958089, + "learning_rate": 1.7433760935746465e-07, + "loss": 0.951, + "step": 11578 + }, + { + "epoch": 0.8702089283030212, + "grad_norm": 2.333686688324964, + "learning_rate": 1.7413886770716935e-07, + "loss": 0.9531, + "step": 11579 + }, + { + "epoch": 0.8702840823688561, + "grad_norm": 1.5100477184561985, + "learning_rate": 1.7394023424613868e-07, + "loss": 0.9982, + "step": 11580 + }, + { + "epoch": 0.8703592364346912, + "grad_norm": 2.0119783993162517, + "learning_rate": 1.7374170898614106e-07, + "loss": 0.9122, + "step": 11581 + }, + { + "epoch": 0.8704343905005261, + "grad_norm": 2.283257319816335, + "learning_rate": 1.7354329193894058e-07, + "loss": 1.0179, + "step": 11582 + }, + { + "epoch": 0.8705095445663611, + "grad_norm": 1.2895453271313568, + "learning_rate": 1.7334498311629385e-07, + "loss": 0.8834, + "step": 11583 + }, + { + "epoch": 0.870584698632196, + "grad_norm": 1.5096313890639557, + "learning_rate": 1.7314678252995173e-07, + "loss": 1.0052, + "step": 11584 + }, + { + "epoch": 0.8706598526980309, + "grad_norm": 1.7620150365588394, + "learning_rate": 1.7294869019165792e-07, + "loss": 0.8662, + "step": 11585 + }, + { + "epoch": 0.870735006763866, + "grad_norm": 1.5518645915413292, + "learning_rate": 1.72750706113151e-07, + "loss": 1.0109, + "step": 11586 + }, + { + "epoch": 0.8708101608297009, + "grad_norm": 2.003122395768161, + "learning_rate": 1.7255283030616142e-07, + "loss": 0.833, + "step": 11587 + }, + { + "epoch": 0.8708853148955359, + "grad_norm": 1.8413432301875232, + "learning_rate": 1.7235506278241463e-07, + "loss": 0.8148, + "step": 11588 + }, + { + "epoch": 0.8709604689613708, + "grad_norm": 1.574494595651325, + "learning_rate": 1.721574035536284e-07, + "loss": 0.9894, + "step": 11589 + }, + { + "epoch": 0.8710356230272057, + "grad_norm": 1.9751491393745984, + "learning_rate": 1.7195985263151603e-07, + "loss": 1.0542, + "step": 11590 + }, + { + "epoch": 0.8711107770930407, + "grad_norm": 2.2162774506755247, + "learning_rate": 1.7176241002778168e-07, + "loss": 1.0757, + "step": 11591 + }, + { + "epoch": 0.8711859311588757, + "grad_norm": 2.5983492048886836, + "learning_rate": 1.7156507575412537e-07, + "loss": 0.8944, + "step": 11592 + }, + { + "epoch": 0.8712610852247107, + "grad_norm": 2.199037588460053, + "learning_rate": 1.7136784982224062e-07, + "loss": 0.9115, + "step": 11593 + }, + { + "epoch": 0.8713362392905456, + "grad_norm": 1.960077530339557, + "learning_rate": 1.711707322438123e-07, + "loss": 0.9393, + "step": 11594 + }, + { + "epoch": 0.8714113933563806, + "grad_norm": 1.6014533028941893, + "learning_rate": 1.709737230305215e-07, + "loss": 0.947, + "step": 11595 + }, + { + "epoch": 0.8714865474222155, + "grad_norm": 0.85049883005554, + "learning_rate": 1.707768221940411e-07, + "loss": 0.9002, + "step": 11596 + }, + { + "epoch": 0.8715617014880505, + "grad_norm": 1.4043415258787204, + "learning_rate": 1.7058002974603936e-07, + "loss": 1.0079, + "step": 11597 + }, + { + "epoch": 0.8716368555538855, + "grad_norm": 2.141011335497037, + "learning_rate": 1.7038334569817536e-07, + "loss": 0.9958, + "step": 11598 + }, + { + "epoch": 0.8717120096197204, + "grad_norm": 1.9081018551804414, + "learning_rate": 1.7018677006210446e-07, + "loss": 0.9025, + "step": 11599 + }, + { + "epoch": 0.8717871636855554, + "grad_norm": 1.9362181453165477, + "learning_rate": 1.6999030284947424e-07, + "loss": 0.9435, + "step": 11600 + }, + { + "epoch": 0.8718623177513903, + "grad_norm": 1.6664390964910307, + "learning_rate": 1.6979394407192625e-07, + "loss": 0.9734, + "step": 11601 + }, + { + "epoch": 0.8719374718172254, + "grad_norm": 3.3536852536543456, + "learning_rate": 1.6959769374109523e-07, + "loss": 1.0316, + "step": 11602 + }, + { + "epoch": 0.8720126258830603, + "grad_norm": 1.5931702849760103, + "learning_rate": 1.6940155186861004e-07, + "loss": 0.9734, + "step": 11603 + }, + { + "epoch": 0.8720877799488952, + "grad_norm": 1.9000000351353692, + "learning_rate": 1.6920551846609276e-07, + "loss": 0.871, + "step": 11604 + }, + { + "epoch": 0.8721629340147302, + "grad_norm": 2.0477463127161712, + "learning_rate": 1.690095935451592e-07, + "loss": 0.9266, + "step": 11605 + }, + { + "epoch": 0.8722380880805651, + "grad_norm": 1.4604079975010542, + "learning_rate": 1.6881377711741807e-07, + "loss": 1.0203, + "step": 11606 + }, + { + "epoch": 0.8723132421464002, + "grad_norm": 1.8297536393353644, + "learning_rate": 1.686180691944734e-07, + "loss": 1.0149, + "step": 11607 + }, + { + "epoch": 0.8723883962122351, + "grad_norm": 1.8404368357826728, + "learning_rate": 1.684224697879204e-07, + "loss": 0.9253, + "step": 11608 + }, + { + "epoch": 0.8724635502780701, + "grad_norm": 1.3812748609005725, + "learning_rate": 1.6822697890935e-07, + "loss": 0.9431, + "step": 11609 + }, + { + "epoch": 0.872538704343905, + "grad_norm": 0.7117871248500459, + "learning_rate": 1.6803159657034537e-07, + "loss": 0.9037, + "step": 11610 + }, + { + "epoch": 0.8726138584097399, + "grad_norm": 1.9048693300222008, + "learning_rate": 1.6783632278248371e-07, + "loss": 0.9179, + "step": 11611 + }, + { + "epoch": 0.872689012475575, + "grad_norm": 2.0565912398361994, + "learning_rate": 1.6764115755733532e-07, + "loss": 0.9188, + "step": 11612 + }, + { + "epoch": 0.8727641665414099, + "grad_norm": 0.7819651191790862, + "learning_rate": 1.6744610090646517e-07, + "loss": 0.8211, + "step": 11613 + }, + { + "epoch": 0.8728393206072449, + "grad_norm": 1.8483053281065287, + "learning_rate": 1.6725115284143132e-07, + "loss": 1.0164, + "step": 11614 + }, + { + "epoch": 0.8729144746730798, + "grad_norm": 1.8628332211206384, + "learning_rate": 1.670563133737841e-07, + "loss": 0.9462, + "step": 11615 + }, + { + "epoch": 0.8729896287389147, + "grad_norm": 1.6392709866742823, + "learning_rate": 1.6686158251506943e-07, + "loss": 0.9511, + "step": 11616 + }, + { + "epoch": 0.8730647828047498, + "grad_norm": 1.5397107458324466, + "learning_rate": 1.6666696027682602e-07, + "loss": 0.9124, + "step": 11617 + }, + { + "epoch": 0.8731399368705847, + "grad_norm": 5.618521286634434, + "learning_rate": 1.664724466705847e-07, + "loss": 0.963, + "step": 11618 + }, + { + "epoch": 0.8732150909364197, + "grad_norm": 1.7085415162804807, + "learning_rate": 1.6627804170787196e-07, + "loss": 0.9567, + "step": 11619 + }, + { + "epoch": 0.8732902450022546, + "grad_norm": 2.088488101668709, + "learning_rate": 1.6608374540020776e-07, + "loss": 0.8401, + "step": 11620 + }, + { + "epoch": 0.8733653990680896, + "grad_norm": 1.3654359769614928, + "learning_rate": 1.658895577591035e-07, + "loss": 1.0197, + "step": 11621 + }, + { + "epoch": 0.8734405531339245, + "grad_norm": 2.3263637357905087, + "learning_rate": 1.656954787960665e-07, + "loss": 0.9901, + "step": 11622 + }, + { + "epoch": 0.8735157071997595, + "grad_norm": 2.4086026666991533, + "learning_rate": 1.6550150852259615e-07, + "loss": 0.9313, + "step": 11623 + }, + { + "epoch": 0.8735908612655945, + "grad_norm": 1.9946436681866537, + "learning_rate": 1.6530764695018684e-07, + "loss": 1.0196, + "step": 11624 + }, + { + "epoch": 0.8736660153314294, + "grad_norm": 1.7033257322396862, + "learning_rate": 1.6511389409032428e-07, + "loss": 0.9456, + "step": 11625 + }, + { + "epoch": 0.8737411693972644, + "grad_norm": 3.2359535677376012, + "learning_rate": 1.6492024995449017e-07, + "loss": 1.0873, + "step": 11626 + }, + { + "epoch": 0.8738163234630993, + "grad_norm": 2.446017227754723, + "learning_rate": 1.6472671455415821e-07, + "loss": 1.0866, + "step": 11627 + }, + { + "epoch": 0.8738914775289344, + "grad_norm": 1.7347217934801957, + "learning_rate": 1.645332879007959e-07, + "loss": 0.9974, + "step": 11628 + }, + { + "epoch": 0.8739666315947693, + "grad_norm": 1.5482051794413187, + "learning_rate": 1.6433997000586475e-07, + "loss": 0.9021, + "step": 11629 + }, + { + "epoch": 0.8740417856606042, + "grad_norm": 1.8756936379737381, + "learning_rate": 1.6414676088081937e-07, + "loss": 0.9109, + "step": 11630 + }, + { + "epoch": 0.8741169397264392, + "grad_norm": 2.676609685140083, + "learning_rate": 1.6395366053710902e-07, + "loss": 0.9575, + "step": 11631 + }, + { + "epoch": 0.8741920937922741, + "grad_norm": 1.5320900928609982, + "learning_rate": 1.637606689861748e-07, + "loss": 0.8963, + "step": 11632 + }, + { + "epoch": 0.8742672478581092, + "grad_norm": 2.1835278004631653, + "learning_rate": 1.63567786239452e-07, + "loss": 1.0288, + "step": 11633 + }, + { + "epoch": 0.8743424019239441, + "grad_norm": 1.4366756231525921, + "learning_rate": 1.6337501230837059e-07, + "loss": 1.0136, + "step": 11634 + }, + { + "epoch": 0.874417555989779, + "grad_norm": 2.293260494043902, + "learning_rate": 1.631823472043521e-07, + "loss": 0.9121, + "step": 11635 + }, + { + "epoch": 0.874492710055614, + "grad_norm": 2.217592286573209, + "learning_rate": 1.6298979093881292e-07, + "loss": 0.9794, + "step": 11636 + }, + { + "epoch": 0.8745678641214489, + "grad_norm": 1.7695131953583627, + "learning_rate": 1.62797343523164e-07, + "loss": 1.0599, + "step": 11637 + }, + { + "epoch": 0.874643018187284, + "grad_norm": 1.8608352312738032, + "learning_rate": 1.626050049688066e-07, + "loss": 0.8858, + "step": 11638 + }, + { + "epoch": 0.8747181722531189, + "grad_norm": 1.9139199962896216, + "learning_rate": 1.6241277528713916e-07, + "loss": 1.0981, + "step": 11639 + }, + { + "epoch": 0.8747933263189539, + "grad_norm": 1.7218215546387716, + "learning_rate": 1.6222065448955081e-07, + "loss": 0.9816, + "step": 11640 + }, + { + "epoch": 0.8748684803847888, + "grad_norm": 2.0925319032048653, + "learning_rate": 1.6202864258742688e-07, + "loss": 0.9307, + "step": 11641 + }, + { + "epoch": 0.8749436344506237, + "grad_norm": 3.8993699176141945, + "learning_rate": 1.6183673959214316e-07, + "loss": 0.8502, + "step": 11642 + }, + { + "epoch": 0.8750187885164588, + "grad_norm": 2.532815613882985, + "learning_rate": 1.6164494551507168e-07, + "loss": 0.8954, + "step": 11643 + }, + { + "epoch": 0.8750939425822937, + "grad_norm": 2.9829198675080297, + "learning_rate": 1.6145326036757667e-07, + "loss": 0.9752, + "step": 11644 + }, + { + "epoch": 0.8751690966481287, + "grad_norm": 2.0404922325324817, + "learning_rate": 1.6126168416101638e-07, + "loss": 1.0018, + "step": 11645 + }, + { + "epoch": 0.8752442507139636, + "grad_norm": 2.085717464657356, + "learning_rate": 1.6107021690674193e-07, + "loss": 1.0438, + "step": 11646 + }, + { + "epoch": 0.8753194047797986, + "grad_norm": 1.6141863263058311, + "learning_rate": 1.608788586160992e-07, + "loss": 0.9638, + "step": 11647 + }, + { + "epoch": 0.8753945588456336, + "grad_norm": 2.1098576876480757, + "learning_rate": 1.6068760930042657e-07, + "loss": 0.9485, + "step": 11648 + }, + { + "epoch": 0.8754697129114685, + "grad_norm": 0.7697788750218475, + "learning_rate": 1.604964689710564e-07, + "loss": 0.796, + "step": 11649 + }, + { + "epoch": 0.8755448669773035, + "grad_norm": 1.4432874359662364, + "learning_rate": 1.6030543763931382e-07, + "loss": 0.9764, + "step": 11650 + }, + { + "epoch": 0.8756200210431384, + "grad_norm": 1.5547949979029874, + "learning_rate": 1.6011451531651953e-07, + "loss": 0.9115, + "step": 11651 + }, + { + "epoch": 0.8756951751089734, + "grad_norm": 1.8186154292813974, + "learning_rate": 1.5992370201398496e-07, + "loss": 0.904, + "step": 11652 + }, + { + "epoch": 0.8757703291748083, + "grad_norm": 1.4302062880836452, + "learning_rate": 1.5973299774301707e-07, + "loss": 1.0064, + "step": 11653 + }, + { + "epoch": 0.8758454832406433, + "grad_norm": 2.33754651844431, + "learning_rate": 1.5954240251491659e-07, + "loss": 0.9068, + "step": 11654 + }, + { + "epoch": 0.8759206373064783, + "grad_norm": 1.916206560333046, + "learning_rate": 1.59351916340976e-07, + "loss": 1.0175, + "step": 11655 + }, + { + "epoch": 0.8759957913723132, + "grad_norm": 1.8772697697674317, + "learning_rate": 1.5916153923248254e-07, + "loss": 1.0368, + "step": 11656 + }, + { + "epoch": 0.8760709454381482, + "grad_norm": 2.0876147210126375, + "learning_rate": 1.5897127120071674e-07, + "loss": 1.0176, + "step": 11657 + }, + { + "epoch": 0.8761460995039831, + "grad_norm": 1.819221267731217, + "learning_rate": 1.5878111225695357e-07, + "loss": 0.9522, + "step": 11658 + }, + { + "epoch": 0.8762212535698182, + "grad_norm": 2.674011892909991, + "learning_rate": 1.5859106241245934e-07, + "loss": 0.9547, + "step": 11659 + }, + { + "epoch": 0.8762964076356531, + "grad_norm": 1.6823295927545936, + "learning_rate": 1.584011216784962e-07, + "loss": 0.9329, + "step": 11660 + }, + { + "epoch": 0.876371561701488, + "grad_norm": 1.6654982842488357, + "learning_rate": 1.582112900663186e-07, + "loss": 0.9129, + "step": 11661 + }, + { + "epoch": 0.876446715767323, + "grad_norm": 1.864268427279563, + "learning_rate": 1.5802156758717478e-07, + "loss": 0.905, + "step": 11662 + }, + { + "epoch": 0.8765218698331579, + "grad_norm": 1.932354832633875, + "learning_rate": 1.578319542523061e-07, + "loss": 1.0136, + "step": 11663 + }, + { + "epoch": 0.876597023898993, + "grad_norm": 0.7107222514175092, + "learning_rate": 1.5764245007294875e-07, + "loss": 0.8427, + "step": 11664 + }, + { + "epoch": 0.8766721779648279, + "grad_norm": 1.6160378908020865, + "learning_rate": 1.574530550603308e-07, + "loss": 0.8688, + "step": 11665 + }, + { + "epoch": 0.8767473320306629, + "grad_norm": 2.954951937263631, + "learning_rate": 1.5726376922567486e-07, + "loss": 0.9845, + "step": 11666 + }, + { + "epoch": 0.8768224860964978, + "grad_norm": 1.5092331118660096, + "learning_rate": 1.5707459258019684e-07, + "loss": 0.9431, + "step": 11667 + }, + { + "epoch": 0.8768976401623327, + "grad_norm": 23.243155179435355, + "learning_rate": 1.5688552513510688e-07, + "loss": 0.9745, + "step": 11668 + }, + { + "epoch": 0.8769727942281678, + "grad_norm": 1.9766746157485808, + "learning_rate": 1.566965669016065e-07, + "loss": 1.0247, + "step": 11669 + }, + { + "epoch": 0.8770479482940027, + "grad_norm": 2.9593551554513007, + "learning_rate": 1.5650771789089358e-07, + "loss": 0.9203, + "step": 11670 + }, + { + "epoch": 0.8771231023598377, + "grad_norm": 2.2301455930044822, + "learning_rate": 1.563189781141574e-07, + "loss": 1.0597, + "step": 11671 + }, + { + "epoch": 0.8771982564256726, + "grad_norm": 1.749498840506083, + "learning_rate": 1.561303475825817e-07, + "loss": 0.9724, + "step": 11672 + }, + { + "epoch": 0.8772734104915076, + "grad_norm": 5.100670729269606, + "learning_rate": 1.5594182630734332e-07, + "loss": 0.8846, + "step": 11673 + }, + { + "epoch": 0.8773485645573426, + "grad_norm": 1.6639838401578293, + "learning_rate": 1.5575341429961286e-07, + "loss": 0.9471, + "step": 11674 + }, + { + "epoch": 0.8774237186231775, + "grad_norm": 1.8744955655887765, + "learning_rate": 1.5556511157055563e-07, + "loss": 0.9074, + "step": 11675 + }, + { + "epoch": 0.8774988726890125, + "grad_norm": 1.7993300542223236, + "learning_rate": 1.5537691813132803e-07, + "loss": 0.9895, + "step": 11676 + }, + { + "epoch": 0.8775740267548474, + "grad_norm": 1.7263273881695815, + "learning_rate": 1.5518883399308112e-07, + "loss": 1.0248, + "step": 11677 + }, + { + "epoch": 0.8776491808206824, + "grad_norm": 2.4743881548218454, + "learning_rate": 1.5500085916696072e-07, + "loss": 0.8698, + "step": 11678 + }, + { + "epoch": 0.8777243348865174, + "grad_norm": 1.9426869518204515, + "learning_rate": 1.548129936641036e-07, + "loss": 0.9756, + "step": 11679 + }, + { + "epoch": 0.8777994889523523, + "grad_norm": 1.663658988635817, + "learning_rate": 1.546252374956425e-07, + "loss": 1.0282, + "step": 11680 + }, + { + "epoch": 0.8778746430181873, + "grad_norm": 5.381082331474354, + "learning_rate": 1.5443759067270313e-07, + "loss": 1.0622, + "step": 11681 + }, + { + "epoch": 0.8779497970840222, + "grad_norm": 1.7298807561715084, + "learning_rate": 1.5425005320640282e-07, + "loss": 0.9446, + "step": 11682 + }, + { + "epoch": 0.8780249511498572, + "grad_norm": 2.5723819460162183, + "learning_rate": 1.5406262510785518e-07, + "loss": 0.925, + "step": 11683 + }, + { + "epoch": 0.8781001052156922, + "grad_norm": 2.063626299731579, + "learning_rate": 1.5387530638816525e-07, + "loss": 0.9953, + "step": 11684 + }, + { + "epoch": 0.8781752592815272, + "grad_norm": 2.7223249586878087, + "learning_rate": 1.5368809705843334e-07, + "loss": 1.0351, + "step": 11685 + }, + { + "epoch": 0.8782504133473621, + "grad_norm": 1.9428835492383167, + "learning_rate": 1.5350099712975095e-07, + "loss": 1.0313, + "step": 11686 + }, + { + "epoch": 0.878325567413197, + "grad_norm": 3.29562339708885, + "learning_rate": 1.533140066132055e-07, + "loss": 0.9743, + "step": 11687 + }, + { + "epoch": 0.878400721479032, + "grad_norm": 1.9566360044053548, + "learning_rate": 1.531271255198767e-07, + "loss": 0.9227, + "step": 11688 + }, + { + "epoch": 0.878475875544867, + "grad_norm": 2.1422732057128098, + "learning_rate": 1.529403538608378e-07, + "loss": 0.989, + "step": 11689 + }, + { + "epoch": 0.878551029610702, + "grad_norm": 3.589420131723095, + "learning_rate": 1.5275369164715547e-07, + "loss": 0.9043, + "step": 11690 + }, + { + "epoch": 0.8786261836765369, + "grad_norm": 1.907904485411065, + "learning_rate": 1.5256713888989082e-07, + "loss": 0.9781, + "step": 11691 + }, + { + "epoch": 0.8787013377423719, + "grad_norm": 2.5679851534940137, + "learning_rate": 1.523806956000977e-07, + "loss": 0.9327, + "step": 11692 + }, + { + "epoch": 0.8787764918082068, + "grad_norm": 1.384311402455239, + "learning_rate": 1.5219436178882305e-07, + "loss": 0.9272, + "step": 11693 + }, + { + "epoch": 0.8788516458740417, + "grad_norm": 1.4821461859713283, + "learning_rate": 1.5200813746710806e-07, + "loss": 1.0137, + "step": 11694 + }, + { + "epoch": 0.8789267999398768, + "grad_norm": 2.7283538792263196, + "learning_rate": 1.5182202264598787e-07, + "loss": 0.9301, + "step": 11695 + }, + { + "epoch": 0.8790019540057117, + "grad_norm": 1.8067165709332873, + "learning_rate": 1.5163601733648945e-07, + "loss": 0.9683, + "step": 11696 + }, + { + "epoch": 0.8790771080715467, + "grad_norm": 1.7901727155121474, + "learning_rate": 1.5145012154963466e-07, + "loss": 0.9886, + "step": 11697 + }, + { + "epoch": 0.8791522621373816, + "grad_norm": 2.9011055318946846, + "learning_rate": 1.5126433529643956e-07, + "loss": 0.9882, + "step": 11698 + }, + { + "epoch": 0.8792274162032165, + "grad_norm": 1.7445644021888098, + "learning_rate": 1.5107865858791157e-07, + "loss": 0.922, + "step": 11699 + }, + { + "epoch": 0.8793025702690516, + "grad_norm": 1.8510289449546202, + "learning_rate": 1.508930914350528e-07, + "loss": 1.0275, + "step": 11700 + }, + { + "epoch": 0.8793777243348865, + "grad_norm": 1.242278857440452, + "learning_rate": 1.5070763384885888e-07, + "loss": 0.9966, + "step": 11701 + }, + { + "epoch": 0.8794528784007215, + "grad_norm": 1.7822471638364106, + "learning_rate": 1.5052228584031969e-07, + "loss": 0.9661, + "step": 11702 + }, + { + "epoch": 0.8795280324665564, + "grad_norm": 1.7476876513099067, + "learning_rate": 1.5033704742041664e-07, + "loss": 0.9555, + "step": 11703 + }, + { + "epoch": 0.8796031865323914, + "grad_norm": 1.7399691155860668, + "learning_rate": 1.5015191860012676e-07, + "loss": 1.0077, + "step": 11704 + }, + { + "epoch": 0.8796783405982264, + "grad_norm": 0.7254075746296968, + "learning_rate": 1.4996689939041907e-07, + "loss": 0.8575, + "step": 11705 + }, + { + "epoch": 0.8797534946640613, + "grad_norm": 2.0944870747422164, + "learning_rate": 1.4978198980225698e-07, + "loss": 0.9407, + "step": 11706 + }, + { + "epoch": 0.8798286487298963, + "grad_norm": 2.1093356658659013, + "learning_rate": 1.4959718984659663e-07, + "loss": 0.9559, + "step": 11707 + }, + { + "epoch": 0.8799038027957312, + "grad_norm": 2.65006388461284, + "learning_rate": 1.4941249953438882e-07, + "loss": 1.1113, + "step": 11708 + }, + { + "epoch": 0.8799789568615662, + "grad_norm": 0.7718824301296413, + "learning_rate": 1.492279188765766e-07, + "loss": 0.8983, + "step": 11709 + }, + { + "epoch": 0.8800541109274012, + "grad_norm": 2.954738761340182, + "learning_rate": 1.490434478840974e-07, + "loss": 0.9545, + "step": 11710 + }, + { + "epoch": 0.8801292649932362, + "grad_norm": 2.1546933733532962, + "learning_rate": 1.4885908656788137e-07, + "loss": 0.8814, + "step": 11711 + }, + { + "epoch": 0.8802044190590711, + "grad_norm": 1.717833048971877, + "learning_rate": 1.4867483493885357e-07, + "loss": 0.9771, + "step": 11712 + }, + { + "epoch": 0.880279573124906, + "grad_norm": 1.5524983544578985, + "learning_rate": 1.4849069300793037e-07, + "loss": 0.8939, + "step": 11713 + }, + { + "epoch": 0.880354727190741, + "grad_norm": 2.2423566694607153, + "learning_rate": 1.4830666078602372e-07, + "loss": 0.9876, + "step": 11714 + }, + { + "epoch": 0.880429881256576, + "grad_norm": 1.7938422441855182, + "learning_rate": 1.4812273828403822e-07, + "loss": 0.9502, + "step": 11715 + }, + { + "epoch": 0.880505035322411, + "grad_norm": 2.2144011739873806, + "learning_rate": 1.4793892551287157e-07, + "loss": 0.9371, + "step": 11716 + }, + { + "epoch": 0.8805801893882459, + "grad_norm": 2.8957293009195633, + "learning_rate": 1.477552224834153e-07, + "loss": 1.0031, + "step": 11717 + }, + { + "epoch": 0.8806553434540809, + "grad_norm": 1.993177280260711, + "learning_rate": 1.4757162920655496e-07, + "loss": 0.9598, + "step": 11718 + }, + { + "epoch": 0.8807304975199158, + "grad_norm": 1.5904416921097662, + "learning_rate": 1.473881456931696e-07, + "loss": 1.062, + "step": 11719 + }, + { + "epoch": 0.8808056515857507, + "grad_norm": 1.9210171529509472, + "learning_rate": 1.4720477195413006e-07, + "loss": 0.9379, + "step": 11720 + }, + { + "epoch": 0.8808808056515858, + "grad_norm": 1.5620783427638592, + "learning_rate": 1.47021508000303e-07, + "loss": 0.9781, + "step": 11721 + }, + { + "epoch": 0.8809559597174207, + "grad_norm": 2.3557590286776553, + "learning_rate": 1.468383538425475e-07, + "loss": 0.9399, + "step": 11722 + }, + { + "epoch": 0.8810311137832557, + "grad_norm": 2.193282631323811, + "learning_rate": 1.466553094917149e-07, + "loss": 1.0308, + "step": 11723 + }, + { + "epoch": 0.8811062678490906, + "grad_norm": 17.61464955388696, + "learning_rate": 1.4647237495865227e-07, + "loss": 1.0407, + "step": 11724 + }, + { + "epoch": 0.8811814219149255, + "grad_norm": 2.9758432876077165, + "learning_rate": 1.4628955025419986e-07, + "loss": 0.9865, + "step": 11725 + }, + { + "epoch": 0.8812565759807606, + "grad_norm": 2.1020290608452443, + "learning_rate": 1.46106835389189e-07, + "loss": 0.9661, + "step": 11726 + }, + { + "epoch": 0.8813317300465955, + "grad_norm": 1.3639279118894725, + "learning_rate": 1.459242303744477e-07, + "loss": 1.037, + "step": 11727 + }, + { + "epoch": 0.8814068841124305, + "grad_norm": 2.0354161892763147, + "learning_rate": 1.4574173522079502e-07, + "loss": 1.0095, + "step": 11728 + }, + { + "epoch": 0.8814820381782654, + "grad_norm": 1.9756979879903571, + "learning_rate": 1.4555934993904572e-07, + "loss": 0.9583, + "step": 11729 + }, + { + "epoch": 0.8815571922441005, + "grad_norm": 1.833847999580441, + "learning_rate": 1.4537707454000536e-07, + "loss": 0.863, + "step": 11730 + }, + { + "epoch": 0.8816323463099354, + "grad_norm": 1.8950295333775269, + "learning_rate": 1.4519490903447528e-07, + "loss": 1.0469, + "step": 11731 + }, + { + "epoch": 0.8817075003757703, + "grad_norm": 2.088255091259656, + "learning_rate": 1.4501285343324975e-07, + "loss": 0.9646, + "step": 11732 + }, + { + "epoch": 0.8817826544416053, + "grad_norm": 2.276899395837385, + "learning_rate": 1.448309077471157e-07, + "loss": 0.9648, + "step": 11733 + }, + { + "epoch": 0.8818578085074402, + "grad_norm": 2.0199107656503505, + "learning_rate": 1.4464907198685382e-07, + "loss": 1.0971, + "step": 11734 + }, + { + "epoch": 0.8819329625732752, + "grad_norm": 1.7880670035009283, + "learning_rate": 1.4446734616323953e-07, + "loss": 0.9548, + "step": 11735 + }, + { + "epoch": 0.8820081166391102, + "grad_norm": 1.8340563287727032, + "learning_rate": 1.4428573028704017e-07, + "loss": 0.9282, + "step": 11736 + }, + { + "epoch": 0.8820832707049452, + "grad_norm": 1.8313785520876762, + "learning_rate": 1.4410422436901736e-07, + "loss": 0.924, + "step": 11737 + }, + { + "epoch": 0.8821584247707801, + "grad_norm": 1.4493409796732126, + "learning_rate": 1.4392282841992566e-07, + "loss": 0.8712, + "step": 11738 + }, + { + "epoch": 0.882233578836615, + "grad_norm": 1.795328154466603, + "learning_rate": 1.437415424505144e-07, + "loss": 1.0361, + "step": 11739 + }, + { + "epoch": 0.88230873290245, + "grad_norm": 2.169052045824705, + "learning_rate": 1.4356036647152413e-07, + "loss": 1.0004, + "step": 11740 + }, + { + "epoch": 0.882383886968285, + "grad_norm": 2.026022066807677, + "learning_rate": 1.4337930049369117e-07, + "loss": 0.9464, + "step": 11741 + }, + { + "epoch": 0.88245904103412, + "grad_norm": 1.9396428594527273, + "learning_rate": 1.4319834452774447e-07, + "loss": 0.9257, + "step": 11742 + }, + { + "epoch": 0.8825341950999549, + "grad_norm": 1.77455379626606, + "learning_rate": 1.4301749858440593e-07, + "loss": 0.9402, + "step": 11743 + }, + { + "epoch": 0.8826093491657898, + "grad_norm": 1.354558985685036, + "learning_rate": 1.4283676267439094e-07, + "loss": 0.9894, + "step": 11744 + }, + { + "epoch": 0.8826845032316248, + "grad_norm": 1.6651373602078554, + "learning_rate": 1.4265613680840938e-07, + "loss": 0.9635, + "step": 11745 + }, + { + "epoch": 0.8827596572974598, + "grad_norm": 2.702192200698196, + "learning_rate": 1.424756209971647e-07, + "loss": 0.8421, + "step": 11746 + }, + { + "epoch": 0.8828348113632948, + "grad_norm": 2.804556075818358, + "learning_rate": 1.4229521525135168e-07, + "loss": 0.9417, + "step": 11747 + }, + { + "epoch": 0.8829099654291297, + "grad_norm": 1.495262452885457, + "learning_rate": 1.4211491958166112e-07, + "loss": 1.0052, + "step": 11748 + }, + { + "epoch": 0.8829851194949647, + "grad_norm": 3.184400959302156, + "learning_rate": 1.4193473399877598e-07, + "loss": 0.9578, + "step": 11749 + }, + { + "epoch": 0.8830602735607996, + "grad_norm": 1.7955295335278878, + "learning_rate": 1.4175465851337266e-07, + "loss": 0.9917, + "step": 11750 + }, + { + "epoch": 0.8831354276266346, + "grad_norm": 1.4206783583052702, + "learning_rate": 1.4157469313612147e-07, + "loss": 0.9478, + "step": 11751 + }, + { + "epoch": 0.8832105816924696, + "grad_norm": 1.8341935986013769, + "learning_rate": 1.4139483787768614e-07, + "loss": 0.8712, + "step": 11752 + }, + { + "epoch": 0.8832857357583045, + "grad_norm": 1.6314776499249353, + "learning_rate": 1.412150927487239e-07, + "loss": 0.9187, + "step": 11753 + }, + { + "epoch": 0.8833608898241395, + "grad_norm": 1.6150283774371794, + "learning_rate": 1.4103545775988512e-07, + "loss": 0.9983, + "step": 11754 + }, + { + "epoch": 0.8834360438899744, + "grad_norm": 0.7883038792608719, + "learning_rate": 1.4085593292181375e-07, + "loss": 0.92, + "step": 11755 + }, + { + "epoch": 0.8835111979558095, + "grad_norm": 9.455660090732156, + "learning_rate": 1.406765182451479e-07, + "loss": 1.0062, + "step": 11756 + }, + { + "epoch": 0.8835863520216444, + "grad_norm": 1.9789195123914054, + "learning_rate": 1.404972137405176e-07, + "loss": 1.0266, + "step": 11757 + }, + { + "epoch": 0.8836615060874793, + "grad_norm": 1.8096062151769257, + "learning_rate": 1.4031801941854827e-07, + "loss": 1.0166, + "step": 11758 + }, + { + "epoch": 0.8837366601533143, + "grad_norm": 1.6217022224963826, + "learning_rate": 1.4013893528985744e-07, + "loss": 1.0195, + "step": 11759 + }, + { + "epoch": 0.8838118142191492, + "grad_norm": 3.3701258183658678, + "learning_rate": 1.3995996136505662e-07, + "loss": 0.8611, + "step": 11760 + }, + { + "epoch": 0.8838869682849843, + "grad_norm": 1.83586334728076, + "learning_rate": 1.3978109765475044e-07, + "loss": 0.9312, + "step": 11761 + }, + { + "epoch": 0.8839621223508192, + "grad_norm": 1.787576583075393, + "learning_rate": 1.396023441695373e-07, + "loss": 0.9464, + "step": 11762 + }, + { + "epoch": 0.8840372764166542, + "grad_norm": 6.1146326880816195, + "learning_rate": 1.3942370092000988e-07, + "loss": 1.0351, + "step": 11763 + }, + { + "epoch": 0.8841124304824891, + "grad_norm": 2.4212578756335277, + "learning_rate": 1.3924516791675212e-07, + "loss": 0.9193, + "step": 11764 + }, + { + "epoch": 0.884187584548324, + "grad_norm": 0.6489264471712276, + "learning_rate": 1.39066745170344e-07, + "loss": 0.7906, + "step": 11765 + }, + { + "epoch": 0.884262738614159, + "grad_norm": 2.4771213803418632, + "learning_rate": 1.3888843269135732e-07, + "loss": 0.8994, + "step": 11766 + }, + { + "epoch": 0.884337892679994, + "grad_norm": 1.7671397640347903, + "learning_rate": 1.3871023049035713e-07, + "loss": 0.9767, + "step": 11767 + }, + { + "epoch": 0.884413046745829, + "grad_norm": 1.6989648388103251, + "learning_rate": 1.3853213857790304e-07, + "loss": 0.9359, + "step": 11768 + }, + { + "epoch": 0.8844882008116639, + "grad_norm": 1.6576202770108623, + "learning_rate": 1.3835415696454856e-07, + "loss": 0.9921, + "step": 11769 + }, + { + "epoch": 0.8845633548774988, + "grad_norm": 1.7297345879545323, + "learning_rate": 1.3817628566083817e-07, + "loss": 1.0576, + "step": 11770 + }, + { + "epoch": 0.8846385089433338, + "grad_norm": 1.5958105492355643, + "learning_rate": 1.3799852467731275e-07, + "loss": 0.9328, + "step": 11771 + }, + { + "epoch": 0.8847136630091688, + "grad_norm": 1.634670897443087, + "learning_rate": 1.3782087402450437e-07, + "loss": 0.9902, + "step": 11772 + }, + { + "epoch": 0.8847888170750038, + "grad_norm": 1.9677656831422896, + "learning_rate": 1.376433337129408e-07, + "loss": 0.9052, + "step": 11773 + }, + { + "epoch": 0.8848639711408387, + "grad_norm": 1.8043019597142307, + "learning_rate": 1.374659037531405e-07, + "loss": 1.0235, + "step": 11774 + }, + { + "epoch": 0.8849391252066737, + "grad_norm": 1.6940385804900433, + "learning_rate": 1.3728858415561772e-07, + "loss": 0.9736, + "step": 11775 + }, + { + "epoch": 0.8850142792725086, + "grad_norm": 1.4380778105087912, + "learning_rate": 1.37111374930879e-07, + "loss": 0.9521, + "step": 11776 + }, + { + "epoch": 0.8850894333383436, + "grad_norm": 2.1902811082508107, + "learning_rate": 1.3693427608942497e-07, + "loss": 0.9625, + "step": 11777 + }, + { + "epoch": 0.8851645874041786, + "grad_norm": 1.7013609963896583, + "learning_rate": 1.3675728764174887e-07, + "loss": 0.9592, + "step": 11778 + }, + { + "epoch": 0.8852397414700135, + "grad_norm": 1.76790431263845, + "learning_rate": 1.3658040959833827e-07, + "loss": 1.0218, + "step": 11779 + }, + { + "epoch": 0.8853148955358485, + "grad_norm": 3.9270805675925207, + "learning_rate": 1.3640364196967459e-07, + "loss": 1.1573, + "step": 11780 + }, + { + "epoch": 0.8853900496016834, + "grad_norm": 1.772685425048793, + "learning_rate": 1.3622698476623097e-07, + "loss": 0.8855, + "step": 11781 + }, + { + "epoch": 0.8854652036675185, + "grad_norm": 1.8507080037866999, + "learning_rate": 1.3605043799847527e-07, + "loss": 0.9909, + "step": 11782 + }, + { + "epoch": 0.8855403577333534, + "grad_norm": 1.9696859526715138, + "learning_rate": 1.3587400167686892e-07, + "loss": 1.0749, + "step": 11783 + }, + { + "epoch": 0.8856155117991883, + "grad_norm": 1.5684147085117432, + "learning_rate": 1.3569767581186574e-07, + "loss": 0.9224, + "step": 11784 + }, + { + "epoch": 0.8856906658650233, + "grad_norm": 0.6910863017526714, + "learning_rate": 1.355214604139141e-07, + "loss": 0.8797, + "step": 11785 + }, + { + "epoch": 0.8857658199308582, + "grad_norm": 9.141514713643774, + "learning_rate": 1.3534535549345626e-07, + "loss": 0.9532, + "step": 11786 + }, + { + "epoch": 0.8858409739966933, + "grad_norm": 2.1254526665944247, + "learning_rate": 1.351693610609257e-07, + "loss": 0.9169, + "step": 11787 + }, + { + "epoch": 0.8859161280625282, + "grad_norm": 2.700181803057299, + "learning_rate": 1.3499347712675158e-07, + "loss": 0.9142, + "step": 11788 + }, + { + "epoch": 0.8859912821283631, + "grad_norm": 2.0257930521358265, + "learning_rate": 1.3481770370135537e-07, + "loss": 0.9817, + "step": 11789 + }, + { + "epoch": 0.8860664361941981, + "grad_norm": 2.1141599284500603, + "learning_rate": 1.3464204079515296e-07, + "loss": 1.0007, + "step": 11790 + }, + { + "epoch": 0.886141590260033, + "grad_norm": 2.097122891777631, + "learning_rate": 1.3446648841855202e-07, + "loss": 0.9839, + "step": 11791 + }, + { + "epoch": 0.886216744325868, + "grad_norm": 2.6259927234882956, + "learning_rate": 1.3429104658195555e-07, + "loss": 0.9137, + "step": 11792 + }, + { + "epoch": 0.886291898391703, + "grad_norm": 1.8420127831097286, + "learning_rate": 1.3411571529575882e-07, + "loss": 0.9635, + "step": 11793 + }, + { + "epoch": 0.886367052457538, + "grad_norm": 1.573283104391101, + "learning_rate": 1.3394049457035105e-07, + "loss": 1.0044, + "step": 11794 + }, + { + "epoch": 0.8864422065233729, + "grad_norm": 1.669056164947869, + "learning_rate": 1.3376538441611396e-07, + "loss": 1.0483, + "step": 11795 + }, + { + "epoch": 0.8865173605892078, + "grad_norm": 1.8484165812813769, + "learning_rate": 1.3359038484342478e-07, + "loss": 0.9697, + "step": 11796 + }, + { + "epoch": 0.8865925146550429, + "grad_norm": 3.605015090720327, + "learning_rate": 1.334154958626521e-07, + "loss": 0.9604, + "step": 11797 + }, + { + "epoch": 0.8866676687208778, + "grad_norm": 2.232467690946363, + "learning_rate": 1.33240717484159e-07, + "loss": 1.0153, + "step": 11798 + }, + { + "epoch": 0.8867428227867128, + "grad_norm": 1.7258175584903177, + "learning_rate": 1.3306604971830115e-07, + "loss": 0.9564, + "step": 11799 + }, + { + "epoch": 0.8868179768525477, + "grad_norm": 1.3147986582811575, + "learning_rate": 1.3289149257542964e-07, + "loss": 0.8268, + "step": 11800 + }, + { + "epoch": 0.8868931309183827, + "grad_norm": 1.5456501128134512, + "learning_rate": 1.3271704606588618e-07, + "loss": 0.9968, + "step": 11801 + }, + { + "epoch": 0.8869682849842176, + "grad_norm": 1.6995967050026197, + "learning_rate": 1.3254271020000852e-07, + "loss": 0.9966, + "step": 11802 + }, + { + "epoch": 0.8870434390500526, + "grad_norm": 1.4806290592574338, + "learning_rate": 1.3236848498812592e-07, + "loss": 1.0214, + "step": 11803 + }, + { + "epoch": 0.8871185931158876, + "grad_norm": 1.7478032629654616, + "learning_rate": 1.3219437044056258e-07, + "loss": 1.0154, + "step": 11804 + }, + { + "epoch": 0.8871937471817225, + "grad_norm": 1.982493191643626, + "learning_rate": 1.320203665676345e-07, + "loss": 1.0518, + "step": 11805 + }, + { + "epoch": 0.8872689012475575, + "grad_norm": 2.0784750048520375, + "learning_rate": 1.318464733796527e-07, + "loss": 0.9926, + "step": 11806 + }, + { + "epoch": 0.8873440553133924, + "grad_norm": 2.1751967900139206, + "learning_rate": 1.3167269088692167e-07, + "loss": 1.0508, + "step": 11807 + }, + { + "epoch": 0.8874192093792275, + "grad_norm": 0.7563633384581042, + "learning_rate": 1.3149901909973738e-07, + "loss": 0.8418, + "step": 11808 + }, + { + "epoch": 0.8874943634450624, + "grad_norm": 1.9306126897881903, + "learning_rate": 1.3132545802839158e-07, + "loss": 0.9673, + "step": 11809 + }, + { + "epoch": 0.8875695175108973, + "grad_norm": 1.6059180603216607, + "learning_rate": 1.3115200768316803e-07, + "loss": 0.932, + "step": 11810 + }, + { + "epoch": 0.8876446715767323, + "grad_norm": 3.41694948530048, + "learning_rate": 1.309786680743441e-07, + "loss": 1.0239, + "step": 11811 + }, + { + "epoch": 0.8877198256425672, + "grad_norm": 0.7460524063239258, + "learning_rate": 1.308054392121909e-07, + "loss": 0.8009, + "step": 11812 + }, + { + "epoch": 0.8877949797084023, + "grad_norm": 1.546968707704583, + "learning_rate": 1.3063232110697375e-07, + "loss": 0.9378, + "step": 11813 + }, + { + "epoch": 0.8878701337742372, + "grad_norm": 2.4979897045903736, + "learning_rate": 1.3045931376894915e-07, + "loss": 0.7752, + "step": 11814 + }, + { + "epoch": 0.8879452878400721, + "grad_norm": 1.6492285977082677, + "learning_rate": 1.3028641720836953e-07, + "loss": 0.9353, + "step": 11815 + }, + { + "epoch": 0.8880204419059071, + "grad_norm": 2.2437239111443543, + "learning_rate": 1.301136314354787e-07, + "loss": 0.9786, + "step": 11816 + }, + { + "epoch": 0.888095595971742, + "grad_norm": 1.8883096057560453, + "learning_rate": 1.299409564605165e-07, + "loss": 1.0041, + "step": 11817 + }, + { + "epoch": 0.8881707500375771, + "grad_norm": 0.8153690521173006, + "learning_rate": 1.2976839229371272e-07, + "loss": 0.8235, + "step": 11818 + }, + { + "epoch": 0.888245904103412, + "grad_norm": 2.4947274876079666, + "learning_rate": 1.2959593894529364e-07, + "loss": 0.9685, + "step": 11819 + }, + { + "epoch": 0.888321058169247, + "grad_norm": 1.4706146092013321, + "learning_rate": 1.294235964254775e-07, + "loss": 0.8685, + "step": 11820 + }, + { + "epoch": 0.8883962122350819, + "grad_norm": 1.5735261590975402, + "learning_rate": 1.2925136474447597e-07, + "loss": 1.0313, + "step": 11821 + }, + { + "epoch": 0.8884713663009168, + "grad_norm": 2.1868142279210057, + "learning_rate": 1.290792439124946e-07, + "loss": 0.9552, + "step": 11822 + }, + { + "epoch": 0.8885465203667519, + "grad_norm": 1.7622717012846116, + "learning_rate": 1.2890723393973213e-07, + "loss": 1.0073, + "step": 11823 + }, + { + "epoch": 0.8886216744325868, + "grad_norm": 1.6167799578950057, + "learning_rate": 1.2873533483638155e-07, + "loss": 0.9209, + "step": 11824 + }, + { + "epoch": 0.8886968284984218, + "grad_norm": 1.850326318571666, + "learning_rate": 1.285635466126278e-07, + "loss": 0.9551, + "step": 11825 + }, + { + "epoch": 0.8887719825642567, + "grad_norm": 2.2381707874888606, + "learning_rate": 1.2839186927864965e-07, + "loss": 0.8717, + "step": 11826 + }, + { + "epoch": 0.8888471366300917, + "grad_norm": 2.556577117807256, + "learning_rate": 1.282203028446207e-07, + "loss": 0.9762, + "step": 11827 + }, + { + "epoch": 0.8889222906959267, + "grad_norm": 0.7350588516107185, + "learning_rate": 1.2804884732070574e-07, + "loss": 0.8556, + "step": 11828 + }, + { + "epoch": 0.8889974447617616, + "grad_norm": 1.6824346743367615, + "learning_rate": 1.2787750271706487e-07, + "loss": 0.9203, + "step": 11829 + }, + { + "epoch": 0.8890725988275966, + "grad_norm": 1.8413774127992304, + "learning_rate": 1.2770626904385128e-07, + "loss": 0.9985, + "step": 11830 + }, + { + "epoch": 0.8891477528934315, + "grad_norm": 2.324147866073767, + "learning_rate": 1.275351463112102e-07, + "loss": 0.9459, + "step": 11831 + }, + { + "epoch": 0.8892229069592665, + "grad_norm": 2.8327568440090785, + "learning_rate": 1.2736413452928218e-07, + "loss": 0.9285, + "step": 11832 + }, + { + "epoch": 0.8892980610251014, + "grad_norm": 7.738494700646476, + "learning_rate": 1.2719323370819955e-07, + "loss": 0.921, + "step": 11833 + }, + { + "epoch": 0.8893732150909364, + "grad_norm": 2.378633730958995, + "learning_rate": 1.2702244385809e-07, + "loss": 0.875, + "step": 11834 + }, + { + "epoch": 0.8894483691567714, + "grad_norm": 1.6451059818622136, + "learning_rate": 1.2685176498907213e-07, + "loss": 1.0353, + "step": 11835 + }, + { + "epoch": 0.8895235232226063, + "grad_norm": 1.807464576680099, + "learning_rate": 1.2668119711126023e-07, + "loss": 0.918, + "step": 11836 + }, + { + "epoch": 0.8895986772884413, + "grad_norm": 2.671582746974513, + "learning_rate": 1.2651074023476095e-07, + "loss": 1.001, + "step": 11837 + }, + { + "epoch": 0.8896738313542762, + "grad_norm": 2.056666708234707, + "learning_rate": 1.2634039436967414e-07, + "loss": 0.8847, + "step": 11838 + }, + { + "epoch": 0.8897489854201113, + "grad_norm": 2.382581155521567, + "learning_rate": 1.2617015952609356e-07, + "loss": 0.9986, + "step": 11839 + }, + { + "epoch": 0.8898241394859462, + "grad_norm": 1.5346336348768355, + "learning_rate": 1.2600003571410668e-07, + "loss": 0.9093, + "step": 11840 + }, + { + "epoch": 0.8898992935517811, + "grad_norm": 1.7545366067793537, + "learning_rate": 1.2583002294379363e-07, + "loss": 0.8865, + "step": 11841 + }, + { + "epoch": 0.8899744476176161, + "grad_norm": 1.5531942586212577, + "learning_rate": 1.2566012122522862e-07, + "loss": 0.9754, + "step": 11842 + }, + { + "epoch": 0.890049601683451, + "grad_norm": 1.5030347800986368, + "learning_rate": 1.2549033056847825e-07, + "loss": 0.9284, + "step": 11843 + }, + { + "epoch": 0.8901247557492861, + "grad_norm": 2.7529782727011485, + "learning_rate": 1.2532065098360445e-07, + "loss": 0.9003, + "step": 11844 + }, + { + "epoch": 0.890199909815121, + "grad_norm": 1.7110679263215496, + "learning_rate": 1.2515108248066008e-07, + "loss": 0.9267, + "step": 11845 + }, + { + "epoch": 0.890275063880956, + "grad_norm": 1.4119139307876634, + "learning_rate": 1.2498162506969312e-07, + "loss": 1.0431, + "step": 11846 + }, + { + "epoch": 0.8903502179467909, + "grad_norm": 4.93259524331265, + "learning_rate": 1.2481227876074575e-07, + "loss": 1.0409, + "step": 11847 + }, + { + "epoch": 0.8904253720126258, + "grad_norm": 1.485568560538046, + "learning_rate": 1.2464304356385102e-07, + "loss": 1.0199, + "step": 11848 + }, + { + "epoch": 0.8905005260784609, + "grad_norm": 1.6039913461504056, + "learning_rate": 1.2447391948903673e-07, + "loss": 0.9523, + "step": 11849 + }, + { + "epoch": 0.8905756801442958, + "grad_norm": 3.101661815062153, + "learning_rate": 1.2430490654632487e-07, + "loss": 0.8952, + "step": 11850 + }, + { + "epoch": 0.8906508342101308, + "grad_norm": 1.5159014420240329, + "learning_rate": 1.2413600474573028e-07, + "loss": 1.036, + "step": 11851 + }, + { + "epoch": 0.8907259882759657, + "grad_norm": 14.163651508274677, + "learning_rate": 1.2396721409725987e-07, + "loss": 1.0851, + "step": 11852 + }, + { + "epoch": 0.8908011423418007, + "grad_norm": 1.8763091285532707, + "learning_rate": 1.2379853461091628e-07, + "loss": 0.959, + "step": 11853 + }, + { + "epoch": 0.8908762964076357, + "grad_norm": 1.7088080963773407, + "learning_rate": 1.2362996629669376e-07, + "loss": 0.9803, + "step": 11854 + }, + { + "epoch": 0.8909514504734706, + "grad_norm": 1.9559733259141399, + "learning_rate": 1.2346150916458098e-07, + "loss": 0.9712, + "step": 11855 + }, + { + "epoch": 0.8910266045393056, + "grad_norm": 1.7667379331909432, + "learning_rate": 1.232931632245593e-07, + "loss": 0.9147, + "step": 11856 + }, + { + "epoch": 0.8911017586051405, + "grad_norm": 4.057366987058945, + "learning_rate": 1.2312492848660448e-07, + "loss": 0.9494, + "step": 11857 + }, + { + "epoch": 0.8911769126709755, + "grad_norm": 1.617409732310351, + "learning_rate": 1.229568049606844e-07, + "loss": 0.9116, + "step": 11858 + }, + { + "epoch": 0.8912520667368105, + "grad_norm": 1.7145156025013897, + "learning_rate": 1.2278879265676122e-07, + "loss": 1.0374, + "step": 11859 + }, + { + "epoch": 0.8913272208026454, + "grad_norm": 3.397495604028728, + "learning_rate": 1.2262089158479038e-07, + "loss": 0.9703, + "step": 11860 + }, + { + "epoch": 0.8914023748684804, + "grad_norm": 2.1275090383478874, + "learning_rate": 1.2245310175472125e-07, + "loss": 0.9777, + "step": 11861 + }, + { + "epoch": 0.8914775289343153, + "grad_norm": 2.289319437338428, + "learning_rate": 1.222854231764947e-07, + "loss": 1.01, + "step": 11862 + }, + { + "epoch": 0.8915526830001503, + "grad_norm": 7.402202288803329, + "learning_rate": 1.2211785586004751e-07, + "loss": 0.9414, + "step": 11863 + }, + { + "epoch": 0.8916278370659853, + "grad_norm": 1.5077292463809462, + "learning_rate": 1.219503998153082e-07, + "loss": 0.8189, + "step": 11864 + }, + { + "epoch": 0.8917029911318203, + "grad_norm": 0.7721935248207443, + "learning_rate": 1.217830550521992e-07, + "loss": 0.8181, + "step": 11865 + }, + { + "epoch": 0.8917781451976552, + "grad_norm": 1.9532949755138873, + "learning_rate": 1.2161582158063622e-07, + "loss": 1.0159, + "step": 11866 + }, + { + "epoch": 0.8918532992634901, + "grad_norm": 1.551608616814875, + "learning_rate": 1.2144869941052837e-07, + "loss": 1.0052, + "step": 11867 + }, + { + "epoch": 0.8919284533293251, + "grad_norm": 1.9749451303706778, + "learning_rate": 1.2128168855177933e-07, + "loss": 0.8818, + "step": 11868 + }, + { + "epoch": 0.89200360739516, + "grad_norm": 1.7144388402942157, + "learning_rate": 1.2111478901428363e-07, + "loss": 0.8651, + "step": 11869 + }, + { + "epoch": 0.8920787614609951, + "grad_norm": 1.6391883736307478, + "learning_rate": 1.2094800080793177e-07, + "loss": 0.9342, + "step": 11870 + }, + { + "epoch": 0.89215391552683, + "grad_norm": 2.391180235766822, + "learning_rate": 1.2078132394260654e-07, + "loss": 1.0032, + "step": 11871 + }, + { + "epoch": 0.892229069592665, + "grad_norm": 2.1035022504942735, + "learning_rate": 1.2061475842818335e-07, + "loss": 1.0224, + "step": 11872 + }, + { + "epoch": 0.8923042236584999, + "grad_norm": 1.8807237998134367, + "learning_rate": 1.2044830427453234e-07, + "loss": 0.9833, + "step": 11873 + }, + { + "epoch": 0.8923793777243348, + "grad_norm": 2.112609756598008, + "learning_rate": 1.2028196149151716e-07, + "loss": 0.8907, + "step": 11874 + }, + { + "epoch": 0.8924545317901699, + "grad_norm": 1.7086937531908997, + "learning_rate": 1.20115730088993e-07, + "loss": 1.0078, + "step": 11875 + }, + { + "epoch": 0.8925296858560048, + "grad_norm": 1.9626167675681674, + "learning_rate": 1.199496100768107e-07, + "loss": 0.8666, + "step": 11876 + }, + { + "epoch": 0.8926048399218398, + "grad_norm": 0.7707694302094762, + "learning_rate": 1.1978360146481281e-07, + "loss": 0.8332, + "step": 11877 + }, + { + "epoch": 0.8926799939876747, + "grad_norm": 1.595026440057475, + "learning_rate": 1.1961770426283723e-07, + "loss": 1.0557, + "step": 11878 + }, + { + "epoch": 0.8927551480535096, + "grad_norm": 4.071969601051163, + "learning_rate": 1.1945191848071234e-07, + "loss": 0.9084, + "step": 11879 + }, + { + "epoch": 0.8928303021193447, + "grad_norm": 0.6687346634933372, + "learning_rate": 1.1928624412826272e-07, + "loss": 0.8024, + "step": 11880 + }, + { + "epoch": 0.8929054561851796, + "grad_norm": 5.5870146879867155, + "learning_rate": 1.1912068121530494e-07, + "loss": 0.9763, + "step": 11881 + }, + { + "epoch": 0.8929806102510146, + "grad_norm": 1.7288921398719572, + "learning_rate": 1.1895522975164918e-07, + "loss": 1.0219, + "step": 11882 + }, + { + "epoch": 0.8930557643168495, + "grad_norm": 1.7189461336067804, + "learning_rate": 1.1878988974709869e-07, + "loss": 0.9366, + "step": 11883 + }, + { + "epoch": 0.8931309183826845, + "grad_norm": 0.8489710722983594, + "learning_rate": 1.1862466121145098e-07, + "loss": 0.8938, + "step": 11884 + }, + { + "epoch": 0.8932060724485195, + "grad_norm": 1.354012402281878, + "learning_rate": 1.1845954415449666e-07, + "loss": 0.9602, + "step": 11885 + }, + { + "epoch": 0.8932812265143544, + "grad_norm": 2.7149637099733335, + "learning_rate": 1.1829453858601901e-07, + "loss": 0.8905, + "step": 11886 + }, + { + "epoch": 0.8933563805801894, + "grad_norm": 1.4475093332625535, + "learning_rate": 1.1812964451579532e-07, + "loss": 0.9384, + "step": 11887 + }, + { + "epoch": 0.8934315346460243, + "grad_norm": 7.362311099467463, + "learning_rate": 1.1796486195359711e-07, + "loss": 0.9829, + "step": 11888 + }, + { + "epoch": 0.8935066887118593, + "grad_norm": 1.8259242578405281, + "learning_rate": 1.178001909091868e-07, + "loss": 1.0052, + "step": 11889 + }, + { + "epoch": 0.8935818427776943, + "grad_norm": 1.4367735312443046, + "learning_rate": 1.1763563139232257e-07, + "loss": 0.9372, + "step": 11890 + }, + { + "epoch": 0.8936569968435293, + "grad_norm": 1.7523046032516447, + "learning_rate": 1.1747118341275597e-07, + "loss": 0.9154, + "step": 11891 + }, + { + "epoch": 0.8937321509093642, + "grad_norm": 1.6500330430392411, + "learning_rate": 1.1730684698023007e-07, + "loss": 0.9879, + "step": 11892 + }, + { + "epoch": 0.8938073049751991, + "grad_norm": 1.642519919218052, + "learning_rate": 1.1714262210448245e-07, + "loss": 1.0432, + "step": 11893 + }, + { + "epoch": 0.8938824590410341, + "grad_norm": 1.8792500964448007, + "learning_rate": 1.1697850879524462e-07, + "loss": 1.0564, + "step": 11894 + }, + { + "epoch": 0.893957613106869, + "grad_norm": 1.7651520492288966, + "learning_rate": 1.1681450706224106e-07, + "loss": 0.9952, + "step": 11895 + }, + { + "epoch": 0.8940327671727041, + "grad_norm": 2.486924692459415, + "learning_rate": 1.1665061691518884e-07, + "loss": 0.983, + "step": 11896 + }, + { + "epoch": 0.894107921238539, + "grad_norm": 1.8369389176905284, + "learning_rate": 1.1648683836379935e-07, + "loss": 0.9415, + "step": 11897 + }, + { + "epoch": 0.894183075304374, + "grad_norm": 2.504001276412005, + "learning_rate": 1.1632317141777748e-07, + "loss": 0.9625, + "step": 11898 + }, + { + "epoch": 0.8942582293702089, + "grad_norm": 1.6316294057148466, + "learning_rate": 1.1615961608682057e-07, + "loss": 1.0269, + "step": 11899 + }, + { + "epoch": 0.8943333834360438, + "grad_norm": 2.1970631417120066, + "learning_rate": 1.1599617238061976e-07, + "loss": 1.0179, + "step": 11900 + }, + { + "epoch": 0.8944085375018789, + "grad_norm": 1.9060629455844074, + "learning_rate": 1.1583284030886087e-07, + "loss": 0.9526, + "step": 11901 + }, + { + "epoch": 0.8944836915677138, + "grad_norm": 0.6734438477679461, + "learning_rate": 1.1566961988122037e-07, + "loss": 0.8175, + "step": 11902 + }, + { + "epoch": 0.8945588456335488, + "grad_norm": 1.4810614292816853, + "learning_rate": 1.1550651110737097e-07, + "loss": 0.976, + "step": 11903 + }, + { + "epoch": 0.8946339996993837, + "grad_norm": 1.6687913068023754, + "learning_rate": 1.153435139969765e-07, + "loss": 0.9041, + "step": 11904 + }, + { + "epoch": 0.8947091537652186, + "grad_norm": 1.45718243916496, + "learning_rate": 1.1518062855969635e-07, + "loss": 0.8743, + "step": 11905 + }, + { + "epoch": 0.8947843078310537, + "grad_norm": 4.486750276040643, + "learning_rate": 1.1501785480518078e-07, + "loss": 0.9585, + "step": 11906 + }, + { + "epoch": 0.8948594618968886, + "grad_norm": 0.6934948641071865, + "learning_rate": 1.1485519274307564e-07, + "loss": 0.8374, + "step": 11907 + }, + { + "epoch": 0.8949346159627236, + "grad_norm": 1.804207147521482, + "learning_rate": 1.1469264238301924e-07, + "loss": 0.9552, + "step": 11908 + }, + { + "epoch": 0.8950097700285585, + "grad_norm": 1.8826067167027132, + "learning_rate": 1.1453020373464295e-07, + "loss": 1.0029, + "step": 11909 + }, + { + "epoch": 0.8950849240943936, + "grad_norm": 1.9309990012997496, + "learning_rate": 1.1436787680757176e-07, + "loss": 0.9396, + "step": 11910 + }, + { + "epoch": 0.8951600781602285, + "grad_norm": 2.008807933757676, + "learning_rate": 1.1420566161142442e-07, + "loss": 0.9231, + "step": 11911 + }, + { + "epoch": 0.8952352322260634, + "grad_norm": 1.9738843664102055, + "learning_rate": 1.1404355815581345e-07, + "loss": 0.9148, + "step": 11912 + }, + { + "epoch": 0.8953103862918984, + "grad_norm": 1.541502978252413, + "learning_rate": 1.1388156645034275e-07, + "loss": 1.0025, + "step": 11913 + }, + { + "epoch": 0.8953855403577333, + "grad_norm": 1.7764261748684325, + "learning_rate": 1.1371968650461216e-07, + "loss": 1.0474, + "step": 11914 + }, + { + "epoch": 0.8954606944235683, + "grad_norm": 3.243305059651487, + "learning_rate": 1.1355791832821338e-07, + "loss": 1.0503, + "step": 11915 + }, + { + "epoch": 0.8955358484894033, + "grad_norm": 1.8249035352530627, + "learning_rate": 1.1339626193073093e-07, + "loss": 0.9852, + "step": 11916 + }, + { + "epoch": 0.8956110025552383, + "grad_norm": 1.6321176026737532, + "learning_rate": 1.132347173217445e-07, + "loss": 0.9884, + "step": 11917 + }, + { + "epoch": 0.8956861566210732, + "grad_norm": 1.783375258561964, + "learning_rate": 1.1307328451082643e-07, + "loss": 0.8706, + "step": 11918 + }, + { + "epoch": 0.8957613106869081, + "grad_norm": 1.8087508131340428, + "learning_rate": 1.1291196350754107e-07, + "loss": 1.0346, + "step": 11919 + }, + { + "epoch": 0.8958364647527431, + "grad_norm": 2.1394617860238125, + "learning_rate": 1.1275075432144831e-07, + "loss": 0.9866, + "step": 11920 + }, + { + "epoch": 0.8959116188185781, + "grad_norm": 2.0632666116792473, + "learning_rate": 1.125896569621001e-07, + "loss": 0.9229, + "step": 11921 + }, + { + "epoch": 0.8959867728844131, + "grad_norm": 2.455019177053101, + "learning_rate": 1.1242867143904233e-07, + "loss": 1.014, + "step": 11922 + }, + { + "epoch": 0.896061926950248, + "grad_norm": 0.6517137694917327, + "learning_rate": 1.1226779776181339e-07, + "loss": 0.7851, + "step": 11923 + }, + { + "epoch": 0.8961370810160829, + "grad_norm": 0.739504371152988, + "learning_rate": 1.121070359399463e-07, + "loss": 0.8239, + "step": 11924 + }, + { + "epoch": 0.8962122350819179, + "grad_norm": 2.04715755929308, + "learning_rate": 1.1194638598296658e-07, + "loss": 0.8415, + "step": 11925 + }, + { + "epoch": 0.8962873891477529, + "grad_norm": 1.3452711145501737, + "learning_rate": 1.1178584790039348e-07, + "loss": 1.0352, + "step": 11926 + }, + { + "epoch": 0.8963625432135879, + "grad_norm": 0.7547627066028789, + "learning_rate": 1.1162542170173873e-07, + "loss": 0.9451, + "step": 11927 + }, + { + "epoch": 0.8964376972794228, + "grad_norm": 2.0690092771385773, + "learning_rate": 1.1146510739650939e-07, + "loss": 1.0129, + "step": 11928 + }, + { + "epoch": 0.8965128513452578, + "grad_norm": 1.945917671167523, + "learning_rate": 1.1130490499420386e-07, + "loss": 1.0206, + "step": 11929 + }, + { + "epoch": 0.8965880054110927, + "grad_norm": 2.458640246441044, + "learning_rate": 1.1114481450431523e-07, + "loss": 1.0411, + "step": 11930 + }, + { + "epoch": 0.8966631594769277, + "grad_norm": 1.736691345739951, + "learning_rate": 1.109848359363288e-07, + "loss": 0.8619, + "step": 11931 + }, + { + "epoch": 0.8967383135427627, + "grad_norm": 2.517809659670548, + "learning_rate": 1.1082496929972496e-07, + "loss": 0.9764, + "step": 11932 + }, + { + "epoch": 0.8968134676085976, + "grad_norm": 1.393358480091198, + "learning_rate": 1.1066521460397527e-07, + "loss": 0.9397, + "step": 11933 + }, + { + "epoch": 0.8968886216744326, + "grad_norm": 1.8387538654175253, + "learning_rate": 1.1050557185854636e-07, + "loss": 1.0044, + "step": 11934 + }, + { + "epoch": 0.8969637757402675, + "grad_norm": 0.7763410839754733, + "learning_rate": 1.1034604107289847e-07, + "loss": 0.8303, + "step": 11935 + }, + { + "epoch": 0.8970389298061026, + "grad_norm": 1.377191963855078, + "learning_rate": 1.1018662225648267e-07, + "loss": 0.9859, + "step": 11936 + }, + { + "epoch": 0.8971140838719375, + "grad_norm": 2.2197744663908145, + "learning_rate": 1.1002731541874654e-07, + "loss": 1.0901, + "step": 11937 + }, + { + "epoch": 0.8971892379377724, + "grad_norm": 1.9177301262303286, + "learning_rate": 1.0986812056912898e-07, + "loss": 0.9228, + "step": 11938 + }, + { + "epoch": 0.8972643920036074, + "grad_norm": 1.641543903053048, + "learning_rate": 1.0970903771706352e-07, + "loss": 0.9771, + "step": 11939 + }, + { + "epoch": 0.8973395460694423, + "grad_norm": 1.7840566775423565, + "learning_rate": 1.0955006687197533e-07, + "loss": 1.0867, + "step": 11940 + }, + { + "epoch": 0.8974147001352774, + "grad_norm": 2.138242969442062, + "learning_rate": 1.0939120804328505e-07, + "loss": 1.0025, + "step": 11941 + }, + { + "epoch": 0.8974898542011123, + "grad_norm": 3.0261636119614987, + "learning_rate": 1.0923246124040542e-07, + "loss": 0.9778, + "step": 11942 + }, + { + "epoch": 0.8975650082669473, + "grad_norm": 1.7620041440553462, + "learning_rate": 1.0907382647274266e-07, + "loss": 0.9532, + "step": 11943 + }, + { + "epoch": 0.8976401623327822, + "grad_norm": 3.080121317616855, + "learning_rate": 1.0891530374969615e-07, + "loss": 0.8835, + "step": 11944 + }, + { + "epoch": 0.8977153163986171, + "grad_norm": 1.9972286693213686, + "learning_rate": 1.0875689308065994e-07, + "loss": 1.0235, + "step": 11945 + }, + { + "epoch": 0.8977904704644522, + "grad_norm": 1.4767785292737452, + "learning_rate": 1.085985944750194e-07, + "loss": 0.9477, + "step": 11946 + }, + { + "epoch": 0.8978656245302871, + "grad_norm": 1.5396943320325434, + "learning_rate": 1.0844040794215503e-07, + "loss": 0.8705, + "step": 11947 + }, + { + "epoch": 0.8979407785961221, + "grad_norm": 1.9649337838933527, + "learning_rate": 1.0828233349143934e-07, + "loss": 0.983, + "step": 11948 + }, + { + "epoch": 0.898015932661957, + "grad_norm": 3.154180017910865, + "learning_rate": 1.0812437113223993e-07, + "loss": 1.0125, + "step": 11949 + }, + { + "epoch": 0.8980910867277919, + "grad_norm": 1.9920620149472943, + "learning_rate": 1.0796652087391556e-07, + "loss": 0.9703, + "step": 11950 + }, + { + "epoch": 0.898166240793627, + "grad_norm": 2.2635824533068596, + "learning_rate": 1.0780878272582006e-07, + "loss": 0.9421, + "step": 11951 + }, + { + "epoch": 0.8982413948594619, + "grad_norm": 2.138643558959237, + "learning_rate": 1.0765115669729974e-07, + "loss": 0.9953, + "step": 11952 + }, + { + "epoch": 0.8983165489252969, + "grad_norm": 3.942347736405764, + "learning_rate": 1.074936427976949e-07, + "loss": 1.0402, + "step": 11953 + }, + { + "epoch": 0.8983917029911318, + "grad_norm": 1.4972892745429578, + "learning_rate": 1.0733624103633831e-07, + "loss": 1.0523, + "step": 11954 + }, + { + "epoch": 0.8984668570569668, + "grad_norm": 1.314281707514703, + "learning_rate": 1.0717895142255672e-07, + "loss": 1.0241, + "step": 11955 + }, + { + "epoch": 0.8985420111228017, + "grad_norm": 1.393187358858761, + "learning_rate": 1.0702177396567114e-07, + "loss": 0.9764, + "step": 11956 + }, + { + "epoch": 0.8986171651886367, + "grad_norm": 1.860704731897084, + "learning_rate": 1.0686470867499342e-07, + "loss": 0.9194, + "step": 11957 + }, + { + "epoch": 0.8986923192544717, + "grad_norm": 1.7653998172265954, + "learning_rate": 1.0670775555983147e-07, + "loss": 0.9126, + "step": 11958 + }, + { + "epoch": 0.8987674733203066, + "grad_norm": 1.8580109537579879, + "learning_rate": 1.0655091462948518e-07, + "loss": 0.9833, + "step": 11959 + }, + { + "epoch": 0.8988426273861416, + "grad_norm": 1.524308019513934, + "learning_rate": 1.063941858932469e-07, + "loss": 0.9726, + "step": 11960 + }, + { + "epoch": 0.8989177814519765, + "grad_norm": 1.9214905765291455, + "learning_rate": 1.0623756936040451e-07, + "loss": 0.9729, + "step": 11961 + }, + { + "epoch": 0.8989929355178116, + "grad_norm": 1.9952046005959228, + "learning_rate": 1.0608106504023817e-07, + "loss": 0.9743, + "step": 11962 + }, + { + "epoch": 0.8990680895836465, + "grad_norm": 1.938624025208215, + "learning_rate": 1.0592467294202046e-07, + "loss": 0.9864, + "step": 11963 + }, + { + "epoch": 0.8991432436494814, + "grad_norm": 1.6641401532461129, + "learning_rate": 1.0576839307501928e-07, + "loss": 1.0236, + "step": 11964 + }, + { + "epoch": 0.8992183977153164, + "grad_norm": 1.8565147904624435, + "learning_rate": 1.056122254484939e-07, + "loss": 1.0249, + "step": 11965 + }, + { + "epoch": 0.8992935517811513, + "grad_norm": 3.2689447640332334, + "learning_rate": 1.0545617007169871e-07, + "loss": 0.955, + "step": 11966 + }, + { + "epoch": 0.8993687058469864, + "grad_norm": 2.0426020393423405, + "learning_rate": 1.0530022695387964e-07, + "loss": 0.9892, + "step": 11967 + }, + { + "epoch": 0.8994438599128213, + "grad_norm": 1.6009216515111986, + "learning_rate": 1.0514439610427772e-07, + "loss": 1.0165, + "step": 11968 + }, + { + "epoch": 0.8995190139786562, + "grad_norm": 2.207135169359311, + "learning_rate": 1.0498867753212626e-07, + "loss": 0.8629, + "step": 11969 + }, + { + "epoch": 0.8995941680444912, + "grad_norm": 1.371779137392488, + "learning_rate": 1.0483307124665208e-07, + "loss": 0.9633, + "step": 11970 + }, + { + "epoch": 0.8996693221103261, + "grad_norm": 2.172911472137572, + "learning_rate": 1.0467757725707538e-07, + "loss": 0.8355, + "step": 11971 + }, + { + "epoch": 0.8997444761761612, + "grad_norm": 2.1676135072829674, + "learning_rate": 1.0452219557260966e-07, + "loss": 1.074, + "step": 11972 + }, + { + "epoch": 0.8998196302419961, + "grad_norm": 1.8886697921031366, + "learning_rate": 1.043669262024629e-07, + "loss": 0.9634, + "step": 11973 + }, + { + "epoch": 0.8998947843078311, + "grad_norm": 1.7815253730870642, + "learning_rate": 1.0421176915583419e-07, + "loss": 0.8563, + "step": 11974 + }, + { + "epoch": 0.899969938373666, + "grad_norm": 1.4516584317440564, + "learning_rate": 1.0405672444191727e-07, + "loss": 0.925, + "step": 11975 + }, + { + "epoch": 0.9000450924395009, + "grad_norm": 1.692856024143339, + "learning_rate": 1.0390179206990012e-07, + "loss": 1.0225, + "step": 11976 + }, + { + "epoch": 0.900120246505336, + "grad_norm": 1.7343043407446352, + "learning_rate": 1.037469720489621e-07, + "loss": 0.96, + "step": 11977 + }, + { + "epoch": 0.9001954005711709, + "grad_norm": 1.8026919871396323, + "learning_rate": 1.0359226438827695e-07, + "loss": 0.984, + "step": 11978 + }, + { + "epoch": 0.9002705546370059, + "grad_norm": 1.7291304190026877, + "learning_rate": 1.0343766909701268e-07, + "loss": 0.9636, + "step": 11979 + }, + { + "epoch": 0.9003457087028408, + "grad_norm": 1.6466309831719497, + "learning_rate": 1.0328318618432819e-07, + "loss": 0.9919, + "step": 11980 + }, + { + "epoch": 0.9004208627686758, + "grad_norm": 2.226315882895082, + "learning_rate": 1.0312881565937837e-07, + "loss": 0.9768, + "step": 11981 + }, + { + "epoch": 0.9004960168345107, + "grad_norm": 2.4091966100375974, + "learning_rate": 1.0297455753130946e-07, + "loss": 0.9927, + "step": 11982 + }, + { + "epoch": 0.9005711709003457, + "grad_norm": 1.6941558830166965, + "learning_rate": 1.0282041180926282e-07, + "loss": 1.0171, + "step": 11983 + }, + { + "epoch": 0.9006463249661807, + "grad_norm": 2.2794113065078534, + "learning_rate": 1.0266637850237115e-07, + "loss": 0.9818, + "step": 11984 + }, + { + "epoch": 0.9007214790320156, + "grad_norm": 2.6073307092852236, + "learning_rate": 1.0251245761976202e-07, + "loss": 0.808, + "step": 11985 + }, + { + "epoch": 0.9007966330978506, + "grad_norm": 1.4056162359728581, + "learning_rate": 1.023586491705557e-07, + "loss": 1.0112, + "step": 11986 + }, + { + "epoch": 0.9008717871636855, + "grad_norm": 0.8737620042622981, + "learning_rate": 1.0220495316386601e-07, + "loss": 0.8594, + "step": 11987 + }, + { + "epoch": 0.9009469412295206, + "grad_norm": 1.357837026218724, + "learning_rate": 1.020513696087999e-07, + "loss": 0.9487, + "step": 11988 + }, + { + "epoch": 0.9010220952953555, + "grad_norm": 3.897998731610017, + "learning_rate": 1.0189789851445829e-07, + "loss": 1.0086, + "step": 11989 + }, + { + "epoch": 0.9010972493611904, + "grad_norm": 1.7401444982402638, + "learning_rate": 1.0174453988993392e-07, + "loss": 0.9297, + "step": 11990 + }, + { + "epoch": 0.9011724034270254, + "grad_norm": 1.5412991145739843, + "learning_rate": 1.0159129374431463e-07, + "loss": 0.986, + "step": 11991 + }, + { + "epoch": 0.9012475574928603, + "grad_norm": 1.9048841617438252, + "learning_rate": 1.0143816008668049e-07, + "loss": 1.011, + "step": 11992 + }, + { + "epoch": 0.9013227115586954, + "grad_norm": 2.091523994149488, + "learning_rate": 1.0128513892610623e-07, + "loss": 1.0227, + "step": 11993 + }, + { + "epoch": 0.9013978656245303, + "grad_norm": 2.2255189826136905, + "learning_rate": 1.0113223027165729e-07, + "loss": 0.9564, + "step": 11994 + }, + { + "epoch": 0.9014730196903652, + "grad_norm": 1.7731133387400317, + "learning_rate": 1.0097943413239507e-07, + "loss": 0.8842, + "step": 11995 + }, + { + "epoch": 0.9015481737562002, + "grad_norm": 2.1156650150633807, + "learning_rate": 1.0082675051737388e-07, + "loss": 0.8389, + "step": 11996 + }, + { + "epoch": 0.9016233278220351, + "grad_norm": 1.55825725546509, + "learning_rate": 1.0067417943563982e-07, + "loss": 0.9795, + "step": 11997 + }, + { + "epoch": 0.9016984818878702, + "grad_norm": 0.6668825743655333, + "learning_rate": 1.0052172089623324e-07, + "loss": 0.8333, + "step": 11998 + }, + { + "epoch": 0.9017736359537051, + "grad_norm": 2.4923108109249084, + "learning_rate": 1.0036937490818842e-07, + "loss": 0.912, + "step": 11999 + }, + { + "epoch": 0.9018487900195401, + "grad_norm": 2.0866382579068063, + "learning_rate": 1.0021714148053262e-07, + "loss": 0.9529, + "step": 12000 + }, + { + "epoch": 0.901923944085375, + "grad_norm": 1.473827837052508, + "learning_rate": 1.0006502062228572e-07, + "loss": 1.0064, + "step": 12001 + }, + { + "epoch": 0.9019990981512099, + "grad_norm": 1.8335347787405272, + "learning_rate": 9.991301234246163e-08, + "loss": 0.9566, + "step": 12002 + }, + { + "epoch": 0.902074252217045, + "grad_norm": 2.0647148031441906, + "learning_rate": 9.976111665006781e-08, + "loss": 0.9711, + "step": 12003 + }, + { + "epoch": 0.9021494062828799, + "grad_norm": 1.8567320046950533, + "learning_rate": 9.960933355410417e-08, + "loss": 0.8533, + "step": 12004 + }, + { + "epoch": 0.9022245603487149, + "grad_norm": 1.7251771227805017, + "learning_rate": 9.945766306356418e-08, + "loss": 1.1034, + "step": 12005 + }, + { + "epoch": 0.9022997144145498, + "grad_norm": 1.7296054314059606, + "learning_rate": 9.930610518743599e-08, + "loss": 0.9169, + "step": 12006 + }, + { + "epoch": 0.9023748684803848, + "grad_norm": 5.2704898083987, + "learning_rate": 9.915465993469884e-08, + "loss": 0.9919, + "step": 12007 + }, + { + "epoch": 0.9024500225462198, + "grad_norm": 1.4564133220984237, + "learning_rate": 9.900332731432715e-08, + "loss": 0.8934, + "step": 12008 + }, + { + "epoch": 0.9025251766120547, + "grad_norm": 1.4661195616008578, + "learning_rate": 9.885210733528748e-08, + "loss": 0.9423, + "step": 12009 + }, + { + "epoch": 0.9026003306778897, + "grad_norm": 2.01561352630425, + "learning_rate": 9.870100000654091e-08, + "loss": 0.9606, + "step": 12010 + }, + { + "epoch": 0.9026754847437246, + "grad_norm": 1.241016387699074, + "learning_rate": 9.855000533704006e-08, + "loss": 1.0076, + "step": 12011 + }, + { + "epoch": 0.9027506388095596, + "grad_norm": 1.6576544367215507, + "learning_rate": 9.839912333573285e-08, + "loss": 0.9242, + "step": 12012 + }, + { + "epoch": 0.9028257928753946, + "grad_norm": 0.8086841551141561, + "learning_rate": 9.824835401155928e-08, + "loss": 0.8017, + "step": 12013 + }, + { + "epoch": 0.9029009469412295, + "grad_norm": 1.5040035067730462, + "learning_rate": 9.809769737345308e-08, + "loss": 0.9772, + "step": 12014 + }, + { + "epoch": 0.9029761010070645, + "grad_norm": 1.7409493190437457, + "learning_rate": 9.794715343034088e-08, + "loss": 1.0146, + "step": 12015 + }, + { + "epoch": 0.9030512550728994, + "grad_norm": 1.9418737205596923, + "learning_rate": 9.779672219114332e-08, + "loss": 1.0354, + "step": 12016 + }, + { + "epoch": 0.9031264091387344, + "grad_norm": 2.895666067292049, + "learning_rate": 9.764640366477439e-08, + "loss": 0.9559, + "step": 12017 + }, + { + "epoch": 0.9032015632045693, + "grad_norm": 0.719942347814487, + "learning_rate": 9.749619786014052e-08, + "loss": 0.8587, + "step": 12018 + }, + { + "epoch": 0.9032767172704044, + "grad_norm": 1.7473115706463318, + "learning_rate": 9.734610478614192e-08, + "loss": 1.0871, + "step": 12019 + }, + { + "epoch": 0.9033518713362393, + "grad_norm": 0.7137317868262234, + "learning_rate": 9.719612445167258e-08, + "loss": 0.8105, + "step": 12020 + }, + { + "epoch": 0.9034270254020742, + "grad_norm": 2.4354889227267305, + "learning_rate": 9.704625686561896e-08, + "loss": 0.8644, + "step": 12021 + }, + { + "epoch": 0.9035021794679092, + "grad_norm": 0.832948313619623, + "learning_rate": 9.689650203686128e-08, + "loss": 0.8192, + "step": 12022 + }, + { + "epoch": 0.9035773335337441, + "grad_norm": 1.8089229539540026, + "learning_rate": 9.6746859974274e-08, + "loss": 1.0092, + "step": 12023 + }, + { + "epoch": 0.9036524875995792, + "grad_norm": 3.8189979585101237, + "learning_rate": 9.659733068672293e-08, + "loss": 0.9628, + "step": 12024 + }, + { + "epoch": 0.9037276416654141, + "grad_norm": 2.3637056332221498, + "learning_rate": 9.644791418306853e-08, + "loss": 0.9178, + "step": 12025 + }, + { + "epoch": 0.9038027957312491, + "grad_norm": 1.881900170600373, + "learning_rate": 9.629861047216436e-08, + "loss": 1.0421, + "step": 12026 + }, + { + "epoch": 0.903877949797084, + "grad_norm": 2.567834743975824, + "learning_rate": 9.614941956285782e-08, + "loss": 0.9345, + "step": 12027 + }, + { + "epoch": 0.9039531038629189, + "grad_norm": 2.077372594018382, + "learning_rate": 9.600034146398806e-08, + "loss": 1.0066, + "step": 12028 + }, + { + "epoch": 0.904028257928754, + "grad_norm": 1.6461575908981665, + "learning_rate": 9.58513761843891e-08, + "loss": 1.0041, + "step": 12029 + }, + { + "epoch": 0.9041034119945889, + "grad_norm": 1.5896477355842995, + "learning_rate": 9.570252373288745e-08, + "loss": 0.9955, + "step": 12030 + }, + { + "epoch": 0.9041785660604239, + "grad_norm": 1.8405361290090179, + "learning_rate": 9.555378411830339e-08, + "loss": 0.9234, + "step": 12031 + }, + { + "epoch": 0.9042537201262588, + "grad_norm": 0.9569232801918481, + "learning_rate": 9.54051573494501e-08, + "loss": 0.8963, + "step": 12032 + }, + { + "epoch": 0.9043288741920938, + "grad_norm": 2.9978661895878256, + "learning_rate": 9.525664343513495e-08, + "loss": 0.9271, + "step": 12033 + }, + { + "epoch": 0.9044040282579288, + "grad_norm": 3.2698528172678105, + "learning_rate": 9.510824238415672e-08, + "loss": 0.9745, + "step": 12034 + }, + { + "epoch": 0.9044791823237637, + "grad_norm": 2.549240230992375, + "learning_rate": 9.495995420530988e-08, + "loss": 0.979, + "step": 12035 + }, + { + "epoch": 0.9045543363895987, + "grad_norm": 1.6130379038387725, + "learning_rate": 9.481177890738012e-08, + "loss": 0.9149, + "step": 12036 + }, + { + "epoch": 0.9046294904554336, + "grad_norm": 1.8130787550227025, + "learning_rate": 9.466371649914883e-08, + "loss": 0.9939, + "step": 12037 + }, + { + "epoch": 0.9047046445212686, + "grad_norm": 1.7325785500470476, + "learning_rate": 9.451576698938746e-08, + "loss": 1.0038, + "step": 12038 + }, + { + "epoch": 0.9047797985871036, + "grad_norm": 0.6825757539368933, + "learning_rate": 9.436793038686364e-08, + "loss": 0.8421, + "step": 12039 + }, + { + "epoch": 0.9048549526529385, + "grad_norm": 1.9434385043956064, + "learning_rate": 9.42202067003377e-08, + "loss": 0.9691, + "step": 12040 + }, + { + "epoch": 0.9049301067187735, + "grad_norm": 1.558499670415996, + "learning_rate": 9.407259593856199e-08, + "loss": 0.9627, + "step": 12041 + }, + { + "epoch": 0.9050052607846084, + "grad_norm": 2.066108090788637, + "learning_rate": 9.392509811028282e-08, + "loss": 0.9952, + "step": 12042 + }, + { + "epoch": 0.9050804148504434, + "grad_norm": 1.4108276459604998, + "learning_rate": 9.377771322424055e-08, + "loss": 0.9924, + "step": 12043 + }, + { + "epoch": 0.9051555689162784, + "grad_norm": 1.2317176898785753, + "learning_rate": 9.363044128916864e-08, + "loss": 0.9848, + "step": 12044 + }, + { + "epoch": 0.9052307229821134, + "grad_norm": 2.4442130350144744, + "learning_rate": 9.348328231379255e-08, + "loss": 1.011, + "step": 12045 + }, + { + "epoch": 0.9053058770479483, + "grad_norm": 2.1894123982190057, + "learning_rate": 9.333623630683285e-08, + "loss": 0.9373, + "step": 12046 + }, + { + "epoch": 0.9053810311137832, + "grad_norm": 1.9405656372426312, + "learning_rate": 9.318930327700214e-08, + "loss": 0.9124, + "step": 12047 + }, + { + "epoch": 0.9054561851796182, + "grad_norm": 1.8978803154499655, + "learning_rate": 9.304248323300701e-08, + "loss": 0.914, + "step": 12048 + }, + { + "epoch": 0.9055313392454531, + "grad_norm": 1.8436647169141587, + "learning_rate": 9.289577618354649e-08, + "loss": 0.9089, + "step": 12049 + }, + { + "epoch": 0.9056064933112882, + "grad_norm": 4.04900760501252, + "learning_rate": 9.274918213731475e-08, + "loss": 0.9303, + "step": 12050 + }, + { + "epoch": 0.9056816473771231, + "grad_norm": 2.0306569334027738, + "learning_rate": 9.260270110299684e-08, + "loss": 0.8734, + "step": 12051 + }, + { + "epoch": 0.9057568014429581, + "grad_norm": 2.112717078979234, + "learning_rate": 9.245633308927293e-08, + "loss": 1.019, + "step": 12052 + }, + { + "epoch": 0.905831955508793, + "grad_norm": 1.4283380028208044, + "learning_rate": 9.231007810481539e-08, + "loss": 0.9804, + "step": 12053 + }, + { + "epoch": 0.905907109574628, + "grad_norm": 3.4658035273614396, + "learning_rate": 9.216393615829154e-08, + "loss": 0.9501, + "step": 12054 + }, + { + "epoch": 0.905982263640463, + "grad_norm": 1.8242575010313018, + "learning_rate": 9.201790725835933e-08, + "loss": 1.0688, + "step": 12055 + }, + { + "epoch": 0.9060574177062979, + "grad_norm": 1.3660498981577858, + "learning_rate": 9.187199141367274e-08, + "loss": 0.9304, + "step": 12056 + }, + { + "epoch": 0.9061325717721329, + "grad_norm": 2.489462004914631, + "learning_rate": 9.17261886328773e-08, + "loss": 0.967, + "step": 12057 + }, + { + "epoch": 0.9062077258379678, + "grad_norm": 1.5916938982459854, + "learning_rate": 9.158049892461228e-08, + "loss": 1.0266, + "step": 12058 + }, + { + "epoch": 0.9062828799038027, + "grad_norm": 1.5814662348242021, + "learning_rate": 9.143492229751038e-08, + "loss": 0.9653, + "step": 12059 + }, + { + "epoch": 0.9063580339696378, + "grad_norm": 1.5739085457394157, + "learning_rate": 9.128945876019777e-08, + "loss": 0.9238, + "step": 12060 + }, + { + "epoch": 0.9064331880354727, + "grad_norm": 2.2362969941883435, + "learning_rate": 9.114410832129404e-08, + "loss": 0.9807, + "step": 12061 + }, + { + "epoch": 0.9065083421013077, + "grad_norm": 1.810393918519672, + "learning_rate": 9.099887098941095e-08, + "loss": 1.0712, + "step": 12062 + }, + { + "epoch": 0.9065834961671426, + "grad_norm": 1.5469223169353297, + "learning_rate": 9.085374677315516e-08, + "loss": 0.9503, + "step": 12063 + }, + { + "epoch": 0.9066586502329776, + "grad_norm": 4.270721855879645, + "learning_rate": 9.070873568112581e-08, + "loss": 0.9522, + "step": 12064 + }, + { + "epoch": 0.9067338042988126, + "grad_norm": 1.2232463740971986, + "learning_rate": 9.056383772191422e-08, + "loss": 0.9267, + "step": 12065 + }, + { + "epoch": 0.9068089583646475, + "grad_norm": 3.2066270989443986, + "learning_rate": 9.041905290410711e-08, + "loss": 0.8472, + "step": 12066 + }, + { + "epoch": 0.9068841124304825, + "grad_norm": 2.416876871459962, + "learning_rate": 9.027438123628406e-08, + "loss": 1.0443, + "step": 12067 + }, + { + "epoch": 0.9069592664963174, + "grad_norm": 2.106241569162582, + "learning_rate": 9.012982272701597e-08, + "loss": 0.9685, + "step": 12068 + }, + { + "epoch": 0.9070344205621524, + "grad_norm": 1.4964650144148628, + "learning_rate": 8.998537738486977e-08, + "loss": 0.9011, + "step": 12069 + }, + { + "epoch": 0.9071095746279874, + "grad_norm": 1.506693924190317, + "learning_rate": 8.984104521840352e-08, + "loss": 0.9864, + "step": 12070 + }, + { + "epoch": 0.9071847286938224, + "grad_norm": 1.8066521721548212, + "learning_rate": 8.969682623617014e-08, + "loss": 0.9628, + "step": 12071 + }, + { + "epoch": 0.9072598827596573, + "grad_norm": 1.7550810212129035, + "learning_rate": 8.955272044671459e-08, + "loss": 0.9542, + "step": 12072 + }, + { + "epoch": 0.9073350368254922, + "grad_norm": 1.9436975232857034, + "learning_rate": 8.940872785857623e-08, + "loss": 0.9646, + "step": 12073 + }, + { + "epoch": 0.9074101908913272, + "grad_norm": 1.690574388574979, + "learning_rate": 8.92648484802867e-08, + "loss": 0.9424, + "step": 12074 + }, + { + "epoch": 0.9074853449571622, + "grad_norm": 1.4825274564313728, + "learning_rate": 8.912108232037163e-08, + "loss": 0.9309, + "step": 12075 + }, + { + "epoch": 0.9075604990229972, + "grad_norm": 1.900344581477483, + "learning_rate": 8.897742938734953e-08, + "loss": 0.9473, + "step": 12076 + }, + { + "epoch": 0.9076356530888321, + "grad_norm": 28.296473803064227, + "learning_rate": 8.883388968973293e-08, + "loss": 0.9194, + "step": 12077 + }, + { + "epoch": 0.9077108071546671, + "grad_norm": 2.554761599711383, + "learning_rate": 8.869046323602636e-08, + "loss": 0.9948, + "step": 12078 + }, + { + "epoch": 0.907785961220502, + "grad_norm": 1.3399288940067913, + "learning_rate": 8.85471500347288e-08, + "loss": 0.966, + "step": 12079 + }, + { + "epoch": 0.907861115286337, + "grad_norm": 1.752229904714861, + "learning_rate": 8.840395009433188e-08, + "loss": 0.9301, + "step": 12080 + }, + { + "epoch": 0.907936269352172, + "grad_norm": 1.8769909143365162, + "learning_rate": 8.82608634233215e-08, + "loss": 0.9208, + "step": 12081 + }, + { + "epoch": 0.9080114234180069, + "grad_norm": 1.60637488863019, + "learning_rate": 8.81178900301749e-08, + "loss": 1.0164, + "step": 12082 + }, + { + "epoch": 0.9080865774838419, + "grad_norm": 8.037164669843202, + "learning_rate": 8.797502992336436e-08, + "loss": 1.0246, + "step": 12083 + }, + { + "epoch": 0.9081617315496768, + "grad_norm": 1.459676187872006, + "learning_rate": 8.783228311135559e-08, + "loss": 1.0014, + "step": 12084 + }, + { + "epoch": 0.9082368856155117, + "grad_norm": 3.8259097793797507, + "learning_rate": 8.768964960260582e-08, + "loss": 0.983, + "step": 12085 + }, + { + "epoch": 0.9083120396813468, + "grad_norm": 1.6010203118307567, + "learning_rate": 8.754712940556675e-08, + "loss": 0.8798, + "step": 12086 + }, + { + "epoch": 0.9083871937471817, + "grad_norm": 2.076795109450073, + "learning_rate": 8.740472252868359e-08, + "loss": 0.9756, + "step": 12087 + }, + { + "epoch": 0.9084623478130167, + "grad_norm": 2.082498065082068, + "learning_rate": 8.726242898039516e-08, + "loss": 0.9819, + "step": 12088 + }, + { + "epoch": 0.9085375018788516, + "grad_norm": 2.6612281376275293, + "learning_rate": 8.712024876913138e-08, + "loss": 0.7816, + "step": 12089 + }, + { + "epoch": 0.9086126559446867, + "grad_norm": 1.7619584083367974, + "learning_rate": 8.697818190331818e-08, + "loss": 0.939, + "step": 12090 + }, + { + "epoch": 0.9086878100105216, + "grad_norm": 1.6501530402908773, + "learning_rate": 8.683622839137306e-08, + "loss": 0.9491, + "step": 12091 + }, + { + "epoch": 0.9087629640763565, + "grad_norm": 2.513847528706347, + "learning_rate": 8.669438824170727e-08, + "loss": 0.9783, + "step": 12092 + }, + { + "epoch": 0.9088381181421915, + "grad_norm": 1.8648197372290463, + "learning_rate": 8.655266146272544e-08, + "loss": 0.9476, + "step": 12093 + }, + { + "epoch": 0.9089132722080264, + "grad_norm": 1.503421616185897, + "learning_rate": 8.641104806282595e-08, + "loss": 0.9369, + "step": 12094 + }, + { + "epoch": 0.9089884262738614, + "grad_norm": 1.742482318651641, + "learning_rate": 8.626954805039921e-08, + "loss": 0.9482, + "step": 12095 + }, + { + "epoch": 0.9090635803396964, + "grad_norm": 3.601572164990082, + "learning_rate": 8.612816143382983e-08, + "loss": 0.926, + "step": 12096 + }, + { + "epoch": 0.9091387344055314, + "grad_norm": 1.9420463995649935, + "learning_rate": 8.598688822149557e-08, + "loss": 0.9309, + "step": 12097 + }, + { + "epoch": 0.9092138884713663, + "grad_norm": 2.53917536558041, + "learning_rate": 8.584572842176774e-08, + "loss": 1.0781, + "step": 12098 + }, + { + "epoch": 0.9092890425372012, + "grad_norm": 1.83875263361739, + "learning_rate": 8.570468204300984e-08, + "loss": 0.949, + "step": 12099 + }, + { + "epoch": 0.9093641966030362, + "grad_norm": 1.6009981230963641, + "learning_rate": 8.556374909358011e-08, + "loss": 0.9445, + "step": 12100 + }, + { + "epoch": 0.9094393506688712, + "grad_norm": 1.7508160187223543, + "learning_rate": 8.542292958182917e-08, + "loss": 1.0443, + "step": 12101 + }, + { + "epoch": 0.9095145047347062, + "grad_norm": 1.800538085937235, + "learning_rate": 8.528222351610104e-08, + "loss": 0.959, + "step": 12102 + }, + { + "epoch": 0.9095896588005411, + "grad_norm": 3.214446366549053, + "learning_rate": 8.514163090473281e-08, + "loss": 0.9182, + "step": 12103 + }, + { + "epoch": 0.909664812866376, + "grad_norm": 2.062867565347588, + "learning_rate": 8.50011517560556e-08, + "loss": 0.9091, + "step": 12104 + }, + { + "epoch": 0.909739966932211, + "grad_norm": 2.3093858318565217, + "learning_rate": 8.486078607839341e-08, + "loss": 1.0441, + "step": 12105 + }, + { + "epoch": 0.909815120998046, + "grad_norm": 1.6133350428002078, + "learning_rate": 8.472053388006295e-08, + "loss": 0.9939, + "step": 12106 + }, + { + "epoch": 0.909890275063881, + "grad_norm": 2.1444995648487493, + "learning_rate": 8.45803951693751e-08, + "loss": 0.9387, + "step": 12107 + }, + { + "epoch": 0.9099654291297159, + "grad_norm": 1.6248793190379949, + "learning_rate": 8.444036995463366e-08, + "loss": 0.906, + "step": 12108 + }, + { + "epoch": 0.9100405831955509, + "grad_norm": 1.72806887268196, + "learning_rate": 8.430045824413512e-08, + "loss": 1.0099, + "step": 12109 + }, + { + "epoch": 0.9101157372613858, + "grad_norm": 2.4874744394800623, + "learning_rate": 8.416066004616995e-08, + "loss": 0.9726, + "step": 12110 + }, + { + "epoch": 0.9101908913272208, + "grad_norm": 1.867087772531247, + "learning_rate": 8.402097536902242e-08, + "loss": 0.963, + "step": 12111 + }, + { + "epoch": 0.9102660453930558, + "grad_norm": 3.4543799972152938, + "learning_rate": 8.388140422096856e-08, + "loss": 0.8568, + "step": 12112 + }, + { + "epoch": 0.9103411994588907, + "grad_norm": 1.8038481374626145, + "learning_rate": 8.374194661027889e-08, + "loss": 0.8357, + "step": 12113 + }, + { + "epoch": 0.9104163535247257, + "grad_norm": 1.8553664651083444, + "learning_rate": 8.360260254521656e-08, + "loss": 0.9387, + "step": 12114 + }, + { + "epoch": 0.9104915075905606, + "grad_norm": 1.9670860282536347, + "learning_rate": 8.346337203403874e-08, + "loss": 0.8727, + "step": 12115 + }, + { + "epoch": 0.9105666616563957, + "grad_norm": 1.7371194107043701, + "learning_rate": 8.332425508499463e-08, + "loss": 0.9272, + "step": 12116 + }, + { + "epoch": 0.9106418157222306, + "grad_norm": 1.460403426360199, + "learning_rate": 8.318525170632829e-08, + "loss": 0.9093, + "step": 12117 + }, + { + "epoch": 0.9107169697880655, + "grad_norm": 1.7523405907543341, + "learning_rate": 8.304636190627557e-08, + "loss": 0.8881, + "step": 12118 + }, + { + "epoch": 0.9107921238539005, + "grad_norm": 1.7008560886134911, + "learning_rate": 8.290758569306633e-08, + "loss": 0.9639, + "step": 12119 + }, + { + "epoch": 0.9108672779197354, + "grad_norm": 1.5431008584168522, + "learning_rate": 8.276892307492356e-08, + "loss": 1.0356, + "step": 12120 + }, + { + "epoch": 0.9109424319855705, + "grad_norm": 1.742058171830593, + "learning_rate": 8.263037406006423e-08, + "loss": 1.0099, + "step": 12121 + }, + { + "epoch": 0.9110175860514054, + "grad_norm": 1.339179761419424, + "learning_rate": 8.249193865669669e-08, + "loss": 0.9527, + "step": 12122 + }, + { + "epoch": 0.9110927401172404, + "grad_norm": 1.6412889862659739, + "learning_rate": 8.235361687302478e-08, + "loss": 0.9157, + "step": 12123 + }, + { + "epoch": 0.9111678941830753, + "grad_norm": 1.5693908567854262, + "learning_rate": 8.221540871724398e-08, + "loss": 0.8387, + "step": 12124 + }, + { + "epoch": 0.9112430482489102, + "grad_norm": 1.5904311236206752, + "learning_rate": 8.207731419754415e-08, + "loss": 0.9306, + "step": 12125 + }, + { + "epoch": 0.9113182023147453, + "grad_norm": 2.3251722415578975, + "learning_rate": 8.193933332210745e-08, + "loss": 0.9963, + "step": 12126 + }, + { + "epoch": 0.9113933563805802, + "grad_norm": 1.709782458739526, + "learning_rate": 8.180146609910998e-08, + "loss": 1.0234, + "step": 12127 + }, + { + "epoch": 0.9114685104464152, + "grad_norm": 0.8173832500278847, + "learning_rate": 8.166371253672122e-08, + "loss": 0.8577, + "step": 12128 + }, + { + "epoch": 0.9115436645122501, + "grad_norm": 1.6731024541180102, + "learning_rate": 8.152607264310308e-08, + "loss": 1.0218, + "step": 12129 + }, + { + "epoch": 0.911618818578085, + "grad_norm": 2.695226872507602, + "learning_rate": 8.138854642641147e-08, + "loss": 0.9893, + "step": 12130 + }, + { + "epoch": 0.91169397264392, + "grad_norm": 1.899045619189657, + "learning_rate": 8.125113389479521e-08, + "loss": 0.8651, + "step": 12131 + }, + { + "epoch": 0.911769126709755, + "grad_norm": 2.146345861268137, + "learning_rate": 8.111383505639713e-08, + "loss": 0.9821, + "step": 12132 + }, + { + "epoch": 0.91184428077559, + "grad_norm": 1.597816739031016, + "learning_rate": 8.09766499193516e-08, + "loss": 0.9118, + "step": 12133 + }, + { + "epoch": 0.9119194348414249, + "grad_norm": 1.9793418070615088, + "learning_rate": 8.083957849178835e-08, + "loss": 1.006, + "step": 12134 + }, + { + "epoch": 0.9119945889072599, + "grad_norm": 1.5241081913318646, + "learning_rate": 8.07026207818291e-08, + "loss": 0.9375, + "step": 12135 + }, + { + "epoch": 0.9120697429730948, + "grad_norm": 2.5495235035320336, + "learning_rate": 8.056577679758891e-08, + "loss": 0.8946, + "step": 12136 + }, + { + "epoch": 0.9121448970389298, + "grad_norm": 2.178603035833775, + "learning_rate": 8.042904654717642e-08, + "loss": 0.8776, + "step": 12137 + }, + { + "epoch": 0.9122200511047648, + "grad_norm": 1.8866174643046383, + "learning_rate": 8.029243003869379e-08, + "loss": 0.9909, + "step": 12138 + }, + { + "epoch": 0.9122952051705997, + "grad_norm": 1.8965391235620577, + "learning_rate": 8.015592728023523e-08, + "loss": 0.9206, + "step": 12139 + }, + { + "epoch": 0.9123703592364347, + "grad_norm": 1.774680957866096, + "learning_rate": 8.001953827988984e-08, + "loss": 0.9743, + "step": 12140 + }, + { + "epoch": 0.9124455133022696, + "grad_norm": 2.388093293455486, + "learning_rate": 7.988326304573844e-08, + "loss": 1.0174, + "step": 12141 + }, + { + "epoch": 0.9125206673681047, + "grad_norm": 0.7465237481848812, + "learning_rate": 7.974710158585706e-08, + "loss": 0.8384, + "step": 12142 + }, + { + "epoch": 0.9125958214339396, + "grad_norm": 2.17483044982917, + "learning_rate": 7.961105390831257e-08, + "loss": 0.9336, + "step": 12143 + }, + { + "epoch": 0.9126709754997745, + "grad_norm": 1.6458292932903424, + "learning_rate": 7.947512002116697e-08, + "loss": 1.0014, + "step": 12144 + }, + { + "epoch": 0.9127461295656095, + "grad_norm": 2.106340953096948, + "learning_rate": 7.933929993247468e-08, + "loss": 0.9251, + "step": 12145 + }, + { + "epoch": 0.9128212836314444, + "grad_norm": 1.5697087887674623, + "learning_rate": 7.92035936502835e-08, + "loss": 0.9305, + "step": 12146 + }, + { + "epoch": 0.9128964376972795, + "grad_norm": 1.6705325432227618, + "learning_rate": 7.906800118263456e-08, + "loss": 0.9001, + "step": 12147 + }, + { + "epoch": 0.9129715917631144, + "grad_norm": 2.5744567668438605, + "learning_rate": 7.893252253756211e-08, + "loss": 0.8665, + "step": 12148 + }, + { + "epoch": 0.9130467458289493, + "grad_norm": 8.00029706403412, + "learning_rate": 7.879715772309458e-08, + "loss": 1.0241, + "step": 12149 + }, + { + "epoch": 0.9131218998947843, + "grad_norm": 1.8341720208808134, + "learning_rate": 7.866190674725159e-08, + "loss": 1.0882, + "step": 12150 + }, + { + "epoch": 0.9131970539606192, + "grad_norm": 4.635621735683658, + "learning_rate": 7.852676961804827e-08, + "loss": 0.9735, + "step": 12151 + }, + { + "epoch": 0.9132722080264543, + "grad_norm": 3.503890599924986, + "learning_rate": 7.839174634349178e-08, + "loss": 0.9752, + "step": 12152 + }, + { + "epoch": 0.9133473620922892, + "grad_norm": 1.708844093014798, + "learning_rate": 7.825683693158236e-08, + "loss": 0.9405, + "step": 12153 + }, + { + "epoch": 0.9134225161581242, + "grad_norm": 2.954358847267084, + "learning_rate": 7.812204139031409e-08, + "loss": 0.965, + "step": 12154 + }, + { + "epoch": 0.9134976702239591, + "grad_norm": 1.8175835188575133, + "learning_rate": 7.798735972767478e-08, + "loss": 0.937, + "step": 12155 + }, + { + "epoch": 0.913572824289794, + "grad_norm": 1.384515995397502, + "learning_rate": 7.785279195164384e-08, + "loss": 1.0048, + "step": 12156 + }, + { + "epoch": 0.913647978355629, + "grad_norm": 2.723984250783555, + "learning_rate": 7.771833807019557e-08, + "loss": 0.9062, + "step": 12157 + }, + { + "epoch": 0.913723132421464, + "grad_norm": 2.364163421675246, + "learning_rate": 7.758399809129645e-08, + "loss": 1.0593, + "step": 12158 + }, + { + "epoch": 0.913798286487299, + "grad_norm": 1.531439088318121, + "learning_rate": 7.744977202290725e-08, + "loss": 1.0095, + "step": 12159 + }, + { + "epoch": 0.9138734405531339, + "grad_norm": 1.9499790337242375, + "learning_rate": 7.73156598729805e-08, + "loss": 1.04, + "step": 12160 + }, + { + "epoch": 0.9139485946189689, + "grad_norm": 1.41971768218866, + "learning_rate": 7.718166164946361e-08, + "loss": 0.8996, + "step": 12161 + }, + { + "epoch": 0.9140237486848038, + "grad_norm": 0.6661535215393959, + "learning_rate": 7.704777736029621e-08, + "loss": 0.8619, + "step": 12162 + }, + { + "epoch": 0.9140989027506388, + "grad_norm": 1.5103532010479845, + "learning_rate": 7.691400701341156e-08, + "loss": 1.0304, + "step": 12163 + }, + { + "epoch": 0.9141740568164738, + "grad_norm": 1.799592658152191, + "learning_rate": 7.678035061673572e-08, + "loss": 0.9807, + "step": 12164 + }, + { + "epoch": 0.9142492108823087, + "grad_norm": 2.025094432748224, + "learning_rate": 7.664680817818925e-08, + "loss": 0.9029, + "step": 12165 + }, + { + "epoch": 0.9143243649481437, + "grad_norm": 2.130282399973697, + "learning_rate": 7.651337970568361e-08, + "loss": 1.0474, + "step": 12166 + }, + { + "epoch": 0.9143995190139786, + "grad_norm": 1.8808868818163205, + "learning_rate": 7.638006520712625e-08, + "loss": 0.8933, + "step": 12167 + }, + { + "epoch": 0.9144746730798137, + "grad_norm": 2.239654279664483, + "learning_rate": 7.624686469041575e-08, + "loss": 1.0544, + "step": 12168 + }, + { + "epoch": 0.9145498271456486, + "grad_norm": 1.6122656481872666, + "learning_rate": 7.611377816344533e-08, + "loss": 0.9507, + "step": 12169 + }, + { + "epoch": 0.9146249812114835, + "grad_norm": 1.3432914040050705, + "learning_rate": 7.598080563410048e-08, + "loss": 0.9758, + "step": 12170 + }, + { + "epoch": 0.9147001352773185, + "grad_norm": 1.8675144819032619, + "learning_rate": 7.584794711026021e-08, + "loss": 0.9208, + "step": 12171 + }, + { + "epoch": 0.9147752893431534, + "grad_norm": 1.930218704560839, + "learning_rate": 7.571520259979757e-08, + "loss": 0.9036, + "step": 12172 + }, + { + "epoch": 0.9148504434089885, + "grad_norm": 1.8048674918908318, + "learning_rate": 7.558257211057739e-08, + "loss": 1.0902, + "step": 12173 + }, + { + "epoch": 0.9149255974748234, + "grad_norm": 1.8760516078698994, + "learning_rate": 7.545005565045914e-08, + "loss": 0.983, + "step": 12174 + }, + { + "epoch": 0.9150007515406583, + "grad_norm": 1.8416057985123357, + "learning_rate": 7.531765322729455e-08, + "loss": 1.0487, + "step": 12175 + }, + { + "epoch": 0.9150759056064933, + "grad_norm": 1.5695708691850028, + "learning_rate": 7.518536484892956e-08, + "loss": 0.8713, + "step": 12176 + }, + { + "epoch": 0.9151510596723282, + "grad_norm": 1.776612854632306, + "learning_rate": 7.50531905232017e-08, + "loss": 1.0103, + "step": 12177 + }, + { + "epoch": 0.9152262137381633, + "grad_norm": 1.7580543182192427, + "learning_rate": 7.492113025794378e-08, + "loss": 0.9907, + "step": 12178 + }, + { + "epoch": 0.9153013678039982, + "grad_norm": 3.536102067233781, + "learning_rate": 7.478918406098044e-08, + "loss": 0.9567, + "step": 12179 + }, + { + "epoch": 0.9153765218698332, + "grad_norm": 1.7062060982781757, + "learning_rate": 7.46573519401299e-08, + "loss": 1.0031, + "step": 12180 + }, + { + "epoch": 0.9154516759356681, + "grad_norm": 1.472793855992255, + "learning_rate": 7.452563390320388e-08, + "loss": 0.8745, + "step": 12181 + }, + { + "epoch": 0.915526830001503, + "grad_norm": 2.0379652089002867, + "learning_rate": 7.43940299580077e-08, + "loss": 0.925, + "step": 12182 + }, + { + "epoch": 0.9156019840673381, + "grad_norm": 2.0989907564320425, + "learning_rate": 7.426254011233823e-08, + "loss": 1.0197, + "step": 12183 + }, + { + "epoch": 0.915677138133173, + "grad_norm": 1.7214280505538542, + "learning_rate": 7.41311643739877e-08, + "loss": 0.9412, + "step": 12184 + }, + { + "epoch": 0.915752292199008, + "grad_norm": 1.5528046214210334, + "learning_rate": 7.399990275074009e-08, + "loss": 1.004, + "step": 12185 + }, + { + "epoch": 0.9158274462648429, + "grad_norm": 2.2639311674053855, + "learning_rate": 7.386875525037384e-08, + "loss": 0.9794, + "step": 12186 + }, + { + "epoch": 0.9159026003306779, + "grad_norm": 1.7067764755526718, + "learning_rate": 7.373772188065919e-08, + "loss": 0.8729, + "step": 12187 + }, + { + "epoch": 0.9159777543965129, + "grad_norm": 1.723862759648277, + "learning_rate": 7.360680264936037e-08, + "loss": 0.9847, + "step": 12188 + }, + { + "epoch": 0.9160529084623478, + "grad_norm": 5.549877460948392, + "learning_rate": 7.347599756423584e-08, + "loss": 0.962, + "step": 12189 + }, + { + "epoch": 0.9161280625281828, + "grad_norm": 1.7196484384982353, + "learning_rate": 7.334530663303539e-08, + "loss": 0.9194, + "step": 12190 + }, + { + "epoch": 0.9162032165940177, + "grad_norm": 2.1410245870490683, + "learning_rate": 7.321472986350286e-08, + "loss": 1.0736, + "step": 12191 + }, + { + "epoch": 0.9162783706598527, + "grad_norm": 1.63981179564712, + "learning_rate": 7.308426726337557e-08, + "loss": 0.9868, + "step": 12192 + }, + { + "epoch": 0.9163535247256877, + "grad_norm": 8.244443467178128, + "learning_rate": 7.295391884038493e-08, + "loss": 1.0102, + "step": 12193 + }, + { + "epoch": 0.9164286787915226, + "grad_norm": 0.7769410897191118, + "learning_rate": 7.282368460225297e-08, + "loss": 0.8567, + "step": 12194 + }, + { + "epoch": 0.9165038328573576, + "grad_norm": 2.171437610521361, + "learning_rate": 7.269356455669772e-08, + "loss": 0.9548, + "step": 12195 + }, + { + "epoch": 0.9165789869231925, + "grad_norm": 1.6129678417152626, + "learning_rate": 7.256355871142883e-08, + "loss": 1.0171, + "step": 12196 + }, + { + "epoch": 0.9166541409890275, + "grad_norm": 2.146930734946509, + "learning_rate": 7.243366707414988e-08, + "loss": 0.9343, + "step": 12197 + }, + { + "epoch": 0.9167292950548624, + "grad_norm": 1.657951416753317, + "learning_rate": 7.230388965255695e-08, + "loss": 0.9718, + "step": 12198 + }, + { + "epoch": 0.9168044491206975, + "grad_norm": 2.1866348735700023, + "learning_rate": 7.217422645434079e-08, + "loss": 0.9617, + "step": 12199 + }, + { + "epoch": 0.9168796031865324, + "grad_norm": 2.0506634778310904, + "learning_rate": 7.204467748718324e-08, + "loss": 0.9779, + "step": 12200 + }, + { + "epoch": 0.9169547572523673, + "grad_norm": 1.9456083386838556, + "learning_rate": 7.191524275876148e-08, + "loss": 1.033, + "step": 12201 + }, + { + "epoch": 0.9170299113182023, + "grad_norm": 1.7427826963779602, + "learning_rate": 7.178592227674451e-08, + "loss": 0.87, + "step": 12202 + }, + { + "epoch": 0.9171050653840372, + "grad_norm": 3.23211753311413, + "learning_rate": 7.165671604879575e-08, + "loss": 0.9501, + "step": 12203 + }, + { + "epoch": 0.9171802194498723, + "grad_norm": 1.7156883019614328, + "learning_rate": 7.152762408257018e-08, + "loss": 0.9401, + "step": 12204 + }, + { + "epoch": 0.9172553735157072, + "grad_norm": 1.9964589122744971, + "learning_rate": 7.139864638571768e-08, + "loss": 0.9157, + "step": 12205 + }, + { + "epoch": 0.9173305275815422, + "grad_norm": 1.9515919276172768, + "learning_rate": 7.12697829658806e-08, + "loss": 0.989, + "step": 12206 + }, + { + "epoch": 0.9174056816473771, + "grad_norm": 0.7380717146695016, + "learning_rate": 7.114103383069436e-08, + "loss": 0.8633, + "step": 12207 + }, + { + "epoch": 0.917480835713212, + "grad_norm": 1.500000635782743, + "learning_rate": 7.101239898778778e-08, + "loss": 1.0011, + "step": 12208 + }, + { + "epoch": 0.9175559897790471, + "grad_norm": 1.7994645408407708, + "learning_rate": 7.088387844478316e-08, + "loss": 0.9644, + "step": 12209 + }, + { + "epoch": 0.917631143844882, + "grad_norm": 1.6404672456010119, + "learning_rate": 7.075547220929624e-08, + "loss": 0.904, + "step": 12210 + }, + { + "epoch": 0.917706297910717, + "grad_norm": 1.4199157462847016, + "learning_rate": 7.062718028893466e-08, + "loss": 0.9599, + "step": 12211 + }, + { + "epoch": 0.9177814519765519, + "grad_norm": 1.3522868337835283, + "learning_rate": 7.049900269130105e-08, + "loss": 0.9269, + "step": 12212 + }, + { + "epoch": 0.917856606042387, + "grad_norm": 2.155227529281401, + "learning_rate": 7.03709394239902e-08, + "loss": 0.9918, + "step": 12213 + }, + { + "epoch": 0.9179317601082219, + "grad_norm": 1.5002264805523329, + "learning_rate": 7.024299049459003e-08, + "loss": 1.0656, + "step": 12214 + }, + { + "epoch": 0.9180069141740568, + "grad_norm": 1.7038653803020631, + "learning_rate": 7.011515591068184e-08, + "loss": 1.0374, + "step": 12215 + }, + { + "epoch": 0.9180820682398918, + "grad_norm": 2.127066448965384, + "learning_rate": 6.998743567984133e-08, + "loss": 0.9357, + "step": 12216 + }, + { + "epoch": 0.9181572223057267, + "grad_norm": 3.958508350451683, + "learning_rate": 6.985982980963534e-08, + "loss": 0.8977, + "step": 12217 + }, + { + "epoch": 0.9182323763715617, + "grad_norm": 2.057279047545027, + "learning_rate": 6.97323383076256e-08, + "loss": 0.9788, + "step": 12218 + }, + { + "epoch": 0.9183075304373967, + "grad_norm": 0.7910196563241217, + "learning_rate": 6.960496118136627e-08, + "loss": 0.9061, + "step": 12219 + }, + { + "epoch": 0.9183826845032316, + "grad_norm": 1.9820704858104379, + "learning_rate": 6.947769843840511e-08, + "loss": 0.9262, + "step": 12220 + }, + { + "epoch": 0.9184578385690666, + "grad_norm": 2.4768859459526618, + "learning_rate": 6.935055008628254e-08, + "loss": 0.7958, + "step": 12221 + }, + { + "epoch": 0.9185329926349015, + "grad_norm": 3.446808021150579, + "learning_rate": 6.922351613253297e-08, + "loss": 0.9054, + "step": 12222 + }, + { + "epoch": 0.9186081467007365, + "grad_norm": 2.0000077485888106, + "learning_rate": 6.909659658468347e-08, + "loss": 0.9115, + "step": 12223 + }, + { + "epoch": 0.9186833007665715, + "grad_norm": 1.832723133090568, + "learning_rate": 6.896979145025472e-08, + "loss": 0.9424, + "step": 12224 + }, + { + "epoch": 0.9187584548324065, + "grad_norm": 1.4775164716547935, + "learning_rate": 6.88431007367598e-08, + "loss": 0.9189, + "step": 12225 + }, + { + "epoch": 0.9188336088982414, + "grad_norm": 1.7179428285982492, + "learning_rate": 6.871652445170672e-08, + "loss": 1.0173, + "step": 12226 + }, + { + "epoch": 0.9189087629640763, + "grad_norm": 1.8269495649386684, + "learning_rate": 6.859006260259437e-08, + "loss": 0.9388, + "step": 12227 + }, + { + "epoch": 0.9189839170299113, + "grad_norm": 1.9918239486180018, + "learning_rate": 6.846371519691673e-08, + "loss": 1.0306, + "step": 12228 + }, + { + "epoch": 0.9190590710957462, + "grad_norm": 3.0208909303282456, + "learning_rate": 6.833748224216029e-08, + "loss": 0.9566, + "step": 12229 + }, + { + "epoch": 0.9191342251615813, + "grad_norm": 1.9064874344838596, + "learning_rate": 6.821136374580528e-08, + "loss": 0.9274, + "step": 12230 + }, + { + "epoch": 0.9192093792274162, + "grad_norm": 2.236640262059421, + "learning_rate": 6.808535971532393e-08, + "loss": 0.897, + "step": 12231 + }, + { + "epoch": 0.9192845332932512, + "grad_norm": 1.9114927519177662, + "learning_rate": 6.79594701581827e-08, + "loss": 1.079, + "step": 12232 + }, + { + "epoch": 0.9193596873590861, + "grad_norm": 1.943154605064494, + "learning_rate": 6.783369508184145e-08, + "loss": 1.0222, + "step": 12233 + }, + { + "epoch": 0.919434841424921, + "grad_norm": 1.6973157025885137, + "learning_rate": 6.770803449375262e-08, + "loss": 0.8355, + "step": 12234 + }, + { + "epoch": 0.9195099954907561, + "grad_norm": 2.2759038911113816, + "learning_rate": 6.75824884013616e-08, + "loss": 0.9581, + "step": 12235 + }, + { + "epoch": 0.919585149556591, + "grad_norm": 1.764823934402437, + "learning_rate": 6.745705681210778e-08, + "loss": 0.9972, + "step": 12236 + }, + { + "epoch": 0.919660303622426, + "grad_norm": 2.005708891248455, + "learning_rate": 6.73317397334241e-08, + "loss": 0.8862, + "step": 12237 + }, + { + "epoch": 0.9197354576882609, + "grad_norm": 1.431636045274266, + "learning_rate": 6.720653717273506e-08, + "loss": 1.0204, + "step": 12238 + }, + { + "epoch": 0.9198106117540958, + "grad_norm": 5.197300489008392, + "learning_rate": 6.708144913746006e-08, + "loss": 0.9498, + "step": 12239 + }, + { + "epoch": 0.9198857658199309, + "grad_norm": 1.637601191182302, + "learning_rate": 6.695647563501072e-08, + "loss": 0.9533, + "step": 12240 + }, + { + "epoch": 0.9199609198857658, + "grad_norm": 1.653809747056359, + "learning_rate": 6.683161667279247e-08, + "loss": 0.9861, + "step": 12241 + }, + { + "epoch": 0.9200360739516008, + "grad_norm": 2.797968986137569, + "learning_rate": 6.670687225820315e-08, + "loss": 0.8195, + "step": 12242 + }, + { + "epoch": 0.9201112280174357, + "grad_norm": 1.5493807293907924, + "learning_rate": 6.65822423986353e-08, + "loss": 0.9467, + "step": 12243 + }, + { + "epoch": 0.9201863820832707, + "grad_norm": 2.021679442203637, + "learning_rate": 6.645772710147279e-08, + "loss": 0.9419, + "step": 12244 + }, + { + "epoch": 0.9202615361491057, + "grad_norm": 2.2106329502591526, + "learning_rate": 6.633332637409395e-08, + "loss": 0.9122, + "step": 12245 + }, + { + "epoch": 0.9203366902149406, + "grad_norm": 1.7579968843182934, + "learning_rate": 6.620904022386997e-08, + "loss": 0.9592, + "step": 12246 + }, + { + "epoch": 0.9204118442807756, + "grad_norm": 0.8099389059668471, + "learning_rate": 6.608486865816587e-08, + "loss": 0.9068, + "step": 12247 + }, + { + "epoch": 0.9204869983466105, + "grad_norm": 1.890821147233282, + "learning_rate": 6.59608116843382e-08, + "loss": 0.9831, + "step": 12248 + }, + { + "epoch": 0.9205621524124455, + "grad_norm": 1.8622598787707105, + "learning_rate": 6.583686930973864e-08, + "loss": 0.8868, + "step": 12249 + }, + { + "epoch": 0.9206373064782805, + "grad_norm": 2.6512558606346928, + "learning_rate": 6.571304154171087e-08, + "loss": 1.0094, + "step": 12250 + }, + { + "epoch": 0.9207124605441155, + "grad_norm": 0.7613547213328764, + "learning_rate": 6.558932838759234e-08, + "loss": 0.841, + "step": 12251 + }, + { + "epoch": 0.9207876146099504, + "grad_norm": 1.2917805078033096, + "learning_rate": 6.546572985471322e-08, + "loss": 0.9964, + "step": 12252 + }, + { + "epoch": 0.9208627686757853, + "grad_norm": 2.2347897231415166, + "learning_rate": 6.534224595039739e-08, + "loss": 0.9656, + "step": 12253 + }, + { + "epoch": 0.9209379227416203, + "grad_norm": 1.996729680416579, + "learning_rate": 6.521887668196213e-08, + "loss": 0.9269, + "step": 12254 + }, + { + "epoch": 0.9210130768074553, + "grad_norm": 1.4810603829231204, + "learning_rate": 6.509562205671692e-08, + "loss": 0.9541, + "step": 12255 + }, + { + "epoch": 0.9210882308732903, + "grad_norm": 2.321277353580587, + "learning_rate": 6.49724820819657e-08, + "loss": 0.9863, + "step": 12256 + }, + { + "epoch": 0.9211633849391252, + "grad_norm": 1.7564789863717214, + "learning_rate": 6.484945676500464e-08, + "loss": 0.9559, + "step": 12257 + }, + { + "epoch": 0.9212385390049602, + "grad_norm": 1.422393620406778, + "learning_rate": 6.4726546113123e-08, + "loss": 1.0313, + "step": 12258 + }, + { + "epoch": 0.9213136930707951, + "grad_norm": 2.518637139226891, + "learning_rate": 6.46037501336043e-08, + "loss": 0.9758, + "step": 12259 + }, + { + "epoch": 0.92138884713663, + "grad_norm": 1.7407866178641587, + "learning_rate": 6.448106883372495e-08, + "loss": 1.0369, + "step": 12260 + }, + { + "epoch": 0.9214640012024651, + "grad_norm": 1.9137684498899838, + "learning_rate": 6.435850222075334e-08, + "loss": 0.9847, + "step": 12261 + }, + { + "epoch": 0.9215391552683, + "grad_norm": 1.7376260615661772, + "learning_rate": 6.423605030195278e-08, + "loss": 0.9847, + "step": 12262 + }, + { + "epoch": 0.921614309334135, + "grad_norm": 1.6270496206987488, + "learning_rate": 6.411371308457857e-08, + "loss": 0.9889, + "step": 12263 + }, + { + "epoch": 0.9216894633999699, + "grad_norm": 1.5946151031021956, + "learning_rate": 6.399149057588027e-08, + "loss": 1.0045, + "step": 12264 + }, + { + "epoch": 0.9217646174658048, + "grad_norm": 1.6232729317134997, + "learning_rate": 6.386938278309916e-08, + "loss": 0.974, + "step": 12265 + }, + { + "epoch": 0.9218397715316399, + "grad_norm": 1.7184364032759205, + "learning_rate": 6.374738971347149e-08, + "loss": 0.9228, + "step": 12266 + }, + { + "epoch": 0.9219149255974748, + "grad_norm": 1.9736532279142647, + "learning_rate": 6.362551137422522e-08, + "loss": 0.936, + "step": 12267 + }, + { + "epoch": 0.9219900796633098, + "grad_norm": 1.8057818376949972, + "learning_rate": 6.350374777258216e-08, + "loss": 0.931, + "step": 12268 + }, + { + "epoch": 0.9220652337291447, + "grad_norm": 1.6805321277565255, + "learning_rate": 6.33820989157574e-08, + "loss": 0.9443, + "step": 12269 + }, + { + "epoch": 0.9221403877949798, + "grad_norm": 1.4556378203401135, + "learning_rate": 6.326056481095942e-08, + "loss": 0.9029, + "step": 12270 + }, + { + "epoch": 0.9222155418608147, + "grad_norm": 2.819339235552929, + "learning_rate": 6.313914546538868e-08, + "loss": 0.9711, + "step": 12271 + }, + { + "epoch": 0.9222906959266496, + "grad_norm": 1.7179102840423752, + "learning_rate": 6.301784088624073e-08, + "loss": 1.028, + "step": 12272 + }, + { + "epoch": 0.9223658499924846, + "grad_norm": 1.6437336126750788, + "learning_rate": 6.289665108070252e-08, + "loss": 0.9608, + "step": 12273 + }, + { + "epoch": 0.9224410040583195, + "grad_norm": 2.0751490114320026, + "learning_rate": 6.277557605595607e-08, + "loss": 0.9589, + "step": 12274 + }, + { + "epoch": 0.9225161581241546, + "grad_norm": 0.746806857460077, + "learning_rate": 6.26546158191743e-08, + "loss": 0.8379, + "step": 12275 + }, + { + "epoch": 0.9225913121899895, + "grad_norm": 2.8152113666761327, + "learning_rate": 6.253377037752506e-08, + "loss": 1.0134, + "step": 12276 + }, + { + "epoch": 0.9226664662558245, + "grad_norm": 2.038837171944438, + "learning_rate": 6.241303973816947e-08, + "loss": 0.9378, + "step": 12277 + }, + { + "epoch": 0.9227416203216594, + "grad_norm": 2.581773430537788, + "learning_rate": 6.229242390826029e-08, + "loss": 0.9677, + "step": 12278 + }, + { + "epoch": 0.9228167743874943, + "grad_norm": 0.7320493030255829, + "learning_rate": 6.217192289494533e-08, + "loss": 0.7999, + "step": 12279 + }, + { + "epoch": 0.9228919284533293, + "grad_norm": 1.5245205672776245, + "learning_rate": 6.2051536705364e-08, + "loss": 1.0661, + "step": 12280 + }, + { + "epoch": 0.9229670825191643, + "grad_norm": 2.1932752394398523, + "learning_rate": 6.193126534665061e-08, + "loss": 0.976, + "step": 12281 + }, + { + "epoch": 0.9230422365849993, + "grad_norm": 1.8345135949983484, + "learning_rate": 6.181110882593054e-08, + "loss": 0.8913, + "step": 12282 + }, + { + "epoch": 0.9231173906508342, + "grad_norm": 3.2494801325592024, + "learning_rate": 6.169106715032435e-08, + "loss": 0.9498, + "step": 12283 + }, + { + "epoch": 0.9231925447166691, + "grad_norm": 2.074233692639797, + "learning_rate": 6.157114032694477e-08, + "loss": 1.031, + "step": 12284 + }, + { + "epoch": 0.9232676987825041, + "grad_norm": 1.2950629938482832, + "learning_rate": 6.14513283628979e-08, + "loss": 0.9578, + "step": 12285 + }, + { + "epoch": 0.9233428528483391, + "grad_norm": 2.5367139077852614, + "learning_rate": 6.133163126528273e-08, + "loss": 0.9751, + "step": 12286 + }, + { + "epoch": 0.9234180069141741, + "grad_norm": 1.9007448844394284, + "learning_rate": 6.121204904119248e-08, + "loss": 0.9075, + "step": 12287 + }, + { + "epoch": 0.923493160980009, + "grad_norm": 2.257597916381653, + "learning_rate": 6.109258169771215e-08, + "loss": 0.9636, + "step": 12288 + }, + { + "epoch": 0.923568315045844, + "grad_norm": 2.0220962851415005, + "learning_rate": 6.097322924192094e-08, + "loss": 0.9385, + "step": 12289 + }, + { + "epoch": 0.9236434691116789, + "grad_norm": 2.3578863436977073, + "learning_rate": 6.085399168089079e-08, + "loss": 0.9583, + "step": 12290 + }, + { + "epoch": 0.9237186231775139, + "grad_norm": 2.066273099074393, + "learning_rate": 6.073486902168756e-08, + "loss": 0.9893, + "step": 12291 + }, + { + "epoch": 0.9237937772433489, + "grad_norm": 1.9073427538971899, + "learning_rate": 6.061586127136875e-08, + "loss": 0.9569, + "step": 12292 + }, + { + "epoch": 0.9238689313091838, + "grad_norm": 1.7329518433278333, + "learning_rate": 6.049696843698692e-08, + "loss": 1.0131, + "step": 12293 + }, + { + "epoch": 0.9239440853750188, + "grad_norm": 1.856215027759059, + "learning_rate": 6.037819052558668e-08, + "loss": 0.9293, + "step": 12294 + }, + { + "epoch": 0.9240192394408537, + "grad_norm": 1.6954233862830526, + "learning_rate": 6.025952754420571e-08, + "loss": 1.0196, + "step": 12295 + }, + { + "epoch": 0.9240943935066888, + "grad_norm": 1.3820675497425556, + "learning_rate": 6.014097949987529e-08, + "loss": 0.9927, + "step": 12296 + }, + { + "epoch": 0.9241695475725237, + "grad_norm": 1.3403838378677495, + "learning_rate": 6.002254639962023e-08, + "loss": 1.0296, + "step": 12297 + }, + { + "epoch": 0.9242447016383586, + "grad_norm": 1.758500096095226, + "learning_rate": 5.990422825045827e-08, + "loss": 1.1183, + "step": 12298 + }, + { + "epoch": 0.9243198557041936, + "grad_norm": 2.5559156539623284, + "learning_rate": 5.978602505939956e-08, + "loss": 1.0431, + "step": 12299 + }, + { + "epoch": 0.9243950097700285, + "grad_norm": 1.5650520844978808, + "learning_rate": 5.96679368334485e-08, + "loss": 0.9759, + "step": 12300 + }, + { + "epoch": 0.9244701638358636, + "grad_norm": 1.6731703544115262, + "learning_rate": 5.954996357960262e-08, + "loss": 0.9313, + "step": 12301 + }, + { + "epoch": 0.9245453179016985, + "grad_norm": 0.7096315902933442, + "learning_rate": 5.943210530485121e-08, + "loss": 0.8531, + "step": 12302 + }, + { + "epoch": 0.9246204719675335, + "grad_norm": 1.72799126381785, + "learning_rate": 5.9314362016178674e-08, + "loss": 1.0206, + "step": 12303 + }, + { + "epoch": 0.9246956260333684, + "grad_norm": 1.3322710733062377, + "learning_rate": 5.9196733720561884e-08, + "loss": 1.0116, + "step": 12304 + }, + { + "epoch": 0.9247707800992033, + "grad_norm": 3.8936962583210044, + "learning_rate": 5.9079220424970154e-08, + "loss": 0.9285, + "step": 12305 + }, + { + "epoch": 0.9248459341650384, + "grad_norm": 0.7395312913305875, + "learning_rate": 5.896182213636702e-08, + "loss": 0.845, + "step": 12306 + }, + { + "epoch": 0.9249210882308733, + "grad_norm": 2.7307589790093942, + "learning_rate": 5.884453886170848e-08, + "loss": 0.9364, + "step": 12307 + }, + { + "epoch": 0.9249962422967083, + "grad_norm": 1.8010050537318059, + "learning_rate": 5.872737060794475e-08, + "loss": 0.929, + "step": 12308 + }, + { + "epoch": 0.9250713963625432, + "grad_norm": 1.41215023335167, + "learning_rate": 5.8610317382017383e-08, + "loss": 0.8312, + "step": 12309 + }, + { + "epoch": 0.9251465504283781, + "grad_norm": 1.9035487813428003, + "learning_rate": 5.849337919086283e-08, + "loss": 1.0255, + "step": 12310 + }, + { + "epoch": 0.9252217044942131, + "grad_norm": 1.8396124886794092, + "learning_rate": 5.837655604141045e-08, + "loss": 0.9766, + "step": 12311 + }, + { + "epoch": 0.9252968585600481, + "grad_norm": 2.415692319339089, + "learning_rate": 5.825984794058181e-08, + "loss": 1.0929, + "step": 12312 + }, + { + "epoch": 0.9253720126258831, + "grad_norm": 2.3653960546773978, + "learning_rate": 5.814325489529248e-08, + "loss": 0.9572, + "step": 12313 + }, + { + "epoch": 0.925447166691718, + "grad_norm": 3.947142767563248, + "learning_rate": 5.8026776912451613e-08, + "loss": 0.9158, + "step": 12314 + }, + { + "epoch": 0.925522320757553, + "grad_norm": 1.4237868482260674, + "learning_rate": 5.791041399896013e-08, + "loss": 1.0126, + "step": 12315 + }, + { + "epoch": 0.925597474823388, + "grad_norm": 1.593415692739969, + "learning_rate": 5.77941661617134e-08, + "loss": 0.9566, + "step": 12316 + }, + { + "epoch": 0.9256726288892229, + "grad_norm": 2.1480408458413525, + "learning_rate": 5.767803340759947e-08, + "loss": 0.9684, + "step": 12317 + }, + { + "epoch": 0.9257477829550579, + "grad_norm": 1.5469137630037066, + "learning_rate": 5.756201574350017e-08, + "loss": 1.0708, + "step": 12318 + }, + { + "epoch": 0.9258229370208928, + "grad_norm": 1.3636695287023732, + "learning_rate": 5.744611317628889e-08, + "loss": 1.0047, + "step": 12319 + }, + { + "epoch": 0.9258980910867278, + "grad_norm": 1.4382586757664138, + "learning_rate": 5.7330325712834135e-08, + "loss": 0.9136, + "step": 12320 + }, + { + "epoch": 0.9259732451525627, + "grad_norm": 1.800209931211495, + "learning_rate": 5.721465335999709e-08, + "loss": 0.9285, + "step": 12321 + }, + { + "epoch": 0.9260483992183978, + "grad_norm": 1.5597635339368967, + "learning_rate": 5.7099096124630705e-08, + "loss": 0.8501, + "step": 12322 + }, + { + "epoch": 0.9261235532842327, + "grad_norm": 2.238678635610484, + "learning_rate": 5.6983654013582846e-08, + "loss": 1.0115, + "step": 12323 + }, + { + "epoch": 0.9261987073500676, + "grad_norm": 2.008444601224172, + "learning_rate": 5.68683270336936e-08, + "loss": 0.8178, + "step": 12324 + }, + { + "epoch": 0.9262738614159026, + "grad_norm": 1.7633730951067579, + "learning_rate": 5.675311519179726e-08, + "loss": 0.9063, + "step": 12325 + }, + { + "epoch": 0.9263490154817375, + "grad_norm": 1.3590903148493245, + "learning_rate": 5.66380184947195e-08, + "loss": 1.0105, + "step": 12326 + }, + { + "epoch": 0.9264241695475726, + "grad_norm": 1.7388827879693387, + "learning_rate": 5.652303694928107e-08, + "loss": 0.9691, + "step": 12327 + }, + { + "epoch": 0.9264993236134075, + "grad_norm": 1.5502812807138144, + "learning_rate": 5.640817056229474e-08, + "loss": 1.08, + "step": 12328 + }, + { + "epoch": 0.9265744776792424, + "grad_norm": 1.9392848868645536, + "learning_rate": 5.629341934056686e-08, + "loss": 0.8545, + "step": 12329 + }, + { + "epoch": 0.9266496317450774, + "grad_norm": 1.7613359191816833, + "learning_rate": 5.617878329089665e-08, + "loss": 0.9807, + "step": 12330 + }, + { + "epoch": 0.9267247858109123, + "grad_norm": 1.554821372857783, + "learning_rate": 5.606426242007734e-08, + "loss": 1.0493, + "step": 12331 + }, + { + "epoch": 0.9267999398767474, + "grad_norm": 1.75922564182546, + "learning_rate": 5.594985673489372e-08, + "loss": 0.8081, + "step": 12332 + }, + { + "epoch": 0.9268750939425823, + "grad_norm": 1.5185405571086945, + "learning_rate": 5.583556624212571e-08, + "loss": 1.0453, + "step": 12333 + }, + { + "epoch": 0.9269502480084173, + "grad_norm": 1.8207128518382236, + "learning_rate": 5.5721390948544775e-08, + "loss": 0.9472, + "step": 12334 + }, + { + "epoch": 0.9270254020742522, + "grad_norm": 2.7837162013372345, + "learning_rate": 5.560733086091707e-08, + "loss": 0.83, + "step": 12335 + }, + { + "epoch": 0.9271005561400871, + "grad_norm": 1.5174844091830557, + "learning_rate": 5.549338598600028e-08, + "loss": 0.8674, + "step": 12336 + }, + { + "epoch": 0.9271757102059222, + "grad_norm": 1.4860554067722611, + "learning_rate": 5.537955633054614e-08, + "loss": 0.8995, + "step": 12337 + }, + { + "epoch": 0.9272508642717571, + "grad_norm": 1.7465474266866536, + "learning_rate": 5.526584190130035e-08, + "loss": 0.9943, + "step": 12338 + }, + { + "epoch": 0.9273260183375921, + "grad_norm": 2.1429319550306776, + "learning_rate": 5.515224270499996e-08, + "loss": 0.9933, + "step": 12339 + }, + { + "epoch": 0.927401172403427, + "grad_norm": 4.005227725904301, + "learning_rate": 5.503875874837649e-08, + "loss": 0.9958, + "step": 12340 + }, + { + "epoch": 0.927476326469262, + "grad_norm": 1.7524731054704774, + "learning_rate": 5.4925390038154105e-08, + "loss": 1.027, + "step": 12341 + }, + { + "epoch": 0.927551480535097, + "grad_norm": 2.523407549134127, + "learning_rate": 5.481213658105121e-08, + "loss": 0.9969, + "step": 12342 + }, + { + "epoch": 0.9276266346009319, + "grad_norm": 1.7780125292453488, + "learning_rate": 5.469899838377734e-08, + "loss": 1.0012, + "step": 12343 + }, + { + "epoch": 0.9277017886667669, + "grad_norm": 1.780228305062514, + "learning_rate": 5.4585975453036894e-08, + "loss": 1.0043, + "step": 12344 + }, + { + "epoch": 0.9277769427326018, + "grad_norm": 2.3297220167421195, + "learning_rate": 5.4473067795526964e-08, + "loss": 0.9253, + "step": 12345 + }, + { + "epoch": 0.9278520967984368, + "grad_norm": 1.673328089053585, + "learning_rate": 5.436027541793775e-08, + "loss": 0.967, + "step": 12346 + }, + { + "epoch": 0.9279272508642717, + "grad_norm": 1.9508415606994824, + "learning_rate": 5.4247598326952357e-08, + "loss": 0.9321, + "step": 12347 + }, + { + "epoch": 0.9280024049301068, + "grad_norm": 1.9472825213189977, + "learning_rate": 5.4135036529248115e-08, + "loss": 1.0074, + "step": 12348 + }, + { + "epoch": 0.9280775589959417, + "grad_norm": 2.7496119138670974, + "learning_rate": 5.4022590031493454e-08, + "loss": 0.9319, + "step": 12349 + }, + { + "epoch": 0.9281527130617766, + "grad_norm": 1.5218638582233, + "learning_rate": 5.3910258840352386e-08, + "loss": 0.9229, + "step": 12350 + }, + { + "epoch": 0.9282278671276116, + "grad_norm": 2.6913327233357367, + "learning_rate": 5.379804296248025e-08, + "loss": 1.0324, + "step": 12351 + }, + { + "epoch": 0.9283030211934465, + "grad_norm": 2.3135586969973128, + "learning_rate": 5.3685942404527063e-08, + "loss": 0.8651, + "step": 12352 + }, + { + "epoch": 0.9283781752592816, + "grad_norm": 2.126660651773459, + "learning_rate": 5.35739571731344e-08, + "loss": 0.9243, + "step": 12353 + }, + { + "epoch": 0.9284533293251165, + "grad_norm": 1.757880451160576, + "learning_rate": 5.3462087274938064e-08, + "loss": 0.8667, + "step": 12354 + }, + { + "epoch": 0.9285284833909514, + "grad_norm": 1.9261064544998965, + "learning_rate": 5.335033271656697e-08, + "loss": 0.9878, + "step": 12355 + }, + { + "epoch": 0.9286036374567864, + "grad_norm": 1.8004732675387274, + "learning_rate": 5.323869350464294e-08, + "loss": 0.9334, + "step": 12356 + }, + { + "epoch": 0.9286787915226213, + "grad_norm": 1.8364966007889785, + "learning_rate": 5.31271696457809e-08, + "loss": 0.9423, + "step": 12357 + }, + { + "epoch": 0.9287539455884564, + "grad_norm": 3.5805352101826093, + "learning_rate": 5.3015761146589345e-08, + "loss": 0.89, + "step": 12358 + }, + { + "epoch": 0.9288290996542913, + "grad_norm": 3.292863620152352, + "learning_rate": 5.290446801366899e-08, + "loss": 1.0315, + "step": 12359 + }, + { + "epoch": 0.9289042537201263, + "grad_norm": 1.7139996860508313, + "learning_rate": 5.279329025361523e-08, + "loss": 0.9308, + "step": 12360 + }, + { + "epoch": 0.9289794077859612, + "grad_norm": 2.068723363791301, + "learning_rate": 5.268222787301502e-08, + "loss": 0.9567, + "step": 12361 + }, + { + "epoch": 0.9290545618517961, + "grad_norm": 1.476615380041655, + "learning_rate": 5.257128087844997e-08, + "loss": 0.9292, + "step": 12362 + }, + { + "epoch": 0.9291297159176312, + "grad_norm": 1.7644944749217604, + "learning_rate": 5.246044927649351e-08, + "loss": 0.902, + "step": 12363 + }, + { + "epoch": 0.9292048699834661, + "grad_norm": 1.4091910231591023, + "learning_rate": 5.2349733073712824e-08, + "loss": 0.9566, + "step": 12364 + }, + { + "epoch": 0.9292800240493011, + "grad_norm": 2.1967991041202484, + "learning_rate": 5.2239132276669096e-08, + "loss": 0.8879, + "step": 12365 + }, + { + "epoch": 0.929355178115136, + "grad_norm": 1.8628774401309531, + "learning_rate": 5.212864689191465e-08, + "loss": 0.8568, + "step": 12366 + }, + { + "epoch": 0.929430332180971, + "grad_norm": 1.6536278751229205, + "learning_rate": 5.201827692599714e-08, + "loss": 0.9996, + "step": 12367 + }, + { + "epoch": 0.929505486246806, + "grad_norm": 1.6300770602118797, + "learning_rate": 5.1908022385455774e-08, + "loss": 0.9092, + "step": 12368 + }, + { + "epoch": 0.9295806403126409, + "grad_norm": 1.563922768371347, + "learning_rate": 5.179788327682444e-08, + "loss": 0.8708, + "step": 12369 + }, + { + "epoch": 0.9296557943784759, + "grad_norm": 2.3176242655962835, + "learning_rate": 5.1687859606627915e-08, + "loss": 1.0216, + "step": 12370 + }, + { + "epoch": 0.9297309484443108, + "grad_norm": 1.7332769131036647, + "learning_rate": 5.1577951381386763e-08, + "loss": 0.8217, + "step": 12371 + }, + { + "epoch": 0.9298061025101458, + "grad_norm": 1.4877524249890883, + "learning_rate": 5.146815860761289e-08, + "loss": 0.9703, + "step": 12372 + }, + { + "epoch": 0.9298812565759808, + "grad_norm": 2.2221139470849254, + "learning_rate": 5.1358481291812194e-08, + "loss": 0.9689, + "step": 12373 + }, + { + "epoch": 0.9299564106418157, + "grad_norm": 1.6607997948449804, + "learning_rate": 5.124891944048282e-08, + "loss": 0.9792, + "step": 12374 + }, + { + "epoch": 0.9300315647076507, + "grad_norm": 1.7248935832494583, + "learning_rate": 5.113947306011801e-08, + "loss": 0.9735, + "step": 12375 + }, + { + "epoch": 0.9301067187734856, + "grad_norm": 2.2950768730902142, + "learning_rate": 5.103014215720147e-08, + "loss": 0.9461, + "step": 12376 + }, + { + "epoch": 0.9301818728393206, + "grad_norm": 1.543391534368381, + "learning_rate": 5.092092673821224e-08, + "loss": 0.9949, + "step": 12377 + }, + { + "epoch": 0.9302570269051555, + "grad_norm": 2.5101822444212134, + "learning_rate": 5.0811826809621596e-08, + "loss": 1.1149, + "step": 12378 + }, + { + "epoch": 0.9303321809709906, + "grad_norm": 1.6777512146501006, + "learning_rate": 5.0702842377894574e-08, + "loss": 0.9481, + "step": 12379 + }, + { + "epoch": 0.9304073350368255, + "grad_norm": 0.8043142026912421, + "learning_rate": 5.059397344948802e-08, + "loss": 0.8933, + "step": 12380 + }, + { + "epoch": 0.9304824891026604, + "grad_norm": 1.6402274967202917, + "learning_rate": 5.0485220030853204e-08, + "loss": 0.96, + "step": 12381 + }, + { + "epoch": 0.9305576431684954, + "grad_norm": 1.9097214654477865, + "learning_rate": 5.0376582128434766e-08, + "loss": 1.0659, + "step": 12382 + }, + { + "epoch": 0.9306327972343303, + "grad_norm": 1.9406325949798227, + "learning_rate": 5.026805974866932e-08, + "loss": 0.9283, + "step": 12383 + }, + { + "epoch": 0.9307079513001654, + "grad_norm": 1.4244485557749413, + "learning_rate": 5.015965289798707e-08, + "loss": 0.9388, + "step": 12384 + }, + { + "epoch": 0.9307831053660003, + "grad_norm": 1.6617391064055658, + "learning_rate": 5.005136158281198e-08, + "loss": 0.9634, + "step": 12385 + }, + { + "epoch": 0.9308582594318353, + "grad_norm": 2.0478669303786745, + "learning_rate": 4.994318580956092e-08, + "loss": 1.0009, + "step": 12386 + }, + { + "epoch": 0.9309334134976702, + "grad_norm": 1.619909235213001, + "learning_rate": 4.983512558464276e-08, + "loss": 0.9716, + "step": 12387 + }, + { + "epoch": 0.9310085675635051, + "grad_norm": 2.338620179578085, + "learning_rate": 4.9727180914461485e-08, + "loss": 1.0471, + "step": 12388 + }, + { + "epoch": 0.9310837216293402, + "grad_norm": 2.149346176801665, + "learning_rate": 4.961935180541288e-08, + "loss": 0.875, + "step": 12389 + }, + { + "epoch": 0.9311588756951751, + "grad_norm": 1.540016459587618, + "learning_rate": 4.9511638263886045e-08, + "loss": 0.9087, + "step": 12390 + }, + { + "epoch": 0.9312340297610101, + "grad_norm": 1.6444097551060648, + "learning_rate": 4.940404029626344e-08, + "loss": 0.9259, + "step": 12391 + }, + { + "epoch": 0.931309183826845, + "grad_norm": 1.6012412144392385, + "learning_rate": 4.929655790892107e-08, + "loss": 0.9498, + "step": 12392 + }, + { + "epoch": 0.93138433789268, + "grad_norm": 2.47924832335132, + "learning_rate": 4.918919110822717e-08, + "loss": 1.0661, + "step": 12393 + }, + { + "epoch": 0.931459491958515, + "grad_norm": 2.3826387936171964, + "learning_rate": 4.908193990054377e-08, + "loss": 0.9936, + "step": 12394 + }, + { + "epoch": 0.9315346460243499, + "grad_norm": 1.801099769739089, + "learning_rate": 4.8974804292226e-08, + "loss": 0.9978, + "step": 12395 + }, + { + "epoch": 0.9316098000901849, + "grad_norm": 1.727456530232461, + "learning_rate": 4.886778428962235e-08, + "loss": 0.9663, + "step": 12396 + }, + { + "epoch": 0.9316849541560198, + "grad_norm": 1.9994228245464085, + "learning_rate": 4.876087989907374e-08, + "loss": 1.0826, + "step": 12397 + }, + { + "epoch": 0.9317601082218548, + "grad_norm": 1.5842083303453327, + "learning_rate": 4.865409112691465e-08, + "loss": 0.9045, + "step": 12398 + }, + { + "epoch": 0.9318352622876898, + "grad_norm": 1.6738582505994872, + "learning_rate": 4.854741797947293e-08, + "loss": 0.9878, + "step": 12399 + }, + { + "epoch": 0.9319104163535247, + "grad_norm": 2.0878942791127986, + "learning_rate": 4.8440860463069496e-08, + "loss": 1.0215, + "step": 12400 + }, + { + "epoch": 0.9319855704193597, + "grad_norm": 1.6474313422384335, + "learning_rate": 4.8334418584017764e-08, + "loss": 0.8733, + "step": 12401 + }, + { + "epoch": 0.9320607244851946, + "grad_norm": 1.1912748670834545, + "learning_rate": 4.822809234862557e-08, + "loss": 0.8278, + "step": 12402 + }, + { + "epoch": 0.9321358785510296, + "grad_norm": 1.9901569265283736, + "learning_rate": 4.812188176319232e-08, + "loss": 0.9743, + "step": 12403 + }, + { + "epoch": 0.9322110326168646, + "grad_norm": 4.721667265851034, + "learning_rate": 4.801578683401186e-08, + "loss": 0.9667, + "step": 12404 + }, + { + "epoch": 0.9322861866826996, + "grad_norm": 2.047051342072914, + "learning_rate": 4.790980756737073e-08, + "loss": 1.0037, + "step": 12405 + }, + { + "epoch": 0.9323613407485345, + "grad_norm": 1.6668010975346386, + "learning_rate": 4.780394396954901e-08, + "loss": 0.9535, + "step": 12406 + }, + { + "epoch": 0.9324364948143694, + "grad_norm": 1.695077343641798, + "learning_rate": 4.769819604681857e-08, + "loss": 1.0076, + "step": 12407 + }, + { + "epoch": 0.9325116488802044, + "grad_norm": 1.7385313389926411, + "learning_rate": 4.759256380544574e-08, + "loss": 0.929, + "step": 12408 + }, + { + "epoch": 0.9325868029460394, + "grad_norm": 1.320409015084837, + "learning_rate": 4.748704725169017e-08, + "loss": 0.9142, + "step": 12409 + }, + { + "epoch": 0.9326619570118744, + "grad_norm": 1.5968048140000304, + "learning_rate": 4.7381646391803534e-08, + "loss": 0.887, + "step": 12410 + }, + { + "epoch": 0.9327371110777093, + "grad_norm": 1.9392970580341544, + "learning_rate": 4.727636123203149e-08, + "loss": 0.8485, + "step": 12411 + }, + { + "epoch": 0.9328122651435443, + "grad_norm": 1.9675111051485978, + "learning_rate": 4.7171191778612396e-08, + "loss": 0.9531, + "step": 12412 + }, + { + "epoch": 0.9328874192093792, + "grad_norm": 2.70321284686104, + "learning_rate": 4.706613803777837e-08, + "loss": 0.8905, + "step": 12413 + }, + { + "epoch": 0.9329625732752141, + "grad_norm": 2.018648113124316, + "learning_rate": 4.6961200015753546e-08, + "loss": 1.0445, + "step": 12414 + }, + { + "epoch": 0.9330377273410492, + "grad_norm": 2.500563558001472, + "learning_rate": 4.68563777187565e-08, + "loss": 0.9105, + "step": 12415 + }, + { + "epoch": 0.9331128814068841, + "grad_norm": 1.847286647097025, + "learning_rate": 4.6751671152998276e-08, + "loss": 1.0482, + "step": 12416 + }, + { + "epoch": 0.9331880354727191, + "grad_norm": 1.8400642926966775, + "learning_rate": 4.664708032468301e-08, + "loss": 1.0316, + "step": 12417 + }, + { + "epoch": 0.933263189538554, + "grad_norm": 1.8175873228816404, + "learning_rate": 4.654260524000797e-08, + "loss": 0.9671, + "step": 12418 + }, + { + "epoch": 0.9333383436043889, + "grad_norm": 1.5864380295627896, + "learning_rate": 4.643824590516399e-08, + "loss": 0.9502, + "step": 12419 + }, + { + "epoch": 0.933413497670224, + "grad_norm": 1.3237036319208777, + "learning_rate": 4.6334002326334554e-08, + "loss": 0.9607, + "step": 12420 + }, + { + "epoch": 0.9334886517360589, + "grad_norm": 1.6075262914141997, + "learning_rate": 4.622987450969651e-08, + "loss": 0.9487, + "step": 12421 + }, + { + "epoch": 0.9335638058018939, + "grad_norm": 3.8055428986598008, + "learning_rate": 4.612586246141981e-08, + "loss": 0.9144, + "step": 12422 + }, + { + "epoch": 0.9336389598677288, + "grad_norm": 1.818645450693892, + "learning_rate": 4.602196618766796e-08, + "loss": 1.0138, + "step": 12423 + }, + { + "epoch": 0.9337141139335638, + "grad_norm": 1.4482587323365892, + "learning_rate": 4.591818569459671e-08, + "loss": 0.9719, + "step": 12424 + }, + { + "epoch": 0.9337892679993988, + "grad_norm": 2.816745097481018, + "learning_rate": 4.581452098835537e-08, + "loss": 0.9548, + "step": 12425 + }, + { + "epoch": 0.9338644220652337, + "grad_norm": 1.7218680102973207, + "learning_rate": 4.571097207508723e-08, + "loss": 0.9441, + "step": 12426 + }, + { + "epoch": 0.9339395761310687, + "grad_norm": 2.5345799712468637, + "learning_rate": 4.560753896092739e-08, + "loss": 1.0308, + "step": 12427 + }, + { + "epoch": 0.9340147301969036, + "grad_norm": 1.783633260669436, + "learning_rate": 4.5504221652004295e-08, + "loss": 0.9684, + "step": 12428 + }, + { + "epoch": 0.9340898842627386, + "grad_norm": 8.152727908563747, + "learning_rate": 4.5401020154440586e-08, + "loss": 0.9602, + "step": 12429 + }, + { + "epoch": 0.9341650383285736, + "grad_norm": 5.440209075612532, + "learning_rate": 4.529793447435137e-08, + "loss": 0.9754, + "step": 12430 + }, + { + "epoch": 0.9342401923944086, + "grad_norm": 4.974572566854402, + "learning_rate": 4.5194964617844225e-08, + "loss": 0.9227, + "step": 12431 + }, + { + "epoch": 0.9343153464602435, + "grad_norm": 1.4267574782896597, + "learning_rate": 4.509211059102092e-08, + "loss": 0.8509, + "step": 12432 + }, + { + "epoch": 0.9343905005260784, + "grad_norm": 1.876077977884065, + "learning_rate": 4.498937239997613e-08, + "loss": 1.0718, + "step": 12433 + }, + { + "epoch": 0.9344656545919134, + "grad_norm": 1.6865187017024201, + "learning_rate": 4.488675005079723e-08, + "loss": 1.0006, + "step": 12434 + }, + { + "epoch": 0.9345408086577484, + "grad_norm": 2.674468538485429, + "learning_rate": 4.478424354956467e-08, + "loss": 0.774, + "step": 12435 + }, + { + "epoch": 0.9346159627235834, + "grad_norm": 2.0601769282460887, + "learning_rate": 4.4681852902353154e-08, + "loss": 0.9452, + "step": 12436 + }, + { + "epoch": 0.9346911167894183, + "grad_norm": 1.8888344601038443, + "learning_rate": 4.4579578115228943e-08, + "loss": 0.9387, + "step": 12437 + }, + { + "epoch": 0.9347662708552533, + "grad_norm": 3.934910604001928, + "learning_rate": 4.447741919425274e-08, + "loss": 1.0417, + "step": 12438 + }, + { + "epoch": 0.9348414249210882, + "grad_norm": 1.7425513464916642, + "learning_rate": 4.437537614547726e-08, + "loss": 0.9706, + "step": 12439 + }, + { + "epoch": 0.9349165789869232, + "grad_norm": 2.257301880009924, + "learning_rate": 4.427344897494989e-08, + "loss": 1.0615, + "step": 12440 + }, + { + "epoch": 0.9349917330527582, + "grad_norm": 2.183040105139384, + "learning_rate": 4.4171637688709354e-08, + "loss": 0.9932, + "step": 12441 + }, + { + "epoch": 0.9350668871185931, + "grad_norm": 1.4971317047251838, + "learning_rate": 4.4069942292788596e-08, + "loss": 0.9212, + "step": 12442 + }, + { + "epoch": 0.9351420411844281, + "grad_norm": 1.5926932028764265, + "learning_rate": 4.39683627932137e-08, + "loss": 0.9605, + "step": 12443 + }, + { + "epoch": 0.935217195250263, + "grad_norm": 1.720429588165256, + "learning_rate": 4.3866899196003393e-08, + "loss": 1.0154, + "step": 12444 + }, + { + "epoch": 0.935292349316098, + "grad_norm": 1.6323879000160393, + "learning_rate": 4.376555150716954e-08, + "loss": 0.9242, + "step": 12445 + }, + { + "epoch": 0.935367503381933, + "grad_norm": 1.8009804307254644, + "learning_rate": 4.3664319732718227e-08, + "loss": 0.9208, + "step": 12446 + }, + { + "epoch": 0.9354426574477679, + "grad_norm": 0.7069852192263059, + "learning_rate": 4.356320387864687e-08, + "loss": 0.8425, + "step": 12447 + }, + { + "epoch": 0.9355178115136029, + "grad_norm": 1.8246236674411123, + "learning_rate": 4.346220395094735e-08, + "loss": 0.9273, + "step": 12448 + }, + { + "epoch": 0.9355929655794378, + "grad_norm": 2.237762870477957, + "learning_rate": 4.336131995560444e-08, + "loss": 0.9769, + "step": 12449 + }, + { + "epoch": 0.9356681196452729, + "grad_norm": 2.490795643703069, + "learning_rate": 4.326055189859601e-08, + "loss": 0.9773, + "step": 12450 + }, + { + "epoch": 0.9357432737111078, + "grad_norm": 2.3092833132990727, + "learning_rate": 4.3159899785892403e-08, + "loss": 1.0049, + "step": 12451 + }, + { + "epoch": 0.9358184277769427, + "grad_norm": 1.6721866352875365, + "learning_rate": 4.305936362345797e-08, + "loss": 1.082, + "step": 12452 + }, + { + "epoch": 0.9358935818427777, + "grad_norm": 1.9522829605780645, + "learning_rate": 4.295894341725037e-08, + "loss": 0.9385, + "step": 12453 + }, + { + "epoch": 0.9359687359086126, + "grad_norm": 1.8984248156967585, + "learning_rate": 4.285863917321886e-08, + "loss": 0.8911, + "step": 12454 + }, + { + "epoch": 0.9360438899744477, + "grad_norm": 2.0891803579816344, + "learning_rate": 4.2758450897307565e-08, + "loss": 0.9356, + "step": 12455 + }, + { + "epoch": 0.9361190440402826, + "grad_norm": 1.589614439227063, + "learning_rate": 4.2658378595452626e-08, + "loss": 0.9935, + "step": 12456 + }, + { + "epoch": 0.9361941981061176, + "grad_norm": 2.0653360692805047, + "learning_rate": 4.2558422273584414e-08, + "loss": 0.9559, + "step": 12457 + }, + { + "epoch": 0.9362693521719525, + "grad_norm": 1.9450433996894299, + "learning_rate": 4.245858193762486e-08, + "loss": 0.9479, + "step": 12458 + }, + { + "epoch": 0.9363445062377874, + "grad_norm": 2.0232952760178873, + "learning_rate": 4.2358857593490337e-08, + "loss": 0.9895, + "step": 12459 + }, + { + "epoch": 0.9364196603036224, + "grad_norm": 2.7600559694724236, + "learning_rate": 4.225924924708968e-08, + "loss": 0.9733, + "step": 12460 + }, + { + "epoch": 0.9364948143694574, + "grad_norm": 1.757951993705591, + "learning_rate": 4.215975690432549e-08, + "loss": 1.059, + "step": 12461 + }, + { + "epoch": 0.9365699684352924, + "grad_norm": 3.1796879498905546, + "learning_rate": 4.20603805710924e-08, + "loss": 0.9848, + "step": 12462 + }, + { + "epoch": 0.9366451225011273, + "grad_norm": 8.438272172421282, + "learning_rate": 4.196112025327969e-08, + "loss": 1.0165, + "step": 12463 + }, + { + "epoch": 0.9367202765669622, + "grad_norm": 2.0113974305357143, + "learning_rate": 4.1861975956767994e-08, + "loss": 1.0059, + "step": 12464 + }, + { + "epoch": 0.9367954306327972, + "grad_norm": 1.5599380852565863, + "learning_rate": 4.1762947687432605e-08, + "loss": 0.9301, + "step": 12465 + }, + { + "epoch": 0.9368705846986322, + "grad_norm": 1.6287646334994204, + "learning_rate": 4.166403545114105e-08, + "loss": 0.9594, + "step": 12466 + }, + { + "epoch": 0.9369457387644672, + "grad_norm": 2.0071247273582915, + "learning_rate": 4.1565239253754655e-08, + "loss": 0.8962, + "step": 12467 + }, + { + "epoch": 0.9370208928303021, + "grad_norm": 2.869993910035374, + "learning_rate": 4.146655910112673e-08, + "loss": 0.9608, + "step": 12468 + }, + { + "epoch": 0.9370960468961371, + "grad_norm": 1.9002287149801809, + "learning_rate": 4.1367994999105036e-08, + "loss": 0.8084, + "step": 12469 + }, + { + "epoch": 0.937171200961972, + "grad_norm": 1.5715706562855372, + "learning_rate": 4.126954695353002e-08, + "loss": 0.9599, + "step": 12470 + }, + { + "epoch": 0.937246355027807, + "grad_norm": 0.7080457374675578, + "learning_rate": 4.117121497023457e-08, + "loss": 0.8076, + "step": 12471 + }, + { + "epoch": 0.937321509093642, + "grad_norm": 1.8366287553385752, + "learning_rate": 4.107299905504558e-08, + "loss": 0.8874, + "step": 12472 + }, + { + "epoch": 0.9373966631594769, + "grad_norm": 2.08510421510775, + "learning_rate": 4.097489921378261e-08, + "loss": 0.8804, + "step": 12473 + }, + { + "epoch": 0.9374718172253119, + "grad_norm": 1.9947870385603508, + "learning_rate": 4.0876915452258577e-08, + "loss": 0.8865, + "step": 12474 + }, + { + "epoch": 0.9375469712911468, + "grad_norm": 1.7973117463450254, + "learning_rate": 4.077904777627905e-08, + "loss": 0.8972, + "step": 12475 + }, + { + "epoch": 0.9376221253569819, + "grad_norm": 1.69638800895211, + "learning_rate": 4.068129619164362e-08, + "loss": 0.9515, + "step": 12476 + }, + { + "epoch": 0.9376972794228168, + "grad_norm": 1.5406157889874244, + "learning_rate": 4.0583660704143874e-08, + "loss": 0.9736, + "step": 12477 + }, + { + "epoch": 0.9377724334886517, + "grad_norm": 2.4017044333198196, + "learning_rate": 4.0486141319565624e-08, + "loss": 1.0968, + "step": 12478 + }, + { + "epoch": 0.9378475875544867, + "grad_norm": 3.370743787508964, + "learning_rate": 4.038873804368648e-08, + "loss": 0.9098, + "step": 12479 + }, + { + "epoch": 0.9379227416203216, + "grad_norm": 1.680275534673187, + "learning_rate": 4.0291450882279144e-08, + "loss": 0.94, + "step": 12480 + }, + { + "epoch": 0.9379978956861567, + "grad_norm": 3.272251542511777, + "learning_rate": 4.0194279841107014e-08, + "loss": 1.0057, + "step": 12481 + }, + { + "epoch": 0.9380730497519916, + "grad_norm": 1.7839513005080123, + "learning_rate": 4.00972249259286e-08, + "loss": 0.9611, + "step": 12482 + }, + { + "epoch": 0.9381482038178266, + "grad_norm": 1.8291890234945911, + "learning_rate": 4.00002861424944e-08, + "loss": 0.9734, + "step": 12483 + }, + { + "epoch": 0.9382233578836615, + "grad_norm": 2.4211192274645694, + "learning_rate": 3.990346349654894e-08, + "loss": 0.9133, + "step": 12484 + }, + { + "epoch": 0.9382985119494964, + "grad_norm": 1.8453017348347744, + "learning_rate": 3.980675699382852e-08, + "loss": 1.0053, + "step": 12485 + }, + { + "epoch": 0.9383736660153315, + "grad_norm": 3.5010172864176097, + "learning_rate": 3.97101666400641e-08, + "loss": 0.9142, + "step": 12486 + }, + { + "epoch": 0.9384488200811664, + "grad_norm": 1.6464964661431407, + "learning_rate": 3.96136924409789e-08, + "loss": 0.8578, + "step": 12487 + }, + { + "epoch": 0.9385239741470014, + "grad_norm": 2.0138209349761444, + "learning_rate": 3.951733440228899e-08, + "loss": 0.9507, + "step": 12488 + }, + { + "epoch": 0.9385991282128363, + "grad_norm": 1.7283247157991428, + "learning_rate": 3.942109252970427e-08, + "loss": 0.969, + "step": 12489 + }, + { + "epoch": 0.9386742822786712, + "grad_norm": 3.531344387177094, + "learning_rate": 3.93249668289275e-08, + "loss": 0.947, + "step": 12490 + }, + { + "epoch": 0.9387494363445062, + "grad_norm": 1.5325612273439067, + "learning_rate": 3.922895730565412e-08, + "loss": 0.9792, + "step": 12491 + }, + { + "epoch": 0.9388245904103412, + "grad_norm": 1.8332402321266419, + "learning_rate": 3.913306396557336e-08, + "loss": 1.0284, + "step": 12492 + }, + { + "epoch": 0.9388997444761762, + "grad_norm": 1.89760266695977, + "learning_rate": 3.903728681436735e-08, + "loss": 0.98, + "step": 12493 + }, + { + "epoch": 0.9389748985420111, + "grad_norm": 1.6659672541153587, + "learning_rate": 3.894162585771132e-08, + "loss": 0.9083, + "step": 12494 + }, + { + "epoch": 0.9390500526078461, + "grad_norm": 1.8289804657841402, + "learning_rate": 3.884608110127319e-08, + "loss": 0.9733, + "step": 12495 + }, + { + "epoch": 0.939125206673681, + "grad_norm": 1.7288856584414714, + "learning_rate": 3.875065255071419e-08, + "loss": 0.8889, + "step": 12496 + }, + { + "epoch": 0.939200360739516, + "grad_norm": 1.7075938818662248, + "learning_rate": 3.8655340211689594e-08, + "loss": 0.9494, + "step": 12497 + }, + { + "epoch": 0.939275514805351, + "grad_norm": 1.7683407313377573, + "learning_rate": 3.856014408984643e-08, + "loss": 0.8241, + "step": 12498 + }, + { + "epoch": 0.9393506688711859, + "grad_norm": 1.7391944115470863, + "learning_rate": 3.846506419082551e-08, + "loss": 0.8573, + "step": 12499 + }, + { + "epoch": 0.9394258229370209, + "grad_norm": 1.7379174698350919, + "learning_rate": 3.837010052026057e-08, + "loss": 1.0578, + "step": 12500 + }, + { + "epoch": 0.9395009770028558, + "grad_norm": 1.872127557952839, + "learning_rate": 3.827525308377932e-08, + "loss": 0.9419, + "step": 12501 + }, + { + "epoch": 0.9395761310686909, + "grad_norm": 1.657848162901157, + "learning_rate": 3.8180521887000825e-08, + "loss": 0.8725, + "step": 12502 + }, + { + "epoch": 0.9396512851345258, + "grad_norm": 1.8900197218819657, + "learning_rate": 3.8085906935538815e-08, + "loss": 0.857, + "step": 12503 + }, + { + "epoch": 0.9397264392003607, + "grad_norm": 2.833674092464532, + "learning_rate": 3.799140823499947e-08, + "loss": 0.986, + "step": 12504 + }, + { + "epoch": 0.9398015932661957, + "grad_norm": 1.9852853924428653, + "learning_rate": 3.7897025790982305e-08, + "loss": 0.9357, + "step": 12505 + }, + { + "epoch": 0.9398767473320306, + "grad_norm": 4.969984560967457, + "learning_rate": 3.780275960907975e-08, + "loss": 1.0056, + "step": 12506 + }, + { + "epoch": 0.9399519013978657, + "grad_norm": 1.754116733894452, + "learning_rate": 3.770860969487755e-08, + "loss": 1.1059, + "step": 12507 + }, + { + "epoch": 0.9400270554637006, + "grad_norm": 1.6142368396058926, + "learning_rate": 3.7614576053954126e-08, + "loss": 1.0156, + "step": 12508 + }, + { + "epoch": 0.9401022095295355, + "grad_norm": 1.4602355086448888, + "learning_rate": 3.752065869188148e-08, + "loss": 0.9578, + "step": 12509 + }, + { + "epoch": 0.9401773635953705, + "grad_norm": 4.450463500154893, + "learning_rate": 3.74268576142247e-08, + "loss": 0.9186, + "step": 12510 + }, + { + "epoch": 0.9402525176612054, + "grad_norm": 3.440890772617177, + "learning_rate": 3.7333172826542025e-08, + "loss": 1.0591, + "step": 12511 + }, + { + "epoch": 0.9403276717270405, + "grad_norm": 1.5854310227628456, + "learning_rate": 3.7239604334384336e-08, + "loss": 0.8823, + "step": 12512 + }, + { + "epoch": 0.9404028257928754, + "grad_norm": 1.8900434371962263, + "learning_rate": 3.714615214329564e-08, + "loss": 0.96, + "step": 12513 + }, + { + "epoch": 0.9404779798587104, + "grad_norm": 1.3489842160759955, + "learning_rate": 3.705281625881418e-08, + "loss": 0.9507, + "step": 12514 + }, + { + "epoch": 0.9405531339245453, + "grad_norm": 0.7717488668450135, + "learning_rate": 3.695959668646975e-08, + "loss": 0.8229, + "step": 12515 + }, + { + "epoch": 0.9406282879903802, + "grad_norm": 0.6957000980814692, + "learning_rate": 3.686649343178616e-08, + "loss": 0.8467, + "step": 12516 + }, + { + "epoch": 0.9407034420562153, + "grad_norm": 2.318475169220177, + "learning_rate": 3.677350650028033e-08, + "loss": 0.95, + "step": 12517 + }, + { + "epoch": 0.9407785961220502, + "grad_norm": 2.0799296867883883, + "learning_rate": 3.668063589746206e-08, + "loss": 1.0207, + "step": 12518 + }, + { + "epoch": 0.9408537501878852, + "grad_norm": 1.9065762850318242, + "learning_rate": 3.658788162883364e-08, + "loss": 0.9479, + "step": 12519 + }, + { + "epoch": 0.9409289042537201, + "grad_norm": 2.0958193900502136, + "learning_rate": 3.649524369989221e-08, + "loss": 0.9167, + "step": 12520 + }, + { + "epoch": 0.9410040583195551, + "grad_norm": 2.210897276819129, + "learning_rate": 3.640272211612605e-08, + "loss": 0.9507, + "step": 12521 + }, + { + "epoch": 0.94107921238539, + "grad_norm": 40.50219720602359, + "learning_rate": 3.631031688301789e-08, + "loss": 0.8852, + "step": 12522 + }, + { + "epoch": 0.941154366451225, + "grad_norm": 1.8813885254409544, + "learning_rate": 3.6218028006042676e-08, + "loss": 0.9048, + "step": 12523 + }, + { + "epoch": 0.94122952051706, + "grad_norm": 1.8112966554897163, + "learning_rate": 3.612585549066938e-08, + "loss": 0.9899, + "step": 12524 + }, + { + "epoch": 0.9413046745828949, + "grad_norm": 1.9941052948365443, + "learning_rate": 3.603379934235917e-08, + "loss": 0.8033, + "step": 12525 + }, + { + "epoch": 0.9413798286487299, + "grad_norm": 2.468110786449225, + "learning_rate": 3.5941859566566816e-08, + "loss": 0.9821, + "step": 12526 + }, + { + "epoch": 0.9414549827145648, + "grad_norm": 2.0725136625532112, + "learning_rate": 3.585003616874016e-08, + "loss": 1.0476, + "step": 12527 + }, + { + "epoch": 0.9415301367803999, + "grad_norm": 0.6913214378279231, + "learning_rate": 3.575832915432042e-08, + "loss": 0.7894, + "step": 12528 + }, + { + "epoch": 0.9416052908462348, + "grad_norm": 1.601119859515493, + "learning_rate": 3.566673852874103e-08, + "loss": 0.9443, + "step": 12529 + }, + { + "epoch": 0.9416804449120697, + "grad_norm": 0.7119614071271665, + "learning_rate": 3.5575264297429185e-08, + "loss": 0.9044, + "step": 12530 + }, + { + "epoch": 0.9417555989779047, + "grad_norm": 2.611970718283918, + "learning_rate": 3.548390646580546e-08, + "loss": 0.9412, + "step": 12531 + }, + { + "epoch": 0.9418307530437396, + "grad_norm": 1.5945995534858188, + "learning_rate": 3.539266503928284e-08, + "loss": 1.0157, + "step": 12532 + }, + { + "epoch": 0.9419059071095747, + "grad_norm": 1.8104050466547206, + "learning_rate": 3.5301540023267684e-08, + "loss": 0.9106, + "step": 12533 + }, + { + "epoch": 0.9419810611754096, + "grad_norm": 1.674046398695676, + "learning_rate": 3.521053142315988e-08, + "loss": 0.9206, + "step": 12534 + }, + { + "epoch": 0.9420562152412445, + "grad_norm": 2.00938846470903, + "learning_rate": 3.5119639244351575e-08, + "loss": 0.9001, + "step": 12535 + }, + { + "epoch": 0.9421313693070795, + "grad_norm": 4.475045418109254, + "learning_rate": 3.502886349222844e-08, + "loss": 1.0121, + "step": 12536 + }, + { + "epoch": 0.9422065233729144, + "grad_norm": 1.8413746290117843, + "learning_rate": 3.493820417216975e-08, + "loss": 1.0146, + "step": 12537 + }, + { + "epoch": 0.9422816774387495, + "grad_norm": 1.6723031894997642, + "learning_rate": 3.4847661289547417e-08, + "loss": 0.9764, + "step": 12538 + }, + { + "epoch": 0.9423568315045844, + "grad_norm": 1.5111521366259908, + "learning_rate": 3.4757234849726036e-08, + "loss": 0.9527, + "step": 12539 + }, + { + "epoch": 0.9424319855704194, + "grad_norm": 1.7823122437666177, + "learning_rate": 3.4666924858063776e-08, + "loss": 0.9058, + "step": 12540 + }, + { + "epoch": 0.9425071396362543, + "grad_norm": 1.5617479421798366, + "learning_rate": 3.4576731319912125e-08, + "loss": 0.9127, + "step": 12541 + }, + { + "epoch": 0.9425822937020892, + "grad_norm": 2.1653033393339065, + "learning_rate": 3.448665424061525e-08, + "loss": 0.9357, + "step": 12542 + }, + { + "epoch": 0.9426574477679243, + "grad_norm": 2.046908662242583, + "learning_rate": 3.439669362551045e-08, + "loss": 0.9531, + "step": 12543 + }, + { + "epoch": 0.9427326018337592, + "grad_norm": 1.7581291083143114, + "learning_rate": 3.4306849479928344e-08, + "loss": 1.0221, + "step": 12544 + }, + { + "epoch": 0.9428077558995942, + "grad_norm": 0.7193959691946218, + "learning_rate": 3.421712180919289e-08, + "loss": 0.8909, + "step": 12545 + }, + { + "epoch": 0.9428829099654291, + "grad_norm": 1.7747659958804898, + "learning_rate": 3.412751061862007e-08, + "loss": 0.9516, + "step": 12546 + }, + { + "epoch": 0.9429580640312641, + "grad_norm": 2.156685494935727, + "learning_rate": 3.403801591352029e-08, + "loss": 0.9437, + "step": 12547 + }, + { + "epoch": 0.9430332180970991, + "grad_norm": 3.028181115290893, + "learning_rate": 3.394863769919598e-08, + "loss": 0.9446, + "step": 12548 + }, + { + "epoch": 0.943108372162934, + "grad_norm": 4.528964183016346, + "learning_rate": 3.3859375980943797e-08, + "loss": 1.0895, + "step": 12549 + }, + { + "epoch": 0.943183526228769, + "grad_norm": 1.4794247451367126, + "learning_rate": 3.3770230764051946e-08, + "loss": 0.9453, + "step": 12550 + }, + { + "epoch": 0.9432586802946039, + "grad_norm": 2.0524897552123043, + "learning_rate": 3.368120205380376e-08, + "loss": 1.0277, + "step": 12551 + }, + { + "epoch": 0.9433338343604389, + "grad_norm": 1.5195769864531967, + "learning_rate": 3.3592289855473244e-08, + "loss": 0.8482, + "step": 12552 + }, + { + "epoch": 0.9434089884262739, + "grad_norm": 1.4619730986535024, + "learning_rate": 3.3503494174329516e-08, + "loss": 1.0621, + "step": 12553 + }, + { + "epoch": 0.9434841424921088, + "grad_norm": 3.6849492189384083, + "learning_rate": 3.341481501563437e-08, + "loss": 0.9331, + "step": 12554 + }, + { + "epoch": 0.9435592965579438, + "grad_norm": 1.8726190072351998, + "learning_rate": 3.332625238464204e-08, + "loss": 0.9315, + "step": 12555 + }, + { + "epoch": 0.9436344506237787, + "grad_norm": 1.7531228131775298, + "learning_rate": 3.3237806286599667e-08, + "loss": 0.9873, + "step": 12556 + }, + { + "epoch": 0.9437096046896137, + "grad_norm": 1.9088862591205897, + "learning_rate": 3.314947672674862e-08, + "loss": 0.9947, + "step": 12557 + }, + { + "epoch": 0.9437847587554486, + "grad_norm": 2.237779917336613, + "learning_rate": 3.3061263710322917e-08, + "loss": 1.0697, + "step": 12558 + }, + { + "epoch": 0.9438599128212837, + "grad_norm": 0.7393086061954792, + "learning_rate": 3.297316724254906e-08, + "loss": 0.8194, + "step": 12559 + }, + { + "epoch": 0.9439350668871186, + "grad_norm": 1.7280954313594068, + "learning_rate": 3.288518732864731e-08, + "loss": 1.0125, + "step": 12560 + }, + { + "epoch": 0.9440102209529535, + "grad_norm": 2.2527891461636513, + "learning_rate": 3.2797323973830834e-08, + "loss": 0.9997, + "step": 12561 + }, + { + "epoch": 0.9440853750187885, + "grad_norm": 2.8729371672948436, + "learning_rate": 3.270957718330591e-08, + "loss": 1.0374, + "step": 12562 + }, + { + "epoch": 0.9441605290846234, + "grad_norm": 3.9076492855088207, + "learning_rate": 3.2621946962271715e-08, + "loss": 1.0613, + "step": 12563 + }, + { + "epoch": 0.9442356831504585, + "grad_norm": 0.69521099592672, + "learning_rate": 3.2534433315920765e-08, + "loss": 0.8088, + "step": 12564 + }, + { + "epoch": 0.9443108372162934, + "grad_norm": 1.9619754916457866, + "learning_rate": 3.2447036249438455e-08, + "loss": 0.9992, + "step": 12565 + }, + { + "epoch": 0.9443859912821284, + "grad_norm": 2.0403426671671823, + "learning_rate": 3.235975576800376e-08, + "loss": 1.0488, + "step": 12566 + }, + { + "epoch": 0.9444611453479633, + "grad_norm": 1.6255829572392482, + "learning_rate": 3.227259187678788e-08, + "loss": 0.9532, + "step": 12567 + }, + { + "epoch": 0.9445362994137982, + "grad_norm": 2.8505255499020756, + "learning_rate": 3.218554458095602e-08, + "loss": 0.9611, + "step": 12568 + }, + { + "epoch": 0.9446114534796333, + "grad_norm": 2.498703238816364, + "learning_rate": 3.2098613885665816e-08, + "loss": 0.9317, + "step": 12569 + }, + { + "epoch": 0.9446866075454682, + "grad_norm": 2.8164044722764414, + "learning_rate": 3.20117997960685e-08, + "loss": 0.9811, + "step": 12570 + }, + { + "epoch": 0.9447617616113032, + "grad_norm": 2.0855031221997913, + "learning_rate": 3.1925102317307716e-08, + "loss": 0.9023, + "step": 12571 + }, + { + "epoch": 0.9448369156771381, + "grad_norm": 2.0456613917716577, + "learning_rate": 3.183852145452115e-08, + "loss": 0.8733, + "step": 12572 + }, + { + "epoch": 0.9449120697429731, + "grad_norm": 0.722781938499668, + "learning_rate": 3.175205721283847e-08, + "loss": 0.8277, + "step": 12573 + }, + { + "epoch": 0.9449872238088081, + "grad_norm": 1.7299185194551383, + "learning_rate": 3.166570959738335e-08, + "loss": 0.9263, + "step": 12574 + }, + { + "epoch": 0.945062377874643, + "grad_norm": 3.9987891271786986, + "learning_rate": 3.1579478613272594e-08, + "loss": 1.0714, + "step": 12575 + }, + { + "epoch": 0.945137531940478, + "grad_norm": 3.320015209576801, + "learning_rate": 3.149336426561522e-08, + "loss": 0.9505, + "step": 12576 + }, + { + "epoch": 0.9452126860063129, + "grad_norm": 1.8602216059431664, + "learning_rate": 3.140736655951359e-08, + "loss": 0.938, + "step": 12577 + }, + { + "epoch": 0.945287840072148, + "grad_norm": 1.5275266884128331, + "learning_rate": 3.1321485500064084e-08, + "loss": 1.059, + "step": 12578 + }, + { + "epoch": 0.9453629941379829, + "grad_norm": 2.035926834043375, + "learning_rate": 3.123572109235484e-08, + "loss": 0.959, + "step": 12579 + }, + { + "epoch": 0.9454381482038178, + "grad_norm": 2.2292025791971266, + "learning_rate": 3.1150073341468016e-08, + "loss": 0.8924, + "step": 12580 + }, + { + "epoch": 0.9455133022696528, + "grad_norm": 1.5475069546889897, + "learning_rate": 3.1064542252478896e-08, + "loss": 1.0099, + "step": 12581 + }, + { + "epoch": 0.9455884563354877, + "grad_norm": 1.9118618517570398, + "learning_rate": 3.097912783045498e-08, + "loss": 0.9844, + "step": 12582 + }, + { + "epoch": 0.9456636104013227, + "grad_norm": 1.4661048445271143, + "learning_rate": 3.0893830080457764e-08, + "loss": 0.9373, + "step": 12583 + }, + { + "epoch": 0.9457387644671577, + "grad_norm": 2.457340578824375, + "learning_rate": 3.080864900754121e-08, + "loss": 0.9903, + "step": 12584 + }, + { + "epoch": 0.9458139185329927, + "grad_norm": 2.0871249473029394, + "learning_rate": 3.072358461675284e-08, + "loss": 0.9722, + "step": 12585 + }, + { + "epoch": 0.9458890725988276, + "grad_norm": 1.769427433294784, + "learning_rate": 3.063863691313284e-08, + "loss": 0.9371, + "step": 12586 + }, + { + "epoch": 0.9459642266646625, + "grad_norm": 1.759876378984749, + "learning_rate": 3.0553805901714745e-08, + "loss": 0.8882, + "step": 12587 + }, + { + "epoch": 0.9460393807304975, + "grad_norm": 1.4512870404020413, + "learning_rate": 3.04690915875252e-08, + "loss": 1.0544, + "step": 12588 + }, + { + "epoch": 0.9461145347963325, + "grad_norm": 1.5547328923178663, + "learning_rate": 3.038449397558396e-08, + "loss": 0.9985, + "step": 12589 + }, + { + "epoch": 0.9461896888621675, + "grad_norm": 2.4772756614699496, + "learning_rate": 3.030001307090346e-08, + "loss": 0.9441, + "step": 12590 + }, + { + "epoch": 0.9462648429280024, + "grad_norm": 2.107444048916014, + "learning_rate": 3.021564887848971e-08, + "loss": 1.0102, + "step": 12591 + }, + { + "epoch": 0.9463399969938374, + "grad_norm": 1.8584018372481055, + "learning_rate": 3.0131401403341584e-08, + "loss": 0.9028, + "step": 12592 + }, + { + "epoch": 0.9464151510596723, + "grad_norm": 2.0567422897845358, + "learning_rate": 3.00472706504511e-08, + "loss": 0.9992, + "step": 12593 + }, + { + "epoch": 0.9464903051255072, + "grad_norm": 1.3875177365105895, + "learning_rate": 2.9963256624803144e-08, + "loss": 0.9982, + "step": 12594 + }, + { + "epoch": 0.9465654591913423, + "grad_norm": 2.2213241259089074, + "learning_rate": 2.987935933137642e-08, + "loss": 1.0303, + "step": 12595 + }, + { + "epoch": 0.9466406132571772, + "grad_norm": 1.6470655906462095, + "learning_rate": 2.979557877514116e-08, + "loss": 0.9547, + "step": 12596 + }, + { + "epoch": 0.9467157673230122, + "grad_norm": 1.684825543957346, + "learning_rate": 2.9711914961062512e-08, + "loss": 0.955, + "step": 12597 + }, + { + "epoch": 0.9467909213888471, + "grad_norm": 1.620736471124096, + "learning_rate": 2.9628367894097615e-08, + "loss": 0.9582, + "step": 12598 + }, + { + "epoch": 0.946866075454682, + "grad_norm": 4.984611481703981, + "learning_rate": 2.9544937579197183e-08, + "loss": 0.7517, + "step": 12599 + }, + { + "epoch": 0.9469412295205171, + "grad_norm": 1.6333449376271434, + "learning_rate": 2.9461624021304366e-08, + "loss": 0.8855, + "step": 12600 + }, + { + "epoch": 0.947016383586352, + "grad_norm": 1.7251430562726422, + "learning_rate": 2.9378427225356107e-08, + "loss": 0.9751, + "step": 12601 + }, + { + "epoch": 0.947091537652187, + "grad_norm": 1.6334719261743844, + "learning_rate": 2.9295347196282015e-08, + "loss": 0.993, + "step": 12602 + }, + { + "epoch": 0.9471666917180219, + "grad_norm": 2.2582577593030413, + "learning_rate": 2.9212383939004827e-08, + "loss": 0.9794, + "step": 12603 + }, + { + "epoch": 0.947241845783857, + "grad_norm": 1.6783975305309984, + "learning_rate": 2.912953745844082e-08, + "loss": 0.9635, + "step": 12604 + }, + { + "epoch": 0.9473169998496919, + "grad_norm": 1.9781560693503537, + "learning_rate": 2.9046807759498303e-08, + "loss": 0.9653, + "step": 12605 + }, + { + "epoch": 0.9473921539155268, + "grad_norm": 2.820543697105921, + "learning_rate": 2.8964194847080238e-08, + "loss": 0.9069, + "step": 12606 + }, + { + "epoch": 0.9474673079813618, + "grad_norm": 1.704428008108177, + "learning_rate": 2.8881698726080705e-08, + "loss": 1.0193, + "step": 12607 + }, + { + "epoch": 0.9475424620471967, + "grad_norm": 1.8838894878096388, + "learning_rate": 2.879931940138869e-08, + "loss": 0.9022, + "step": 12608 + }, + { + "epoch": 0.9476176161130317, + "grad_norm": 1.4427180743204215, + "learning_rate": 2.8717056877885394e-08, + "loss": 0.9774, + "step": 12609 + }, + { + "epoch": 0.9476927701788667, + "grad_norm": 1.5140770154637093, + "learning_rate": 2.8634911160444696e-08, + "loss": 0.9439, + "step": 12610 + }, + { + "epoch": 0.9477679242447017, + "grad_norm": 1.7013492951576112, + "learning_rate": 2.8552882253934485e-08, + "loss": 0.9704, + "step": 12611 + }, + { + "epoch": 0.9478430783105366, + "grad_norm": 2.0026462452952507, + "learning_rate": 2.8470970163215312e-08, + "loss": 0.9352, + "step": 12612 + }, + { + "epoch": 0.9479182323763715, + "grad_norm": 1.6186976361734502, + "learning_rate": 2.838917489314041e-08, + "loss": 1.0146, + "step": 12613 + }, + { + "epoch": 0.9479933864422065, + "grad_norm": 1.6714542608184948, + "learning_rate": 2.830749644855679e-08, + "loss": 1.0085, + "step": 12614 + }, + { + "epoch": 0.9480685405080415, + "grad_norm": 1.62512991459255, + "learning_rate": 2.8225934834304133e-08, + "loss": 1.0159, + "step": 12615 + }, + { + "epoch": 0.9481436945738765, + "grad_norm": 1.8751579218163168, + "learning_rate": 2.8144490055215465e-08, + "loss": 0.9674, + "step": 12616 + }, + { + "epoch": 0.9482188486397114, + "grad_norm": 1.7818541589733734, + "learning_rate": 2.8063162116116256e-08, + "loss": 0.9454, + "step": 12617 + }, + { + "epoch": 0.9482940027055464, + "grad_norm": 1.5873904032684953, + "learning_rate": 2.7981951021825544e-08, + "loss": 1.0047, + "step": 12618 + }, + { + "epoch": 0.9483691567713813, + "grad_norm": 1.9158876397709048, + "learning_rate": 2.7900856777156147e-08, + "loss": 0.9452, + "step": 12619 + }, + { + "epoch": 0.9484443108372163, + "grad_norm": 1.858682647798284, + "learning_rate": 2.7819879386912214e-08, + "loss": 1.0237, + "step": 12620 + }, + { + "epoch": 0.9485194649030513, + "grad_norm": 1.625195124722078, + "learning_rate": 2.773901885589258e-08, + "loss": 0.9382, + "step": 12621 + }, + { + "epoch": 0.9485946189688862, + "grad_norm": 1.5582518238367247, + "learning_rate": 2.7658275188888526e-08, + "loss": 1.0575, + "step": 12622 + }, + { + "epoch": 0.9486697730347212, + "grad_norm": 0.6720157741954552, + "learning_rate": 2.7577648390683995e-08, + "loss": 0.8303, + "step": 12623 + }, + { + "epoch": 0.9487449271005561, + "grad_norm": 1.768567224770926, + "learning_rate": 2.7497138466056724e-08, + "loss": 1.0471, + "step": 12624 + }, + { + "epoch": 0.948820081166391, + "grad_norm": 2.3679443760859975, + "learning_rate": 2.7416745419777344e-08, + "loss": 0.9073, + "step": 12625 + }, + { + "epoch": 0.9488952352322261, + "grad_norm": 1.4259064083818676, + "learning_rate": 2.7336469256609152e-08, + "loss": 0.9898, + "step": 12626 + }, + { + "epoch": 0.948970389298061, + "grad_norm": 1.4641632533951852, + "learning_rate": 2.7256309981309234e-08, + "loss": 0.8976, + "step": 12627 + }, + { + "epoch": 0.949045543363896, + "grad_norm": 2.4632170277120022, + "learning_rate": 2.71762675986269e-08, + "loss": 0.9336, + "step": 12628 + }, + { + "epoch": 0.9491206974297309, + "grad_norm": 2.1601176198088594, + "learning_rate": 2.709634211330547e-08, + "loss": 1.0602, + "step": 12629 + }, + { + "epoch": 0.949195851495566, + "grad_norm": 2.072303591855896, + "learning_rate": 2.7016533530080044e-08, + "loss": 0.9136, + "step": 12630 + }, + { + "epoch": 0.9492710055614009, + "grad_norm": 1.5128682815237422, + "learning_rate": 2.6936841853680393e-08, + "loss": 0.9536, + "step": 12631 + }, + { + "epoch": 0.9493461596272358, + "grad_norm": 1.5056595527481431, + "learning_rate": 2.6857267088828073e-08, + "loss": 0.9063, + "step": 12632 + }, + { + "epoch": 0.9494213136930708, + "grad_norm": 2.5200633825172525, + "learning_rate": 2.6777809240238425e-08, + "loss": 0.9611, + "step": 12633 + }, + { + "epoch": 0.9494964677589057, + "grad_norm": 6.717558147376091, + "learning_rate": 2.669846831261946e-08, + "loss": 1.0022, + "step": 12634 + }, + { + "epoch": 0.9495716218247408, + "grad_norm": 1.567644500277757, + "learning_rate": 2.661924431067275e-08, + "loss": 0.891, + "step": 12635 + }, + { + "epoch": 0.9496467758905757, + "grad_norm": 1.5698503411998415, + "learning_rate": 2.6540137239092098e-08, + "loss": 0.9478, + "step": 12636 + }, + { + "epoch": 0.9497219299564107, + "grad_norm": 1.9167893273729917, + "learning_rate": 2.6461147102565527e-08, + "loss": 1.0008, + "step": 12637 + }, + { + "epoch": 0.9497970840222456, + "grad_norm": 1.4680230797036962, + "learning_rate": 2.6382273905772858e-08, + "loss": 0.9061, + "step": 12638 + }, + { + "epoch": 0.9498722380880805, + "grad_norm": 13.193650099729853, + "learning_rate": 2.630351765338812e-08, + "loss": 0.9081, + "step": 12639 + }, + { + "epoch": 0.9499473921539155, + "grad_norm": 6.1360946133915, + "learning_rate": 2.6224878350077585e-08, + "loss": 0.9717, + "step": 12640 + }, + { + "epoch": 0.9500225462197505, + "grad_norm": 0.739956041009838, + "learning_rate": 2.614635600050108e-08, + "loss": 0.8613, + "step": 12641 + }, + { + "epoch": 0.9500977002855855, + "grad_norm": 2.2534725095816968, + "learning_rate": 2.6067950609311552e-08, + "loss": 0.9667, + "step": 12642 + }, + { + "epoch": 0.9501728543514204, + "grad_norm": 1.9327007659218984, + "learning_rate": 2.5989662181154835e-08, + "loss": 0.972, + "step": 12643 + }, + { + "epoch": 0.9502480084172553, + "grad_norm": 2.0347288418757272, + "learning_rate": 2.5911490720669227e-08, + "loss": 0.9595, + "step": 12644 + }, + { + "epoch": 0.9503231624830903, + "grad_norm": 2.586419098139492, + "learning_rate": 2.5833436232487238e-08, + "loss": 0.9847, + "step": 12645 + }, + { + "epoch": 0.9503983165489253, + "grad_norm": 3.1505062604707876, + "learning_rate": 2.575549872123384e-08, + "loss": 1.0536, + "step": 12646 + }, + { + "epoch": 0.9504734706147603, + "grad_norm": 2.1434747146599316, + "learning_rate": 2.5677678191526885e-08, + "loss": 0.9032, + "step": 12647 + }, + { + "epoch": 0.9505486246805952, + "grad_norm": 1.7908465556275956, + "learning_rate": 2.5599974647977805e-08, + "loss": 0.8369, + "step": 12648 + }, + { + "epoch": 0.9506237787464302, + "grad_norm": 2.9766669467813403, + "learning_rate": 2.5522388095190472e-08, + "loss": 0.9361, + "step": 12649 + }, + { + "epoch": 0.9506989328122651, + "grad_norm": 2.1814323196072016, + "learning_rate": 2.544491853776276e-08, + "loss": 1.0319, + "step": 12650 + }, + { + "epoch": 0.9507740868781, + "grad_norm": 2.0891776190847136, + "learning_rate": 2.5367565980284332e-08, + "loss": 0.93, + "step": 12651 + }, + { + "epoch": 0.9508492409439351, + "grad_norm": 0.8181775603520741, + "learning_rate": 2.52903304273393e-08, + "loss": 0.9187, + "step": 12652 + }, + { + "epoch": 0.95092439500977, + "grad_norm": 1.7996229969036561, + "learning_rate": 2.5213211883503784e-08, + "loss": 0.9418, + "step": 12653 + }, + { + "epoch": 0.950999549075605, + "grad_norm": 1.8990183126240565, + "learning_rate": 2.5136210353347452e-08, + "loss": 0.9964, + "step": 12654 + }, + { + "epoch": 0.9510747031414399, + "grad_norm": 1.4640347701045051, + "learning_rate": 2.5059325841432667e-08, + "loss": 0.9908, + "step": 12655 + }, + { + "epoch": 0.951149857207275, + "grad_norm": 2.300843009702716, + "learning_rate": 2.4982558352315775e-08, + "loss": 0.9581, + "step": 12656 + }, + { + "epoch": 0.9512250112731099, + "grad_norm": 2.5817381538096433, + "learning_rate": 2.490590789054492e-08, + "loss": 0.8241, + "step": 12657 + }, + { + "epoch": 0.9513001653389448, + "grad_norm": 1.864300335198808, + "learning_rate": 2.4829374460662244e-08, + "loss": 0.8618, + "step": 12658 + }, + { + "epoch": 0.9513753194047798, + "grad_norm": 5.027081581280885, + "learning_rate": 2.4752958067202347e-08, + "loss": 1.0097, + "step": 12659 + }, + { + "epoch": 0.9514504734706147, + "grad_norm": 1.9341355306520756, + "learning_rate": 2.467665871469382e-08, + "loss": 0.9166, + "step": 12660 + }, + { + "epoch": 0.9515256275364498, + "grad_norm": 2.0876711381213338, + "learning_rate": 2.4600476407656835e-08, + "loss": 1.0885, + "step": 12661 + }, + { + "epoch": 0.9516007816022847, + "grad_norm": 1.6698646223281324, + "learning_rate": 2.4524411150605995e-08, + "loss": 1.0339, + "step": 12662 + }, + { + "epoch": 0.9516759356681197, + "grad_norm": 1.7308947725492458, + "learning_rate": 2.444846294804881e-08, + "loss": 0.9078, + "step": 12663 + }, + { + "epoch": 0.9517510897339546, + "grad_norm": 1.456526190635221, + "learning_rate": 2.4372631804484567e-08, + "loss": 0.9264, + "step": 12664 + }, + { + "epoch": 0.9518262437997895, + "grad_norm": 1.5412348408815946, + "learning_rate": 2.429691772440745e-08, + "loss": 0.9266, + "step": 12665 + }, + { + "epoch": 0.9519013978656246, + "grad_norm": 2.3449140582573156, + "learning_rate": 2.422132071230343e-08, + "loss": 0.9605, + "step": 12666 + }, + { + "epoch": 0.9519765519314595, + "grad_norm": 1.8706164298735892, + "learning_rate": 2.414584077265158e-08, + "loss": 0.8952, + "step": 12667 + }, + { + "epoch": 0.9520517059972945, + "grad_norm": 1.9552492110691349, + "learning_rate": 2.407047790992478e-08, + "loss": 1.0418, + "step": 12668 + }, + { + "epoch": 0.9521268600631294, + "grad_norm": 1.731550167241537, + "learning_rate": 2.3995232128588782e-08, + "loss": 1.0258, + "step": 12669 + }, + { + "epoch": 0.9522020141289643, + "grad_norm": 1.9889295678918713, + "learning_rate": 2.39201034331018e-08, + "loss": 0.9372, + "step": 12670 + }, + { + "epoch": 0.9522771681947994, + "grad_norm": 1.8441011773277733, + "learning_rate": 2.3845091827915608e-08, + "loss": 0.9488, + "step": 12671 + }, + { + "epoch": 0.9523523222606343, + "grad_norm": 0.7481155563179116, + "learning_rate": 2.377019731747465e-08, + "loss": 0.7794, + "step": 12672 + }, + { + "epoch": 0.9524274763264693, + "grad_norm": 2.2685072882007393, + "learning_rate": 2.3695419906217594e-08, + "loss": 0.8937, + "step": 12673 + }, + { + "epoch": 0.9525026303923042, + "grad_norm": 2.5838892451368793, + "learning_rate": 2.3620759598574013e-08, + "loss": 0.8971, + "step": 12674 + }, + { + "epoch": 0.9525777844581392, + "grad_norm": 1.731397392526386, + "learning_rate": 2.3546216398969033e-08, + "loss": 1.0613, + "step": 12675 + }, + { + "epoch": 0.9526529385239741, + "grad_norm": 1.6100751964707751, + "learning_rate": 2.3471790311818675e-08, + "loss": 0.9923, + "step": 12676 + }, + { + "epoch": 0.9527280925898091, + "grad_norm": 1.7643361071179167, + "learning_rate": 2.3397481341533632e-08, + "loss": 0.9406, + "step": 12677 + }, + { + "epoch": 0.9528032466556441, + "grad_norm": 1.9275395102601378, + "learning_rate": 2.3323289492516607e-08, + "loss": 1.052, + "step": 12678 + }, + { + "epoch": 0.952878400721479, + "grad_norm": 2.319878739936955, + "learning_rate": 2.324921476916386e-08, + "loss": 0.9003, + "step": 12679 + }, + { + "epoch": 0.952953554787314, + "grad_norm": 2.3298833978316287, + "learning_rate": 2.3175257175864772e-08, + "loss": 0.9228, + "step": 12680 + }, + { + "epoch": 0.9530287088531489, + "grad_norm": 1.4507651086257498, + "learning_rate": 2.310141671700139e-08, + "loss": 0.9733, + "step": 12681 + }, + { + "epoch": 0.953103862918984, + "grad_norm": 2.0852660051990886, + "learning_rate": 2.30276933969491e-08, + "loss": 0.9142, + "step": 12682 + }, + { + "epoch": 0.9531790169848189, + "grad_norm": 1.60204847452497, + "learning_rate": 2.295408722007641e-08, + "loss": 0.904, + "step": 12683 + }, + { + "epoch": 0.9532541710506538, + "grad_norm": 1.6824878857702215, + "learning_rate": 2.2880598190744503e-08, + "loss": 0.9543, + "step": 12684 + }, + { + "epoch": 0.9533293251164888, + "grad_norm": 1.3888578008775232, + "learning_rate": 2.280722631330789e-08, + "loss": 0.9251, + "step": 12685 + }, + { + "epoch": 0.9534044791823237, + "grad_norm": 1.5621849505378322, + "learning_rate": 2.2733971592114654e-08, + "loss": 0.9366, + "step": 12686 + }, + { + "epoch": 0.9534796332481588, + "grad_norm": 3.3408158878496717, + "learning_rate": 2.266083403150487e-08, + "loss": 0.9365, + "step": 12687 + }, + { + "epoch": 0.9535547873139937, + "grad_norm": 7.1852976202861525, + "learning_rate": 2.2587813635812414e-08, + "loss": 1.0342, + "step": 12688 + }, + { + "epoch": 0.9536299413798286, + "grad_norm": 2.256598017262098, + "learning_rate": 2.251491040936404e-08, + "loss": 0.9661, + "step": 12689 + }, + { + "epoch": 0.9537050954456636, + "grad_norm": 4.778973205606236, + "learning_rate": 2.244212435647963e-08, + "loss": 0.978, + "step": 12690 + }, + { + "epoch": 0.9537802495114985, + "grad_norm": 2.624501408046607, + "learning_rate": 2.236945548147173e-08, + "loss": 0.8819, + "step": 12691 + }, + { + "epoch": 0.9538554035773336, + "grad_norm": 1.8539961904385798, + "learning_rate": 2.229690378864668e-08, + "loss": 0.9954, + "step": 12692 + }, + { + "epoch": 0.9539305576431685, + "grad_norm": 1.7368293785316435, + "learning_rate": 2.2224469282303037e-08, + "loss": 0.9664, + "step": 12693 + }, + { + "epoch": 0.9540057117090035, + "grad_norm": 4.275368764933836, + "learning_rate": 2.2152151966733146e-08, + "loss": 0.9626, + "step": 12694 + }, + { + "epoch": 0.9540808657748384, + "grad_norm": 2.122053909062809, + "learning_rate": 2.20799518462218e-08, + "loss": 0.8457, + "step": 12695 + }, + { + "epoch": 0.9541560198406733, + "grad_norm": 2.1866316025333927, + "learning_rate": 2.2007868925047135e-08, + "loss": 0.9655, + "step": 12696 + }, + { + "epoch": 0.9542311739065084, + "grad_norm": 3.321208375232772, + "learning_rate": 2.1935903207480844e-08, + "loss": 1.0268, + "step": 12697 + }, + { + "epoch": 0.9543063279723433, + "grad_norm": 1.7441359089513158, + "learning_rate": 2.1864054697786626e-08, + "loss": 0.9838, + "step": 12698 + }, + { + "epoch": 0.9543814820381783, + "grad_norm": 1.7241979780099665, + "learning_rate": 2.1792323400221745e-08, + "loss": 1.0479, + "step": 12699 + }, + { + "epoch": 0.9544566361040132, + "grad_norm": 1.7205098938883354, + "learning_rate": 2.1720709319037024e-08, + "loss": 0.9523, + "step": 12700 + }, + { + "epoch": 0.9545317901698482, + "grad_norm": 21.15162488679209, + "learning_rate": 2.1649212458475508e-08, + "loss": 0.9581, + "step": 12701 + }, + { + "epoch": 0.9546069442356832, + "grad_norm": 1.725606316973425, + "learning_rate": 2.157783282277381e-08, + "loss": 0.9183, + "step": 12702 + }, + { + "epoch": 0.9546820983015181, + "grad_norm": 1.5939797628498926, + "learning_rate": 2.1506570416161217e-08, + "loss": 0.921, + "step": 12703 + }, + { + "epoch": 0.9547572523673531, + "grad_norm": 1.6806859796880398, + "learning_rate": 2.1435425242861015e-08, + "loss": 0.9984, + "step": 12704 + }, + { + "epoch": 0.954832406433188, + "grad_norm": 2.4052806734881513, + "learning_rate": 2.1364397307087834e-08, + "loss": 1.0041, + "step": 12705 + }, + { + "epoch": 0.954907560499023, + "grad_norm": 1.6058592681031039, + "learning_rate": 2.129348661305075e-08, + "loss": 1.0224, + "step": 12706 + }, + { + "epoch": 0.954982714564858, + "grad_norm": 1.4983703980124181, + "learning_rate": 2.1222693164951956e-08, + "loss": 0.9791, + "step": 12707 + }, + { + "epoch": 0.955057868630693, + "grad_norm": 2.1247017595026123, + "learning_rate": 2.115201696698543e-08, + "loss": 0.9985, + "step": 12708 + }, + { + "epoch": 0.9551330226965279, + "grad_norm": 1.875504680105994, + "learning_rate": 2.1081458023339605e-08, + "loss": 0.875, + "step": 12709 + }, + { + "epoch": 0.9552081767623628, + "grad_norm": 1.6301621825619308, + "learning_rate": 2.101101633819513e-08, + "loss": 0.9793, + "step": 12710 + }, + { + "epoch": 0.9552833308281978, + "grad_norm": 1.650286031264081, + "learning_rate": 2.0940691915726005e-08, + "loss": 0.9353, + "step": 12711 + }, + { + "epoch": 0.9553584848940327, + "grad_norm": 2.0286756194200555, + "learning_rate": 2.0870484760099117e-08, + "loss": 0.9393, + "step": 12712 + }, + { + "epoch": 0.9554336389598678, + "grad_norm": 1.8715169981404425, + "learning_rate": 2.0800394875474915e-08, + "loss": 0.9349, + "step": 12713 + }, + { + "epoch": 0.9555087930257027, + "grad_norm": 1.615812444973216, + "learning_rate": 2.0730422266005853e-08, + "loss": 0.8753, + "step": 12714 + }, + { + "epoch": 0.9555839470915376, + "grad_norm": 2.2395582567333974, + "learning_rate": 2.066056693583862e-08, + "loss": 0.9679, + "step": 12715 + }, + { + "epoch": 0.9556591011573726, + "grad_norm": 2.4707034144947952, + "learning_rate": 2.05908288891119e-08, + "loss": 0.8906, + "step": 12716 + }, + { + "epoch": 0.9557342552232075, + "grad_norm": 1.635964303949321, + "learning_rate": 2.0521208129958613e-08, + "loss": 0.9784, + "step": 12717 + }, + { + "epoch": 0.9558094092890426, + "grad_norm": 1.6397060181968108, + "learning_rate": 2.0451704662503456e-08, + "loss": 0.9887, + "step": 12718 + }, + { + "epoch": 0.9558845633548775, + "grad_norm": 1.5437323503605023, + "learning_rate": 2.0382318490865134e-08, + "loss": 1.0057, + "step": 12719 + }, + { + "epoch": 0.9559597174207125, + "grad_norm": 1.6760881049639982, + "learning_rate": 2.031304961915459e-08, + "loss": 1.0399, + "step": 12720 + }, + { + "epoch": 0.9560348714865474, + "grad_norm": 1.3713225258049138, + "learning_rate": 2.024389805147697e-08, + "loss": 0.9741, + "step": 12721 + }, + { + "epoch": 0.9561100255523823, + "grad_norm": 1.3443273146936248, + "learning_rate": 2.017486379192901e-08, + "loss": 1.0011, + "step": 12722 + }, + { + "epoch": 0.9561851796182174, + "grad_norm": 1.487478685627025, + "learning_rate": 2.0105946844601874e-08, + "loss": 0.9859, + "step": 12723 + }, + { + "epoch": 0.9562603336840523, + "grad_norm": 1.7929337774452114, + "learning_rate": 2.0037147213578964e-08, + "loss": 0.9549, + "step": 12724 + }, + { + "epoch": 0.9563354877498873, + "grad_norm": 1.5836897833621133, + "learning_rate": 1.99684649029368e-08, + "loss": 0.922, + "step": 12725 + }, + { + "epoch": 0.9564106418157222, + "grad_norm": 2.0613633694074567, + "learning_rate": 1.989989991674501e-08, + "loss": 1.0125, + "step": 12726 + }, + { + "epoch": 0.9564857958815572, + "grad_norm": 2.7253709173180245, + "learning_rate": 1.983145225906657e-08, + "loss": 0.8272, + "step": 12727 + }, + { + "epoch": 0.9565609499473922, + "grad_norm": 1.9380171916147975, + "learning_rate": 1.9763121933957128e-08, + "loss": 0.8976, + "step": 12728 + }, + { + "epoch": 0.9566361040132271, + "grad_norm": 1.490766797628898, + "learning_rate": 1.9694908945465438e-08, + "loss": 0.948, + "step": 12729 + }, + { + "epoch": 0.9567112580790621, + "grad_norm": 5.608673646907498, + "learning_rate": 1.9626813297633604e-08, + "loss": 1.0692, + "step": 12730 + }, + { + "epoch": 0.956786412144897, + "grad_norm": 2.069651256982926, + "learning_rate": 1.9558834994496397e-08, + "loss": 1.0003, + "step": 12731 + }, + { + "epoch": 0.956861566210732, + "grad_norm": 1.611898499310508, + "learning_rate": 1.949097404008193e-08, + "loss": 0.8876, + "step": 12732 + }, + { + "epoch": 0.956936720276567, + "grad_norm": 1.5657236033929103, + "learning_rate": 1.9423230438410987e-08, + "loss": 0.9335, + "step": 12733 + }, + { + "epoch": 0.9570118743424019, + "grad_norm": 1.6935819591763221, + "learning_rate": 1.935560419349791e-08, + "loss": 0.9985, + "step": 12734 + }, + { + "epoch": 0.9570870284082369, + "grad_norm": 2.02350678196424, + "learning_rate": 1.9288095309349718e-08, + "loss": 1.0047, + "step": 12735 + }, + { + "epoch": 0.9571621824740718, + "grad_norm": 3.022758624526351, + "learning_rate": 1.9220703789966318e-08, + "loss": 0.9297, + "step": 12736 + }, + { + "epoch": 0.9572373365399068, + "grad_norm": 1.7581651122029405, + "learning_rate": 1.915342963934119e-08, + "loss": 1.0008, + "step": 12737 + }, + { + "epoch": 0.9573124906057418, + "grad_norm": 2.1761636449342894, + "learning_rate": 1.9086272861460695e-08, + "loss": 0.9471, + "step": 12738 + }, + { + "epoch": 0.9573876446715768, + "grad_norm": 1.5983327165641563, + "learning_rate": 1.9019233460303652e-08, + "loss": 1.0248, + "step": 12739 + }, + { + "epoch": 0.9574627987374117, + "grad_norm": 2.3961304742954654, + "learning_rate": 1.8952311439843106e-08, + "loss": 1.0033, + "step": 12740 + }, + { + "epoch": 0.9575379528032466, + "grad_norm": 0.7410863236327263, + "learning_rate": 1.8885506804043884e-08, + "loss": 0.8472, + "step": 12741 + }, + { + "epoch": 0.9576131068690816, + "grad_norm": 2.3334754037748016, + "learning_rate": 1.8818819556864374e-08, + "loss": 0.9695, + "step": 12742 + }, + { + "epoch": 0.9576882609349165, + "grad_norm": 1.675711534735034, + "learning_rate": 1.8752249702256307e-08, + "loss": 0.9855, + "step": 12743 + }, + { + "epoch": 0.9577634150007516, + "grad_norm": 0.6818051289985141, + "learning_rate": 1.8685797244164524e-08, + "loss": 0.7603, + "step": 12744 + }, + { + "epoch": 0.9578385690665865, + "grad_norm": 2.7886895109325116, + "learning_rate": 1.861946218652588e-08, + "loss": 0.9191, + "step": 12745 + }, + { + "epoch": 0.9579137231324215, + "grad_norm": 2.2972875730584663, + "learning_rate": 1.8553244533271227e-08, + "loss": 0.9115, + "step": 12746 + }, + { + "epoch": 0.9579888771982564, + "grad_norm": 2.0035131117161074, + "learning_rate": 1.848714428832454e-08, + "loss": 0.9465, + "step": 12747 + }, + { + "epoch": 0.9580640312640913, + "grad_norm": 1.7437059431082076, + "learning_rate": 1.8421161455602242e-08, + "loss": 1.0667, + "step": 12748 + }, + { + "epoch": 0.9581391853299264, + "grad_norm": 2.7857625122227514, + "learning_rate": 1.8355296039013867e-08, + "loss": 0.9426, + "step": 12749 + }, + { + "epoch": 0.9582143393957613, + "grad_norm": 1.8701112592717721, + "learning_rate": 1.828954804246252e-08, + "loss": 0.8589, + "step": 12750 + }, + { + "epoch": 0.9582894934615963, + "grad_norm": 0.7662672346457615, + "learning_rate": 1.8223917469844198e-08, + "loss": 0.8183, + "step": 12751 + }, + { + "epoch": 0.9583646475274312, + "grad_norm": 1.7367188021667994, + "learning_rate": 1.8158404325047338e-08, + "loss": 0.9879, + "step": 12752 + }, + { + "epoch": 0.9584398015932662, + "grad_norm": 0.8272502886004159, + "learning_rate": 1.8093008611953952e-08, + "loss": 0.8651, + "step": 12753 + }, + { + "epoch": 0.9585149556591012, + "grad_norm": 1.7298581529358308, + "learning_rate": 1.802773033443894e-08, + "loss": 1.0189, + "step": 12754 + }, + { + "epoch": 0.9585901097249361, + "grad_norm": 1.467347307205966, + "learning_rate": 1.796256949637054e-08, + "loss": 0.9479, + "step": 12755 + }, + { + "epoch": 0.9586652637907711, + "grad_norm": 1.4439134579558992, + "learning_rate": 1.789752610160944e-08, + "loss": 0.9894, + "step": 12756 + }, + { + "epoch": 0.958740417856606, + "grad_norm": 2.864272629042397, + "learning_rate": 1.7832600154010114e-08, + "loss": 0.9138, + "step": 12757 + }, + { + "epoch": 0.958815571922441, + "grad_norm": 1.6558324179465238, + "learning_rate": 1.7767791657419484e-08, + "loss": 0.9255, + "step": 12758 + }, + { + "epoch": 0.958890725988276, + "grad_norm": 1.3919893065896858, + "learning_rate": 1.770310061567759e-08, + "loss": 0.9648, + "step": 12759 + }, + { + "epoch": 0.9589658800541109, + "grad_norm": 2.090465189437976, + "learning_rate": 1.763852703261759e-08, + "loss": 1.0086, + "step": 12760 + }, + { + "epoch": 0.9590410341199459, + "grad_norm": 1.5633681365140246, + "learning_rate": 1.7574070912065975e-08, + "loss": 0.8639, + "step": 12761 + }, + { + "epoch": 0.9591161881857808, + "grad_norm": 1.727896472796656, + "learning_rate": 1.7509732257841693e-08, + "loss": 0.9944, + "step": 12762 + }, + { + "epoch": 0.9591913422516158, + "grad_norm": 1.5960313513121658, + "learning_rate": 1.7445511073757468e-08, + "loss": 0.9974, + "step": 12763 + }, + { + "epoch": 0.9592664963174508, + "grad_norm": 2.6454675276383224, + "learning_rate": 1.738140736361826e-08, + "loss": 0.967, + "step": 12764 + }, + { + "epoch": 0.9593416503832858, + "grad_norm": 2.176771503609086, + "learning_rate": 1.7317421131222808e-08, + "loss": 0.767, + "step": 12765 + }, + { + "epoch": 0.9594168044491207, + "grad_norm": 1.7831737770765754, + "learning_rate": 1.725355238036208e-08, + "loss": 0.9662, + "step": 12766 + }, + { + "epoch": 0.9594919585149556, + "grad_norm": 1.8758079695317855, + "learning_rate": 1.7189801114820825e-08, + "loss": 1.0167, + "step": 12767 + }, + { + "epoch": 0.9595671125807906, + "grad_norm": 1.9146169676816744, + "learning_rate": 1.7126167338376908e-08, + "loss": 0.9259, + "step": 12768 + }, + { + "epoch": 0.9596422666466256, + "grad_norm": 2.6421334737694253, + "learning_rate": 1.7062651054800203e-08, + "loss": 0.9799, + "step": 12769 + }, + { + "epoch": 0.9597174207124606, + "grad_norm": 1.6222070019732304, + "learning_rate": 1.6999252267854592e-08, + "loss": 0.9496, + "step": 12770 + }, + { + "epoch": 0.9597925747782955, + "grad_norm": 1.5492503106715734, + "learning_rate": 1.6935970981297067e-08, + "loss": 1.0291, + "step": 12771 + }, + { + "epoch": 0.9598677288441305, + "grad_norm": 1.6598570071530097, + "learning_rate": 1.6872807198876404e-08, + "loss": 0.8744, + "step": 12772 + }, + { + "epoch": 0.9599428829099654, + "grad_norm": 1.8185108748607182, + "learning_rate": 1.680976092433606e-08, + "loss": 0.9919, + "step": 12773 + }, + { + "epoch": 0.9600180369758003, + "grad_norm": 2.551547766302474, + "learning_rate": 1.6746832161411482e-08, + "loss": 0.8863, + "step": 12774 + }, + { + "epoch": 0.9600931910416354, + "grad_norm": 2.019512360888452, + "learning_rate": 1.6684020913831476e-08, + "loss": 0.9632, + "step": 12775 + }, + { + "epoch": 0.9601683451074703, + "grad_norm": 1.2892592771210056, + "learning_rate": 1.6621327185317947e-08, + "loss": 0.9772, + "step": 12776 + }, + { + "epoch": 0.9602434991733053, + "grad_norm": 2.0213147668128397, + "learning_rate": 1.655875097958548e-08, + "loss": 1.0053, + "step": 12777 + }, + { + "epoch": 0.9603186532391402, + "grad_norm": 2.5240327587692057, + "learning_rate": 1.649629230034244e-08, + "loss": 0.994, + "step": 12778 + }, + { + "epoch": 0.9603938073049751, + "grad_norm": 2.3121987223903853, + "learning_rate": 1.6433951151288983e-08, + "loss": 0.994, + "step": 12779 + }, + { + "epoch": 0.9604689613708102, + "grad_norm": 1.7509283600622592, + "learning_rate": 1.6371727536119705e-08, + "loss": 0.9766, + "step": 12780 + }, + { + "epoch": 0.9605441154366451, + "grad_norm": 0.8054374977361057, + "learning_rate": 1.6309621458521437e-08, + "loss": 0.8508, + "step": 12781 + }, + { + "epoch": 0.9606192695024801, + "grad_norm": 2.372350720933824, + "learning_rate": 1.6247632922174348e-08, + "loss": 0.8978, + "step": 12782 + }, + { + "epoch": 0.960694423568315, + "grad_norm": 1.6555470108320915, + "learning_rate": 1.618576193075083e-08, + "loss": 0.8794, + "step": 12783 + }, + { + "epoch": 0.96076957763415, + "grad_norm": 2.652710295168927, + "learning_rate": 1.6124008487917727e-08, + "loss": 1.0607, + "step": 12784 + }, + { + "epoch": 0.960844731699985, + "grad_norm": 1.7323624214713766, + "learning_rate": 1.6062372597333896e-08, + "loss": 0.9132, + "step": 12785 + }, + { + "epoch": 0.9609198857658199, + "grad_norm": 1.4282437732146143, + "learning_rate": 1.600085426265152e-08, + "loss": 0.921, + "step": 12786 + }, + { + "epoch": 0.9609950398316549, + "grad_norm": 1.4871888644730604, + "learning_rate": 1.5939453487515686e-08, + "loss": 1.0705, + "step": 12787 + }, + { + "epoch": 0.9610701938974898, + "grad_norm": 2.9340171457633475, + "learning_rate": 1.5878170275564818e-08, + "loss": 0.9687, + "step": 12788 + }, + { + "epoch": 0.9611453479633248, + "grad_norm": 2.8960429785790223, + "learning_rate": 1.581700463043001e-08, + "loss": 0.9616, + "step": 12789 + }, + { + "epoch": 0.9612205020291598, + "grad_norm": 2.025765982914957, + "learning_rate": 1.5755956555735473e-08, + "loss": 0.9774, + "step": 12790 + }, + { + "epoch": 0.9612956560949948, + "grad_norm": 2.2140623190912727, + "learning_rate": 1.569502605509876e-08, + "loss": 0.9035, + "step": 12791 + }, + { + "epoch": 0.9613708101608297, + "grad_norm": 3.006579971445101, + "learning_rate": 1.5634213132130537e-08, + "loss": 0.9493, + "step": 12792 + }, + { + "epoch": 0.9614459642266646, + "grad_norm": 1.7625282609986972, + "learning_rate": 1.5573517790433255e-08, + "loss": 0.8823, + "step": 12793 + }, + { + "epoch": 0.9615211182924996, + "grad_norm": 2.122491140897525, + "learning_rate": 1.551294003360426e-08, + "loss": 0.918, + "step": 12794 + }, + { + "epoch": 0.9615962723583346, + "grad_norm": 1.6295027338258437, + "learning_rate": 1.5452479865232683e-08, + "loss": 0.9692, + "step": 12795 + }, + { + "epoch": 0.9616714264241696, + "grad_norm": 2.5963317706877653, + "learning_rate": 1.5392137288900764e-08, + "loss": 0.94, + "step": 12796 + }, + { + "epoch": 0.9617465804900045, + "grad_norm": 1.9698590379232013, + "learning_rate": 1.5331912308184537e-08, + "loss": 0.9858, + "step": 12797 + }, + { + "epoch": 0.9618217345558395, + "grad_norm": 1.961563496767927, + "learning_rate": 1.5271804926652032e-08, + "loss": 0.9897, + "step": 12798 + }, + { + "epoch": 0.9618968886216744, + "grad_norm": 2.701881607827642, + "learning_rate": 1.5211815147865514e-08, + "loss": 1.0194, + "step": 12799 + }, + { + "epoch": 0.9619720426875094, + "grad_norm": 1.725174289688861, + "learning_rate": 1.5151942975378808e-08, + "loss": 0.9797, + "step": 12800 + }, + { + "epoch": 0.9620471967533444, + "grad_norm": 1.7452227235549684, + "learning_rate": 1.509218841274018e-08, + "loss": 1.004, + "step": 12801 + }, + { + "epoch": 0.9621223508191793, + "grad_norm": 2.3907615273578795, + "learning_rate": 1.503255146349014e-08, + "loss": 0.8973, + "step": 12802 + }, + { + "epoch": 0.9621975048850143, + "grad_norm": 1.7925083342162913, + "learning_rate": 1.49730321311623e-08, + "loss": 1.0179, + "step": 12803 + }, + { + "epoch": 0.9622726589508492, + "grad_norm": 1.6776842813707156, + "learning_rate": 1.4913630419283617e-08, + "loss": 1.0143, + "step": 12804 + }, + { + "epoch": 0.9623478130166842, + "grad_norm": 2.7966906097836075, + "learning_rate": 1.4854346331373725e-08, + "loss": 0.9593, + "step": 12805 + }, + { + "epoch": 0.9624229670825192, + "grad_norm": 1.9633834599743574, + "learning_rate": 1.4795179870945141e-08, + "loss": 0.9122, + "step": 12806 + }, + { + "epoch": 0.9624981211483541, + "grad_norm": 1.6145148560143812, + "learning_rate": 1.4736131041504173e-08, + "loss": 0.9032, + "step": 12807 + }, + { + "epoch": 0.9625732752141891, + "grad_norm": 0.6914540505992559, + "learning_rate": 1.4677199846549581e-08, + "loss": 0.8583, + "step": 12808 + }, + { + "epoch": 0.962648429280024, + "grad_norm": 1.8537276601956498, + "learning_rate": 1.4618386289573237e-08, + "loss": 0.9727, + "step": 12809 + }, + { + "epoch": 0.9627235833458591, + "grad_norm": 1.9283098871008566, + "learning_rate": 1.4559690374059907e-08, + "loss": 0.979, + "step": 12810 + }, + { + "epoch": 0.962798737411694, + "grad_norm": 1.5742174626574683, + "learning_rate": 1.45011121034877e-08, + "loss": 0.9714, + "step": 12811 + }, + { + "epoch": 0.9628738914775289, + "grad_norm": 1.7448335405325863, + "learning_rate": 1.4442651481327839e-08, + "loss": 0.9762, + "step": 12812 + }, + { + "epoch": 0.9629490455433639, + "grad_norm": 2.1131217177105213, + "learning_rate": 1.4384308511043775e-08, + "loss": 0.973, + "step": 12813 + }, + { + "epoch": 0.9630241996091988, + "grad_norm": 1.63832519646943, + "learning_rate": 1.4326083196092963e-08, + "loss": 0.8325, + "step": 12814 + }, + { + "epoch": 0.9630993536750339, + "grad_norm": 1.4363157120853776, + "learning_rate": 1.4267975539925536e-08, + "loss": 1.0478, + "step": 12815 + }, + { + "epoch": 0.9631745077408688, + "grad_norm": 1.501704995403223, + "learning_rate": 1.4209985545984294e-08, + "loss": 1.0244, + "step": 12816 + }, + { + "epoch": 0.9632496618067038, + "grad_norm": 2.822216458535838, + "learning_rate": 1.4152113217705375e-08, + "loss": 0.9046, + "step": 12817 + }, + { + "epoch": 0.9633248158725387, + "grad_norm": 1.5924580891602176, + "learning_rate": 1.409435855851826e-08, + "loss": 0.9703, + "step": 12818 + }, + { + "epoch": 0.9633999699383736, + "grad_norm": 1.368741705185652, + "learning_rate": 1.4036721571844879e-08, + "loss": 0.9512, + "step": 12819 + }, + { + "epoch": 0.9634751240042086, + "grad_norm": 1.7722007707648177, + "learning_rate": 1.3979202261100497e-08, + "loss": 0.9329, + "step": 12820 + }, + { + "epoch": 0.9635502780700436, + "grad_norm": 1.617059380067959, + "learning_rate": 1.392180062969328e-08, + "loss": 0.8745, + "step": 12821 + }, + { + "epoch": 0.9636254321358786, + "grad_norm": 1.4690990439070866, + "learning_rate": 1.386451668102473e-08, + "loss": 1.0646, + "step": 12822 + }, + { + "epoch": 0.9637005862017135, + "grad_norm": 2.5308278991654274, + "learning_rate": 1.3807350418488795e-08, + "loss": 0.9861, + "step": 12823 + }, + { + "epoch": 0.9637757402675484, + "grad_norm": 2.751176062443569, + "learning_rate": 1.3750301845473211e-08, + "loss": 1.0003, + "step": 12824 + }, + { + "epoch": 0.9638508943333834, + "grad_norm": 1.607030178843036, + "learning_rate": 1.3693370965357942e-08, + "loss": 0.7703, + "step": 12825 + }, + { + "epoch": 0.9639260483992184, + "grad_norm": 1.8366807447725868, + "learning_rate": 1.3636557781516733e-08, + "loss": 0.9318, + "step": 12826 + }, + { + "epoch": 0.9640012024650534, + "grad_norm": 2.024561860522246, + "learning_rate": 1.3579862297315558e-08, + "loss": 0.9306, + "step": 12827 + }, + { + "epoch": 0.9640763565308883, + "grad_norm": 2.0266622526137485, + "learning_rate": 1.3523284516113953e-08, + "loss": 0.9875, + "step": 12828 + }, + { + "epoch": 0.9641515105967233, + "grad_norm": 1.7764767722563481, + "learning_rate": 1.3466824441264791e-08, + "loss": 0.9774, + "step": 12829 + }, + { + "epoch": 0.9642266646625582, + "grad_norm": 1.690909720259444, + "learning_rate": 1.341048207611295e-08, + "loss": 0.9611, + "step": 12830 + }, + { + "epoch": 0.9643018187283932, + "grad_norm": 2.583457472084803, + "learning_rate": 1.3354257423997318e-08, + "loss": 1.0212, + "step": 12831 + }, + { + "epoch": 0.9643769727942282, + "grad_norm": 1.3435174607738356, + "learning_rate": 1.3298150488249449e-08, + "loss": 0.9923, + "step": 12832 + }, + { + "epoch": 0.9644521268600631, + "grad_norm": 1.8213982352500702, + "learning_rate": 1.3242161272193575e-08, + "loss": 0.8575, + "step": 12833 + }, + { + "epoch": 0.9645272809258981, + "grad_norm": 1.9719553693785266, + "learning_rate": 1.3186289779147264e-08, + "loss": 0.8603, + "step": 12834 + }, + { + "epoch": 0.964602434991733, + "grad_norm": 1.7556632232579135, + "learning_rate": 1.3130536012421645e-08, + "loss": 1.0297, + "step": 12835 + }, + { + "epoch": 0.9646775890575681, + "grad_norm": 1.5017806451425848, + "learning_rate": 1.3074899975319853e-08, + "loss": 1.0079, + "step": 12836 + }, + { + "epoch": 0.964752743123403, + "grad_norm": 1.9156026443745173, + "learning_rate": 1.3019381671138806e-08, + "loss": 0.9869, + "step": 12837 + }, + { + "epoch": 0.9648278971892379, + "grad_norm": 2.280120883715214, + "learning_rate": 1.2963981103167875e-08, + "loss": 0.9094, + "step": 12838 + }, + { + "epoch": 0.9649030512550729, + "grad_norm": 2.0312107962713175, + "learning_rate": 1.2908698274689989e-08, + "loss": 0.9854, + "step": 12839 + }, + { + "epoch": 0.9649782053209078, + "grad_norm": 0.6442908098664571, + "learning_rate": 1.2853533188980747e-08, + "loss": 0.8164, + "step": 12840 + }, + { + "epoch": 0.9650533593867429, + "grad_norm": 2.0447977022874486, + "learning_rate": 1.2798485849309092e-08, + "loss": 0.9749, + "step": 12841 + }, + { + "epoch": 0.9651285134525778, + "grad_norm": 1.4854087843647925, + "learning_rate": 1.2743556258936639e-08, + "loss": 0.9018, + "step": 12842 + }, + { + "epoch": 0.9652036675184128, + "grad_norm": 3.0671267689518946, + "learning_rate": 1.2688744421118115e-08, + "loss": 0.9007, + "step": 12843 + }, + { + "epoch": 0.9652788215842477, + "grad_norm": 2.2168579630900185, + "learning_rate": 1.2634050339101366e-08, + "loss": 0.9771, + "step": 12844 + }, + { + "epoch": 0.9653539756500826, + "grad_norm": 1.761804413668755, + "learning_rate": 1.2579474016127355e-08, + "loss": 0.9789, + "step": 12845 + }, + { + "epoch": 0.9654291297159177, + "grad_norm": 2.436969992896272, + "learning_rate": 1.2525015455429943e-08, + "loss": 0.9617, + "step": 12846 + }, + { + "epoch": 0.9655042837817526, + "grad_norm": 1.6370167608876967, + "learning_rate": 1.247067466023588e-08, + "loss": 0.8342, + "step": 12847 + }, + { + "epoch": 0.9655794378475876, + "grad_norm": 1.856329017781188, + "learning_rate": 1.2416451633764813e-08, + "loss": 0.9168, + "step": 12848 + }, + { + "epoch": 0.9656545919134225, + "grad_norm": 2.0415086360676793, + "learning_rate": 1.2362346379230393e-08, + "loss": 0.9483, + "step": 12849 + }, + { + "epoch": 0.9657297459792574, + "grad_norm": 1.7075192519927567, + "learning_rate": 1.2308358899837833e-08, + "loss": 0.8534, + "step": 12850 + }, + { + "epoch": 0.9658049000450925, + "grad_norm": 1.8008399672461808, + "learning_rate": 1.2254489198786354e-08, + "loss": 0.9382, + "step": 12851 + }, + { + "epoch": 0.9658800541109274, + "grad_norm": 2.3928267369239737, + "learning_rate": 1.220073727926807e-08, + "loss": 0.9766, + "step": 12852 + }, + { + "epoch": 0.9659552081767624, + "grad_norm": 1.8826309053214298, + "learning_rate": 1.2147103144467985e-08, + "loss": 0.9586, + "step": 12853 + }, + { + "epoch": 0.9660303622425973, + "grad_norm": 2.0507536968015123, + "learning_rate": 1.2093586797564004e-08, + "loss": 0.8786, + "step": 12854 + }, + { + "epoch": 0.9661055163084323, + "grad_norm": 2.388342771527031, + "learning_rate": 1.2040188241726924e-08, + "loss": 0.9487, + "step": 12855 + }, + { + "epoch": 0.9661806703742672, + "grad_norm": 1.988803997157922, + "learning_rate": 1.1986907480121545e-08, + "loss": 1.0236, + "step": 12856 + }, + { + "epoch": 0.9662558244401022, + "grad_norm": 1.7110711311166953, + "learning_rate": 1.1933744515904232e-08, + "loss": 0.9052, + "step": 12857 + }, + { + "epoch": 0.9663309785059372, + "grad_norm": 2.249225695149815, + "learning_rate": 1.1880699352225354e-08, + "loss": 0.9826, + "step": 12858 + }, + { + "epoch": 0.9664061325717721, + "grad_norm": 1.6278254081403223, + "learning_rate": 1.1827771992228175e-08, + "loss": 1.0868, + "step": 12859 + }, + { + "epoch": 0.9664812866376071, + "grad_norm": 1.744856223055568, + "learning_rate": 1.177496243904863e-08, + "loss": 1.0178, + "step": 12860 + }, + { + "epoch": 0.966556440703442, + "grad_norm": 1.7242499006021612, + "learning_rate": 1.1722270695815994e-08, + "loss": 0.9332, + "step": 12861 + }, + { + "epoch": 0.9666315947692771, + "grad_norm": 1.9075185588368804, + "learning_rate": 1.1669696765652659e-08, + "loss": 1.0551, + "step": 12862 + }, + { + "epoch": 0.966706748835112, + "grad_norm": 1.6847811094333565, + "learning_rate": 1.1617240651673243e-08, + "loss": 0.8994, + "step": 12863 + }, + { + "epoch": 0.9667819029009469, + "grad_norm": 1.6841879944329377, + "learning_rate": 1.1564902356986595e-08, + "loss": 0.9939, + "step": 12864 + }, + { + "epoch": 0.9668570569667819, + "grad_norm": 1.5477306427038595, + "learning_rate": 1.1512681884693565e-08, + "loss": 0.9395, + "step": 12865 + }, + { + "epoch": 0.9669322110326168, + "grad_norm": 1.3763122798815044, + "learning_rate": 1.146057923788879e-08, + "loss": 1.0063, + "step": 12866 + }, + { + "epoch": 0.9670073650984519, + "grad_norm": 2.37153684193294, + "learning_rate": 1.1408594419659135e-08, + "loss": 1.0315, + "step": 12867 + }, + { + "epoch": 0.9670825191642868, + "grad_norm": 2.2539585047966644, + "learning_rate": 1.1356727433085245e-08, + "loss": 0.9641, + "step": 12868 + }, + { + "epoch": 0.9671576732301217, + "grad_norm": 3.5547742895912484, + "learning_rate": 1.1304978281239996e-08, + "loss": 0.9522, + "step": 12869 + }, + { + "epoch": 0.9672328272959567, + "grad_norm": 1.5554825173366984, + "learning_rate": 1.125334696719027e-08, + "loss": 0.9301, + "step": 12870 + }, + { + "epoch": 0.9673079813617916, + "grad_norm": 1.6511125599495537, + "learning_rate": 1.120183349399495e-08, + "loss": 0.9541, + "step": 12871 + }, + { + "epoch": 0.9673831354276267, + "grad_norm": 1.521195547090042, + "learning_rate": 1.1150437864706708e-08, + "loss": 1.0033, + "step": 12872 + }, + { + "epoch": 0.9674582894934616, + "grad_norm": 1.747731577819111, + "learning_rate": 1.1099160082371106e-08, + "loss": 1.03, + "step": 12873 + }, + { + "epoch": 0.9675334435592966, + "grad_norm": 1.828067550408576, + "learning_rate": 1.1048000150025939e-08, + "loss": 0.9372, + "step": 12874 + }, + { + "epoch": 0.9676085976251315, + "grad_norm": 0.8394485608416595, + "learning_rate": 1.0996958070703e-08, + "loss": 0.856, + "step": 12875 + }, + { + "epoch": 0.9676837516909664, + "grad_norm": 1.5818219081691272, + "learning_rate": 1.0946033847426761e-08, + "loss": 1.0634, + "step": 12876 + }, + { + "epoch": 0.9677589057568015, + "grad_norm": 2.2530775215977052, + "learning_rate": 1.0895227483214587e-08, + "loss": 0.8952, + "step": 12877 + }, + { + "epoch": 0.9678340598226364, + "grad_norm": 2.5758896818469643, + "learning_rate": 1.0844538981076956e-08, + "loss": 0.9653, + "step": 12878 + }, + { + "epoch": 0.9679092138884714, + "grad_norm": 10.066363807147622, + "learning_rate": 1.0793968344017467e-08, + "loss": 0.9871, + "step": 12879 + }, + { + "epoch": 0.9679843679543063, + "grad_norm": 1.8567306564143196, + "learning_rate": 1.0743515575032392e-08, + "loss": 0.9309, + "step": 12880 + }, + { + "epoch": 0.9680595220201413, + "grad_norm": 2.898337802046288, + "learning_rate": 1.0693180677111557e-08, + "loss": 0.8936, + "step": 12881 + }, + { + "epoch": 0.9681346760859763, + "grad_norm": 1.8757561748310978, + "learning_rate": 1.0642963653237246e-08, + "loss": 0.9375, + "step": 12882 + }, + { + "epoch": 0.9682098301518112, + "grad_norm": 1.6117709204569108, + "learning_rate": 1.0592864506385079e-08, + "loss": 0.9352, + "step": 12883 + }, + { + "epoch": 0.9682849842176462, + "grad_norm": 1.6611948870722795, + "learning_rate": 1.054288323952357e-08, + "loss": 1.0642, + "step": 12884 + }, + { + "epoch": 0.9683601382834811, + "grad_norm": 1.6936684648659033, + "learning_rate": 1.0493019855614572e-08, + "loss": 0.959, + "step": 12885 + }, + { + "epoch": 0.9684352923493161, + "grad_norm": 2.3679737761823927, + "learning_rate": 1.0443274357612386e-08, + "loss": 1.0266, + "step": 12886 + }, + { + "epoch": 0.968510446415151, + "grad_norm": 2.0659610723057478, + "learning_rate": 1.0393646748464658e-08, + "loss": 0.8843, + "step": 12887 + }, + { + "epoch": 0.9685856004809861, + "grad_norm": 1.8229924068156778, + "learning_rate": 1.0344137031112143e-08, + "loss": 0.946, + "step": 12888 + }, + { + "epoch": 0.968660754546821, + "grad_norm": 2.199939778977581, + "learning_rate": 1.0294745208488276e-08, + "loss": 0.9309, + "step": 12889 + }, + { + "epoch": 0.9687359086126559, + "grad_norm": 1.4196321173413329, + "learning_rate": 1.0245471283520046e-08, + "loss": 1.0275, + "step": 12890 + }, + { + "epoch": 0.9688110626784909, + "grad_norm": 2.576069700399886, + "learning_rate": 1.0196315259126897e-08, + "loss": 0.9105, + "step": 12891 + }, + { + "epoch": 0.9688862167443258, + "grad_norm": 0.7839846623852121, + "learning_rate": 1.0147277138221388e-08, + "loss": 0.8886, + "step": 12892 + }, + { + "epoch": 0.9689613708101609, + "grad_norm": 2.083061378530665, + "learning_rate": 1.0098356923709417e-08, + "loss": 0.9795, + "step": 12893 + }, + { + "epoch": 0.9690365248759958, + "grad_norm": 1.647560935159819, + "learning_rate": 1.0049554618489552e-08, + "loss": 0.996, + "step": 12894 + }, + { + "epoch": 0.9691116789418307, + "grad_norm": 1.5705040202849923, + "learning_rate": 1.0000870225453705e-08, + "loss": 0.9693, + "step": 12895 + }, + { + "epoch": 0.9691868330076657, + "grad_norm": 1.5240094798386463, + "learning_rate": 9.952303747486678e-09, + "loss": 1.0025, + "step": 12896 + }, + { + "epoch": 0.9692619870735006, + "grad_norm": 1.9650918185886401, + "learning_rate": 9.903855187465948e-09, + "loss": 0.9215, + "step": 12897 + }, + { + "epoch": 0.9693371411393357, + "grad_norm": 1.6300260139214928, + "learning_rate": 9.855524548262106e-09, + "loss": 0.9803, + "step": 12898 + }, + { + "epoch": 0.9694122952051706, + "grad_norm": 1.4127731768006937, + "learning_rate": 9.807311832739529e-09, + "loss": 1.0064, + "step": 12899 + }, + { + "epoch": 0.9694874492710056, + "grad_norm": 1.9586060787219173, + "learning_rate": 9.759217043754597e-09, + "loss": 0.9707, + "step": 12900 + }, + { + "epoch": 0.9695626033368405, + "grad_norm": 2.090518564385878, + "learning_rate": 9.711240184157255e-09, + "loss": 0.8881, + "step": 12901 + }, + { + "epoch": 0.9696377574026754, + "grad_norm": 2.131376572595224, + "learning_rate": 9.663381256790116e-09, + "loss": 0.8773, + "step": 12902 + }, + { + "epoch": 0.9697129114685105, + "grad_norm": 2.1098393812108203, + "learning_rate": 9.615640264489134e-09, + "loss": 1.0111, + "step": 12903 + }, + { + "epoch": 0.9697880655343454, + "grad_norm": 1.617853147465089, + "learning_rate": 9.568017210083379e-09, + "loss": 1.0075, + "step": 12904 + }, + { + "epoch": 0.9698632196001804, + "grad_norm": 1.5504920363087913, + "learning_rate": 9.52051209639415e-09, + "loss": 0.8931, + "step": 12905 + }, + { + "epoch": 0.9699383736660153, + "grad_norm": 3.0594378362637804, + "learning_rate": 9.473124926236975e-09, + "loss": 0.9889, + "step": 12906 + }, + { + "epoch": 0.9700135277318503, + "grad_norm": 1.5965597778464358, + "learning_rate": 9.42585570241916e-09, + "loss": 0.801, + "step": 12907 + }, + { + "epoch": 0.9700886817976853, + "grad_norm": 2.345232685314317, + "learning_rate": 9.378704427742024e-09, + "loss": 1.0239, + "step": 12908 + }, + { + "epoch": 0.9701638358635202, + "grad_norm": 1.5796244079487307, + "learning_rate": 9.331671104998884e-09, + "loss": 0.9432, + "step": 12909 + }, + { + "epoch": 0.9702389899293552, + "grad_norm": 2.5039871845465855, + "learning_rate": 9.284755736977513e-09, + "loss": 0.9707, + "step": 12910 + }, + { + "epoch": 0.9703141439951901, + "grad_norm": 1.6335322057878723, + "learning_rate": 9.237958326457018e-09, + "loss": 1.0578, + "step": 12911 + }, + { + "epoch": 0.9703892980610251, + "grad_norm": 1.7034435849176428, + "learning_rate": 9.191278876210518e-09, + "loss": 1.0173, + "step": 12912 + }, + { + "epoch": 0.97046445212686, + "grad_norm": 1.755406814657411, + "learning_rate": 9.144717389004241e-09, + "loss": 0.9761, + "step": 12913 + }, + { + "epoch": 0.970539606192695, + "grad_norm": 2.0016238058023874, + "learning_rate": 9.098273867596873e-09, + "loss": 0.9197, + "step": 12914 + }, + { + "epoch": 0.97061476025853, + "grad_norm": 1.8268054209981845, + "learning_rate": 9.051948314740432e-09, + "loss": 0.916, + "step": 12915 + }, + { + "epoch": 0.9706899143243649, + "grad_norm": 1.830786048001882, + "learning_rate": 9.005740733180055e-09, + "loss": 0.8914, + "step": 12916 + }, + { + "epoch": 0.9707650683901999, + "grad_norm": 1.9635819313784597, + "learning_rate": 8.959651125653556e-09, + "loss": 0.8131, + "step": 12917 + }, + { + "epoch": 0.9708402224560349, + "grad_norm": 3.3274506794679906, + "learning_rate": 8.913679494891857e-09, + "loss": 1.0276, + "step": 12918 + }, + { + "epoch": 0.9709153765218699, + "grad_norm": 1.3979854172667767, + "learning_rate": 8.867825843618782e-09, + "loss": 0.9737, + "step": 12919 + }, + { + "epoch": 0.9709905305877048, + "grad_norm": 2.3283242774009945, + "learning_rate": 8.822090174551933e-09, + "loss": 1.1261, + "step": 12920 + }, + { + "epoch": 0.9710656846535397, + "grad_norm": 0.7791998092892001, + "learning_rate": 8.776472490400922e-09, + "loss": 0.8415, + "step": 12921 + }, + { + "epoch": 0.9711408387193747, + "grad_norm": 1.8268769397392506, + "learning_rate": 8.730972793868696e-09, + "loss": 1.0368, + "step": 12922 + }, + { + "epoch": 0.9712159927852096, + "grad_norm": 1.6840496817755932, + "learning_rate": 8.685591087651323e-09, + "loss": 0.9357, + "step": 12923 + }, + { + "epoch": 0.9712911468510447, + "grad_norm": 0.7298031798977413, + "learning_rate": 8.640327374438205e-09, + "loss": 0.8495, + "step": 12924 + }, + { + "epoch": 0.9713663009168796, + "grad_norm": 3.0846301436094037, + "learning_rate": 8.595181656910978e-09, + "loss": 0.9347, + "step": 12925 + }, + { + "epoch": 0.9714414549827146, + "grad_norm": 1.626426144325326, + "learning_rate": 8.55015393774483e-09, + "loss": 0.8294, + "step": 12926 + }, + { + "epoch": 0.9715166090485495, + "grad_norm": 1.5144329298954151, + "learning_rate": 8.505244219607854e-09, + "loss": 0.8663, + "step": 12927 + }, + { + "epoch": 0.9715917631143844, + "grad_norm": 1.861220461762804, + "learning_rate": 8.460452505161031e-09, + "loss": 1.0627, + "step": 12928 + }, + { + "epoch": 0.9716669171802195, + "grad_norm": 2.1040479292904, + "learning_rate": 8.415778797058681e-09, + "loss": 1.0001, + "step": 12929 + }, + { + "epoch": 0.9717420712460544, + "grad_norm": 1.7925564825929845, + "learning_rate": 8.371223097947356e-09, + "loss": 0.9502, + "step": 12930 + }, + { + "epoch": 0.9718172253118894, + "grad_norm": 2.097440399487107, + "learning_rate": 8.326785410468052e-09, + "loss": 1.0301, + "step": 12931 + }, + { + "epoch": 0.9718923793777243, + "grad_norm": 1.3308956193929102, + "learning_rate": 8.282465737252887e-09, + "loss": 0.9953, + "step": 12932 + }, + { + "epoch": 0.9719675334435594, + "grad_norm": 1.9726547996591204, + "learning_rate": 8.238264080928647e-09, + "loss": 0.9936, + "step": 12933 + }, + { + "epoch": 0.9720426875093943, + "grad_norm": 0.6817068597918209, + "learning_rate": 8.19418044411413e-09, + "loss": 0.8491, + "step": 12934 + }, + { + "epoch": 0.9721178415752292, + "grad_norm": 1.3839584171575179, + "learning_rate": 8.150214829421687e-09, + "loss": 0.9243, + "step": 12935 + }, + { + "epoch": 0.9721929956410642, + "grad_norm": 1.756576103262288, + "learning_rate": 8.106367239456124e-09, + "loss": 0.9425, + "step": 12936 + }, + { + "epoch": 0.9722681497068991, + "grad_norm": 3.9719309637057307, + "learning_rate": 8.062637676816031e-09, + "loss": 1.0559, + "step": 12937 + }, + { + "epoch": 0.9723433037727341, + "grad_norm": 3.4789386301441954, + "learning_rate": 8.019026144092001e-09, + "loss": 0.9456, + "step": 12938 + }, + { + "epoch": 0.9724184578385691, + "grad_norm": 0.7857209257984732, + "learning_rate": 7.975532643868632e-09, + "loss": 0.8704, + "step": 12939 + }, + { + "epoch": 0.972493611904404, + "grad_norm": 2.1333789562274266, + "learning_rate": 7.932157178722976e-09, + "loss": 0.979, + "step": 12940 + }, + { + "epoch": 0.972568765970239, + "grad_norm": 1.6571267164787893, + "learning_rate": 7.888899751224976e-09, + "loss": 0.9071, + "step": 12941 + }, + { + "epoch": 0.9726439200360739, + "grad_norm": 2.686796584226235, + "learning_rate": 7.845760363938136e-09, + "loss": 0.8923, + "step": 12942 + }, + { + "epoch": 0.9727190741019089, + "grad_norm": 1.708678264995796, + "learning_rate": 7.802739019418192e-09, + "loss": 0.9688, + "step": 12943 + }, + { + "epoch": 0.9727942281677439, + "grad_norm": 1.5630564652419137, + "learning_rate": 7.75983572021488e-09, + "loss": 0.9138, + "step": 12944 + }, + { + "epoch": 0.9728693822335789, + "grad_norm": 1.744979263772697, + "learning_rate": 7.717050468870168e-09, + "loss": 1.007, + "step": 12945 + }, + { + "epoch": 0.9729445362994138, + "grad_norm": 2.1423157189614868, + "learning_rate": 7.674383267918916e-09, + "loss": 0.992, + "step": 12946 + }, + { + "epoch": 0.9730196903652487, + "grad_norm": 0.7070997215541809, + "learning_rate": 7.631834119889768e-09, + "loss": 0.8264, + "step": 12947 + }, + { + "epoch": 0.9730948444310837, + "grad_norm": 2.661143294847474, + "learning_rate": 7.589403027303598e-09, + "loss": 0.9757, + "step": 12948 + }, + { + "epoch": 0.9731699984969187, + "grad_norm": 1.7125955847356606, + "learning_rate": 7.547089992674838e-09, + "loss": 0.9387, + "step": 12949 + }, + { + "epoch": 0.9732451525627537, + "grad_norm": 1.7969690712892545, + "learning_rate": 7.504895018510593e-09, + "loss": 0.9157, + "step": 12950 + }, + { + "epoch": 0.9733203066285886, + "grad_norm": 1.8226561296563533, + "learning_rate": 7.462818107311086e-09, + "loss": 1.0313, + "step": 12951 + }, + { + "epoch": 0.9733954606944236, + "grad_norm": 1.2925513017164811, + "learning_rate": 7.420859261569434e-09, + "loss": 0.9612, + "step": 12952 + }, + { + "epoch": 0.9734706147602585, + "grad_norm": 1.8709107630055162, + "learning_rate": 7.379018483772092e-09, + "loss": 1.0489, + "step": 12953 + }, + { + "epoch": 0.9735457688260934, + "grad_norm": 1.627148748133823, + "learning_rate": 7.337295776398189e-09, + "loss": 1.0114, + "step": 12954 + }, + { + "epoch": 0.9736209228919285, + "grad_norm": 2.287833342731452, + "learning_rate": 7.295691141919746e-09, + "loss": 1.0182, + "step": 12955 + }, + { + "epoch": 0.9736960769577634, + "grad_norm": 2.06178144017783, + "learning_rate": 7.254204582802348e-09, + "loss": 0.9045, + "step": 12956 + }, + { + "epoch": 0.9737712310235984, + "grad_norm": 1.8304225471321656, + "learning_rate": 7.2128361015040274e-09, + "loss": 1.0184, + "step": 12957 + }, + { + "epoch": 0.9738463850894333, + "grad_norm": 1.5957500399506312, + "learning_rate": 7.171585700476157e-09, + "loss": 0.9122, + "step": 12958 + }, + { + "epoch": 0.9739215391552682, + "grad_norm": 1.7867642095376264, + "learning_rate": 7.13045338216256e-09, + "loss": 0.9862, + "step": 12959 + }, + { + "epoch": 0.9739966932211033, + "grad_norm": 1.3235069770095864, + "learning_rate": 7.0894391490010644e-09, + "loss": 0.9008, + "step": 12960 + }, + { + "epoch": 0.9740718472869382, + "grad_norm": 2.217895343206701, + "learning_rate": 7.048543003421725e-09, + "loss": 0.9064, + "step": 12961 + }, + { + "epoch": 0.9741470013527732, + "grad_norm": 2.106333369294557, + "learning_rate": 7.007764947847494e-09, + "loss": 0.871, + "step": 12962 + }, + { + "epoch": 0.9742221554186081, + "grad_norm": 3.482220723605515, + "learning_rate": 6.967104984695105e-09, + "loss": 1.0382, + "step": 12963 + }, + { + "epoch": 0.9742973094844432, + "grad_norm": 1.7597040468337957, + "learning_rate": 6.9265631163735186e-09, + "loss": 0.9533, + "step": 12964 + }, + { + "epoch": 0.9743724635502781, + "grad_norm": 1.6639887117314756, + "learning_rate": 6.8861393452848134e-09, + "loss": 0.9861, + "step": 12965 + }, + { + "epoch": 0.974447617616113, + "grad_norm": 2.243832825218509, + "learning_rate": 6.84583367382463e-09, + "loss": 1.0242, + "step": 12966 + }, + { + "epoch": 0.974522771681948, + "grad_norm": 2.2643292108718467, + "learning_rate": 6.80564610438128e-09, + "loss": 0.9026, + "step": 12967 + }, + { + "epoch": 0.9745979257477829, + "grad_norm": 1.6045933028137356, + "learning_rate": 6.765576639335746e-09, + "loss": 0.9518, + "step": 12968 + }, + { + "epoch": 0.974673079813618, + "grad_norm": 1.740271982114166, + "learning_rate": 6.725625281062352e-09, + "loss": 0.9511, + "step": 12969 + }, + { + "epoch": 0.9747482338794529, + "grad_norm": 3.096526624252501, + "learning_rate": 6.6857920319283165e-09, + "loss": 0.9977, + "step": 12970 + }, + { + "epoch": 0.9748233879452879, + "grad_norm": 3.1125600786040937, + "learning_rate": 6.646076894294195e-09, + "loss": 0.9545, + "step": 12971 + }, + { + "epoch": 0.9748985420111228, + "grad_norm": 2.271500014220936, + "learning_rate": 6.606479870512993e-09, + "loss": 0.9764, + "step": 12972 + }, + { + "epoch": 0.9749736960769577, + "grad_norm": 2.2290348715161588, + "learning_rate": 6.5670009629312794e-09, + "loss": 1.0427, + "step": 12973 + }, + { + "epoch": 0.9750488501427927, + "grad_norm": 2.127444376055851, + "learning_rate": 6.5276401738878495e-09, + "loss": 1.0007, + "step": 12974 + }, + { + "epoch": 0.9751240042086277, + "grad_norm": 1.5557829098464189, + "learning_rate": 6.488397505715504e-09, + "loss": 0.979, + "step": 12975 + }, + { + "epoch": 0.9751991582744627, + "grad_norm": 1.8238204096379496, + "learning_rate": 6.44927296073905e-09, + "loss": 0.9824, + "step": 12976 + }, + { + "epoch": 0.9752743123402976, + "grad_norm": 2.084639356054666, + "learning_rate": 6.410266541277077e-09, + "loss": 0.8564, + "step": 12977 + }, + { + "epoch": 0.9753494664061326, + "grad_norm": 1.5516150704781178, + "learning_rate": 6.371378249640624e-09, + "loss": 0.9666, + "step": 12978 + }, + { + "epoch": 0.9754246204719675, + "grad_norm": 1.4966764186946393, + "learning_rate": 6.332608088134295e-09, + "loss": 0.9519, + "step": 12979 + }, + { + "epoch": 0.9754997745378025, + "grad_norm": 1.7364617930261312, + "learning_rate": 6.293956059055139e-09, + "loss": 1.018, + "step": 12980 + }, + { + "epoch": 0.9755749286036375, + "grad_norm": 1.7147950185051601, + "learning_rate": 6.255422164693547e-09, + "loss": 0.9561, + "step": 12981 + }, + { + "epoch": 0.9756500826694724, + "grad_norm": 1.953870585229345, + "learning_rate": 6.217006407332803e-09, + "loss": 0.9106, + "step": 12982 + }, + { + "epoch": 0.9757252367353074, + "grad_norm": 1.7461996366410253, + "learning_rate": 6.178708789248866e-09, + "loss": 0.9905, + "step": 12983 + }, + { + "epoch": 0.9758003908011423, + "grad_norm": 2.5810418476849097, + "learning_rate": 6.140529312711473e-09, + "loss": 0.9113, + "step": 12984 + }, + { + "epoch": 0.9758755448669773, + "grad_norm": 1.5901212332949075, + "learning_rate": 6.102467979982817e-09, + "loss": 0.9888, + "step": 12985 + }, + { + "epoch": 0.9759506989328123, + "grad_norm": 5.101090273535625, + "learning_rate": 6.064524793317982e-09, + "loss": 0.9537, + "step": 12986 + }, + { + "epoch": 0.9760258529986472, + "grad_norm": 1.8477927030537982, + "learning_rate": 6.026699754965392e-09, + "loss": 0.9769, + "step": 12987 + }, + { + "epoch": 0.9761010070644822, + "grad_norm": 0.7231714010559155, + "learning_rate": 5.988992867166143e-09, + "loss": 0.8642, + "step": 12988 + }, + { + "epoch": 0.9761761611303171, + "grad_norm": 2.6369614326975244, + "learning_rate": 5.951404132154669e-09, + "loss": 0.9443, + "step": 12989 + }, + { + "epoch": 0.9762513151961522, + "grad_norm": 1.4925309030729674, + "learning_rate": 5.9139335521583015e-09, + "loss": 0.9344, + "step": 12990 + }, + { + "epoch": 0.9763264692619871, + "grad_norm": 1.8099990559148698, + "learning_rate": 5.876581129397262e-09, + "loss": 0.9454, + "step": 12991 + }, + { + "epoch": 0.976401623327822, + "grad_norm": 2.2041328235280337, + "learning_rate": 5.839346866084893e-09, + "loss": 0.9176, + "step": 12992 + }, + { + "epoch": 0.976476777393657, + "grad_norm": 1.9071393674264394, + "learning_rate": 5.802230764426985e-09, + "loss": 0.9473, + "step": 12993 + }, + { + "epoch": 0.9765519314594919, + "grad_norm": 2.43022154798865, + "learning_rate": 5.765232826623556e-09, + "loss": 1.03, + "step": 12994 + }, + { + "epoch": 0.976627085525327, + "grad_norm": 1.5697226104363822, + "learning_rate": 5.728353054866408e-09, + "loss": 1.028, + "step": 12995 + }, + { + "epoch": 0.9767022395911619, + "grad_norm": 1.6541613959061112, + "learning_rate": 5.691591451340905e-09, + "loss": 0.9956, + "step": 12996 + }, + { + "epoch": 0.9767773936569969, + "grad_norm": 1.971929737425566, + "learning_rate": 5.654948018225303e-09, + "loss": 0.9745, + "step": 12997 + }, + { + "epoch": 0.9768525477228318, + "grad_norm": 1.7100571409795169, + "learning_rate": 5.6184227576909774e-09, + "loss": 0.9818, + "step": 12998 + }, + { + "epoch": 0.9769277017886667, + "grad_norm": 2.7359136121786527, + "learning_rate": 5.582015671901974e-09, + "loss": 0.999, + "step": 12999 + }, + { + "epoch": 0.9770028558545018, + "grad_norm": 1.7567289954944216, + "learning_rate": 5.5457267630159014e-09, + "loss": 0.9425, + "step": 13000 + }, + { + "epoch": 0.9770780099203367, + "grad_norm": 3.2104811146823358, + "learning_rate": 5.509556033182372e-09, + "loss": 1.0262, + "step": 13001 + }, + { + "epoch": 0.9771531639861717, + "grad_norm": 1.3941071022900786, + "learning_rate": 5.47350348454545e-09, + "loss": 0.9116, + "step": 13002 + }, + { + "epoch": 0.9772283180520066, + "grad_norm": 2.666376624625975, + "learning_rate": 5.437569119240981e-09, + "loss": 0.9701, + "step": 13003 + }, + { + "epoch": 0.9773034721178415, + "grad_norm": 0.8271993105758214, + "learning_rate": 5.40175293939793e-09, + "loss": 0.8665, + "step": 13004 + }, + { + "epoch": 0.9773786261836765, + "grad_norm": 1.4732260980173806, + "learning_rate": 5.3660549471392645e-09, + "loss": 1.0261, + "step": 13005 + }, + { + "epoch": 0.9774537802495115, + "grad_norm": 3.3550898701076584, + "learning_rate": 5.330475144579516e-09, + "loss": 0.8856, + "step": 13006 + }, + { + "epoch": 0.9775289343153465, + "grad_norm": 2.1685601055108847, + "learning_rate": 5.295013533827219e-09, + "loss": 0.7498, + "step": 13007 + }, + { + "epoch": 0.9776040883811814, + "grad_norm": 2.6753110152219173, + "learning_rate": 5.259670116983805e-09, + "loss": 0.9772, + "step": 13008 + }, + { + "epoch": 0.9776792424470164, + "grad_norm": 1.200076712699453, + "learning_rate": 5.224444896143154e-09, + "loss": 0.9141, + "step": 13009 + }, + { + "epoch": 0.9777543965128513, + "grad_norm": 2.4728499543067493, + "learning_rate": 5.189337873392485e-09, + "loss": 0.9762, + "step": 13010 + }, + { + "epoch": 0.9778295505786863, + "grad_norm": 1.6267510296619159, + "learning_rate": 5.1543490508123565e-09, + "loss": 0.8943, + "step": 13011 + }, + { + "epoch": 0.9779047046445213, + "grad_norm": 2.0901094356888392, + "learning_rate": 5.119478430475999e-09, + "loss": 0.957, + "step": 13012 + }, + { + "epoch": 0.9779798587103562, + "grad_norm": 2.473205312601449, + "learning_rate": 5.0847260144490926e-09, + "loss": 0.9566, + "step": 13013 + }, + { + "epoch": 0.9780550127761912, + "grad_norm": 3.1368648357268487, + "learning_rate": 5.0500918047915455e-09, + "loss": 0.9907, + "step": 13014 + }, + { + "epoch": 0.9781301668420261, + "grad_norm": 1.5822720568079849, + "learning_rate": 5.01557580355505e-09, + "loss": 1.066, + "step": 13015 + }, + { + "epoch": 0.9782053209078612, + "grad_norm": 1.5802413045133683, + "learning_rate": 4.981178012785081e-09, + "loss": 0.885, + "step": 13016 + }, + { + "epoch": 0.9782804749736961, + "grad_norm": 0.7895923006727829, + "learning_rate": 4.946898434519564e-09, + "loss": 0.7755, + "step": 13017 + }, + { + "epoch": 0.978355629039531, + "grad_norm": 2.539985371645157, + "learning_rate": 4.912737070789985e-09, + "loss": 0.9666, + "step": 13018 + }, + { + "epoch": 0.978430783105366, + "grad_norm": 2.217068424481804, + "learning_rate": 4.878693923620725e-09, + "loss": 0.8981, + "step": 13019 + }, + { + "epoch": 0.9785059371712009, + "grad_norm": 1.8217238819924604, + "learning_rate": 4.84476899502817e-09, + "loss": 0.9888, + "step": 13020 + }, + { + "epoch": 0.978581091237036, + "grad_norm": 2.04011561066144, + "learning_rate": 4.810962287023379e-09, + "loss": 0.9173, + "step": 13021 + }, + { + "epoch": 0.9786562453028709, + "grad_norm": 1.6571518943568386, + "learning_rate": 4.777273801608972e-09, + "loss": 1.0572, + "step": 13022 + }, + { + "epoch": 0.9787313993687059, + "grad_norm": 2.185042935959961, + "learning_rate": 4.743703540781574e-09, + "loss": 1.0715, + "step": 13023 + }, + { + "epoch": 0.9788065534345408, + "grad_norm": 2.8445979939514094, + "learning_rate": 4.710251506529816e-09, + "loss": 0.968, + "step": 13024 + }, + { + "epoch": 0.9788817075003757, + "grad_norm": 2.960552260971153, + "learning_rate": 4.6769177008363355e-09, + "loss": 1.0057, + "step": 13025 + }, + { + "epoch": 0.9789568615662108, + "grad_norm": 1.5523875490948897, + "learning_rate": 4.643702125675775e-09, + "loss": 1.0059, + "step": 13026 + }, + { + "epoch": 0.9790320156320457, + "grad_norm": 1.9627796654155738, + "learning_rate": 4.610604783016781e-09, + "loss": 0.8934, + "step": 13027 + }, + { + "epoch": 0.9791071696978807, + "grad_norm": 2.587019403025822, + "learning_rate": 4.577625674820451e-09, + "loss": 0.8982, + "step": 13028 + }, + { + "epoch": 0.9791823237637156, + "grad_norm": 1.8982526901147683, + "learning_rate": 4.544764803040557e-09, + "loss": 0.9604, + "step": 13029 + }, + { + "epoch": 0.9792574778295505, + "grad_norm": 1.7723934780766206, + "learning_rate": 4.512022169624652e-09, + "loss": 1.0214, + "step": 13030 + }, + { + "epoch": 0.9793326318953856, + "grad_norm": 1.766285325525231, + "learning_rate": 4.479397776512517e-09, + "loss": 0.8709, + "step": 13031 + }, + { + "epoch": 0.9794077859612205, + "grad_norm": 1.762581759781702, + "learning_rate": 4.446891625637495e-09, + "loss": 0.9065, + "step": 13032 + }, + { + "epoch": 0.9794829400270555, + "grad_norm": 1.8577036678946808, + "learning_rate": 4.4145037189255995e-09, + "loss": 0.8744, + "step": 13033 + }, + { + "epoch": 0.9795580940928904, + "grad_norm": 1.771905110959462, + "learning_rate": 4.382234058295964e-09, + "loss": 0.9939, + "step": 13034 + }, + { + "epoch": 0.9796332481587254, + "grad_norm": 3.737963624094503, + "learning_rate": 4.350082645660613e-09, + "loss": 0.8389, + "step": 13035 + }, + { + "epoch": 0.9797084022245603, + "grad_norm": 2.0045606351508396, + "learning_rate": 4.318049482924913e-09, + "loss": 0.9085, + "step": 13036 + }, + { + "epoch": 0.9797835562903953, + "grad_norm": 0.6889583552194904, + "learning_rate": 4.286134571986455e-09, + "loss": 0.8284, + "step": 13037 + }, + { + "epoch": 0.9798587103562303, + "grad_norm": 2.4084153774062633, + "learning_rate": 4.254337914736839e-09, + "loss": 0.9806, + "step": 13038 + }, + { + "epoch": 0.9799338644220652, + "grad_norm": 1.7113377351192804, + "learning_rate": 4.22265951305989e-09, + "loss": 0.9638, + "step": 13039 + }, + { + "epoch": 0.9800090184879002, + "grad_norm": 2.1054117906328886, + "learning_rate": 4.191099368832774e-09, + "loss": 0.9271, + "step": 13040 + }, + { + "epoch": 0.9800841725537351, + "grad_norm": 3.200755959863171, + "learning_rate": 4.159657483925328e-09, + "loss": 0.9394, + "step": 13041 + }, + { + "epoch": 0.9801593266195702, + "grad_norm": 5.59795592331783, + "learning_rate": 4.12833386020095e-09, + "loss": 0.8732, + "step": 13042 + }, + { + "epoch": 0.9802344806854051, + "grad_norm": 2.062291105846875, + "learning_rate": 4.097128499515268e-09, + "loss": 1.0284, + "step": 13043 + }, + { + "epoch": 0.98030963475124, + "grad_norm": 1.9013774697173187, + "learning_rate": 4.06604140371769e-09, + "loss": 0.9314, + "step": 13044 + }, + { + "epoch": 0.980384788817075, + "grad_norm": 13.144434899349694, + "learning_rate": 4.035072574650078e-09, + "loss": 1.0398, + "step": 13045 + }, + { + "epoch": 0.9804599428829099, + "grad_norm": 2.1855216753532143, + "learning_rate": 4.004222014147629e-09, + "loss": 0.9833, + "step": 13046 + }, + { + "epoch": 0.980535096948745, + "grad_norm": 2.9719492488855965, + "learning_rate": 3.973489724037993e-09, + "loss": 1.0186, + "step": 13047 + }, + { + "epoch": 0.9806102510145799, + "grad_norm": 2.0165657391264253, + "learning_rate": 3.942875706142379e-09, + "loss": 0.9709, + "step": 13048 + }, + { + "epoch": 0.9806854050804148, + "grad_norm": 1.5724979205337626, + "learning_rate": 3.912379962274892e-09, + "loss": 0.9355, + "step": 13049 + }, + { + "epoch": 0.9807605591462498, + "grad_norm": 1.6688918121316239, + "learning_rate": 3.882002494242309e-09, + "loss": 0.9634, + "step": 13050 + }, + { + "epoch": 0.9808357132120847, + "grad_norm": 1.8112796753110283, + "learning_rate": 3.8517433038449675e-09, + "loss": 0.9897, + "step": 13051 + }, + { + "epoch": 0.9809108672779198, + "grad_norm": 1.6329373398653464, + "learning_rate": 3.821602392875434e-09, + "loss": 1.0439, + "step": 13052 + }, + { + "epoch": 0.9809860213437547, + "grad_norm": 2.5041731336597928, + "learning_rate": 3.791579763119834e-09, + "loss": 0.8802, + "step": 13053 + }, + { + "epoch": 0.9810611754095897, + "grad_norm": 2.1592088384226207, + "learning_rate": 3.761675416356969e-09, + "loss": 0.9351, + "step": 13054 + }, + { + "epoch": 0.9811363294754246, + "grad_norm": 1.7757843218943403, + "learning_rate": 3.7318893543591966e-09, + "loss": 0.9815, + "step": 13055 + }, + { + "epoch": 0.9812114835412595, + "grad_norm": 2.046766205437064, + "learning_rate": 3.702221578891107e-09, + "loss": 0.7816, + "step": 13056 + }, + { + "epoch": 0.9812866376070946, + "grad_norm": 3.468483442082503, + "learning_rate": 3.6726720917106268e-09, + "loss": 0.8579, + "step": 13057 + }, + { + "epoch": 0.9813617916729295, + "grad_norm": 2.5597131973781813, + "learning_rate": 3.643240894569022e-09, + "loss": 0.9243, + "step": 13058 + }, + { + "epoch": 0.9814369457387645, + "grad_norm": 2.86321613399056, + "learning_rate": 3.613927989209786e-09, + "loss": 0.849, + "step": 13059 + }, + { + "epoch": 0.9815120998045994, + "grad_norm": 3.199873301858875, + "learning_rate": 3.584733377369975e-09, + "loss": 0.9289, + "step": 13060 + }, + { + "epoch": 0.9815872538704344, + "grad_norm": 2.6436468586743778, + "learning_rate": 3.5556570607795377e-09, + "loss": 0.9295, + "step": 13061 + }, + { + "epoch": 0.9816624079362694, + "grad_norm": 1.9658577882744932, + "learning_rate": 3.5266990411613183e-09, + "loss": 0.9928, + "step": 13062 + }, + { + "epoch": 0.9817375620021043, + "grad_norm": 1.7996059728200693, + "learning_rate": 3.4978593202312777e-09, + "loss": 0.955, + "step": 13063 + }, + { + "epoch": 0.9818127160679393, + "grad_norm": 1.617147325394517, + "learning_rate": 3.4691378996980493e-09, + "loss": 1.0201, + "step": 13064 + }, + { + "epoch": 0.9818878701337742, + "grad_norm": 1.3227727416236947, + "learning_rate": 3.440534781263604e-09, + "loss": 0.9528, + "step": 13065 + }, + { + "epoch": 0.9819630241996092, + "grad_norm": 2.2429431877951695, + "learning_rate": 3.41204996662281e-09, + "loss": 0.9843, + "step": 13066 + }, + { + "epoch": 0.9820381782654442, + "grad_norm": 7.831130861196583, + "learning_rate": 3.3836834574636485e-09, + "loss": 0.9586, + "step": 13067 + }, + { + "epoch": 0.9821133323312792, + "grad_norm": 2.4123257312191897, + "learning_rate": 3.3554352554665545e-09, + "loss": 0.9929, + "step": 13068 + }, + { + "epoch": 0.9821884863971141, + "grad_norm": 1.6847062473460475, + "learning_rate": 3.3273053623059655e-09, + "loss": 0.9714, + "step": 13069 + }, + { + "epoch": 0.982263640462949, + "grad_norm": 1.4724352440576514, + "learning_rate": 3.2992937796478824e-09, + "loss": 0.9356, + "step": 13070 + }, + { + "epoch": 0.982338794528784, + "grad_norm": 4.155028428991496, + "learning_rate": 3.2714005091527554e-09, + "loss": 0.9012, + "step": 13071 + }, + { + "epoch": 0.982413948594619, + "grad_norm": 2.149776083353131, + "learning_rate": 3.2436255524730394e-09, + "loss": 0.9653, + "step": 13072 + }, + { + "epoch": 0.982489102660454, + "grad_norm": 2.33793039694397, + "learning_rate": 3.215968911254752e-09, + "loss": 0.9342, + "step": 13073 + }, + { + "epoch": 0.9825642567262889, + "grad_norm": 1.6921000478307706, + "learning_rate": 3.1884305871363593e-09, + "loss": 1.0212, + "step": 13074 + }, + { + "epoch": 0.9826394107921238, + "grad_norm": 0.7343974414399609, + "learning_rate": 3.16101058174989e-09, + "loss": 0.8418, + "step": 13075 + }, + { + "epoch": 0.9827145648579588, + "grad_norm": 1.710835911073274, + "learning_rate": 3.1337088967198223e-09, + "loss": 0.9942, + "step": 13076 + }, + { + "epoch": 0.9827897189237937, + "grad_norm": 0.6550928085765125, + "learning_rate": 3.1065255336639727e-09, + "loss": 0.8452, + "step": 13077 + }, + { + "epoch": 0.9828648729896288, + "grad_norm": 1.6327129863000829, + "learning_rate": 3.0794604941932754e-09, + "loss": 1.0146, + "step": 13078 + }, + { + "epoch": 0.9829400270554637, + "grad_norm": 0.70028813596778, + "learning_rate": 3.0525137799111146e-09, + "loss": 0.8619, + "step": 13079 + }, + { + "epoch": 0.9830151811212987, + "grad_norm": 4.785387430640086, + "learning_rate": 3.0256853924144344e-09, + "loss": 0.9846, + "step": 13080 + }, + { + "epoch": 0.9830903351871336, + "grad_norm": 2.098968152448406, + "learning_rate": 2.9989753332928526e-09, + "loss": 1.0976, + "step": 13081 + }, + { + "epoch": 0.9831654892529685, + "grad_norm": 2.1742806045116994, + "learning_rate": 2.9723836041288806e-09, + "loss": 0.941, + "step": 13082 + }, + { + "epoch": 0.9832406433188036, + "grad_norm": 10.089756792840543, + "learning_rate": 2.94591020649837e-09, + "loss": 0.9873, + "step": 13083 + }, + { + "epoch": 0.9833157973846385, + "grad_norm": 2.20937796542604, + "learning_rate": 2.9195551419698426e-09, + "loss": 1.0493, + "step": 13084 + }, + { + "epoch": 0.9833909514504735, + "grad_norm": 1.5396735048364512, + "learning_rate": 2.8933184121051613e-09, + "loss": 0.9757, + "step": 13085 + }, + { + "epoch": 0.9834661055163084, + "grad_norm": 2.071827920462227, + "learning_rate": 2.8672000184586377e-09, + "loss": 0.8478, + "step": 13086 + }, + { + "epoch": 0.9835412595821434, + "grad_norm": 1.7234874977475083, + "learning_rate": 2.841199962578145e-09, + "loss": 0.9738, + "step": 13087 + }, + { + "epoch": 0.9836164136479784, + "grad_norm": 1.6300919788831367, + "learning_rate": 2.815318246004006e-09, + "loss": 1.016, + "step": 13088 + }, + { + "epoch": 0.9836915677138133, + "grad_norm": 1.9028444982310673, + "learning_rate": 2.7895548702703277e-09, + "loss": 0.8936, + "step": 13089 + }, + { + "epoch": 0.9837667217796483, + "grad_norm": 2.6272377285421644, + "learning_rate": 2.7639098369032222e-09, + "loss": 0.8229, + "step": 13090 + }, + { + "epoch": 0.9838418758454832, + "grad_norm": 1.8417034752505959, + "learning_rate": 2.738383147422141e-09, + "loss": 0.9694, + "step": 13091 + }, + { + "epoch": 0.9839170299113182, + "grad_norm": 2.496786436341224, + "learning_rate": 2.7129748033400956e-09, + "loss": 1.0101, + "step": 13092 + }, + { + "epoch": 0.9839921839771532, + "grad_norm": 1.6569173655926803, + "learning_rate": 2.687684806162549e-09, + "loss": 0.998, + "step": 13093 + }, + { + "epoch": 0.9840673380429881, + "grad_norm": 2.77714115053065, + "learning_rate": 2.662513157387636e-09, + "loss": 1.0242, + "step": 13094 + }, + { + "epoch": 0.9841424921088231, + "grad_norm": 1.5817273260100952, + "learning_rate": 2.637459858507274e-09, + "loss": 0.9023, + "step": 13095 + }, + { + "epoch": 0.984217646174658, + "grad_norm": 1.7515682958801368, + "learning_rate": 2.612524911005831e-09, + "loss": 0.8996, + "step": 13096 + }, + { + "epoch": 0.984292800240493, + "grad_norm": 2.230924275876441, + "learning_rate": 2.5877083163607927e-09, + "loss": 0.938, + "step": 13097 + }, + { + "epoch": 0.984367954306328, + "grad_norm": 2.245100302547696, + "learning_rate": 2.5630100760425378e-09, + "loss": 0.992, + "step": 13098 + }, + { + "epoch": 0.984443108372163, + "grad_norm": 3.5693647934025106, + "learning_rate": 2.5384301915145624e-09, + "loss": 1.01, + "step": 13099 + }, + { + "epoch": 0.9845182624379979, + "grad_norm": 1.6618513720648613, + "learning_rate": 2.513968664233701e-09, + "loss": 0.998, + "step": 13100 + }, + { + "epoch": 0.9845934165038328, + "grad_norm": 1.7700775179079238, + "learning_rate": 2.489625495648795e-09, + "loss": 1.056, + "step": 13101 + }, + { + "epoch": 0.9846685705696678, + "grad_norm": 2.0051069622095286, + "learning_rate": 2.46540068720269e-09, + "loss": 0.9521, + "step": 13102 + }, + { + "epoch": 0.9847437246355027, + "grad_norm": 2.1909729772352504, + "learning_rate": 2.4412942403306826e-09, + "loss": 0.9818, + "step": 13103 + }, + { + "epoch": 0.9848188787013378, + "grad_norm": 1.7881088713712219, + "learning_rate": 2.4173061564609632e-09, + "loss": 0.9564, + "step": 13104 + }, + { + "epoch": 0.9848940327671727, + "grad_norm": 1.5749363174885158, + "learning_rate": 2.3934364370152836e-09, + "loss": 1.0405, + "step": 13105 + }, + { + "epoch": 0.9849691868330077, + "grad_norm": 1.5341556773015752, + "learning_rate": 2.3696850834078463e-09, + "loss": 0.9759, + "step": 13106 + }, + { + "epoch": 0.9850443408988426, + "grad_norm": 1.7188775448725158, + "learning_rate": 2.3460520970459697e-09, + "loss": 1.0329, + "step": 13107 + }, + { + "epoch": 0.9851194949646775, + "grad_norm": 2.488851676484861, + "learning_rate": 2.322537479330089e-09, + "loss": 1.0054, + "step": 13108 + }, + { + "epoch": 0.9851946490305126, + "grad_norm": 1.8677560108380666, + "learning_rate": 2.2991412316533122e-09, + "loss": 0.8991, + "step": 13109 + }, + { + "epoch": 0.9852698030963475, + "grad_norm": 2.1261367001585834, + "learning_rate": 2.2758633554023078e-09, + "loss": 0.9764, + "step": 13110 + }, + { + "epoch": 0.9853449571621825, + "grad_norm": 1.3684537418897347, + "learning_rate": 2.2527038519561948e-09, + "loss": 1.0337, + "step": 13111 + }, + { + "epoch": 0.9854201112280174, + "grad_norm": 1.4412250353196931, + "learning_rate": 2.2296627226872088e-09, + "loss": 0.9574, + "step": 13112 + }, + { + "epoch": 0.9854952652938525, + "grad_norm": 2.8427247934690776, + "learning_rate": 2.2067399689607024e-09, + "loss": 1.0001, + "step": 13113 + }, + { + "epoch": 0.9855704193596874, + "grad_norm": 2.419869657111653, + "learning_rate": 2.1839355921349224e-09, + "loss": 0.8651, + "step": 13114 + }, + { + "epoch": 0.9856455734255223, + "grad_norm": 1.8817845621108842, + "learning_rate": 2.16124959356101e-09, + "loss": 0.9709, + "step": 13115 + }, + { + "epoch": 0.9857207274913573, + "grad_norm": 1.515406012691246, + "learning_rate": 2.138681974583223e-09, + "loss": 0.9654, + "step": 13116 + }, + { + "epoch": 0.9857958815571922, + "grad_norm": 2.0287668161876344, + "learning_rate": 2.1162327365391587e-09, + "loss": 0.9774, + "step": 13117 + }, + { + "epoch": 0.9858710356230272, + "grad_norm": 2.2654368355321965, + "learning_rate": 2.0939018807584196e-09, + "loss": 0.9864, + "step": 13118 + }, + { + "epoch": 0.9859461896888622, + "grad_norm": 0.8362944010421127, + "learning_rate": 2.071689408564614e-09, + "loss": 0.8332, + "step": 13119 + }, + { + "epoch": 0.9860213437546971, + "grad_norm": 0.6854845980535411, + "learning_rate": 2.0495953212738005e-09, + "loss": 0.8001, + "step": 13120 + }, + { + "epoch": 0.9860964978205321, + "grad_norm": 0.764741231810082, + "learning_rate": 2.0276196201951535e-09, + "loss": 0.8585, + "step": 13121 + }, + { + "epoch": 0.986171651886367, + "grad_norm": 1.4835946692232376, + "learning_rate": 2.005762306630743e-09, + "loss": 0.9472, + "step": 13122 + }, + { + "epoch": 0.986246805952202, + "grad_norm": 1.9520248977019143, + "learning_rate": 1.9840233818757546e-09, + "loss": 1.0033, + "step": 13123 + }, + { + "epoch": 0.986321960018037, + "grad_norm": 1.5414285819615399, + "learning_rate": 1.9624028472182696e-09, + "loss": 1.0573, + "step": 13124 + }, + { + "epoch": 0.986397114083872, + "grad_norm": 1.7899588792222687, + "learning_rate": 1.9409007039392632e-09, + "loss": 1.0235, + "step": 13125 + }, + { + "epoch": 0.9864722681497069, + "grad_norm": 1.5593084164395912, + "learning_rate": 1.9195169533132714e-09, + "loss": 0.9869, + "step": 13126 + }, + { + "epoch": 0.9865474222155418, + "grad_norm": 1.640863383095324, + "learning_rate": 1.8982515966068367e-09, + "loss": 0.8549, + "step": 13127 + }, + { + "epoch": 0.9866225762813768, + "grad_norm": 2.318954121532687, + "learning_rate": 1.8771046350805063e-09, + "loss": 1.0392, + "step": 13128 + }, + { + "epoch": 0.9866977303472118, + "grad_norm": 1.6483638078027998, + "learning_rate": 1.856076069986834e-09, + "loss": 0.938, + "step": 13129 + }, + { + "epoch": 0.9867728844130468, + "grad_norm": 3.87812260974254, + "learning_rate": 1.8351659025721555e-09, + "loss": 0.9192, + "step": 13130 + }, + { + "epoch": 0.9868480384788817, + "grad_norm": 1.7795745599426682, + "learning_rate": 1.8143741340752583e-09, + "loss": 1.0813, + "step": 13131 + }, + { + "epoch": 0.9869231925447167, + "grad_norm": 1.6827923135069132, + "learning_rate": 1.7937007657282677e-09, + "loss": 0.8994, + "step": 13132 + }, + { + "epoch": 0.9869983466105516, + "grad_norm": 1.4900642987913209, + "learning_rate": 1.7731457987562038e-09, + "loss": 0.9904, + "step": 13133 + }, + { + "epoch": 0.9870735006763866, + "grad_norm": 1.6564595071971318, + "learning_rate": 1.752709234376981e-09, + "loss": 1.0882, + "step": 13134 + }, + { + "epoch": 0.9871486547422216, + "grad_norm": 1.439300901731072, + "learning_rate": 1.732391073801409e-09, + "loss": 1.0466, + "step": 13135 + }, + { + "epoch": 0.9872238088080565, + "grad_norm": 1.3497932628997598, + "learning_rate": 1.7121913182336356e-09, + "loss": 0.8767, + "step": 13136 + }, + { + "epoch": 0.9872989628738915, + "grad_norm": 1.8929602332716673, + "learning_rate": 1.692109968870703e-09, + "loss": 0.9195, + "step": 13137 + }, + { + "epoch": 0.9873741169397264, + "grad_norm": 0.7909078642358306, + "learning_rate": 1.6721470269021042e-09, + "loss": 0.8791, + "step": 13138 + }, + { + "epoch": 0.9874492710055613, + "grad_norm": 3.923823628629517, + "learning_rate": 1.6523024935108931e-09, + "loss": 0.9573, + "step": 13139 + }, + { + "epoch": 0.9875244250713964, + "grad_norm": 1.8085508403260515, + "learning_rate": 1.6325763698727957e-09, + "loss": 0.9805, + "step": 13140 + }, + { + "epoch": 0.9875995791372313, + "grad_norm": 1.9155953633690201, + "learning_rate": 1.6129686571570988e-09, + "loss": 1.0132, + "step": 13141 + }, + { + "epoch": 0.9876747332030663, + "grad_norm": 12.61889198529425, + "learning_rate": 1.593479356525096e-09, + "loss": 1.0152, + "step": 13142 + }, + { + "epoch": 0.9877498872689012, + "grad_norm": 2.3890723877546876, + "learning_rate": 1.5741084691318628e-09, + "loss": 1.0028, + "step": 13143 + }, + { + "epoch": 0.9878250413347363, + "grad_norm": 1.6100666078567152, + "learning_rate": 1.5548559961253705e-09, + "loss": 0.8636, + "step": 13144 + }, + { + "epoch": 0.9879001954005712, + "grad_norm": 1.6344036763521916, + "learning_rate": 1.5357219386460397e-09, + "loss": 1.0063, + "step": 13145 + }, + { + "epoch": 0.9879753494664061, + "grad_norm": 1.925363119977193, + "learning_rate": 1.5167062978278521e-09, + "loss": 0.9335, + "step": 13146 + }, + { + "epoch": 0.9880505035322411, + "grad_norm": 2.5511388367391628, + "learning_rate": 1.4978090747976846e-09, + "loss": 0.9486, + "step": 13147 + }, + { + "epoch": 0.988125657598076, + "grad_norm": 1.8154887189291373, + "learning_rate": 1.4790302706750856e-09, + "loss": 1.0726, + "step": 13148 + }, + { + "epoch": 0.988200811663911, + "grad_norm": 2.038963812416671, + "learning_rate": 1.4603698865724989e-09, + "loss": 0.9543, + "step": 13149 + }, + { + "epoch": 0.988275965729746, + "grad_norm": 1.581538747727078, + "learning_rate": 1.4418279235961506e-09, + "loss": 0.8643, + "step": 13150 + }, + { + "epoch": 0.988351119795581, + "grad_norm": 12.48551406724759, + "learning_rate": 1.4234043828444952e-09, + "loss": 0.9698, + "step": 13151 + }, + { + "epoch": 0.9884262738614159, + "grad_norm": 1.6817165450659657, + "learning_rate": 1.4050992654091043e-09, + "loss": 0.9999, + "step": 13152 + }, + { + "epoch": 0.9885014279272508, + "grad_norm": 2.0027862452826093, + "learning_rate": 1.3869125723746655e-09, + "loss": 0.9427, + "step": 13153 + }, + { + "epoch": 0.9885765819930858, + "grad_norm": 2.295670381591296, + "learning_rate": 1.3688443048189836e-09, + "loss": 1.0023, + "step": 13154 + }, + { + "epoch": 0.9886517360589208, + "grad_norm": 4.381760958971005, + "learning_rate": 1.3508944638125353e-09, + "loss": 0.9647, + "step": 13155 + }, + { + "epoch": 0.9887268901247558, + "grad_norm": 2.036600080339179, + "learning_rate": 1.3330630504189143e-09, + "loss": 0.9287, + "step": 13156 + }, + { + "epoch": 0.9888020441905907, + "grad_norm": 2.7368840504435896, + "learning_rate": 1.3153500656948313e-09, + "loss": 0.9302, + "step": 13157 + }, + { + "epoch": 0.9888771982564257, + "grad_norm": 2.7234635107272314, + "learning_rate": 1.2977555106894467e-09, + "loss": 0.9212, + "step": 13158 + }, + { + "epoch": 0.9889523523222606, + "grad_norm": 3.3543225996524337, + "learning_rate": 1.280279386445704e-09, + "loss": 0.8376, + "step": 13159 + }, + { + "epoch": 0.9890275063880956, + "grad_norm": 1.90173035431693, + "learning_rate": 1.2629216939992194e-09, + "loss": 1.0031, + "step": 13160 + }, + { + "epoch": 0.9891026604539306, + "grad_norm": 1.703791076881461, + "learning_rate": 1.2456824343780592e-09, + "loss": 1.0461, + "step": 13161 + }, + { + "epoch": 0.9891778145197655, + "grad_norm": 7.503447948084876, + "learning_rate": 1.2285616086040728e-09, + "loss": 1.1096, + "step": 13162 + }, + { + "epoch": 0.9892529685856005, + "grad_norm": 1.4838377130769353, + "learning_rate": 1.2115592176915601e-09, + "loss": 0.8485, + "step": 13163 + }, + { + "epoch": 0.9893281226514354, + "grad_norm": 1.838785113966017, + "learning_rate": 1.1946752626481594e-09, + "loss": 0.9686, + "step": 13164 + }, + { + "epoch": 0.9894032767172704, + "grad_norm": 1.9171253222243692, + "learning_rate": 1.1779097444739594e-09, + "loss": 0.9183, + "step": 13165 + }, + { + "epoch": 0.9894784307831054, + "grad_norm": 1.8611406550110567, + "learning_rate": 1.161262664162832e-09, + "loss": 1.0118, + "step": 13166 + }, + { + "epoch": 0.9895535848489403, + "grad_norm": 1.592947963786649, + "learning_rate": 1.1447340227008772e-09, + "loss": 1.0029, + "step": 13167 + }, + { + "epoch": 0.9896287389147753, + "grad_norm": 2.524761691036819, + "learning_rate": 1.1283238210675338e-09, + "loss": 0.9943, + "step": 13168 + }, + { + "epoch": 0.9897038929806102, + "grad_norm": 3.072112897507072, + "learning_rate": 1.1120320602351352e-09, + "loss": 1.0888, + "step": 13169 + }, + { + "epoch": 0.9897790470464453, + "grad_norm": 1.989037570362633, + "learning_rate": 1.095858741169131e-09, + "loss": 0.9407, + "step": 13170 + }, + { + "epoch": 0.9898542011122802, + "grad_norm": 1.5814455808526464, + "learning_rate": 1.0798038648278663e-09, + "loss": 0.9837, + "step": 13171 + }, + { + "epoch": 0.9899293551781151, + "grad_norm": 0.6559046336059631, + "learning_rate": 1.0638674321625796e-09, + "loss": 0.8396, + "step": 13172 + }, + { + "epoch": 0.9900045092439501, + "grad_norm": 1.7299093543543942, + "learning_rate": 1.0480494441174047e-09, + "loss": 0.9428, + "step": 13173 + }, + { + "epoch": 0.990079663309785, + "grad_norm": 1.8937190566351856, + "learning_rate": 1.0323499016300364e-09, + "loss": 1.0712, + "step": 13174 + }, + { + "epoch": 0.99015481737562, + "grad_norm": 2.0263411848858763, + "learning_rate": 1.016768805630397e-09, + "loss": 0.9568, + "step": 13175 + }, + { + "epoch": 0.990229971441455, + "grad_norm": 1.5504431367969203, + "learning_rate": 1.00130615704197e-09, + "loss": 0.8992, + "step": 13176 + }, + { + "epoch": 0.99030512550729, + "grad_norm": 1.5202734539353768, + "learning_rate": 9.859619567806898e-10, + "loss": 1.0677, + "step": 13177 + }, + { + "epoch": 0.9903802795731249, + "grad_norm": 2.1569261458287277, + "learning_rate": 9.707362057558289e-10, + "loss": 1.0131, + "step": 13178 + }, + { + "epoch": 0.9904554336389598, + "grad_norm": 2.571035266119492, + "learning_rate": 9.556289048697763e-10, + "loss": 0.9622, + "step": 13179 + }, + { + "epoch": 0.9905305877047949, + "grad_norm": 1.4575933032000281, + "learning_rate": 9.40640055017594e-10, + "loss": 0.8997, + "step": 13180 + }, + { + "epoch": 0.9906057417706298, + "grad_norm": 3.6739563582418375, + "learning_rate": 9.257696570872386e-10, + "loss": 0.873, + "step": 13181 + }, + { + "epoch": 0.9906808958364648, + "grad_norm": 1.604919562418661, + "learning_rate": 9.110177119600048e-10, + "loss": 0.8915, + "step": 13182 + }, + { + "epoch": 0.9907560499022997, + "grad_norm": 1.6410426017071207, + "learning_rate": 8.963842205100824e-10, + "loss": 1.0338, + "step": 13183 + }, + { + "epoch": 0.9908312039681346, + "grad_norm": 1.5863321499185195, + "learning_rate": 8.818691836045556e-10, + "loss": 0.9798, + "step": 13184 + }, + { + "epoch": 0.9909063580339696, + "grad_norm": 2.340775700646541, + "learning_rate": 8.674726021034028e-10, + "loss": 0.8615, + "step": 13185 + }, + { + "epoch": 0.9909815120998046, + "grad_norm": 2.3913798418303602, + "learning_rate": 8.531944768594979e-10, + "loss": 1.0152, + "step": 13186 + }, + { + "epoch": 0.9910566661656396, + "grad_norm": 1.674252824530028, + "learning_rate": 8.390348087192745e-10, + "loss": 0.9972, + "step": 13187 + }, + { + "epoch": 0.9911318202314745, + "grad_norm": 2.0699387698861726, + "learning_rate": 8.249935985213952e-10, + "loss": 0.8792, + "step": 13188 + }, + { + "epoch": 0.9912069742973095, + "grad_norm": 5.028649078426041, + "learning_rate": 8.110708470980831e-10, + "loss": 0.8995, + "step": 13189 + }, + { + "epoch": 0.9912821283631444, + "grad_norm": 4.755815660091699, + "learning_rate": 7.972665552742342e-10, + "loss": 1.0601, + "step": 13190 + }, + { + "epoch": 0.9913572824289794, + "grad_norm": 1.5556129499852278, + "learning_rate": 7.835807238676384e-10, + "loss": 1.0287, + "step": 13191 + }, + { + "epoch": 0.9914324364948144, + "grad_norm": 1.6886519279775523, + "learning_rate": 7.700133536896469e-10, + "loss": 1.015, + "step": 13192 + }, + { + "epoch": 0.9915075905606493, + "grad_norm": 2.6086284917327034, + "learning_rate": 7.565644455436171e-10, + "loss": 0.8697, + "step": 13193 + }, + { + "epoch": 0.9915827446264843, + "grad_norm": 1.5511743157895208, + "learning_rate": 7.432340002269111e-10, + "loss": 0.9272, + "step": 13194 + }, + { + "epoch": 0.9916578986923192, + "grad_norm": 0.8732856918825357, + "learning_rate": 7.300220185293416e-10, + "loss": 0.813, + "step": 13195 + }, + { + "epoch": 0.9917330527581543, + "grad_norm": 1.7587320592815168, + "learning_rate": 7.169285012336157e-10, + "loss": 0.9959, + "step": 13196 + }, + { + "epoch": 0.9918082068239892, + "grad_norm": 1.8309068299677576, + "learning_rate": 7.039534491155574e-10, + "loss": 0.9597, + "step": 13197 + }, + { + "epoch": 0.9918833608898241, + "grad_norm": 1.7613680674477603, + "learning_rate": 6.910968629443292e-10, + "loss": 1.0424, + "step": 13198 + }, + { + "epoch": 0.9919585149556591, + "grad_norm": 1.870256654802753, + "learning_rate": 6.783587434813221e-10, + "loss": 0.9969, + "step": 13199 + }, + { + "epoch": 0.992033669021494, + "grad_norm": 2.32799637202344, + "learning_rate": 6.657390914814875e-10, + "loss": 0.9259, + "step": 13200 + }, + { + "epoch": 0.9921088230873291, + "grad_norm": 1.6284230931328667, + "learning_rate": 6.532379076924499e-10, + "loss": 1.0111, + "step": 13201 + }, + { + "epoch": 0.992183977153164, + "grad_norm": 1.867440306336321, + "learning_rate": 6.40855192855172e-10, + "loss": 1.0513, + "step": 13202 + }, + { + "epoch": 0.992259131218999, + "grad_norm": 3.7304204328513646, + "learning_rate": 6.285909477032892e-10, + "loss": 0.9418, + "step": 13203 + }, + { + "epoch": 0.9923342852848339, + "grad_norm": 2.7583998792137154, + "learning_rate": 6.164451729635534e-10, + "loss": 0.971, + "step": 13204 + }, + { + "epoch": 0.9924094393506688, + "grad_norm": 7.961200084604668, + "learning_rate": 6.044178693553892e-10, + "loss": 0.8636, + "step": 13205 + }, + { + "epoch": 0.9924845934165039, + "grad_norm": 1.5573240380978683, + "learning_rate": 5.925090375917818e-10, + "loss": 0.9105, + "step": 13206 + }, + { + "epoch": 0.9925597474823388, + "grad_norm": 1.6448634303718304, + "learning_rate": 5.807186783783891e-10, + "loss": 1.0354, + "step": 13207 + }, + { + "epoch": 0.9926349015481738, + "grad_norm": 2.12192391295652, + "learning_rate": 5.690467924135412e-10, + "loss": 0.9264, + "step": 13208 + }, + { + "epoch": 0.9927100556140087, + "grad_norm": 1.5240147988459432, + "learning_rate": 5.574933803891291e-10, + "loss": 1.0283, + "step": 13209 + }, + { + "epoch": 0.9927852096798436, + "grad_norm": 1.583914006148276, + "learning_rate": 5.460584429894944e-10, + "loss": 1.0124, + "step": 13210 + }, + { + "epoch": 0.9928603637456787, + "grad_norm": 1.8366940501964664, + "learning_rate": 5.34741980892317e-10, + "loss": 0.9277, + "step": 13211 + }, + { + "epoch": 0.9929355178115136, + "grad_norm": 1.6739544637176789, + "learning_rate": 5.235439947681719e-10, + "loss": 0.9966, + "step": 13212 + }, + { + "epoch": 0.9930106718773486, + "grad_norm": 2.8946286080855965, + "learning_rate": 5.124644852805282e-10, + "loss": 0.9146, + "step": 13213 + }, + { + "epoch": 0.9930858259431835, + "grad_norm": 1.8340142748328125, + "learning_rate": 5.015034530859719e-10, + "loss": 1.0218, + "step": 13214 + }, + { + "epoch": 0.9931609800090185, + "grad_norm": 2.1341428974004195, + "learning_rate": 4.906608988339833e-10, + "loss": 0.9112, + "step": 13215 + }, + { + "epoch": 0.9932361340748534, + "grad_norm": 1.8238504760967456, + "learning_rate": 4.799368231669376e-10, + "loss": 0.9106, + "step": 13216 + }, + { + "epoch": 0.9933112881406884, + "grad_norm": 2.95381261500092, + "learning_rate": 4.693312267201044e-10, + "loss": 1.0513, + "step": 13217 + }, + { + "epoch": 0.9933864422065234, + "grad_norm": 1.851501850148883, + "learning_rate": 4.5884411012231395e-10, + "loss": 0.9869, + "step": 13218 + }, + { + "epoch": 0.9934615962723583, + "grad_norm": 2.3690188534631726, + "learning_rate": 4.484754739948471e-10, + "loss": 0.8898, + "step": 13219 + }, + { + "epoch": 0.9935367503381933, + "grad_norm": 2.605486319281093, + "learning_rate": 4.382253189518792e-10, + "loss": 0.9671, + "step": 13220 + }, + { + "epoch": 0.9936119044040282, + "grad_norm": 2.1890772991238356, + "learning_rate": 4.2809364560070225e-10, + "loss": 0.9127, + "step": 13221 + }, + { + "epoch": 0.9936870584698633, + "grad_norm": 1.6234952855771585, + "learning_rate": 4.18080454542169e-10, + "loss": 0.9689, + "step": 13222 + }, + { + "epoch": 0.9937622125356982, + "grad_norm": 1.7823644130121357, + "learning_rate": 4.081857463691385e-10, + "loss": 0.98, + "step": 13223 + }, + { + "epoch": 0.9938373666015331, + "grad_norm": 1.6383372023036493, + "learning_rate": 3.9840952166803054e-10, + "loss": 0.9765, + "step": 13224 + }, + { + "epoch": 0.9939125206673681, + "grad_norm": 1.9108190342262343, + "learning_rate": 3.8875178101815955e-10, + "loss": 1.0246, + "step": 13225 + }, + { + "epoch": 0.993987674733203, + "grad_norm": 4.459966390166636, + "learning_rate": 3.792125249917344e-10, + "loss": 1.081, + "step": 13226 + }, + { + "epoch": 0.9940628287990381, + "grad_norm": 0.8476955914487029, + "learning_rate": 3.697917541540807e-10, + "loss": 0.9388, + "step": 13227 + }, + { + "epoch": 0.994137982864873, + "grad_norm": 1.4776414431079885, + "learning_rate": 3.604894690634186e-10, + "loss": 1.0031, + "step": 13228 + }, + { + "epoch": 0.9942131369307079, + "grad_norm": 1.963600812134193, + "learning_rate": 3.5130567027086277e-10, + "loss": 1.0255, + "step": 13229 + }, + { + "epoch": 0.9942882909965429, + "grad_norm": 1.8610137001406535, + "learning_rate": 3.4224035832042254e-10, + "loss": 0.9794, + "step": 13230 + }, + { + "epoch": 0.9943634450623778, + "grad_norm": 1.6575116084916406, + "learning_rate": 3.3329353374966783e-10, + "loss": 0.9955, + "step": 13231 + }, + { + "epoch": 0.9944385991282129, + "grad_norm": 2.56726217875228, + "learning_rate": 3.2446519708839713e-10, + "loss": 1.0072, + "step": 13232 + }, + { + "epoch": 0.9945137531940478, + "grad_norm": 2.0131458738773196, + "learning_rate": 3.1575534885996957e-10, + "loss": 0.8919, + "step": 13233 + }, + { + "epoch": 0.9945889072598828, + "grad_norm": 2.151678796089353, + "learning_rate": 3.071639895801947e-10, + "loss": 1.0396, + "step": 13234 + }, + { + "epoch": 0.9946640613257177, + "grad_norm": 3.635563422067345, + "learning_rate": 2.986911197582209e-10, + "loss": 0.8769, + "step": 13235 + }, + { + "epoch": 0.9947392153915526, + "grad_norm": 1.6142194850684746, + "learning_rate": 2.90336739896313e-10, + "loss": 0.8354, + "step": 13236 + }, + { + "epoch": 0.9948143694573877, + "grad_norm": 1.7308705984899742, + "learning_rate": 2.8210085048940844e-10, + "loss": 1.0044, + "step": 13237 + }, + { + "epoch": 0.9948895235232226, + "grad_norm": 2.1542654441553473, + "learning_rate": 2.7398345202533925e-10, + "loss": 1.096, + "step": 13238 + }, + { + "epoch": 0.9949646775890576, + "grad_norm": 1.2897662929149842, + "learning_rate": 2.65984544985276e-10, + "loss": 1.0457, + "step": 13239 + }, + { + "epoch": 0.9950398316548925, + "grad_norm": 0.7863050568291035, + "learning_rate": 2.58104129843062e-10, + "loss": 0.8846, + "step": 13240 + }, + { + "epoch": 0.9951149857207275, + "grad_norm": 1.7251441618899501, + "learning_rate": 2.503422070656569e-10, + "loss": 1.0527, + "step": 13241 + }, + { + "epoch": 0.9951901397865625, + "grad_norm": 1.60531853114609, + "learning_rate": 2.426987771131372e-10, + "loss": 0.9916, + "step": 13242 + }, + { + "epoch": 0.9952652938523974, + "grad_norm": 1.588816617924122, + "learning_rate": 2.3517384043825194e-10, + "loss": 1.0377, + "step": 13243 + }, + { + "epoch": 0.9953404479182324, + "grad_norm": 1.7814080017686176, + "learning_rate": 2.277673974868666e-10, + "loss": 1.0572, + "step": 13244 + }, + { + "epoch": 0.9954156019840673, + "grad_norm": 1.4213588742012448, + "learning_rate": 2.204794486979633e-10, + "loss": 1.0011, + "step": 13245 + }, + { + "epoch": 0.9954907560499023, + "grad_norm": 2.553697866840355, + "learning_rate": 2.133099945034189e-10, + "loss": 0.7746, + "step": 13246 + }, + { + "epoch": 0.9955659101157373, + "grad_norm": 0.740852681505554, + "learning_rate": 2.0625903532778266e-10, + "loss": 0.8438, + "step": 13247 + }, + { + "epoch": 0.9956410641815723, + "grad_norm": 15.658809612529469, + "learning_rate": 1.9932657158916455e-10, + "loss": 1.014, + "step": 13248 + }, + { + "epoch": 0.9957162182474072, + "grad_norm": 2.2837498175960773, + "learning_rate": 1.9251260369812506e-10, + "loss": 0.9158, + "step": 13249 + }, + { + "epoch": 0.9957913723132421, + "grad_norm": 2.5650325448869737, + "learning_rate": 1.8581713205834126e-10, + "loss": 0.9813, + "step": 13250 + }, + { + "epoch": 0.9958665263790771, + "grad_norm": 1.7474583151761298, + "learning_rate": 1.7924015706682893e-10, + "loss": 0.9444, + "step": 13251 + }, + { + "epoch": 0.995941680444912, + "grad_norm": 2.20299805620396, + "learning_rate": 1.7278167911327635e-10, + "loss": 0.9474, + "step": 13252 + }, + { + "epoch": 0.9960168345107471, + "grad_norm": 1.7149872947867302, + "learning_rate": 1.664416985800443e-10, + "loss": 0.8469, + "step": 13253 + }, + { + "epoch": 0.996091988576582, + "grad_norm": 1.3971309418856968, + "learning_rate": 1.6022021584327638e-10, + "loss": 0.9356, + "step": 13254 + }, + { + "epoch": 0.9961671426424169, + "grad_norm": 2.5193432168510137, + "learning_rate": 1.5411723127112253e-10, + "loss": 0.9583, + "step": 13255 + }, + { + "epoch": 0.9962422967082519, + "grad_norm": 1.8640909734759217, + "learning_rate": 1.4813274522551545e-10, + "loss": 0.9902, + "step": 13256 + }, + { + "epoch": 0.9963174507740868, + "grad_norm": 1.998608343413929, + "learning_rate": 1.4226675806106037e-10, + "loss": 1.0195, + "step": 13257 + }, + { + "epoch": 0.9963926048399219, + "grad_norm": 2.072577737930919, + "learning_rate": 1.3651927012503506e-10, + "loss": 0.9337, + "step": 13258 + }, + { + "epoch": 0.9964677589057568, + "grad_norm": 2.2223127532108293, + "learning_rate": 1.3089028175850004e-10, + "loss": 0.9116, + "step": 13259 + }, + { + "epoch": 0.9965429129715918, + "grad_norm": 1.6241599625980334, + "learning_rate": 1.2537979329474424e-10, + "loss": 0.8844, + "step": 13260 + }, + { + "epoch": 0.9966180670374267, + "grad_norm": 2.0913096767924175, + "learning_rate": 1.1998780505995122e-10, + "loss": 0.9589, + "step": 13261 + }, + { + "epoch": 0.9966932211032616, + "grad_norm": 1.7026237660148624, + "learning_rate": 1.1471431737430926e-10, + "loss": 1.0292, + "step": 13262 + }, + { + "epoch": 0.9967683751690967, + "grad_norm": 1.6225324382390072, + "learning_rate": 1.0955933054956901e-10, + "loss": 0.9003, + "step": 13263 + }, + { + "epoch": 0.9968435292349316, + "grad_norm": 1.6425443775098538, + "learning_rate": 1.0452284489170793e-10, + "loss": 1.0453, + "step": 13264 + }, + { + "epoch": 0.9969186833007666, + "grad_norm": 2.454665362515052, + "learning_rate": 9.960486069915396e-11, + "loss": 0.9607, + "step": 13265 + }, + { + "epoch": 0.9969938373666015, + "grad_norm": 1.498818886186342, + "learning_rate": 9.480537826278556e-11, + "loss": 0.9608, + "step": 13266 + }, + { + "epoch": 0.9970689914324365, + "grad_norm": 1.7686259330667786, + "learning_rate": 9.012439786770798e-11, + "loss": 0.9678, + "step": 13267 + }, + { + "epoch": 0.9971441454982715, + "grad_norm": 2.049655452709796, + "learning_rate": 8.55619197905888e-11, + "loss": 0.9533, + "step": 13268 + }, + { + "epoch": 0.9972192995641064, + "grad_norm": 2.0737391490318977, + "learning_rate": 8.111794430232244e-11, + "loss": 0.976, + "step": 13269 + }, + { + "epoch": 0.9972944536299414, + "grad_norm": 2.0916410614310723, + "learning_rate": 7.679247166603175e-11, + "loss": 1.0033, + "step": 13270 + }, + { + "epoch": 0.9973696076957763, + "grad_norm": 1.4525959323318214, + "learning_rate": 7.258550213795622e-11, + "loss": 1.0254, + "step": 13271 + }, + { + "epoch": 0.9974447617616113, + "grad_norm": 2.1751590845969044, + "learning_rate": 6.849703596722989e-11, + "loss": 0.9057, + "step": 13272 + }, + { + "epoch": 0.9975199158274463, + "grad_norm": 2.630223934517597, + "learning_rate": 6.452707339654751e-11, + "loss": 0.9741, + "step": 13273 + }, + { + "epoch": 0.9975950698932812, + "grad_norm": 2.0246431153960094, + "learning_rate": 6.067561466083227e-11, + "loss": 0.9602, + "step": 13274 + }, + { + "epoch": 0.9976702239591162, + "grad_norm": 1.9699037592272561, + "learning_rate": 5.694265998834602e-11, + "loss": 0.937, + "step": 13275 + }, + { + "epoch": 0.9977453780249511, + "grad_norm": 2.5742986779546566, + "learning_rate": 5.33282096002452e-11, + "loss": 0.9321, + "step": 13276 + }, + { + "epoch": 0.9978205320907861, + "grad_norm": 1.813767483270091, + "learning_rate": 4.9832263710802845e-11, + "loss": 0.9923, + "step": 13277 + }, + { + "epoch": 0.997895686156621, + "grad_norm": 1.70303372479756, + "learning_rate": 4.645482252718658e-11, + "loss": 0.9506, + "step": 13278 + }, + { + "epoch": 0.9979708402224561, + "grad_norm": 1.5302144655745764, + "learning_rate": 4.3195886249458577e-11, + "loss": 0.8785, + "step": 13279 + }, + { + "epoch": 0.998045994288291, + "grad_norm": 1.998827769548013, + "learning_rate": 4.0055455070575614e-11, + "loss": 0.8633, + "step": 13280 + }, + { + "epoch": 0.9981211483541259, + "grad_norm": 1.5067316638520105, + "learning_rate": 3.7033529177055156e-11, + "loss": 0.913, + "step": 13281 + }, + { + "epoch": 0.9981963024199609, + "grad_norm": 1.0794084621735995, + "learning_rate": 3.413010874742106e-11, + "loss": 0.7512, + "step": 13282 + }, + { + "epoch": 0.9982714564857958, + "grad_norm": 0.6478749328206973, + "learning_rate": 3.134519395397994e-11, + "loss": 0.8101, + "step": 13283 + }, + { + "epoch": 0.9983466105516309, + "grad_norm": 1.7880410689303303, + "learning_rate": 2.8678784961710945e-11, + "loss": 0.9021, + "step": 13284 + }, + { + "epoch": 0.9984217646174658, + "grad_norm": 3.7525933517240206, + "learning_rate": 2.6130881928709823e-11, + "loss": 0.9544, + "step": 13285 + }, + { + "epoch": 0.9984969186833008, + "grad_norm": 2.189883650036184, + "learning_rate": 2.370148500574487e-11, + "loss": 0.9871, + "step": 13286 + }, + { + "epoch": 0.9985720727491357, + "grad_norm": 1.63487711791715, + "learning_rate": 2.1390594337145073e-11, + "loss": 0.9036, + "step": 13287 + }, + { + "epoch": 0.9986472268149706, + "grad_norm": 2.35048217290834, + "learning_rate": 1.9198210059245822e-11, + "loss": 0.9384, + "step": 13288 + }, + { + "epoch": 0.9987223808808057, + "grad_norm": 1.7692627692597962, + "learning_rate": 1.7124332302609346e-11, + "loss": 0.8892, + "step": 13289 + }, + { + "epoch": 0.9987975349466406, + "grad_norm": 1.5441132369098676, + "learning_rate": 1.5168961189582218e-11, + "loss": 1.0211, + "step": 13290 + }, + { + "epoch": 0.9988726890124756, + "grad_norm": 4.600498147611491, + "learning_rate": 1.3332096836293772e-11, + "loss": 0.985, + "step": 13291 + }, + { + "epoch": 0.9989478430783105, + "grad_norm": 1.707908771450422, + "learning_rate": 1.161373935154586e-11, + "loss": 1.0056, + "step": 13292 + }, + { + "epoch": 0.9990229971441456, + "grad_norm": 1.506898672265236, + "learning_rate": 1.0013888837256957e-11, + "loss": 0.9441, + "step": 13293 + }, + { + "epoch": 0.9990981512099805, + "grad_norm": 1.4543309283822812, + "learning_rate": 8.532545388018064e-12, + "loss": 0.8533, + "step": 13294 + }, + { + "epoch": 0.9991733052758154, + "grad_norm": 1.6330084432183927, + "learning_rate": 7.169709091536802e-12, + "loss": 0.9464, + "step": 13295 + }, + { + "epoch": 0.9992484593416504, + "grad_norm": 2.9802660391930913, + "learning_rate": 5.9253800290814945e-12, + "loss": 0.9548, + "step": 13296 + }, + { + "epoch": 0.9993236134074853, + "grad_norm": 2.433911347215826, + "learning_rate": 4.799558273704818e-12, + "loss": 0.9727, + "step": 13297 + }, + { + "epoch": 0.9993987674733203, + "grad_norm": 5.023543052691606, + "learning_rate": 3.7922438926862866e-12, + "loss": 1.0654, + "step": 13298 + }, + { + "epoch": 0.9994739215391553, + "grad_norm": 1.4156280618596595, + "learning_rate": 2.9034369453118103e-12, + "loss": 0.963, + "step": 13299 + }, + { + "epoch": 0.9995490756049902, + "grad_norm": 4.42439182218481, + "learning_rate": 2.1331374846500495e-12, + "loss": 1.0374, + "step": 13300 + }, + { + "epoch": 0.9996242296708252, + "grad_norm": 1.7935080787130047, + "learning_rate": 1.481345555776059e-12, + "loss": 1.0212, + "step": 13301 + }, + { + "epoch": 0.9996993837366601, + "grad_norm": 1.9229200669654989, + "learning_rate": 9.480611977696894e-13, + "loss": 1.0815, + "step": 13302 + }, + { + "epoch": 0.9997745378024951, + "grad_norm": 1.8800301788829727, + "learning_rate": 5.332844421612748e-13, + "loss": 0.9347, + "step": 13303 + }, + { + "epoch": 0.9998496918683301, + "grad_norm": 1.9423610250671293, + "learning_rate": 2.3701531359776597e-13, + "loss": 1.0381, + "step": 13304 + }, + { + "epoch": 0.9999248459341651, + "grad_norm": 1.592331647852202, + "learning_rate": 5.925382917659761e-14, + "loss": 0.9552, + "step": 13305 + }, + { + "epoch": 1.0, + "grad_norm": 1.7559513666243929, + "learning_rate": 0.0, + "loss": 0.9136, + "step": 13306 + }, + { + "epoch": 1.0, + "step": 13306, + "total_flos": 1.3375485825840579e+18, + "train_loss": 0.9802125646721129, + "train_runtime": 189864.5773, + "train_samples_per_second": 3.504, + "train_steps_per_second": 0.07 + } + ], + "logging_steps": 1.0, + "max_steps": 13306, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 1109, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1.3375485825840579e+18, + "train_batch_size": 5, + "trial_name": null, + "trial_params": null +}