diff --git "a/sft_full/smoe/trainer_state.json" "b/sft_full/smoe/trainer_state.json" new file mode 100644--- /dev/null +++ "b/sft_full/smoe/trainer_state.json" @@ -0,0 +1,68791 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.66676687208778, + "eval_steps": 500, + "global_step": 8872, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 7.515406583496167e-05, + "grad_norm": 22.88061321157062, + "learning_rate": 0.0, + "loss": 1.996, + "step": 1 + }, + { + "epoch": 0.00015030813166992335, + "grad_norm": 19.826708972289754, + "learning_rate": 4.6275642631951835e-07, + "loss": 2.1064, + "step": 2 + }, + { + "epoch": 0.000225462197504885, + "grad_norm": 15.949310115615111, + "learning_rate": 7.334515826841693e-07, + "loss": 2.0213, + "step": 3 + }, + { + "epoch": 0.0003006162633398467, + "grad_norm": 17.254419617532506, + "learning_rate": 9.255128526390367e-07, + "loss": 1.9655, + "step": 4 + }, + { + "epoch": 0.00037577032917480833, + "grad_norm": 15.457248920472658, + "learning_rate": 1.0744871473609632e-06, + "loss": 2.0031, + "step": 5 + }, + { + "epoch": 0.00045092439500977, + "grad_norm": 16.858378496673676, + "learning_rate": 1.1962080090036876e-06, + "loss": 1.9809, + "step": 6 + }, + { + "epoch": 0.0005260784608447317, + "grad_norm": 31.694277427947267, + "learning_rate": 1.2991215311418868e-06, + "loss": 1.8592, + "step": 7 + }, + { + "epoch": 0.0006012325266796934, + "grad_norm": 13.69743091413353, + "learning_rate": 1.3882692789585548e-06, + "loss": 1.7705, + "step": 8 + }, + { + "epoch": 0.000676386592514655, + "grad_norm": 13.465153685329078, + "learning_rate": 1.4669031653683387e-06, + "loss": 1.9088, + "step": 9 + }, + { + "epoch": 0.0007515406583496167, + "grad_norm": 13.055181323881108, + "learning_rate": 1.5372435736804818e-06, + "loss": 1.7528, + "step": 10 + }, + { + "epoch": 0.0008266947241845784, + "grad_norm": 12.127818615686829, + "learning_rate": 1.6008742129373428e-06, + "loss": 1.3647, + "step": 11 + }, + { + "epoch": 0.00090184879001954, + "grad_norm": 10.993373261936917, + "learning_rate": 1.658964435323206e-06, + "loss": 1.8207, + "step": 12 + }, + { + "epoch": 0.0009770028558545017, + "grad_norm": 13.877426897480738, + "learning_rate": 1.7124022597777776e-06, + "loss": 1.8128, + "step": 13 + }, + { + "epoch": 0.0010521569216894633, + "grad_norm": 8.932788380564778, + "learning_rate": 1.761877957461405e-06, + "loss": 1.6142, + "step": 14 + }, + { + "epoch": 0.0011273109875244252, + "grad_norm": 9.289672728852167, + "learning_rate": 1.8079387300451324e-06, + "loss": 1.6661, + "step": 15 + }, + { + "epoch": 0.0012024650533593868, + "grad_norm": 8.82989516613983, + "learning_rate": 1.8510257052780734e-06, + "loss": 1.6166, + "step": 16 + }, + { + "epoch": 0.0012776191191943484, + "grad_norm": 8.988833387675328, + "learning_rate": 1.891499697130832e-06, + "loss": 1.6704, + "step": 17 + }, + { + "epoch": 0.00135277318502931, + "grad_norm": 8.125978851042971, + "learning_rate": 1.929659591687857e-06, + "loss": 1.6605, + "step": 18 + }, + { + "epoch": 0.0014279272508642717, + "grad_norm": 7.833439224284976, + "learning_rate": 1.9657557553855114e-06, + "loss": 1.5232, + "step": 19 + }, + { + "epoch": 0.0015030813166992333, + "grad_norm": 7.798469823740865, + "learning_rate": 2e-06, + "loss": 1.484, + "step": 20 + }, + { + "epoch": 0.0015782353825341952, + "grad_norm": 7.1636082828677266, + "learning_rate": 2.032573113826056e-06, + "loss": 1.4991, + "step": 21 + }, + { + "epoch": 0.0016533894483691568, + "grad_norm": 5.719685128888898, + "learning_rate": 2.063630639256861e-06, + "loss": 1.3558, + "step": 22 + }, + { + "epoch": 0.0017285435142041184, + "grad_norm": 6.102615246088916, + "learning_rate": 2.093307365019873e-06, + "loss": 1.4155, + "step": 23 + }, + { + "epoch": 0.00180369758003908, + "grad_norm": 5.150861664324695, + "learning_rate": 2.1217208616427245e-06, + "loss": 1.4968, + "step": 24 + }, + { + "epoch": 0.0018788516458740417, + "grad_norm": 4.75292537352768, + "learning_rate": 2.1489742947219264e-06, + "loss": 1.455, + "step": 25 + }, + { + "epoch": 0.0019540057117090033, + "grad_norm": 4.819329308347771, + "learning_rate": 2.175158686097296e-06, + "loss": 1.2833, + "step": 26 + }, + { + "epoch": 0.002029159777543965, + "grad_norm": 4.547677614875836, + "learning_rate": 2.200354748052508e-06, + "loss": 1.3188, + "step": 27 + }, + { + "epoch": 0.0021043138433789266, + "grad_norm": 4.578846116282114, + "learning_rate": 2.2246343837809235e-06, + "loss": 1.3518, + "step": 28 + }, + { + "epoch": 0.0021794679092138887, + "grad_norm": 4.52045876259439, + "learning_rate": 2.2480619244333726e-06, + "loss": 1.3212, + "step": 29 + }, + { + "epoch": 0.0022546219750488503, + "grad_norm": 3.890940948281366, + "learning_rate": 2.270695156364651e-06, + "loss": 1.2898, + "step": 30 + }, + { + "epoch": 0.002329776040883812, + "grad_norm": 4.229375949936742, + "learning_rate": 2.2925861798799734e-06, + "loss": 1.4416, + "step": 31 + }, + { + "epoch": 0.0024049301067187736, + "grad_norm": 3.9512215958623953, + "learning_rate": 2.3137821315975918e-06, + "loss": 1.3013, + "step": 32 + }, + { + "epoch": 0.002480084172553735, + "grad_norm": 3.4610769273904407, + "learning_rate": 2.334325795621512e-06, + "loss": 1.2535, + "step": 33 + }, + { + "epoch": 0.002555238238388697, + "grad_norm": 3.056081780482817, + "learning_rate": 2.3542561234503503e-06, + "loss": 1.2305, + "step": 34 + }, + { + "epoch": 0.0026303923042236585, + "grad_norm": 2.5069422652874107, + "learning_rate": 2.3736086785028504e-06, + "loss": 1.3483, + "step": 35 + }, + { + "epoch": 0.00270554637005862, + "grad_norm": 2.5250793406796737, + "learning_rate": 2.392416018007375e-06, + "loss": 1.2737, + "step": 36 + }, + { + "epoch": 0.0027807004358935817, + "grad_norm": 3.072687603877581, + "learning_rate": 2.41070802255664e-06, + "loss": 1.0224, + "step": 37 + }, + { + "epoch": 0.0028558545017285434, + "grad_norm": 2.4613035868269275, + "learning_rate": 2.4285121817050297e-06, + "loss": 1.2869, + "step": 38 + }, + { + "epoch": 0.002931008567563505, + "grad_norm": 2.282873229381562, + "learning_rate": 2.445853842461947e-06, + "loss": 1.2089, + "step": 39 + }, + { + "epoch": 0.0030061626333984666, + "grad_norm": 2.2374276602362007, + "learning_rate": 2.4627564263195183e-06, + "loss": 1.2453, + "step": 40 + }, + { + "epoch": 0.0030813166992334287, + "grad_norm": 2.9307206837041146, + "learning_rate": 2.4792416194780364e-06, + "loss": 1.1545, + "step": 41 + }, + { + "epoch": 0.0031564707650683903, + "grad_norm": 1.9274650470617238, + "learning_rate": 2.4953295401455745e-06, + "loss": 1.2268, + "step": 42 + }, + { + "epoch": 0.003231624830903352, + "grad_norm": 1.8669304451309339, + "learning_rate": 2.511038886149501e-06, + "loss": 1.2473, + "step": 43 + }, + { + "epoch": 0.0033067788967383136, + "grad_norm": 2.7033426715330497, + "learning_rate": 2.526387065576379e-06, + "loss": 1.194, + "step": 44 + }, + { + "epoch": 0.0033819329625732752, + "grad_norm": 3.5415724049908226, + "learning_rate": 2.5413903127293017e-06, + "loss": 1.0963, + "step": 45 + }, + { + "epoch": 0.003457087028408237, + "grad_norm": 1.9218388375702538, + "learning_rate": 2.5560637913393917e-06, + "loss": 1.2618, + "step": 46 + }, + { + "epoch": 0.0035322410942431985, + "grad_norm": 1.8048537536930378, + "learning_rate": 2.5704216866765804e-06, + "loss": 1.1582, + "step": 47 + }, + { + "epoch": 0.00360739516007816, + "grad_norm": 1.8532088159451503, + "learning_rate": 2.584477287962243e-06, + "loss": 1.2729, + "step": 48 + }, + { + "epoch": 0.0036825492259131218, + "grad_norm": 1.8793439137279488, + "learning_rate": 2.5982430622837735e-06, + "loss": 1.1633, + "step": 49 + }, + { + "epoch": 0.0037577032917480834, + "grad_norm": 1.8239196270710545, + "learning_rate": 2.6117307210414448e-06, + "loss": 1.1298, + "step": 50 + }, + { + "epoch": 0.003832857357583045, + "grad_norm": 3.145026164469115, + "learning_rate": 2.624951279815001e-06, + "loss": 1.1343, + "step": 51 + }, + { + "epoch": 0.003908011423418007, + "grad_norm": 2.4970831066963, + "learning_rate": 2.6379151124168143e-06, + "loss": 1.177, + "step": 52 + }, + { + "epoch": 0.003983165489252968, + "grad_norm": 2.0725798085564024, + "learning_rate": 2.650631999796137e-06, + "loss": 1.2133, + "step": 53 + }, + { + "epoch": 0.00405831955508793, + "grad_norm": 5.004296364269449, + "learning_rate": 2.6631111743720262e-06, + "loss": 1.0359, + "step": 54 + }, + { + "epoch": 0.004133473620922892, + "grad_norm": 1.940157267891346, + "learning_rate": 2.675361360298306e-06, + "loss": 1.1628, + "step": 55 + }, + { + "epoch": 0.004208627686757853, + "grad_norm": 2.6788426616127103, + "learning_rate": 2.6873908101004422e-06, + "loss": 1.2235, + "step": 56 + }, + { + "epoch": 0.004283781752592815, + "grad_norm": 1.6860384086270397, + "learning_rate": 2.6992073380696804e-06, + "loss": 1.1812, + "step": 57 + }, + { + "epoch": 0.004358935818427777, + "grad_norm": 1.5297608432974603, + "learning_rate": 2.710818350752891e-06, + "loss": 1.1343, + "step": 58 + }, + { + "epoch": 0.004434089884262739, + "grad_norm": 1.771168538699075, + "learning_rate": 2.7222308748360397e-06, + "loss": 1.0894, + "step": 59 + }, + { + "epoch": 0.004509243950097701, + "grad_norm": 2.292473032718904, + "learning_rate": 2.733451582684169e-06, + "loss": 1.1554, + "step": 60 + }, + { + "epoch": 0.004584398015932662, + "grad_norm": 2.173283730722902, + "learning_rate": 2.744486815770336e-06, + "loss": 1.1615, + "step": 61 + }, + { + "epoch": 0.004659552081767624, + "grad_norm": 1.99114969186478, + "learning_rate": 2.755342606199492e-06, + "loss": 1.168, + "step": 62 + }, + { + "epoch": 0.0047347061476025855, + "grad_norm": 2.7597056690784703, + "learning_rate": 2.766024696510225e-06, + "loss": 1.1125, + "step": 63 + }, + { + "epoch": 0.004809860213437547, + "grad_norm": 2.092321335469726, + "learning_rate": 2.7765385579171097e-06, + "loss": 1.1896, + "step": 64 + }, + { + "epoch": 0.004885014279272509, + "grad_norm": 2.19674440432559, + "learning_rate": 2.7868894071387408e-06, + "loss": 1.1301, + "step": 65 + }, + { + "epoch": 0.00496016834510747, + "grad_norm": 2.420756715578682, + "learning_rate": 2.79708222194103e-06, + "loss": 1.1607, + "step": 66 + }, + { + "epoch": 0.005035322410942432, + "grad_norm": 1.9540559914921456, + "learning_rate": 2.807121755511699e-06, + "loss": 1.1849, + "step": 67 + }, + { + "epoch": 0.005110476476777394, + "grad_norm": 1.8520889499551965, + "learning_rate": 2.8170125497698686e-06, + "loss": 1.1476, + "step": 68 + }, + { + "epoch": 0.005185630542612355, + "grad_norm": 2.0188525477957056, + "learning_rate": 2.826758947704043e-06, + "loss": 1.1312, + "step": 69 + }, + { + "epoch": 0.005260784608447317, + "grad_norm": 2.079481671563706, + "learning_rate": 2.8363651048223687e-06, + "loss": 1.1478, + "step": 70 + }, + { + "epoch": 0.005335938674282279, + "grad_norm": 1.7368716578882109, + "learning_rate": 2.8458349997907386e-06, + "loss": 1.1532, + "step": 71 + }, + { + "epoch": 0.00541109274011724, + "grad_norm": 1.854038434112744, + "learning_rate": 2.8551724443268935e-06, + "loss": 0.9295, + "step": 72 + }, + { + "epoch": 0.005486246805952202, + "grad_norm": 1.729679039414966, + "learning_rate": 2.8643810924121057e-06, + "loss": 1.1479, + "step": 73 + }, + { + "epoch": 0.0055614008717871635, + "grad_norm": 1.9422390109264553, + "learning_rate": 2.8734644488761585e-06, + "loss": 1.2267, + "step": 74 + }, + { + "epoch": 0.005636554937622125, + "grad_norm": 1.8481077001572759, + "learning_rate": 2.882425877406096e-06, + "loss": 1.1569, + "step": 75 + }, + { + "epoch": 0.005711709003457087, + "grad_norm": 1.73104001641257, + "learning_rate": 2.891268608024548e-06, + "loss": 1.2287, + "step": 76 + }, + { + "epoch": 0.005786863069292048, + "grad_norm": 1.5643317357663484, + "learning_rate": 2.8999957440792298e-06, + "loss": 0.9078, + "step": 77 + }, + { + "epoch": 0.00586201713512701, + "grad_norm": 1.8383023229607875, + "learning_rate": 2.9086102687814654e-06, + "loss": 1.0149, + "step": 78 + }, + { + "epoch": 0.005937171200961972, + "grad_norm": 1.6232902628976766, + "learning_rate": 2.9171150513282156e-06, + "loss": 1.1288, + "step": 79 + }, + { + "epoch": 0.006012325266796933, + "grad_norm": 1.5699238462262521, + "learning_rate": 2.9255128526390366e-06, + "loss": 1.1631, + "step": 80 + }, + { + "epoch": 0.006087479332631896, + "grad_norm": 1.5432624633918584, + "learning_rate": 2.9338063307366773e-06, + "loss": 1.1712, + "step": 81 + }, + { + "epoch": 0.006162633398466857, + "grad_norm": 1.9306630128308666, + "learning_rate": 2.9419980457975543e-06, + "loss": 1.0652, + "step": 82 + }, + { + "epoch": 0.006237787464301819, + "grad_norm": 5.256641002359354, + "learning_rate": 2.9500904648961173e-06, + "loss": 1.044, + "step": 83 + }, + { + "epoch": 0.006312941530136781, + "grad_norm": 1.528926955654497, + "learning_rate": 2.958085966465093e-06, + "loss": 1.1444, + "step": 84 + }, + { + "epoch": 0.006388095595971742, + "grad_norm": 2.4393202025435508, + "learning_rate": 2.965986844491795e-06, + "loss": 1.1667, + "step": 85 + }, + { + "epoch": 0.006463249661806704, + "grad_norm": 1.8684206924409852, + "learning_rate": 2.973795312469019e-06, + "loss": 1.1657, + "step": 86 + }, + { + "epoch": 0.006538403727641666, + "grad_norm": 1.918936801726515, + "learning_rate": 2.981513507117542e-06, + "loss": 1.1731, + "step": 87 + }, + { + "epoch": 0.006613557793476627, + "grad_norm": 1.8885020798077592, + "learning_rate": 2.989143491895898e-06, + "loss": 1.1242, + "step": 88 + }, + { + "epoch": 0.006688711859311589, + "grad_norm": 1.9065750345250432, + "learning_rate": 2.9966872603118436e-06, + "loss": 1.1072, + "step": 89 + }, + { + "epoch": 0.0067638659251465505, + "grad_norm": 1.697666625663976, + "learning_rate": 3.00414673904882e-06, + "loss": 1.0563, + "step": 90 + }, + { + "epoch": 0.006839019990981512, + "grad_norm": 1.953292900498056, + "learning_rate": 3.0115237909196643e-06, + "loss": 1.1495, + "step": 91 + }, + { + "epoch": 0.006914174056816474, + "grad_norm": 2.3748255213844747, + "learning_rate": 3.01882021765891e-06, + "loss": 1.1186, + "step": 92 + }, + { + "epoch": 0.006989328122651435, + "grad_norm": 1.7997263965354497, + "learning_rate": 3.0260377625641433e-06, + "loss": 1.1663, + "step": 93 + }, + { + "epoch": 0.007064482188486397, + "grad_norm": 3.4987467838230324, + "learning_rate": 3.033178112996099e-06, + "loss": 1.0756, + "step": 94 + }, + { + "epoch": 0.007139636254321359, + "grad_norm": 2.0508107791772856, + "learning_rate": 3.0402429027464746e-06, + "loss": 1.1472, + "step": 95 + }, + { + "epoch": 0.00721479032015632, + "grad_norm": 1.8076849810862308, + "learning_rate": 3.047233714281761e-06, + "loss": 1.1306, + "step": 96 + }, + { + "epoch": 0.007289944385991282, + "grad_norm": 2.401806679835208, + "learning_rate": 3.0541520808708106e-06, + "loss": 1.1261, + "step": 97 + }, + { + "epoch": 0.0073650984518262436, + "grad_norm": 2.0754278362376697, + "learning_rate": 3.0609994886032923e-06, + "loss": 1.0191, + "step": 98 + }, + { + "epoch": 0.007440252517661205, + "grad_norm": 2.233799039860609, + "learning_rate": 3.067777378305681e-06, + "loss": 1.0664, + "step": 99 + }, + { + "epoch": 0.007515406583496167, + "grad_norm": 1.5882280467476804, + "learning_rate": 3.0744871473609635e-06, + "loss": 1.0688, + "step": 100 + }, + { + "epoch": 0.0075905606493311285, + "grad_norm": 1.121895905196312, + "learning_rate": 3.081130151437802e-06, + "loss": 0.8129, + "step": 101 + }, + { + "epoch": 0.00766571471516609, + "grad_norm": 2.1899248035358663, + "learning_rate": 3.0877077061345193e-06, + "loss": 1.0184, + "step": 102 + }, + { + "epoch": 0.007740868781001052, + "grad_norm": 0.9502034358493814, + "learning_rate": 3.0942210885428814e-06, + "loss": 0.8414, + "step": 103 + }, + { + "epoch": 0.007816022846836013, + "grad_norm": 0.9955821799285441, + "learning_rate": 3.1006715387363326e-06, + "loss": 0.8457, + "step": 104 + }, + { + "epoch": 0.007891176912670975, + "grad_norm": 1.7430107327085391, + "learning_rate": 3.107060261187019e-06, + "loss": 1.0534, + "step": 105 + }, + { + "epoch": 0.007966330978505937, + "grad_norm": 2.636564846492174, + "learning_rate": 3.113388426115655e-06, + "loss": 1.0776, + "step": 106 + }, + { + "epoch": 0.008041485044340898, + "grad_norm": 3.3660202658502736, + "learning_rate": 3.119657170778007e-06, + "loss": 1.0524, + "step": 107 + }, + { + "epoch": 0.00811663911017586, + "grad_norm": 1.612582297775527, + "learning_rate": 3.1258676006915446e-06, + "loss": 1.109, + "step": 108 + }, + { + "epoch": 0.008191793176010822, + "grad_norm": 2.568433451581457, + "learning_rate": 3.1320207908055525e-06, + "loss": 1.0611, + "step": 109 + }, + { + "epoch": 0.008266947241845783, + "grad_norm": 2.8553642118133533, + "learning_rate": 3.138117786617824e-06, + "loss": 1.1739, + "step": 110 + }, + { + "epoch": 0.008342101307680745, + "grad_norm": 1.4558255931966564, + "learning_rate": 3.144159605240809e-06, + "loss": 1.0646, + "step": 111 + }, + { + "epoch": 0.008417255373515706, + "grad_norm": 1.0431625814736256, + "learning_rate": 3.1501472364199597e-06, + "loss": 0.8261, + "step": 112 + }, + { + "epoch": 0.008492409439350668, + "grad_norm": 1.897558817397927, + "learning_rate": 3.156081643506813e-06, + "loss": 1.1135, + "step": 113 + }, + { + "epoch": 0.00856756350518563, + "grad_norm": 2.1908877706263787, + "learning_rate": 3.161963764389199e-06, + "loss": 1.1831, + "step": 114 + }, + { + "epoch": 0.008642717571020593, + "grad_norm": 1.6247549605850726, + "learning_rate": 3.167794512380837e-06, + "loss": 1.0892, + "step": 115 + }, + { + "epoch": 0.008717871636855555, + "grad_norm": 0.9416421618409901, + "learning_rate": 3.1735747770724093e-06, + "loss": 0.8499, + "step": 116 + }, + { + "epoch": 0.008793025702690516, + "grad_norm": 1.8114875728213555, + "learning_rate": 3.179305425146116e-06, + "loss": 1.0746, + "step": 117 + }, + { + "epoch": 0.008868179768525478, + "grad_norm": 2.0886765687774544, + "learning_rate": 3.184987301155558e-06, + "loss": 1.0714, + "step": 118 + }, + { + "epoch": 0.00894333383436044, + "grad_norm": 1.86086783867275, + "learning_rate": 3.190621228272719e-06, + "loss": 0.9911, + "step": 119 + }, + { + "epoch": 0.009018487900195401, + "grad_norm": 1.7017597010652468, + "learning_rate": 3.1962080090036873e-06, + "loss": 1.2081, + "step": 120 + }, + { + "epoch": 0.009093641966030363, + "grad_norm": 2.800962561959466, + "learning_rate": 3.2017484258746856e-06, + "loss": 1.1035, + "step": 121 + }, + { + "epoch": 0.009168796031865325, + "grad_norm": 2.0144891421663833, + "learning_rate": 3.207243242089855e-06, + "loss": 1.1702, + "step": 122 + }, + { + "epoch": 0.009243950097700286, + "grad_norm": 1.668102893304108, + "learning_rate": 3.212693202162205e-06, + "loss": 1.1202, + "step": 123 + }, + { + "epoch": 0.009319104163535248, + "grad_norm": 1.50320457834721, + "learning_rate": 3.2180990325190106e-06, + "loss": 1.0587, + "step": 124 + }, + { + "epoch": 0.00939425822937021, + "grad_norm": 2.9488242182299595, + "learning_rate": 3.22346144208289e-06, + "loss": 1.0679, + "step": 125 + }, + { + "epoch": 0.009469412295205171, + "grad_norm": 1.7888936458679725, + "learning_rate": 3.2287811228297436e-06, + "loss": 1.0944, + "step": 126 + }, + { + "epoch": 0.009544566361040133, + "grad_norm": 2.6171958809334095, + "learning_rate": 3.2340587503246298e-06, + "loss": 1.162, + "step": 127 + }, + { + "epoch": 0.009619720426875094, + "grad_norm": 2.26180645339454, + "learning_rate": 3.239294984236628e-06, + "loss": 1.1451, + "step": 128 + }, + { + "epoch": 0.009694874492710056, + "grad_norm": 1.6661142943882097, + "learning_rate": 3.24449046883367e-06, + "loss": 1.0449, + "step": 129 + }, + { + "epoch": 0.009770028558545018, + "grad_norm": 1.6301346133527614, + "learning_rate": 3.249645833458259e-06, + "loss": 0.988, + "step": 130 + }, + { + "epoch": 0.00984518262437998, + "grad_norm": 1.9348555946187787, + "learning_rate": 3.2547616929849703e-06, + "loss": 1.1746, + "step": 131 + }, + { + "epoch": 0.00992033669021494, + "grad_norm": 1.6896251436540932, + "learning_rate": 3.2598386482605483e-06, + "loss": 1.0965, + "step": 132 + }, + { + "epoch": 0.009995490756049902, + "grad_norm": 1.829862437367386, + "learning_rate": 3.2648772865273986e-06, + "loss": 0.9921, + "step": 133 + }, + { + "epoch": 0.010070644821884864, + "grad_norm": 1.7610776288686802, + "learning_rate": 3.269878181831217e-06, + "loss": 1.0569, + "step": 134 + }, + { + "epoch": 0.010145798887719826, + "grad_norm": 2.049525052388639, + "learning_rate": 3.274841895413471e-06, + "loss": 1.1787, + "step": 135 + }, + { + "epoch": 0.010220952953554787, + "grad_norm": 2.00079103562041, + "learning_rate": 3.279768976089387e-06, + "loss": 1.1127, + "step": 136 + }, + { + "epoch": 0.010296107019389749, + "grad_norm": 1.856441973382715, + "learning_rate": 3.2846599606121004e-06, + "loss": 1.042, + "step": 137 + }, + { + "epoch": 0.01037126108522471, + "grad_norm": 1.1085545366918161, + "learning_rate": 3.289515374023561e-06, + "loss": 0.8658, + "step": 138 + }, + { + "epoch": 0.010446415151059672, + "grad_norm": 1.912046219091542, + "learning_rate": 3.2943357299927686e-06, + "loss": 1.0955, + "step": 139 + }, + { + "epoch": 0.010521569216894634, + "grad_norm": 1.808687673152632, + "learning_rate": 3.2991215311418867e-06, + "loss": 1.0428, + "step": 140 + }, + { + "epoch": 0.010596723282729596, + "grad_norm": 2.0977917277596076, + "learning_rate": 3.30387326936075e-06, + "loss": 1.0804, + "step": 141 + }, + { + "epoch": 0.010671877348564557, + "grad_norm": 1.503249384121346, + "learning_rate": 3.308591426110257e-06, + "loss": 1.0478, + "step": 142 + }, + { + "epoch": 0.010747031414399519, + "grad_norm": 2.1380335582458248, + "learning_rate": 3.3132764727151197e-06, + "loss": 1.0202, + "step": 143 + }, + { + "epoch": 0.01082218548023448, + "grad_norm": 2.4110712079829995, + "learning_rate": 3.317928870646412e-06, + "loss": 1.1087, + "step": 144 + }, + { + "epoch": 0.010897339546069442, + "grad_norm": 2.1066355683449767, + "learning_rate": 3.3225490717943362e-06, + "loss": 1.1236, + "step": 145 + }, + { + "epoch": 0.010972493611904404, + "grad_norm": 1.8811985553791875, + "learning_rate": 3.327137518731624e-06, + "loss": 1.1344, + "step": 146 + }, + { + "epoch": 0.011047647677739365, + "grad_norm": 1.9045508269169875, + "learning_rate": 3.3316946449679425e-06, + "loss": 1.1192, + "step": 147 + }, + { + "epoch": 0.011122801743574327, + "grad_norm": 1.77457421797183, + "learning_rate": 3.336220875195677e-06, + "loss": 1.1818, + "step": 148 + }, + { + "epoch": 0.011197955809409289, + "grad_norm": 2.01062254436273, + "learning_rate": 3.3407166255274344e-06, + "loss": 1.0976, + "step": 149 + }, + { + "epoch": 0.01127310987524425, + "grad_norm": 2.0765279498645337, + "learning_rate": 3.345182303725614e-06, + "loss": 1.0168, + "step": 150 + }, + { + "epoch": 0.011348263941079212, + "grad_norm": 1.4307658813490998, + "learning_rate": 3.3496183094243384e-06, + "loss": 1.1125, + "step": 151 + }, + { + "epoch": 0.011423418006914174, + "grad_norm": 1.6470559645074987, + "learning_rate": 3.3540250343440664e-06, + "loss": 1.1553, + "step": 152 + }, + { + "epoch": 0.011498572072749135, + "grad_norm": 3.1146990564253825, + "learning_rate": 3.35840286249917e-06, + "loss": 1.1853, + "step": 153 + }, + { + "epoch": 0.011573726138584097, + "grad_norm": 1.6293775576449603, + "learning_rate": 3.3627521703987477e-06, + "loss": 1.1328, + "step": 154 + }, + { + "epoch": 0.011648880204419058, + "grad_norm": 1.9201314817549766, + "learning_rate": 3.367073327240937e-06, + "loss": 1.0518, + "step": 155 + }, + { + "epoch": 0.01172403427025402, + "grad_norm": 2.4381512481051035, + "learning_rate": 3.3713666951009833e-06, + "loss": 1.1113, + "step": 156 + }, + { + "epoch": 0.011799188336088982, + "grad_norm": 2.05408053988125, + "learning_rate": 3.375632629113298e-06, + "loss": 1.0883, + "step": 157 + }, + { + "epoch": 0.011874342401923943, + "grad_norm": 2.092849423893197, + "learning_rate": 3.3798714776477344e-06, + "loss": 1.1002, + "step": 158 + }, + { + "epoch": 0.011949496467758905, + "grad_norm": 2.4718375877124443, + "learning_rate": 3.3840835824803065e-06, + "loss": 1.1405, + "step": 159 + }, + { + "epoch": 0.012024650533593867, + "grad_norm": 1.9298728205230402, + "learning_rate": 3.388269278958555e-06, + "loss": 0.9672, + "step": 160 + }, + { + "epoch": 0.01209980459942883, + "grad_norm": 1.0805600627673326, + "learning_rate": 3.3924288961617605e-06, + "loss": 0.8467, + "step": 161 + }, + { + "epoch": 0.012174958665263792, + "grad_norm": 1.8818819271024767, + "learning_rate": 3.3965627570561953e-06, + "loss": 0.9933, + "step": 162 + }, + { + "epoch": 0.012250112731098753, + "grad_norm": 2.7449367902249264, + "learning_rate": 3.4006711786456036e-06, + "loss": 1.0506, + "step": 163 + }, + { + "epoch": 0.012325266796933715, + "grad_norm": 1.7362767695758783, + "learning_rate": 3.404754472117073e-06, + "loss": 1.0539, + "step": 164 + }, + { + "epoch": 0.012400420862768676, + "grad_norm": 2.125401851580233, + "learning_rate": 3.408812942982475e-06, + "loss": 1.1042, + "step": 165 + }, + { + "epoch": 0.012475574928603638, + "grad_norm": 1.7911592548648045, + "learning_rate": 3.4128468912156357e-06, + "loss": 1.1101, + "step": 166 + }, + { + "epoch": 0.0125507289944386, + "grad_norm": 1.4946893619347903, + "learning_rate": 3.4168566113853806e-06, + "loss": 1.0075, + "step": 167 + }, + { + "epoch": 0.012625883060273561, + "grad_norm": 2.2199551170554606, + "learning_rate": 3.420842392784611e-06, + "loss": 1.0727, + "step": 168 + }, + { + "epoch": 0.012701037126108523, + "grad_norm": 2.5121139761221665, + "learning_rate": 3.424804519555555e-06, + "loss": 1.0524, + "step": 169 + }, + { + "epoch": 0.012776191191943485, + "grad_norm": 1.62342582162396, + "learning_rate": 3.4287432708113135e-06, + "loss": 1.1143, + "step": 170 + }, + { + "epoch": 0.012851345257778446, + "grad_norm": 1.812902011354736, + "learning_rate": 3.4326589207538503e-06, + "loss": 1.0435, + "step": 171 + }, + { + "epoch": 0.012926499323613408, + "grad_norm": 4.896806119764244, + "learning_rate": 3.436551738788537e-06, + "loss": 0.92, + "step": 172 + }, + { + "epoch": 0.01300165338944837, + "grad_norm": 2.210768622428974, + "learning_rate": 3.440421989635386e-06, + "loss": 1.1258, + "step": 173 + }, + { + "epoch": 0.013076807455283331, + "grad_norm": 1.6035055143941297, + "learning_rate": 3.44426993343706e-06, + "loss": 1.1115, + "step": 174 + }, + { + "epoch": 0.013151961521118293, + "grad_norm": 2.863571672685944, + "learning_rate": 3.4480958258638136e-06, + "loss": 1.0884, + "step": 175 + }, + { + "epoch": 0.013227115586953254, + "grad_norm": 2.085635274073576, + "learning_rate": 3.4518999182154156e-06, + "loss": 1.0761, + "step": 176 + }, + { + "epoch": 0.013302269652788216, + "grad_norm": 1.4069396976760336, + "learning_rate": 3.4556824575202087e-06, + "loss": 1.0916, + "step": 177 + }, + { + "epoch": 0.013377423718623178, + "grad_norm": 2.1552672427131183, + "learning_rate": 3.4594436866313616e-06, + "loss": 1.083, + "step": 178 + }, + { + "epoch": 0.01345257778445814, + "grad_norm": 3.1323235717677624, + "learning_rate": 3.463183844320436e-06, + "loss": 1.1226, + "step": 179 + }, + { + "epoch": 0.013527731850293101, + "grad_norm": 1.6559401888158722, + "learning_rate": 3.4669031653683388e-06, + "loss": 0.9651, + "step": 180 + }, + { + "epoch": 0.013602885916128063, + "grad_norm": 1.6731722068468171, + "learning_rate": 3.4706018806537624e-06, + "loss": 1.1362, + "step": 181 + }, + { + "epoch": 0.013678039981963024, + "grad_norm": 1.5806597003619687, + "learning_rate": 3.4742802172391827e-06, + "loss": 1.0204, + "step": 182 + }, + { + "epoch": 0.013753194047797986, + "grad_norm": 1.6374459847430234, + "learning_rate": 3.4779383984545055e-06, + "loss": 1.1859, + "step": 183 + }, + { + "epoch": 0.013828348113632948, + "grad_norm": 1.945553806765763, + "learning_rate": 3.481576643978429e-06, + "loss": 1.024, + "step": 184 + }, + { + "epoch": 0.01390350217946791, + "grad_norm": 1.650076245944908, + "learning_rate": 3.485195169917603e-06, + "loss": 1.0165, + "step": 185 + }, + { + "epoch": 0.01397865624530287, + "grad_norm": 1.7459463445583836, + "learning_rate": 3.4887941888836612e-06, + "loss": 1.0648, + "step": 186 + }, + { + "epoch": 0.014053810311137832, + "grad_norm": 1.8064810027668556, + "learning_rate": 3.4923739100681745e-06, + "loss": 1.0348, + "step": 187 + }, + { + "epoch": 0.014128964376972794, + "grad_norm": 1.647711571406351, + "learning_rate": 3.4959345393156175e-06, + "loss": 1.1214, + "step": 188 + }, + { + "epoch": 0.014204118442807756, + "grad_norm": 1.7471786644537703, + "learning_rate": 3.4994762791943946e-06, + "loss": 1.0807, + "step": 189 + }, + { + "epoch": 0.014279272508642717, + "grad_norm": 1.8381885120946393, + "learning_rate": 3.502999329065993e-06, + "loss": 1.0328, + "step": 190 + }, + { + "epoch": 0.014354426574477679, + "grad_norm": 2.0407748577712055, + "learning_rate": 3.506503885152319e-06, + "loss": 0.9767, + "step": 191 + }, + { + "epoch": 0.01442958064031264, + "grad_norm": 1.8806845642170376, + "learning_rate": 3.5099901406012796e-06, + "loss": 1.1412, + "step": 192 + }, + { + "epoch": 0.014504734706147602, + "grad_norm": 2.2519650884940785, + "learning_rate": 3.513458285550655e-06, + "loss": 1.0949, + "step": 193 + }, + { + "epoch": 0.014579888771982564, + "grad_norm": 1.7294699238270914, + "learning_rate": 3.516908507190329e-06, + "loss": 1.0166, + "step": 194 + }, + { + "epoch": 0.014655042837817525, + "grad_norm": 2.6582607960944107, + "learning_rate": 3.5203409898229102e-06, + "loss": 1.0162, + "step": 195 + }, + { + "epoch": 0.014730196903652487, + "grad_norm": 1.9901862770385645, + "learning_rate": 3.52375591492281e-06, + "loss": 1.0319, + "step": 196 + }, + { + "epoch": 0.014805350969487449, + "grad_norm": 1.218251933515311, + "learning_rate": 3.527153461193815e-06, + "loss": 0.8738, + "step": 197 + }, + { + "epoch": 0.01488050503532241, + "grad_norm": 1.9265148939699979, + "learning_rate": 3.5305338046251994e-06, + "loss": 1.0784, + "step": 198 + }, + { + "epoch": 0.014955659101157372, + "grad_norm": 1.9056290412477135, + "learning_rate": 3.533897118546427e-06, + "loss": 1.1741, + "step": 199 + }, + { + "epoch": 0.015030813166992334, + "grad_norm": 1.6982655625053538, + "learning_rate": 3.5372435736804815e-06, + "loss": 1.1301, + "step": 200 + }, + { + "epoch": 0.015105967232827295, + "grad_norm": 2.1308974167376475, + "learning_rate": 3.5405733381958684e-06, + "loss": 1.0256, + "step": 201 + }, + { + "epoch": 0.015181121298662257, + "grad_norm": 1.6434103437154126, + "learning_rate": 3.5438865777573207e-06, + "loss": 1.1064, + "step": 202 + }, + { + "epoch": 0.015256275364497219, + "grad_norm": 2.141653106389242, + "learning_rate": 3.5471834555752594e-06, + "loss": 1.1508, + "step": 203 + }, + { + "epoch": 0.01533142943033218, + "grad_norm": 3.289793101011208, + "learning_rate": 3.5504641324540377e-06, + "loss": 1.1414, + "step": 204 + }, + { + "epoch": 0.015406583496167142, + "grad_norm": 1.4820315685281928, + "learning_rate": 3.5537287668389996e-06, + "loss": 1.0709, + "step": 205 + }, + { + "epoch": 0.015481737562002103, + "grad_norm": 1.5854750837607623, + "learning_rate": 3.5569775148623998e-06, + "loss": 1.0734, + "step": 206 + }, + { + "epoch": 0.015556891627837067, + "grad_norm": 1.7370256668416981, + "learning_rate": 3.5602105303882114e-06, + "loss": 1.0182, + "step": 207 + }, + { + "epoch": 0.015632045693672027, + "grad_norm": 2.346864182692385, + "learning_rate": 3.563427965055851e-06, + "loss": 1.0278, + "step": 208 + }, + { + "epoch": 0.01570719975950699, + "grad_norm": 1.842791809410936, + "learning_rate": 3.566629968322854e-06, + "loss": 1.0987, + "step": 209 + }, + { + "epoch": 0.01578235382534195, + "grad_norm": 1.6189043450139522, + "learning_rate": 3.5698166875065377e-06, + "loss": 0.9972, + "step": 210 + }, + { + "epoch": 0.015857507891176913, + "grad_norm": 1.639721939775053, + "learning_rate": 3.5729882678246694e-06, + "loss": 1.0882, + "step": 211 + }, + { + "epoch": 0.015932661957011873, + "grad_norm": 2.050297794577108, + "learning_rate": 3.5761448524351738e-06, + "loss": 0.9046, + "step": 212 + }, + { + "epoch": 0.016007816022846837, + "grad_norm": 1.5668433285933288, + "learning_rate": 3.579286582474908e-06, + "loss": 1.125, + "step": 213 + }, + { + "epoch": 0.016082970088681797, + "grad_norm": 2.2456863166470686, + "learning_rate": 3.582413597097526e-06, + "loss": 1.0502, + "step": 214 + }, + { + "epoch": 0.01615812415451676, + "grad_norm": 2.495278668637087, + "learning_rate": 3.5855260335104637e-06, + "loss": 1.1062, + "step": 215 + }, + { + "epoch": 0.01623327822035172, + "grad_norm": 2.5865981073568296, + "learning_rate": 3.588624027011063e-06, + "loss": 1.0702, + "step": 216 + }, + { + "epoch": 0.016308432286186683, + "grad_norm": 1.273553105823012, + "learning_rate": 3.5917077110218606e-06, + "loss": 0.9709, + "step": 217 + }, + { + "epoch": 0.016383586352021643, + "grad_norm": 2.274713217297351, + "learning_rate": 3.5947772171250713e-06, + "loss": 1.0692, + "step": 218 + }, + { + "epoch": 0.016458740417856606, + "grad_norm": 2.1718533741265396, + "learning_rate": 3.597832675096275e-06, + "loss": 1.0781, + "step": 219 + }, + { + "epoch": 0.016533894483691566, + "grad_norm": 1.6830822373378587, + "learning_rate": 3.600874212937343e-06, + "loss": 1.0677, + "step": 220 + }, + { + "epoch": 0.01660904854952653, + "grad_norm": 1.5116491178498455, + "learning_rate": 3.603901956908609e-06, + "loss": 1.0352, + "step": 221 + }, + { + "epoch": 0.01668420261536149, + "grad_norm": 1.8478112831156874, + "learning_rate": 3.6069160315603275e-06, + "loss": 1.1283, + "step": 222 + }, + { + "epoch": 0.016759356681196453, + "grad_norm": 2.2688923394395624, + "learning_rate": 3.6099165597634083e-06, + "loss": 1.1178, + "step": 223 + }, + { + "epoch": 0.016834510747031413, + "grad_norm": 2.710638200139754, + "learning_rate": 3.6129036627394785e-06, + "loss": 1.0997, + "step": 224 + }, + { + "epoch": 0.016909664812866376, + "grad_norm": 2.052923917866619, + "learning_rate": 3.615877460090265e-06, + "loss": 1.0209, + "step": 225 + }, + { + "epoch": 0.016984818878701336, + "grad_norm": 2.116613106037429, + "learning_rate": 3.6188380698263314e-06, + "loss": 1.1177, + "step": 226 + }, + { + "epoch": 0.0170599729445363, + "grad_norm": 1.8509903035705924, + "learning_rate": 3.6217856083951765e-06, + "loss": 1.1224, + "step": 227 + }, + { + "epoch": 0.01713512701037126, + "grad_norm": 1.9189978051225236, + "learning_rate": 3.6247201907087175e-06, + "loss": 1.0801, + "step": 228 + }, + { + "epoch": 0.017210281076206223, + "grad_norm": 1.7173374266773116, + "learning_rate": 3.627641930170173e-06, + "loss": 1.1179, + "step": 229 + }, + { + "epoch": 0.017285435142041186, + "grad_norm": 1.423650283067792, + "learning_rate": 3.630550938700355e-06, + "loss": 1.0527, + "step": 230 + }, + { + "epoch": 0.017360589207876146, + "grad_norm": 1.6883920501686307, + "learning_rate": 3.6334473267633984e-06, + "loss": 0.9857, + "step": 231 + }, + { + "epoch": 0.01743574327371111, + "grad_norm": 2.0852690922377946, + "learning_rate": 3.6363312033919277e-06, + "loss": 1.0798, + "step": 232 + }, + { + "epoch": 0.01751089733954607, + "grad_norm": 2.026081022715506, + "learning_rate": 3.639202676211685e-06, + "loss": 1.189, + "step": 233 + }, + { + "epoch": 0.017586051405381033, + "grad_norm": 1.5200433726145832, + "learning_rate": 3.642061851465635e-06, + "loss": 1.1103, + "step": 234 + }, + { + "epoch": 0.017661205471215993, + "grad_norm": 1.6359583287702961, + "learning_rate": 3.644908834037544e-06, + "loss": 1.043, + "step": 235 + }, + { + "epoch": 0.017736359537050956, + "grad_norm": 1.8153460289946302, + "learning_rate": 3.647743727475077e-06, + "loss": 1.1258, + "step": 236 + }, + { + "epoch": 0.017811513602885916, + "grad_norm": 1.9032732124431355, + "learning_rate": 3.650566634012385e-06, + "loss": 1.1169, + "step": 237 + }, + { + "epoch": 0.01788666766872088, + "grad_norm": 1.6277785388364665, + "learning_rate": 3.653377654592237e-06, + "loss": 1.068, + "step": 238 + }, + { + "epoch": 0.01796182173455584, + "grad_norm": 2.294792840454891, + "learning_rate": 3.6561768888876717e-06, + "loss": 1.0521, + "step": 239 + }, + { + "epoch": 0.018036975800390802, + "grad_norm": 1.6593283350358237, + "learning_rate": 3.658964435323206e-06, + "loss": 1.0053, + "step": 240 + }, + { + "epoch": 0.018112129866225762, + "grad_norm": 1.6820277744227183, + "learning_rate": 3.6617403910956026e-06, + "loss": 1.0679, + "step": 241 + }, + { + "epoch": 0.018187283932060726, + "grad_norm": 1.5944794873999275, + "learning_rate": 3.6645048521942035e-06, + "loss": 1.1065, + "step": 242 + }, + { + "epoch": 0.018262437997895686, + "grad_norm": 1.4382970507581119, + "learning_rate": 3.667257913420846e-06, + "loss": 1.0898, + "step": 243 + }, + { + "epoch": 0.01833759206373065, + "grad_norm": 3.064912896466594, + "learning_rate": 3.6699996684093732e-06, + "loss": 1.0421, + "step": 244 + }, + { + "epoch": 0.01841274612956561, + "grad_norm": 2.1980280795191156, + "learning_rate": 3.6727302096447376e-06, + "loss": 1.1292, + "step": 245 + }, + { + "epoch": 0.018487900195400572, + "grad_norm": 1.5520847329084329, + "learning_rate": 3.6754496284817233e-06, + "loss": 1.1462, + "step": 246 + }, + { + "epoch": 0.018563054261235532, + "grad_norm": 1.6734789703388149, + "learning_rate": 3.678158015163289e-06, + "loss": 1.0464, + "step": 247 + }, + { + "epoch": 0.018638208327070496, + "grad_norm": 2.5160504093972706, + "learning_rate": 3.680855458838529e-06, + "loss": 1.0539, + "step": 248 + }, + { + "epoch": 0.018713362392905455, + "grad_norm": 2.3431326498139247, + "learning_rate": 3.6835420475802863e-06, + "loss": 1.0706, + "step": 249 + }, + { + "epoch": 0.01878851645874042, + "grad_norm": 1.6352473452764766, + "learning_rate": 3.686217868402408e-06, + "loss": 0.9923, + "step": 250 + }, + { + "epoch": 0.01886367052457538, + "grad_norm": 1.8787846833236383, + "learning_rate": 3.688883007276652e-06, + "loss": 1.0818, + "step": 251 + }, + { + "epoch": 0.018938824590410342, + "grad_norm": 1.8466661312547772, + "learning_rate": 3.6915375491492623e-06, + "loss": 1.0112, + "step": 252 + }, + { + "epoch": 0.019013978656245302, + "grad_norm": 1.6046488726608885, + "learning_rate": 3.694181577957216e-06, + "loss": 1.0967, + "step": 253 + }, + { + "epoch": 0.019089132722080265, + "grad_norm": 2.023586429726741, + "learning_rate": 3.6968151766441486e-06, + "loss": 1.043, + "step": 254 + }, + { + "epoch": 0.019164286787915225, + "grad_norm": 1.6156996364803935, + "learning_rate": 3.699438427175964e-06, + "loss": 1.1116, + "step": 255 + }, + { + "epoch": 0.01923944085375019, + "grad_norm": 1.8195368215132468, + "learning_rate": 3.702051410556147e-06, + "loss": 1.0735, + "step": 256 + }, + { + "epoch": 0.01931459491958515, + "grad_norm": 5.715934018415613, + "learning_rate": 3.7046542068407645e-06, + "loss": 0.9036, + "step": 257 + }, + { + "epoch": 0.019389748985420112, + "grad_norm": 2.4715278545695343, + "learning_rate": 3.7072468951531883e-06, + "loss": 1.1386, + "step": 258 + }, + { + "epoch": 0.019464903051255072, + "grad_norm": 2.1494577655569524, + "learning_rate": 3.7098295536985265e-06, + "loss": 1.0342, + "step": 259 + }, + { + "epoch": 0.019540057117090035, + "grad_norm": 1.9366022306524555, + "learning_rate": 3.7124022597777775e-06, + "loss": 0.9913, + "step": 260 + }, + { + "epoch": 0.019615211182924995, + "grad_norm": 2.1419066887129037, + "learning_rate": 3.7149650898017115e-06, + "loss": 1.0774, + "step": 261 + }, + { + "epoch": 0.01969036524875996, + "grad_norm": 2.379938312019569, + "learning_rate": 3.717518119304489e-06, + "loss": 1.1334, + "step": 262 + }, + { + "epoch": 0.01976551931459492, + "grad_norm": 1.5834220141957331, + "learning_rate": 3.7200614229570204e-06, + "loss": 0.9923, + "step": 263 + }, + { + "epoch": 0.01984067338042988, + "grad_norm": 1.639779227097354, + "learning_rate": 3.7225950745800667e-06, + "loss": 1.0755, + "step": 264 + }, + { + "epoch": 0.01991582744626484, + "grad_norm": 1.7885491580024941, + "learning_rate": 3.7251191471571003e-06, + "loss": 1.057, + "step": 265 + }, + { + "epoch": 0.019990981512099805, + "grad_norm": 1.9693847117522012, + "learning_rate": 3.7276337128469165e-06, + "loss": 1.0386, + "step": 266 + }, + { + "epoch": 0.020066135577934765, + "grad_norm": 1.8144202747799956, + "learning_rate": 3.730138842996013e-06, + "loss": 1.0771, + "step": 267 + }, + { + "epoch": 0.020141289643769728, + "grad_norm": 1.8623622331294964, + "learning_rate": 3.7326346081507353e-06, + "loss": 1.0744, + "step": 268 + }, + { + "epoch": 0.020216443709604688, + "grad_norm": 1.298889617611525, + "learning_rate": 3.7351210780691997e-06, + "loss": 1.0639, + "step": 269 + }, + { + "epoch": 0.02029159777543965, + "grad_norm": 1.897479345337402, + "learning_rate": 3.7375983217329894e-06, + "loss": 0.9978, + "step": 270 + }, + { + "epoch": 0.02036675184127461, + "grad_norm": 1.977626227824348, + "learning_rate": 3.7400664073586386e-06, + "loss": 0.9834, + "step": 271 + }, + { + "epoch": 0.020441905907109575, + "grad_norm": 1.6394345278590647, + "learning_rate": 3.7425254024089058e-06, + "loss": 1.0414, + "step": 272 + }, + { + "epoch": 0.020517059972944535, + "grad_norm": 1.7476425641018107, + "learning_rate": 3.7449753736038338e-06, + "loss": 1.1089, + "step": 273 + }, + { + "epoch": 0.020592214038779498, + "grad_norm": 1.6249735903427858, + "learning_rate": 3.7474163869316188e-06, + "loss": 1.1634, + "step": 274 + }, + { + "epoch": 0.020667368104614458, + "grad_norm": 2.2098176634124784, + "learning_rate": 3.749848507659269e-06, + "loss": 1.1461, + "step": 275 + }, + { + "epoch": 0.02074252217044942, + "grad_norm": 2.5148315126015266, + "learning_rate": 3.752271800343079e-06, + "loss": 1.0657, + "step": 276 + }, + { + "epoch": 0.020817676236284385, + "grad_norm": 1.8419366097865708, + "learning_rate": 3.7546863288389105e-06, + "loss": 1.0274, + "step": 277 + }, + { + "epoch": 0.020892830302119345, + "grad_norm": 1.94849720292139, + "learning_rate": 3.757092156312287e-06, + "loss": 0.945, + "step": 278 + }, + { + "epoch": 0.020967984367954308, + "grad_norm": 1.518923915120279, + "learning_rate": 3.759489345248312e-06, + "loss": 1.0478, + "step": 279 + }, + { + "epoch": 0.021043138433789268, + "grad_norm": 1.93089553147809, + "learning_rate": 3.7618779574614046e-06, + "loss": 1.0224, + "step": 280 + }, + { + "epoch": 0.02111829249962423, + "grad_norm": 1.4965795619626856, + "learning_rate": 3.7642580541048654e-06, + "loss": 1.0378, + "step": 281 + }, + { + "epoch": 0.02119344656545919, + "grad_norm": 1.6475676641620987, + "learning_rate": 3.766629695680268e-06, + "loss": 1.1255, + "step": 282 + }, + { + "epoch": 0.021268600631294154, + "grad_norm": 1.2215673210495859, + "learning_rate": 3.7689929420466896e-06, + "loss": 0.8176, + "step": 283 + }, + { + "epoch": 0.021343754697129114, + "grad_norm": 0.8106365475975487, + "learning_rate": 3.7713478524297754e-06, + "loss": 0.8015, + "step": 284 + }, + { + "epoch": 0.021418908762964078, + "grad_norm": 1.7074883937860135, + "learning_rate": 3.7736944854306444e-06, + "loss": 1.1493, + "step": 285 + }, + { + "epoch": 0.021494062828799038, + "grad_norm": 3.4600361840886342, + "learning_rate": 3.7760328990346385e-06, + "loss": 1.0831, + "step": 286 + }, + { + "epoch": 0.021569216894634, + "grad_norm": 2.0014450098317393, + "learning_rate": 3.778363150619923e-06, + "loss": 1.0228, + "step": 287 + }, + { + "epoch": 0.02164437096046896, + "grad_norm": 1.563581550118266, + "learning_rate": 3.78068529696593e-06, + "loss": 0.9141, + "step": 288 + }, + { + "epoch": 0.021719525026303924, + "grad_norm": 1.9551258672016367, + "learning_rate": 3.782999394261664e-06, + "loss": 1.1056, + "step": 289 + }, + { + "epoch": 0.021794679092138884, + "grad_norm": 1.8696340071179438, + "learning_rate": 3.785305498113854e-06, + "loss": 1.0831, + "step": 290 + }, + { + "epoch": 0.021869833157973848, + "grad_norm": 1.6512616449215471, + "learning_rate": 3.78760366355498e-06, + "loss": 1.1163, + "step": 291 + }, + { + "epoch": 0.021944987223808807, + "grad_norm": 1.755492039361743, + "learning_rate": 3.789893945051143e-06, + "loss": 1.0075, + "step": 292 + }, + { + "epoch": 0.02202014128964377, + "grad_norm": 1.7286008652181424, + "learning_rate": 3.792176396509817e-06, + "loss": 1.049, + "step": 293 + }, + { + "epoch": 0.02209529535547873, + "grad_norm": 3.540569329897322, + "learning_rate": 3.7944510712874613e-06, + "loss": 1.0755, + "step": 294 + }, + { + "epoch": 0.022170449421313694, + "grad_norm": 1.949410775000985, + "learning_rate": 3.796718022197003e-06, + "loss": 1.0226, + "step": 295 + }, + { + "epoch": 0.022245603487148654, + "grad_norm": 1.9809369318212515, + "learning_rate": 3.7989773015151948e-06, + "loss": 1.094, + "step": 296 + }, + { + "epoch": 0.022320757552983617, + "grad_norm": 1.7634208221701457, + "learning_rate": 3.80122896098985e-06, + "loss": 1.047, + "step": 297 + }, + { + "epoch": 0.022395911618818577, + "grad_norm": 1.9625272955484478, + "learning_rate": 3.803473051846953e-06, + "loss": 0.9851, + "step": 298 + }, + { + "epoch": 0.02247106568465354, + "grad_norm": 2.0166234345439182, + "learning_rate": 3.805709624797651e-06, + "loss": 1.1726, + "step": 299 + }, + { + "epoch": 0.0225462197504885, + "grad_norm": 2.0577388486665464, + "learning_rate": 3.8079387300451326e-06, + "loss": 1.0716, + "step": 300 + }, + { + "epoch": 0.022621373816323464, + "grad_norm": 0.9749639932390531, + "learning_rate": 3.8101604172913872e-06, + "loss": 0.8147, + "step": 301 + }, + { + "epoch": 0.022696527882158424, + "grad_norm": 1.7381807791145254, + "learning_rate": 3.8123747357438563e-06, + "loss": 0.9973, + "step": 302 + }, + { + "epoch": 0.022771681947993387, + "grad_norm": 1.488859209592008, + "learning_rate": 3.814581734121971e-06, + "loss": 1.0086, + "step": 303 + }, + { + "epoch": 0.022846836013828347, + "grad_norm": 1.5461167297660219, + "learning_rate": 3.816781460663585e-06, + "loss": 1.0802, + "step": 304 + }, + { + "epoch": 0.02292199007966331, + "grad_norm": 1.9744703355204454, + "learning_rate": 3.818973963131299e-06, + "loss": 1.0929, + "step": 305 + }, + { + "epoch": 0.02299714414549827, + "grad_norm": 2.1613307195165614, + "learning_rate": 3.821159288818688e-06, + "loss": 1.1092, + "step": 306 + }, + { + "epoch": 0.023072298211333234, + "grad_norm": 1.9206275302859124, + "learning_rate": 3.823337484556417e-06, + "loss": 1.0304, + "step": 307 + }, + { + "epoch": 0.023147452277168194, + "grad_norm": 2.0853197418869174, + "learning_rate": 3.825508596718266e-06, + "loss": 1.1293, + "step": 308 + }, + { + "epoch": 0.023222606343003157, + "grad_norm": 1.8400990173400242, + "learning_rate": 3.827672671227052e-06, + "loss": 1.0314, + "step": 309 + }, + { + "epoch": 0.023297760408838117, + "grad_norm": 1.9097640994325888, + "learning_rate": 3.8298297535604554e-06, + "loss": 0.9407, + "step": 310 + }, + { + "epoch": 0.02337291447467308, + "grad_norm": 1.681991133009936, + "learning_rate": 3.831979888756763e-06, + "loss": 1.0531, + "step": 311 + }, + { + "epoch": 0.02344806854050804, + "grad_norm": 2.317551328220272, + "learning_rate": 3.834123121420502e-06, + "loss": 1.0723, + "step": 312 + }, + { + "epoch": 0.023523222606343003, + "grad_norm": 1.8287153024569422, + "learning_rate": 3.836259495727992e-06, + "loss": 1.0237, + "step": 313 + }, + { + "epoch": 0.023598376672177963, + "grad_norm": 1.8141031738768252, + "learning_rate": 3.838389055432816e-06, + "loss": 1.0729, + "step": 314 + }, + { + "epoch": 0.023673530738012927, + "grad_norm": 1.9284661632652222, + "learning_rate": 3.840511843871188e-06, + "loss": 1.1389, + "step": 315 + }, + { + "epoch": 0.023748684803847887, + "grad_norm": 2.0686757653507857, + "learning_rate": 3.842627903967253e-06, + "loss": 0.9897, + "step": 316 + }, + { + "epoch": 0.02382383886968285, + "grad_norm": 3.3716725087991266, + "learning_rate": 3.844737278238285e-06, + "loss": 0.9601, + "step": 317 + }, + { + "epoch": 0.02389899293551781, + "grad_norm": 1.7659880678372117, + "learning_rate": 3.846840008799825e-06, + "loss": 1.1368, + "step": 318 + }, + { + "epoch": 0.023974147001352773, + "grad_norm": 1.9475623911901063, + "learning_rate": 3.848936137370715e-06, + "loss": 0.9763, + "step": 319 + }, + { + "epoch": 0.024049301067187733, + "grad_norm": 1.1299166266311476, + "learning_rate": 3.851025705278073e-06, + "loss": 0.7995, + "step": 320 + }, + { + "epoch": 0.024124455133022697, + "grad_norm": 2.5202841883072518, + "learning_rate": 3.853108753462177e-06, + "loss": 1.004, + "step": 321 + }, + { + "epoch": 0.02419960919885766, + "grad_norm": 0.8666423768222083, + "learning_rate": 3.855185322481279e-06, + "loss": 0.8091, + "step": 322 + }, + { + "epoch": 0.02427476326469262, + "grad_norm": 2.0854923759180246, + "learning_rate": 3.857255452516343e-06, + "loss": 1.0582, + "step": 323 + }, + { + "epoch": 0.024349917330527583, + "grad_norm": 2.013543643809014, + "learning_rate": 3.859319183375714e-06, + "loss": 1.0533, + "step": 324 + }, + { + "epoch": 0.024425071396362543, + "grad_norm": 1.9348055654616454, + "learning_rate": 3.861376554499704e-06, + "loss": 1.1347, + "step": 325 + }, + { + "epoch": 0.024500225462197506, + "grad_norm": 1.7576594307269022, + "learning_rate": 3.863427604965122e-06, + "loss": 1.0592, + "step": 326 + }, + { + "epoch": 0.024575379528032466, + "grad_norm": 1.48845777691203, + "learning_rate": 3.865472373489722e-06, + "loss": 1.0167, + "step": 327 + }, + { + "epoch": 0.02465053359386743, + "grad_norm": 3.9820494802745094, + "learning_rate": 3.8675108984365914e-06, + "loss": 1.0815, + "step": 328 + }, + { + "epoch": 0.02472568765970239, + "grad_norm": 1.5483720259388047, + "learning_rate": 3.869543217818467e-06, + "loss": 1.0184, + "step": 329 + }, + { + "epoch": 0.024800841725537353, + "grad_norm": 1.792800529749876, + "learning_rate": 3.871569369301993e-06, + "loss": 1.0871, + "step": 330 + }, + { + "epoch": 0.024875995791372313, + "grad_norm": 1.593885378604175, + "learning_rate": 3.873589390211904e-06, + "loss": 1.0663, + "step": 331 + }, + { + "epoch": 0.024951149857207276, + "grad_norm": 1.859170277328553, + "learning_rate": 3.875603317535154e-06, + "loss": 1.0168, + "step": 332 + }, + { + "epoch": 0.025026303923042236, + "grad_norm": 1.5146530023403972, + "learning_rate": 3.877611187924979e-06, + "loss": 1.1186, + "step": 333 + }, + { + "epoch": 0.0251014579888772, + "grad_norm": 1.5430266164238093, + "learning_rate": 3.879613037704899e-06, + "loss": 0.9468, + "step": 334 + }, + { + "epoch": 0.02517661205471216, + "grad_norm": 1.8021971458995818, + "learning_rate": 3.881608902872662e-06, + "loss": 0.9506, + "step": 335 + }, + { + "epoch": 0.025251766120547123, + "grad_norm": 2.0578732469263112, + "learning_rate": 3.88359881910413e-06, + "loss": 1.0782, + "step": 336 + }, + { + "epoch": 0.025326920186382083, + "grad_norm": 2.7589968982829807, + "learning_rate": 3.885582821757098e-06, + "loss": 0.9073, + "step": 337 + }, + { + "epoch": 0.025402074252217046, + "grad_norm": 1.7773093419360018, + "learning_rate": 3.887560945875073e-06, + "loss": 1.0619, + "step": 338 + }, + { + "epoch": 0.025477228318052006, + "grad_norm": 1.0377100697377588, + "learning_rate": 3.889533226190982e-06, + "loss": 0.779, + "step": 339 + }, + { + "epoch": 0.02555238238388697, + "grad_norm": 1.8038197202155015, + "learning_rate": 3.891499697130832e-06, + "loss": 0.9958, + "step": 340 + }, + { + "epoch": 0.02562753644972193, + "grad_norm": 12.046223307397206, + "learning_rate": 3.893460392817316e-06, + "loss": 1.1148, + "step": 341 + }, + { + "epoch": 0.025702690515556893, + "grad_norm": 1.8058574236694858, + "learning_rate": 3.895415347073368e-06, + "loss": 1.0537, + "step": 342 + }, + { + "epoch": 0.025777844581391852, + "grad_norm": 2.150347384400063, + "learning_rate": 3.89736459342566e-06, + "loss": 1.0273, + "step": 343 + }, + { + "epoch": 0.025852998647226816, + "grad_norm": 4.008930964868203, + "learning_rate": 3.899308165108055e-06, + "loss": 1.1486, + "step": 344 + }, + { + "epoch": 0.025928152713061776, + "grad_norm": 3.751172708886453, + "learning_rate": 3.9012460950650064e-06, + "loss": 1.0495, + "step": 345 + }, + { + "epoch": 0.02600330677889674, + "grad_norm": 1.7449955911225419, + "learning_rate": 3.903178415954904e-06, + "loss": 1.0685, + "step": 346 + }, + { + "epoch": 0.0260784608447317, + "grad_norm": 1.8765288795426107, + "learning_rate": 3.905105160153384e-06, + "loss": 1.1227, + "step": 347 + }, + { + "epoch": 0.026153614910566662, + "grad_norm": 1.1130165722410934, + "learning_rate": 3.907026359756579e-06, + "loss": 0.836, + "step": 348 + }, + { + "epoch": 0.026228768976401622, + "grad_norm": 1.8473364006773254, + "learning_rate": 3.908942046584326e-06, + "loss": 1.0618, + "step": 349 + }, + { + "epoch": 0.026303923042236586, + "grad_norm": 2.3982441479410608, + "learning_rate": 3.910852252183332e-06, + "loss": 1.1317, + "step": 350 + }, + { + "epoch": 0.026379077108071546, + "grad_norm": 1.0073221477537182, + "learning_rate": 3.912757007830285e-06, + "loss": 0.8716, + "step": 351 + }, + { + "epoch": 0.02645423117390651, + "grad_norm": 1.686899572859865, + "learning_rate": 3.914656344534934e-06, + "loss": 1.0687, + "step": 352 + }, + { + "epoch": 0.02652938523974147, + "grad_norm": 1.9736603551393361, + "learning_rate": 3.916550293043113e-06, + "loss": 1.0218, + "step": 353 + }, + { + "epoch": 0.026604539305576432, + "grad_norm": 1.5845278032342671, + "learning_rate": 3.9184388838397275e-06, + "loss": 0.9916, + "step": 354 + }, + { + "epoch": 0.026679693371411392, + "grad_norm": 1.4552873499431447, + "learning_rate": 3.9203221471517014e-06, + "loss": 0.9853, + "step": 355 + }, + { + "epoch": 0.026754847437246355, + "grad_norm": 1.4352790426968072, + "learning_rate": 3.92220011295088e-06, + "loss": 1.0978, + "step": 356 + }, + { + "epoch": 0.026830001503081315, + "grad_norm": 1.4701920592773277, + "learning_rate": 3.924072810956888e-06, + "loss": 1.025, + "step": 357 + }, + { + "epoch": 0.02690515556891628, + "grad_norm": 1.5692928666742398, + "learning_rate": 3.925940270639954e-06, + "loss": 1.0147, + "step": 358 + }, + { + "epoch": 0.02698030963475124, + "grad_norm": 1.9799270641495375, + "learning_rate": 3.927802521223695e-06, + "loss": 1.0214, + "step": 359 + }, + { + "epoch": 0.027055463700586202, + "grad_norm": 1.8944213835330963, + "learning_rate": 3.929659591687857e-06, + "loss": 1.0396, + "step": 360 + }, + { + "epoch": 0.027130617766421162, + "grad_norm": 1.7768733146087887, + "learning_rate": 3.931511510771023e-06, + "loss": 1.0273, + "step": 361 + }, + { + "epoch": 0.027205771832256125, + "grad_norm": 1.516236486665617, + "learning_rate": 3.93335830697328e-06, + "loss": 1.1444, + "step": 362 + }, + { + "epoch": 0.027280925898091085, + "grad_norm": 1.585058635866551, + "learning_rate": 3.935200008558854e-06, + "loss": 1.027, + "step": 363 + }, + { + "epoch": 0.02735607996392605, + "grad_norm": 1.6383430232825105, + "learning_rate": 3.937036643558701e-06, + "loss": 0.9529, + "step": 364 + }, + { + "epoch": 0.02743123402976101, + "grad_norm": 1.7694563355072093, + "learning_rate": 3.938868239773069e-06, + "loss": 1.0096, + "step": 365 + }, + { + "epoch": 0.027506388095595972, + "grad_norm": 1.6592798410490814, + "learning_rate": 3.940694824774024e-06, + "loss": 1.0889, + "step": 366 + }, + { + "epoch": 0.02758154216143093, + "grad_norm": 1.7290609932594208, + "learning_rate": 3.942516425907938e-06, + "loss": 1.0125, + "step": 367 + }, + { + "epoch": 0.027656696227265895, + "grad_norm": 2.3159688758603507, + "learning_rate": 3.944333070297947e-06, + "loss": 1.0621, + "step": 368 + }, + { + "epoch": 0.02773185029310086, + "grad_norm": 1.9285672909323897, + "learning_rate": 3.946144784846375e-06, + "loss": 1.0049, + "step": 369 + }, + { + "epoch": 0.02780700435893582, + "grad_norm": 1.571515206326821, + "learning_rate": 3.947951596237121e-06, + "loss": 0.9635, + "step": 370 + }, + { + "epoch": 0.02788215842477078, + "grad_norm": 1.4846985313794179, + "learning_rate": 3.949753530938024e-06, + "loss": 0.9869, + "step": 371 + }, + { + "epoch": 0.02795731249060574, + "grad_norm": 1.5136344658935796, + "learning_rate": 3.951550615203179e-06, + "loss": 1.0444, + "step": 372 + }, + { + "epoch": 0.028032466556440705, + "grad_norm": 1.8542683462773406, + "learning_rate": 3.953342875075244e-06, + "loss": 1.0023, + "step": 373 + }, + { + "epoch": 0.028107620622275665, + "grad_norm": 2.3212041201394364, + "learning_rate": 3.955130336387693e-06, + "loss": 0.9372, + "step": 374 + }, + { + "epoch": 0.028182774688110628, + "grad_norm": 1.5788479650310663, + "learning_rate": 3.956913024767059e-06, + "loss": 1.0389, + "step": 375 + }, + { + "epoch": 0.028257928753945588, + "grad_norm": 1.6777683383403834, + "learning_rate": 3.958690965635136e-06, + "loss": 1.0527, + "step": 376 + }, + { + "epoch": 0.02833308281978055, + "grad_norm": 1.8986395583137414, + "learning_rate": 3.96046418421115e-06, + "loss": 0.9492, + "step": 377 + }, + { + "epoch": 0.02840823688561551, + "grad_norm": 1.522026543249198, + "learning_rate": 3.962232705513913e-06, + "loss": 1.0089, + "step": 378 + }, + { + "epoch": 0.028483390951450475, + "grad_norm": 4.060961622788106, + "learning_rate": 3.963996554363933e-06, + "loss": 1.0915, + "step": 379 + }, + { + "epoch": 0.028558545017285435, + "grad_norm": 2.0199274084059287, + "learning_rate": 3.965755755385512e-06, + "loss": 1.0536, + "step": 380 + }, + { + "epoch": 0.028633699083120398, + "grad_norm": 1.59627519902807, + "learning_rate": 3.9675103330088e-06, + "loss": 0.9604, + "step": 381 + }, + { + "epoch": 0.028708853148955358, + "grad_norm": 1.6968243445603686, + "learning_rate": 3.969260311471838e-06, + "loss": 1.0565, + "step": 382 + }, + { + "epoch": 0.02878400721479032, + "grad_norm": 1.9047256378286117, + "learning_rate": 3.971005714822564e-06, + "loss": 1.0357, + "step": 383 + }, + { + "epoch": 0.02885916128062528, + "grad_norm": 1.5660209657573714, + "learning_rate": 3.9727465669207975e-06, + "loss": 1.0583, + "step": 384 + }, + { + "epoch": 0.028934315346460245, + "grad_norm": 1.798404341324898, + "learning_rate": 3.974482891440193e-06, + "loss": 0.9701, + "step": 385 + }, + { + "epoch": 0.029009469412295204, + "grad_norm": 1.5624740598432685, + "learning_rate": 3.976214711870174e-06, + "loss": 0.9913, + "step": 386 + }, + { + "epoch": 0.029084623478130168, + "grad_norm": 3.5638232534232985, + "learning_rate": 3.977942051517839e-06, + "loss": 1.0346, + "step": 387 + }, + { + "epoch": 0.029159777543965128, + "grad_norm": 1.625024428550796, + "learning_rate": 3.979664933509847e-06, + "loss": 1.0346, + "step": 388 + }, + { + "epoch": 0.02923493160980009, + "grad_norm": 3.2913258211100227, + "learning_rate": 3.9813833807942695e-06, + "loss": 0.9764, + "step": 389 + }, + { + "epoch": 0.02931008567563505, + "grad_norm": 3.021849538070568, + "learning_rate": 3.9830974161424286e-06, + "loss": 1.0454, + "step": 390 + }, + { + "epoch": 0.029385239741470014, + "grad_norm": 3.7880793918318028, + "learning_rate": 3.984807062150705e-06, + "loss": 1.0832, + "step": 391 + }, + { + "epoch": 0.029460393807304974, + "grad_norm": 2.05575754129798, + "learning_rate": 3.986512341242329e-06, + "loss": 1.0712, + "step": 392 + }, + { + "epoch": 0.029535547873139938, + "grad_norm": 1.6463205928048197, + "learning_rate": 3.98821327566914e-06, + "loss": 1.0054, + "step": 393 + }, + { + "epoch": 0.029610701938974898, + "grad_norm": 2.1399989639707537, + "learning_rate": 3.989909887513334e-06, + "loss": 1.1151, + "step": 394 + }, + { + "epoch": 0.02968585600480986, + "grad_norm": 3.0444302654477595, + "learning_rate": 3.991602198689179e-06, + "loss": 1.0439, + "step": 395 + }, + { + "epoch": 0.02976101007064482, + "grad_norm": 1.6610726733032704, + "learning_rate": 3.993290230944718e-06, + "loss": 1.0991, + "step": 396 + }, + { + "epoch": 0.029836164136479784, + "grad_norm": 1.6945222339944275, + "learning_rate": 3.994974005863441e-06, + "loss": 1.0218, + "step": 397 + }, + { + "epoch": 0.029911318202314744, + "grad_norm": 2.2579321384081705, + "learning_rate": 3.996653544865945e-06, + "loss": 0.9531, + "step": 398 + }, + { + "epoch": 0.029986472268149707, + "grad_norm": 1.737482809077782, + "learning_rate": 3.9983288692115676e-06, + "loss": 1.045, + "step": 399 + }, + { + "epoch": 0.030061626333984667, + "grad_norm": 1.7291403465872286, + "learning_rate": 4e-06, + "loss": 1.0652, + "step": 400 + }, + { + "epoch": 0.03013678039981963, + "grad_norm": 2.43779214917752, + "learning_rate": 3.999999940746171e-06, + "loss": 1.1364, + "step": 401 + }, + { + "epoch": 0.03021193446565459, + "grad_norm": 1.945369305987703, + "learning_rate": 3.999999762984686e-06, + "loss": 1.0956, + "step": 402 + }, + { + "epoch": 0.030287088531489554, + "grad_norm": 1.8076095374804355, + "learning_rate": 3.9999994667155576e-06, + "loss": 1.0058, + "step": 403 + }, + { + "epoch": 0.030362242597324514, + "grad_norm": 1.4746447022103182, + "learning_rate": 3.999999051938802e-06, + "loss": 0.9496, + "step": 404 + }, + { + "epoch": 0.030437396663159477, + "grad_norm": 1.5264912833681208, + "learning_rate": 3.9999985186544445e-06, + "loss": 0.9277, + "step": 405 + }, + { + "epoch": 0.030512550728994437, + "grad_norm": 1.9381015213002788, + "learning_rate": 3.999997866862515e-06, + "loss": 1.0455, + "step": 406 + }, + { + "epoch": 0.0305877047948294, + "grad_norm": 2.184852960383397, + "learning_rate": 3.9999970965630544e-06, + "loss": 1.1252, + "step": 407 + }, + { + "epoch": 0.03066285886066436, + "grad_norm": 1.3690235125944692, + "learning_rate": 3.9999962077561075e-06, + "loss": 1.0505, + "step": 408 + }, + { + "epoch": 0.030738012926499324, + "grad_norm": 1.9490669512457064, + "learning_rate": 3.999995200441726e-06, + "loss": 0.9904, + "step": 409 + }, + { + "epoch": 0.030813166992334284, + "grad_norm": 1.9265993558725971, + "learning_rate": 3.999994074619971e-06, + "loss": 1.0907, + "step": 410 + }, + { + "epoch": 0.030888321058169247, + "grad_norm": 1.8628656655621463, + "learning_rate": 3.999992830290909e-06, + "loss": 1.0212, + "step": 411 + }, + { + "epoch": 0.030963475124004207, + "grad_norm": 1.6300041468982953, + "learning_rate": 3.999991467454612e-06, + "loss": 1.0611, + "step": 412 + }, + { + "epoch": 0.03103862918983917, + "grad_norm": 1.8263701795321579, + "learning_rate": 3.999989986111163e-06, + "loss": 0.9874, + "step": 413 + }, + { + "epoch": 0.031113783255674134, + "grad_norm": 1.8603985958211353, + "learning_rate": 3.999988386260648e-06, + "loss": 1.0476, + "step": 414 + }, + { + "epoch": 0.031188937321509094, + "grad_norm": 2.027097242359248, + "learning_rate": 3.999986667903163e-06, + "loss": 0.9841, + "step": 415 + }, + { + "epoch": 0.03126409138734405, + "grad_norm": 1.4595882284503596, + "learning_rate": 3.999984831038811e-06, + "loss": 1.0567, + "step": 416 + }, + { + "epoch": 0.03133924545317902, + "grad_norm": 2.1287350087021566, + "learning_rate": 3.999982875667697e-06, + "loss": 0.9922, + "step": 417 + }, + { + "epoch": 0.03141439951901398, + "grad_norm": 2.0914768006170976, + "learning_rate": 3.999980801789941e-06, + "loss": 1.0022, + "step": 418 + }, + { + "epoch": 0.031489553584848944, + "grad_norm": 1.6346059924067289, + "learning_rate": 3.999978609405662e-06, + "loss": 1.0062, + "step": 419 + }, + { + "epoch": 0.0315647076506839, + "grad_norm": 1.653376696053853, + "learning_rate": 3.999976298514994e-06, + "loss": 1.0691, + "step": 420 + }, + { + "epoch": 0.03163986171651886, + "grad_norm": 1.9920522008092312, + "learning_rate": 3.999973869118071e-06, + "loss": 1.183, + "step": 421 + }, + { + "epoch": 0.03171501578235383, + "grad_norm": 1.8768523920598579, + "learning_rate": 3.999971321215038e-06, + "loss": 1.1097, + "step": 422 + }, + { + "epoch": 0.03179016984818879, + "grad_norm": 1.6836531555599357, + "learning_rate": 3.999968654806046e-06, + "loss": 1.0282, + "step": 423 + }, + { + "epoch": 0.03186532391402375, + "grad_norm": 1.9454319315943487, + "learning_rate": 3.999965869891253e-06, + "loss": 0.8539, + "step": 424 + }, + { + "epoch": 0.03194047797985871, + "grad_norm": 1.2449075918871475, + "learning_rate": 3.999962966470823e-06, + "loss": 0.8887, + "step": 425 + }, + { + "epoch": 0.03201563204569367, + "grad_norm": 1.9794284714323085, + "learning_rate": 3.9999599445449295e-06, + "loss": 1.0639, + "step": 426 + }, + { + "epoch": 0.03209078611152864, + "grad_norm": 1.7962988551749013, + "learning_rate": 3.999956804113751e-06, + "loss": 1.0019, + "step": 427 + }, + { + "epoch": 0.03216594017736359, + "grad_norm": 1.9855007072453235, + "learning_rate": 3.999953545177472e-06, + "loss": 1.1091, + "step": 428 + }, + { + "epoch": 0.032241094243198556, + "grad_norm": 1.664827842403691, + "learning_rate": 3.9999501677362885e-06, + "loss": 1.0018, + "step": 429 + }, + { + "epoch": 0.03231624830903352, + "grad_norm": 1.5525520265618393, + "learning_rate": 3.9999466717903995e-06, + "loss": 0.9939, + "step": 430 + }, + { + "epoch": 0.03239140237486848, + "grad_norm": 1.7630094887515435, + "learning_rate": 3.999943057340012e-06, + "loss": 1.0579, + "step": 431 + }, + { + "epoch": 0.03246655644070344, + "grad_norm": 1.7490924798148741, + "learning_rate": 3.999939324385339e-06, + "loss": 1.0437, + "step": 432 + }, + { + "epoch": 0.0325417105065384, + "grad_norm": 1.6089793339238445, + "learning_rate": 3.999935472926604e-06, + "loss": 1.015, + "step": 433 + }, + { + "epoch": 0.032616864572373366, + "grad_norm": 25.069410479643814, + "learning_rate": 3.9999315029640325e-06, + "loss": 1.1092, + "step": 434 + }, + { + "epoch": 0.03269201863820833, + "grad_norm": 1.7870418681751248, + "learning_rate": 3.999927414497862e-06, + "loss": 1.0041, + "step": 435 + }, + { + "epoch": 0.032767172704043286, + "grad_norm": 1.6398171025147004, + "learning_rate": 3.999923207528334e-06, + "loss": 0.9685, + "step": 436 + }, + { + "epoch": 0.03284232676987825, + "grad_norm": 2.049231766758661, + "learning_rate": 3.999918882055698e-06, + "loss": 1.0333, + "step": 437 + }, + { + "epoch": 0.03291748083571321, + "grad_norm": 1.6321020451553179, + "learning_rate": 3.9999144380802095e-06, + "loss": 1.0932, + "step": 438 + }, + { + "epoch": 0.032992634901548176, + "grad_norm": 2.052013325021471, + "learning_rate": 3.999909875602132e-06, + "loss": 1.0163, + "step": 439 + }, + { + "epoch": 0.03306778896738313, + "grad_norm": 0.9066139674181418, + "learning_rate": 3.999905194621737e-06, + "loss": 0.8184, + "step": 440 + }, + { + "epoch": 0.033142943033218096, + "grad_norm": 1.7181348219917223, + "learning_rate": 3.999900395139301e-06, + "loss": 1.0476, + "step": 441 + }, + { + "epoch": 0.03321809709905306, + "grad_norm": 2.096602702759632, + "learning_rate": 3.999895477155108e-06, + "loss": 0.9952, + "step": 442 + }, + { + "epoch": 0.03329325116488802, + "grad_norm": 1.630965506932089, + "learning_rate": 3.9998904406694504e-06, + "loss": 1.0871, + "step": 443 + }, + { + "epoch": 0.03336840523072298, + "grad_norm": 1.3998858234986158, + "learning_rate": 3.999885285682626e-06, + "loss": 1.1237, + "step": 444 + }, + { + "epoch": 0.03344355929655794, + "grad_norm": 1.5367977306275737, + "learning_rate": 3.99988001219494e-06, + "loss": 1.0975, + "step": 445 + }, + { + "epoch": 0.033518713362392906, + "grad_norm": 1.558173407373807, + "learning_rate": 3.999874620206705e-06, + "loss": 1.0427, + "step": 446 + }, + { + "epoch": 0.03359386742822787, + "grad_norm": 1.5112455351884264, + "learning_rate": 3.999869109718242e-06, + "loss": 1.0234, + "step": 447 + }, + { + "epoch": 0.033669021494062826, + "grad_norm": 2.3757799775139947, + "learning_rate": 3.999863480729875e-06, + "loss": 1.0348, + "step": 448 + }, + { + "epoch": 0.03374417555989779, + "grad_norm": 1.9828015795347607, + "learning_rate": 3.999857733241938e-06, + "loss": 0.9979, + "step": 449 + }, + { + "epoch": 0.03381932962573275, + "grad_norm": 1.5251477935966147, + "learning_rate": 3.999851867254774e-06, + "loss": 1.0178, + "step": 450 + }, + { + "epoch": 0.033894483691567716, + "grad_norm": 2.2571923483308374, + "learning_rate": 3.9998458827687286e-06, + "loss": 0.9825, + "step": 451 + }, + { + "epoch": 0.03396963775740267, + "grad_norm": 0.7749450325700165, + "learning_rate": 3.999839779784157e-06, + "loss": 0.8038, + "step": 452 + }, + { + "epoch": 0.034044791823237636, + "grad_norm": 1.7134796486929704, + "learning_rate": 3.999833558301419e-06, + "loss": 0.9731, + "step": 453 + }, + { + "epoch": 0.0341199458890726, + "grad_norm": 1.6485414878986206, + "learning_rate": 3.999827218320886e-06, + "loss": 1.0113, + "step": 454 + }, + { + "epoch": 0.03419509995490756, + "grad_norm": 0.9399696246778816, + "learning_rate": 3.999820759842933e-06, + "loss": 0.8339, + "step": 455 + }, + { + "epoch": 0.03427025402074252, + "grad_norm": 2.3114928036824667, + "learning_rate": 3.999814182867941e-06, + "loss": 1.1062, + "step": 456 + }, + { + "epoch": 0.03434540808657748, + "grad_norm": 1.5161507864473351, + "learning_rate": 3.999807487396301e-06, + "loss": 1.1242, + "step": 457 + }, + { + "epoch": 0.034420562152412446, + "grad_norm": 1.6054198914307476, + "learning_rate": 3.999800673428411e-06, + "loss": 0.9795, + "step": 458 + }, + { + "epoch": 0.03449571621824741, + "grad_norm": 1.598347409451224, + "learning_rate": 3.999793740964672e-06, + "loss": 1.0416, + "step": 459 + }, + { + "epoch": 0.03457087028408237, + "grad_norm": 1.634697733828642, + "learning_rate": 3.999786690005496e-06, + "loss": 1.0572, + "step": 460 + }, + { + "epoch": 0.03464602434991733, + "grad_norm": 1.9137465235643063, + "learning_rate": 3.999779520551302e-06, + "loss": 0.8958, + "step": 461 + }, + { + "epoch": 0.03472117841575229, + "grad_norm": 1.027621213626881, + "learning_rate": 3.9997722326025135e-06, + "loss": 0.8134, + "step": 462 + }, + { + "epoch": 0.034796332481587255, + "grad_norm": 1.8322593042733644, + "learning_rate": 3.999764826159562e-06, + "loss": 0.9804, + "step": 463 + }, + { + "epoch": 0.03487148654742222, + "grad_norm": 1.911834167064376, + "learning_rate": 3.999757301222887e-06, + "loss": 1.0262, + "step": 464 + }, + { + "epoch": 0.034946640613257175, + "grad_norm": 1.913288131084662, + "learning_rate": 3.999749657792934e-06, + "loss": 1.0014, + "step": 465 + }, + { + "epoch": 0.03502179467909214, + "grad_norm": 1.461960296817121, + "learning_rate": 3.999741895870157e-06, + "loss": 0.9991, + "step": 466 + }, + { + "epoch": 0.0350969487449271, + "grad_norm": 1.8766141936767875, + "learning_rate": 3.9997340154550145e-06, + "loss": 1.0185, + "step": 467 + }, + { + "epoch": 0.035172102810762065, + "grad_norm": 1.3772474042113572, + "learning_rate": 3.999726016547974e-06, + "loss": 1.0618, + "step": 468 + }, + { + "epoch": 0.03524725687659702, + "grad_norm": 1.3016174652839874, + "learning_rate": 3.9997178991495105e-06, + "loss": 1.032, + "step": 469 + }, + { + "epoch": 0.035322410942431985, + "grad_norm": 1.486776397116659, + "learning_rate": 3.9997096632601035e-06, + "loss": 1.0469, + "step": 470 + }, + { + "epoch": 0.03539756500826695, + "grad_norm": 1.318796088330846, + "learning_rate": 3.999701308880242e-06, + "loss": 0.988, + "step": 471 + }, + { + "epoch": 0.03547271907410191, + "grad_norm": 1.39314992328733, + "learning_rate": 3.999692836010419e-06, + "loss": 1.0618, + "step": 472 + }, + { + "epoch": 0.03554787313993687, + "grad_norm": 1.6151742981666877, + "learning_rate": 3.99968424465114e-06, + "loss": 1.1669, + "step": 473 + }, + { + "epoch": 0.03562302720577183, + "grad_norm": 1.7526388027931, + "learning_rate": 3.999675534802911e-06, + "loss": 1.0008, + "step": 474 + }, + { + "epoch": 0.035698181271606795, + "grad_norm": 1.4502798961286294, + "learning_rate": 3.99966670646625e-06, + "loss": 0.922, + "step": 475 + }, + { + "epoch": 0.03577333533744176, + "grad_norm": 1.3962849507956316, + "learning_rate": 3.999657759641679e-06, + "loss": 1.0494, + "step": 476 + }, + { + "epoch": 0.035848489403276715, + "grad_norm": 1.8322185103440083, + "learning_rate": 3.999648694329729e-06, + "loss": 1.1096, + "step": 477 + }, + { + "epoch": 0.03592364346911168, + "grad_norm": 1.2823975703957116, + "learning_rate": 3.9996395105309365e-06, + "loss": 1.0036, + "step": 478 + }, + { + "epoch": 0.03599879753494664, + "grad_norm": 1.4449587827301726, + "learning_rate": 3.999630208245846e-06, + "loss": 0.9557, + "step": 479 + }, + { + "epoch": 0.036073951600781605, + "grad_norm": 1.4698094544211873, + "learning_rate": 3.9996207874750075e-06, + "loss": 1.0568, + "step": 480 + }, + { + "epoch": 0.03614910566661656, + "grad_norm": 1.4318228861909017, + "learning_rate": 3.999611248218982e-06, + "loss": 1.0759, + "step": 481 + }, + { + "epoch": 0.036224259732451525, + "grad_norm": 2.6961002691339884, + "learning_rate": 3.999601590478332e-06, + "loss": 0.895, + "step": 482 + }, + { + "epoch": 0.03629941379828649, + "grad_norm": 1.6151116357962985, + "learning_rate": 3.99959181425363e-06, + "loss": 1.1191, + "step": 483 + }, + { + "epoch": 0.03637456786412145, + "grad_norm": 1.7844066423326286, + "learning_rate": 3.999581919545458e-06, + "loss": 1.0173, + "step": 484 + }, + { + "epoch": 0.03644972192995641, + "grad_norm": 1.7201940539032525, + "learning_rate": 3.999571906354399e-06, + "loss": 1.1319, + "step": 485 + }, + { + "epoch": 0.03652487599579137, + "grad_norm": 1.9115334755024063, + "learning_rate": 3.999561774681048e-06, + "loss": 1.0803, + "step": 486 + }, + { + "epoch": 0.036600030061626335, + "grad_norm": 1.3031972265924738, + "learning_rate": 3.999551524526005e-06, + "loss": 0.9165, + "step": 487 + }, + { + "epoch": 0.0366751841274613, + "grad_norm": 1.7106888821220578, + "learning_rate": 3.9995411558898775e-06, + "loss": 0.9951, + "step": 488 + }, + { + "epoch": 0.036750338193296254, + "grad_norm": 1.7683509106830575, + "learning_rate": 3.9995306687732795e-06, + "loss": 1.0104, + "step": 489 + }, + { + "epoch": 0.03682549225913122, + "grad_norm": 1.106999876968253, + "learning_rate": 3.9995200631768326e-06, + "loss": 0.8899, + "step": 490 + }, + { + "epoch": 0.03690064632496618, + "grad_norm": 2.0315763944870406, + "learning_rate": 3.999509339101166e-06, + "loss": 1.0384, + "step": 491 + }, + { + "epoch": 0.036975800390801145, + "grad_norm": 1.5271550132647946, + "learning_rate": 3.999498496546914e-06, + "loss": 1.0115, + "step": 492 + }, + { + "epoch": 0.0370509544566361, + "grad_norm": 1.5594725461734416, + "learning_rate": 3.99948753551472e-06, + "loss": 0.9934, + "step": 493 + }, + { + "epoch": 0.037126108522471064, + "grad_norm": 4.606497380356834, + "learning_rate": 3.999476456005232e-06, + "loss": 1.0733, + "step": 494 + }, + { + "epoch": 0.03720126258830603, + "grad_norm": 1.690959633587353, + "learning_rate": 3.999465258019108e-06, + "loss": 0.9242, + "step": 495 + }, + { + "epoch": 0.03727641665414099, + "grad_norm": 1.6887172264911534, + "learning_rate": 3.999453941557011e-06, + "loss": 0.9485, + "step": 496 + }, + { + "epoch": 0.03735157071997595, + "grad_norm": 2.373727859305427, + "learning_rate": 3.9994425066196105e-06, + "loss": 1.0347, + "step": 497 + }, + { + "epoch": 0.03742672478581091, + "grad_norm": 1.6437084468332326, + "learning_rate": 3.999430953207586e-06, + "loss": 1.0655, + "step": 498 + }, + { + "epoch": 0.037501878851645874, + "grad_norm": 1.6009150242982304, + "learning_rate": 3.999419281321621e-06, + "loss": 1.0409, + "step": 499 + }, + { + "epoch": 0.03757703291748084, + "grad_norm": 2.066548275889245, + "learning_rate": 3.999407490962408e-06, + "loss": 0.9587, + "step": 500 + }, + { + "epoch": 0.037652186983315794, + "grad_norm": 1.5349481261990812, + "learning_rate": 3.999395582130644e-06, + "loss": 0.9413, + "step": 501 + }, + { + "epoch": 0.03772734104915076, + "grad_norm": 1.6954382221348212, + "learning_rate": 3.999383554827037e-06, + "loss": 1.0078, + "step": 502 + }, + { + "epoch": 0.03780249511498572, + "grad_norm": 1.5181787743119992, + "learning_rate": 3.999371409052297e-06, + "loss": 1.0955, + "step": 503 + }, + { + "epoch": 0.037877649180820684, + "grad_norm": 1.9270376938922706, + "learning_rate": 3.999359144807145e-06, + "loss": 1.0707, + "step": 504 + }, + { + "epoch": 0.03795280324665565, + "grad_norm": 1.559123166835187, + "learning_rate": 3.999346762092307e-06, + "loss": 1.1243, + "step": 505 + }, + { + "epoch": 0.038027957312490604, + "grad_norm": 1.5812954922491156, + "learning_rate": 3.999334260908518e-06, + "loss": 1.0058, + "step": 506 + }, + { + "epoch": 0.03810311137832557, + "grad_norm": 1.341590165786663, + "learning_rate": 3.999321641256519e-06, + "loss": 1.0213, + "step": 507 + }, + { + "epoch": 0.03817826544416053, + "grad_norm": 1.6139804165584475, + "learning_rate": 3.999308903137056e-06, + "loss": 0.9943, + "step": 508 + }, + { + "epoch": 0.038253419509995494, + "grad_norm": 1.6181905832920238, + "learning_rate": 3.999296046550884e-06, + "loss": 0.9815, + "step": 509 + }, + { + "epoch": 0.03832857357583045, + "grad_norm": 1.5852001712405286, + "learning_rate": 3.999283071498766e-06, + "loss": 1.0501, + "step": 510 + }, + { + "epoch": 0.038403727641665414, + "grad_norm": 2.0160090354156863, + "learning_rate": 3.9992699779814704e-06, + "loss": 1.0334, + "step": 511 + }, + { + "epoch": 0.03847888170750038, + "grad_norm": 1.5895530941121359, + "learning_rate": 3.999256765999773e-06, + "loss": 0.9981, + "step": 512 + }, + { + "epoch": 0.03855403577333534, + "grad_norm": 1.676573950452001, + "learning_rate": 3.999243435554456e-06, + "loss": 1.0193, + "step": 513 + }, + { + "epoch": 0.0386291898391703, + "grad_norm": 1.686423558832284, + "learning_rate": 3.999229986646311e-06, + "loss": 1.042, + "step": 514 + }, + { + "epoch": 0.03870434390500526, + "grad_norm": 1.6232914378865164, + "learning_rate": 3.999216419276132e-06, + "loss": 1.1034, + "step": 515 + }, + { + "epoch": 0.038779497970840224, + "grad_norm": 1.6769780507024659, + "learning_rate": 3.999202733444726e-06, + "loss": 0.9859, + "step": 516 + }, + { + "epoch": 0.03885465203667519, + "grad_norm": 1.8121618415817167, + "learning_rate": 3.999188929152902e-06, + "loss": 1.0648, + "step": 517 + }, + { + "epoch": 0.038929806102510144, + "grad_norm": 1.780041401735218, + "learning_rate": 3.999175006401478e-06, + "loss": 0.9883, + "step": 518 + }, + { + "epoch": 0.03900496016834511, + "grad_norm": 1.9128227734426178, + "learning_rate": 3.999160965191281e-06, + "loss": 1.0158, + "step": 519 + }, + { + "epoch": 0.03908011423418007, + "grad_norm": 1.6984941713481856, + "learning_rate": 3.99914680552314e-06, + "loss": 0.9725, + "step": 520 + }, + { + "epoch": 0.039155268300015034, + "grad_norm": 2.079723689726679, + "learning_rate": 3.999132527397897e-06, + "loss": 1.1001, + "step": 521 + }, + { + "epoch": 0.03923042236584999, + "grad_norm": 1.5757690822602466, + "learning_rate": 3.999118130816395e-06, + "loss": 0.9693, + "step": 522 + }, + { + "epoch": 0.03930557643168495, + "grad_norm": 1.8235929342801986, + "learning_rate": 3.999103615779489e-06, + "loss": 1.0089, + "step": 523 + }, + { + "epoch": 0.03938073049751992, + "grad_norm": 0.8959631640960085, + "learning_rate": 3.99908898228804e-06, + "loss": 0.8316, + "step": 524 + }, + { + "epoch": 0.03945588456335488, + "grad_norm": 1.4175505218383264, + "learning_rate": 3.999074230342913e-06, + "loss": 1.0258, + "step": 525 + }, + { + "epoch": 0.03953103862918984, + "grad_norm": 1.6097583499365455, + "learning_rate": 3.999059359944982e-06, + "loss": 0.9763, + "step": 526 + }, + { + "epoch": 0.0396061926950248, + "grad_norm": 1.6334817783213966, + "learning_rate": 3.99904437109513e-06, + "loss": 1.0594, + "step": 527 + }, + { + "epoch": 0.03968134676085976, + "grad_norm": 1.522981233006529, + "learning_rate": 3.999029263794244e-06, + "loss": 1.0436, + "step": 528 + }, + { + "epoch": 0.03975650082669473, + "grad_norm": 1.5512932765681469, + "learning_rate": 3.999014038043219e-06, + "loss": 1.0294, + "step": 529 + }, + { + "epoch": 0.03983165489252968, + "grad_norm": 3.925235235855211, + "learning_rate": 3.9989986938429574e-06, + "loss": 1.0917, + "step": 530 + }, + { + "epoch": 0.03990680895836465, + "grad_norm": 1.9990121070529532, + "learning_rate": 3.9989832311943695e-06, + "loss": 1.0259, + "step": 531 + }, + { + "epoch": 0.03998196302419961, + "grad_norm": 1.5609602398595435, + "learning_rate": 3.99896765009837e-06, + "loss": 0.9858, + "step": 532 + }, + { + "epoch": 0.04005711709003457, + "grad_norm": 1.5217881101521058, + "learning_rate": 3.998951950555883e-06, + "loss": 0.9678, + "step": 533 + }, + { + "epoch": 0.04013227115586953, + "grad_norm": 1.008688553329571, + "learning_rate": 3.998936132567837e-06, + "loss": 0.8212, + "step": 534 + }, + { + "epoch": 0.04020742522170449, + "grad_norm": 1.5230629484977534, + "learning_rate": 3.998920196135172e-06, + "loss": 1.0131, + "step": 535 + }, + { + "epoch": 0.040282579287539456, + "grad_norm": 1.449434906912843, + "learning_rate": 3.998904141258831e-06, + "loss": 1.0432, + "step": 536 + }, + { + "epoch": 0.04035773335337442, + "grad_norm": 1.3227028511251395, + "learning_rate": 3.9988879679397644e-06, + "loss": 1.0699, + "step": 537 + }, + { + "epoch": 0.040432887419209376, + "grad_norm": 1.5174888083785254, + "learning_rate": 3.9988716761789324e-06, + "loss": 1.076, + "step": 538 + }, + { + "epoch": 0.04050804148504434, + "grad_norm": 1.8387298127498353, + "learning_rate": 3.998855265977299e-06, + "loss": 1.0398, + "step": 539 + }, + { + "epoch": 0.0405831955508793, + "grad_norm": 1.5780158430060829, + "learning_rate": 3.998838737335837e-06, + "loss": 1.0148, + "step": 540 + }, + { + "epoch": 0.040658349616714266, + "grad_norm": 2.0247235427335535, + "learning_rate": 3.998822090255526e-06, + "loss": 1.0554, + "step": 541 + }, + { + "epoch": 0.04073350368254922, + "grad_norm": 1.6993085464778939, + "learning_rate": 3.9988053247373515e-06, + "loss": 1.0013, + "step": 542 + }, + { + "epoch": 0.040808657748384186, + "grad_norm": 1.8602462137986326, + "learning_rate": 3.998788440782309e-06, + "loss": 0.9981, + "step": 543 + }, + { + "epoch": 0.04088381181421915, + "grad_norm": 2.0903820449773347, + "learning_rate": 3.998771438391396e-06, + "loss": 0.9987, + "step": 544 + }, + { + "epoch": 0.04095896588005411, + "grad_norm": 1.6984814677649394, + "learning_rate": 3.9987543175656214e-06, + "loss": 0.9909, + "step": 545 + }, + { + "epoch": 0.04103411994588907, + "grad_norm": 1.5247895330059817, + "learning_rate": 3.998737078306001e-06, + "loss": 1.0911, + "step": 546 + }, + { + "epoch": 0.04110927401172403, + "grad_norm": 1.8471706147342601, + "learning_rate": 3.998719720613554e-06, + "loss": 1.0563, + "step": 547 + }, + { + "epoch": 0.041184428077558996, + "grad_norm": 1.0603830340934473, + "learning_rate": 3.99870224448931e-06, + "loss": 0.9039, + "step": 548 + }, + { + "epoch": 0.04125958214339396, + "grad_norm": 1.5261551312479342, + "learning_rate": 3.998684649934305e-06, + "loss": 1.0409, + "step": 549 + }, + { + "epoch": 0.041334736209228916, + "grad_norm": 1.5288262940377852, + "learning_rate": 3.9986669369495805e-06, + "loss": 1.0182, + "step": 550 + }, + { + "epoch": 0.04140989027506388, + "grad_norm": 1.6728170004958185, + "learning_rate": 3.998649105536187e-06, + "loss": 1.0478, + "step": 551 + }, + { + "epoch": 0.04148504434089884, + "grad_norm": 1.7333108197486295, + "learning_rate": 3.998631155695181e-06, + "loss": 1.0575, + "step": 552 + }, + { + "epoch": 0.041560198406733806, + "grad_norm": 2.5909855829777846, + "learning_rate": 3.9986130874276244e-06, + "loss": 1.0369, + "step": 553 + }, + { + "epoch": 0.04163535247256877, + "grad_norm": 1.567236929786206, + "learning_rate": 3.998594900734591e-06, + "loss": 1.0581, + "step": 554 + }, + { + "epoch": 0.041710506538403726, + "grad_norm": 1.923340029691061, + "learning_rate": 3.998576595617155e-06, + "loss": 0.9822, + "step": 555 + }, + { + "epoch": 0.04178566060423869, + "grad_norm": 1.8117034082392347, + "learning_rate": 3.998558172076404e-06, + "loss": 1.126, + "step": 556 + }, + { + "epoch": 0.04186081467007365, + "grad_norm": 2.6954727788621637, + "learning_rate": 3.998539630113427e-06, + "loss": 1.0153, + "step": 557 + }, + { + "epoch": 0.041935968735908616, + "grad_norm": 1.7578031412511284, + "learning_rate": 3.998520969729325e-06, + "loss": 1.0742, + "step": 558 + }, + { + "epoch": 0.04201112280174357, + "grad_norm": 1.703772325600369, + "learning_rate": 3.998502190925202e-06, + "loss": 1.0738, + "step": 559 + }, + { + "epoch": 0.042086276867578536, + "grad_norm": 1.3727839558684192, + "learning_rate": 3.998483293702172e-06, + "loss": 0.9959, + "step": 560 + }, + { + "epoch": 0.0421614309334135, + "grad_norm": 0.8146167837691205, + "learning_rate": 3.998464278061353e-06, + "loss": 0.8181, + "step": 561 + }, + { + "epoch": 0.04223658499924846, + "grad_norm": 2.652673355319722, + "learning_rate": 3.998445144003874e-06, + "loss": 1.0083, + "step": 562 + }, + { + "epoch": 0.04231173906508342, + "grad_norm": 1.4183971924232799, + "learning_rate": 3.9984258915308674e-06, + "loss": 1.0784, + "step": 563 + }, + { + "epoch": 0.04238689313091838, + "grad_norm": 0.848320380007685, + "learning_rate": 3.998406520643475e-06, + "loss": 0.7649, + "step": 564 + }, + { + "epoch": 0.042462047196753346, + "grad_norm": 1.4026827601659646, + "learning_rate": 3.998387031342843e-06, + "loss": 1.0117, + "step": 565 + }, + { + "epoch": 0.04253720126258831, + "grad_norm": 1.5917846679219436, + "learning_rate": 3.998367423630127e-06, + "loss": 1.0269, + "step": 566 + }, + { + "epoch": 0.042612355328423265, + "grad_norm": 1.6429450207065521, + "learning_rate": 3.9983476975064885e-06, + "loss": 1.0596, + "step": 567 + }, + { + "epoch": 0.04268750939425823, + "grad_norm": 1.7034007557902104, + "learning_rate": 3.998327852973098e-06, + "loss": 1.0142, + "step": 568 + }, + { + "epoch": 0.04276266346009319, + "grad_norm": 1.5807127933397882, + "learning_rate": 3.998307890031129e-06, + "loss": 1.0497, + "step": 569 + }, + { + "epoch": 0.042837817525928155, + "grad_norm": 1.6003119820175122, + "learning_rate": 3.998287808681766e-06, + "loss": 1.0518, + "step": 570 + }, + { + "epoch": 0.04291297159176311, + "grad_norm": 1.0757885369612954, + "learning_rate": 3.998267608926198e-06, + "loss": 0.8702, + "step": 571 + }, + { + "epoch": 0.042988125657598075, + "grad_norm": 1.5947545943217187, + "learning_rate": 3.998247290765623e-06, + "loss": 1.108, + "step": 572 + }, + { + "epoch": 0.04306327972343304, + "grad_norm": 1.8212092077835607, + "learning_rate": 3.9982268542012435e-06, + "loss": 0.9797, + "step": 573 + }, + { + "epoch": 0.043138433789268, + "grad_norm": 1.883719388109343, + "learning_rate": 3.998206299234272e-06, + "loss": 0.9094, + "step": 574 + }, + { + "epoch": 0.04321358785510296, + "grad_norm": 1.6265459776230666, + "learning_rate": 3.998185625865924e-06, + "loss": 1.0727, + "step": 575 + }, + { + "epoch": 0.04328874192093792, + "grad_norm": 1.9716366389555093, + "learning_rate": 3.998164834097428e-06, + "loss": 0.9985, + "step": 576 + }, + { + "epoch": 0.043363895986772885, + "grad_norm": 1.0848691886063173, + "learning_rate": 3.998143923930013e-06, + "loss": 0.8436, + "step": 577 + }, + { + "epoch": 0.04343905005260785, + "grad_norm": 1.5780745394353717, + "learning_rate": 3.998122895364919e-06, + "loss": 1.0599, + "step": 578 + }, + { + "epoch": 0.043514204118442805, + "grad_norm": 2.7396754743700034, + "learning_rate": 3.998101748403393e-06, + "loss": 1.016, + "step": 579 + }, + { + "epoch": 0.04358935818427777, + "grad_norm": 1.591375489422198, + "learning_rate": 3.998080483046687e-06, + "loss": 0.9779, + "step": 580 + }, + { + "epoch": 0.04366451225011273, + "grad_norm": 2.132513898690156, + "learning_rate": 3.998059099296061e-06, + "loss": 1.0358, + "step": 581 + }, + { + "epoch": 0.043739666315947695, + "grad_norm": 1.6418001598295342, + "learning_rate": 3.9980375971527814e-06, + "loss": 1.0912, + "step": 582 + }, + { + "epoch": 0.04381482038178265, + "grad_norm": 1.6446211330469545, + "learning_rate": 3.998015976618124e-06, + "loss": 1.0636, + "step": 583 + }, + { + "epoch": 0.043889974447617615, + "grad_norm": 1.4817863780872849, + "learning_rate": 3.997994237693369e-06, + "loss": 1.0912, + "step": 584 + }, + { + "epoch": 0.04396512851345258, + "grad_norm": 0.9018014045871905, + "learning_rate": 3.997972380379804e-06, + "loss": 0.8115, + "step": 585 + }, + { + "epoch": 0.04404028257928754, + "grad_norm": 1.5576499433272117, + "learning_rate": 3.997950404678726e-06, + "loss": 1.0934, + "step": 586 + }, + { + "epoch": 0.0441154366451225, + "grad_norm": 1.250480559480503, + "learning_rate": 3.997928310591435e-06, + "loss": 0.9474, + "step": 587 + }, + { + "epoch": 0.04419059071095746, + "grad_norm": 2.317303694360269, + "learning_rate": 3.997906098119241e-06, + "loss": 1.0298, + "step": 588 + }, + { + "epoch": 0.044265744776792425, + "grad_norm": 1.9935923690426358, + "learning_rate": 3.997883767263461e-06, + "loss": 1.0213, + "step": 589 + }, + { + "epoch": 0.04434089884262739, + "grad_norm": 1.5800047151277903, + "learning_rate": 3.997861318025417e-06, + "loss": 0.9567, + "step": 590 + }, + { + "epoch": 0.044416052908462345, + "grad_norm": 1.503488378557885, + "learning_rate": 3.997838750406439e-06, + "loss": 0.9216, + "step": 591 + }, + { + "epoch": 0.04449120697429731, + "grad_norm": 1.5901088634146296, + "learning_rate": 3.997816064407865e-06, + "loss": 1.0277, + "step": 592 + }, + { + "epoch": 0.04456636104013227, + "grad_norm": 2.0805358669615335, + "learning_rate": 3.997793260031039e-06, + "loss": 1.0173, + "step": 593 + }, + { + "epoch": 0.044641515105967235, + "grad_norm": 1.550646413326977, + "learning_rate": 3.997770337277313e-06, + "loss": 1.0407, + "step": 594 + }, + { + "epoch": 0.04471666917180219, + "grad_norm": 1.3016363317607598, + "learning_rate": 3.997747296148044e-06, + "loss": 0.8667, + "step": 595 + }, + { + "epoch": 0.044791823237637154, + "grad_norm": 1.7194349831338098, + "learning_rate": 3.997724136644597e-06, + "loss": 1.0395, + "step": 596 + }, + { + "epoch": 0.04486697730347212, + "grad_norm": 1.868560350165684, + "learning_rate": 3.997700858768346e-06, + "loss": 0.9956, + "step": 597 + }, + { + "epoch": 0.04494213136930708, + "grad_norm": 1.502803646005244, + "learning_rate": 3.99767746252067e-06, + "loss": 1.0146, + "step": 598 + }, + { + "epoch": 0.045017285435142045, + "grad_norm": 1.8471173071008258, + "learning_rate": 3.997653947902954e-06, + "loss": 1.053, + "step": 599 + }, + { + "epoch": 0.045092439500977, + "grad_norm": 1.4710061621053891, + "learning_rate": 3.997630314916592e-06, + "loss": 1.1035, + "step": 600 + }, + { + "epoch": 0.045167593566811964, + "grad_norm": 1.7973876221906884, + "learning_rate": 3.9976065635629845e-06, + "loss": 0.9968, + "step": 601 + }, + { + "epoch": 0.04524274763264693, + "grad_norm": 0.9843937705536138, + "learning_rate": 3.997582693843539e-06, + "loss": 0.8018, + "step": 602 + }, + { + "epoch": 0.04531790169848189, + "grad_norm": 2.2601218903003644, + "learning_rate": 3.997558705759669e-06, + "loss": 1.0485, + "step": 603 + }, + { + "epoch": 0.04539305576431685, + "grad_norm": 1.4481012606331656, + "learning_rate": 3.9975345993127975e-06, + "loss": 1.0415, + "step": 604 + }, + { + "epoch": 0.04546820983015181, + "grad_norm": 3.4307457892165445, + "learning_rate": 3.997510374504351e-06, + "loss": 1.0669, + "step": 605 + }, + { + "epoch": 0.045543363895986774, + "grad_norm": 2.216055509980612, + "learning_rate": 3.9974860313357665e-06, + "loss": 1.0709, + "step": 606 + }, + { + "epoch": 0.04561851796182174, + "grad_norm": 1.6012464257944137, + "learning_rate": 3.997461569808485e-06, + "loss": 0.9552, + "step": 607 + }, + { + "epoch": 0.045693672027656694, + "grad_norm": 1.1706002995644296, + "learning_rate": 3.997436989923957e-06, + "loss": 0.8099, + "step": 608 + }, + { + "epoch": 0.04576882609349166, + "grad_norm": 1.3656420883401497, + "learning_rate": 3.997412291683639e-06, + "loss": 1.0135, + "step": 609 + }, + { + "epoch": 0.04584398015932662, + "grad_norm": 1.4297526193599117, + "learning_rate": 3.997387475088994e-06, + "loss": 1.0272, + "step": 610 + }, + { + "epoch": 0.045919134225161584, + "grad_norm": 1.8949962556545334, + "learning_rate": 3.997362540141493e-06, + "loss": 1.0094, + "step": 611 + }, + { + "epoch": 0.04599428829099654, + "grad_norm": 1.6511408618607246, + "learning_rate": 3.997337486842612e-06, + "loss": 0.975, + "step": 612 + }, + { + "epoch": 0.046069442356831504, + "grad_norm": 1.978509419046443, + "learning_rate": 3.997312315193837e-06, + "loss": 0.9271, + "step": 613 + }, + { + "epoch": 0.04614459642266647, + "grad_norm": 1.812050204535212, + "learning_rate": 3.9972870251966595e-06, + "loss": 0.9867, + "step": 614 + }, + { + "epoch": 0.04621975048850143, + "grad_norm": 1.9130114102220217, + "learning_rate": 3.997261616852578e-06, + "loss": 0.9479, + "step": 615 + }, + { + "epoch": 0.04629490455433639, + "grad_norm": 1.4472466800906636, + "learning_rate": 3.997236090163097e-06, + "loss": 1.0053, + "step": 616 + }, + { + "epoch": 0.04637005862017135, + "grad_norm": 0.9510355700951688, + "learning_rate": 3.997210445129729e-06, + "loss": 0.8498, + "step": 617 + }, + { + "epoch": 0.046445212686006314, + "grad_norm": 1.4186467843174264, + "learning_rate": 3.997184681753996e-06, + "loss": 1.0017, + "step": 618 + }, + { + "epoch": 0.04652036675184128, + "grad_norm": 1.7005989169721532, + "learning_rate": 3.997158800037422e-06, + "loss": 1.0064, + "step": 619 + }, + { + "epoch": 0.046595520817676234, + "grad_norm": 1.7963790209046129, + "learning_rate": 3.997132799981541e-06, + "loss": 0.9697, + "step": 620 + }, + { + "epoch": 0.0466706748835112, + "grad_norm": 1.750762432768376, + "learning_rate": 3.997106681587895e-06, + "loss": 0.875, + "step": 621 + }, + { + "epoch": 0.04674582894934616, + "grad_norm": 1.4518966405194231, + "learning_rate": 3.99708044485803e-06, + "loss": 1.0207, + "step": 622 + }, + { + "epoch": 0.046820983015181124, + "grad_norm": 1.8734592465116084, + "learning_rate": 3.997054089793501e-06, + "loss": 0.9413, + "step": 623 + }, + { + "epoch": 0.04689613708101608, + "grad_norm": 1.707032786353356, + "learning_rate": 3.997027616395871e-06, + "loss": 1.0565, + "step": 624 + }, + { + "epoch": 0.046971291146851044, + "grad_norm": 2.2058961362969107, + "learning_rate": 3.997001024666707e-06, + "loss": 0.9566, + "step": 625 + }, + { + "epoch": 0.04704644521268601, + "grad_norm": 1.584424053087377, + "learning_rate": 3.996974314607585e-06, + "loss": 1.0447, + "step": 626 + }, + { + "epoch": 0.04712159927852097, + "grad_norm": 1.702051716965719, + "learning_rate": 3.996947486220088e-06, + "loss": 1.014, + "step": 627 + }, + { + "epoch": 0.04719675334435593, + "grad_norm": 1.0996734437859903, + "learning_rate": 3.9969205395058064e-06, + "loss": 0.8729, + "step": 628 + }, + { + "epoch": 0.04727190741019089, + "grad_norm": 1.6535813046177819, + "learning_rate": 3.996893474466336e-06, + "loss": 1.0038, + "step": 629 + }, + { + "epoch": 0.04734706147602585, + "grad_norm": 1.6331102747642015, + "learning_rate": 3.99686629110328e-06, + "loss": 1.0387, + "step": 630 + }, + { + "epoch": 0.04742221554186082, + "grad_norm": 1.5973746277709977, + "learning_rate": 3.99683898941825e-06, + "loss": 1.0467, + "step": 631 + }, + { + "epoch": 0.04749736960769577, + "grad_norm": 1.495532216797991, + "learning_rate": 3.996811569412864e-06, + "loss": 1.0128, + "step": 632 + }, + { + "epoch": 0.04757252367353074, + "grad_norm": 1.5482131102540675, + "learning_rate": 3.996784031088745e-06, + "loss": 1.0611, + "step": 633 + }, + { + "epoch": 0.0476476777393657, + "grad_norm": 1.443612248178021, + "learning_rate": 3.996756374447526e-06, + "loss": 1.0322, + "step": 634 + }, + { + "epoch": 0.04772283180520066, + "grad_norm": 1.4564522009358356, + "learning_rate": 3.996728599490847e-06, + "loss": 1.0484, + "step": 635 + }, + { + "epoch": 0.04779798587103562, + "grad_norm": 1.466668622420192, + "learning_rate": 3.996700706220352e-06, + "loss": 1.0208, + "step": 636 + }, + { + "epoch": 0.04787313993687058, + "grad_norm": 1.400460128442694, + "learning_rate": 3.996672694637694e-06, + "loss": 1.0011, + "step": 637 + }, + { + "epoch": 0.04794829400270555, + "grad_norm": 1.784451869504711, + "learning_rate": 3.996644564744534e-06, + "loss": 1.0245, + "step": 638 + }, + { + "epoch": 0.04802344806854051, + "grad_norm": 1.6780527365698596, + "learning_rate": 3.996616316542537e-06, + "loss": 0.9876, + "step": 639 + }, + { + "epoch": 0.048098602134375466, + "grad_norm": 2.105211684576504, + "learning_rate": 3.996587950033377e-06, + "loss": 1.0195, + "step": 640 + }, + { + "epoch": 0.04817375620021043, + "grad_norm": 1.568370928312054, + "learning_rate": 3.996559465218736e-06, + "loss": 0.9567, + "step": 641 + }, + { + "epoch": 0.04824891026604539, + "grad_norm": 2.040660013129225, + "learning_rate": 3.996530862100302e-06, + "loss": 1.0972, + "step": 642 + }, + { + "epoch": 0.048324064331880356, + "grad_norm": 1.7850805888429633, + "learning_rate": 3.996502140679769e-06, + "loss": 0.9364, + "step": 643 + }, + { + "epoch": 0.04839921839771532, + "grad_norm": 1.5801699391249924, + "learning_rate": 3.996473300958839e-06, + "loss": 1.1282, + "step": 644 + }, + { + "epoch": 0.048474372463550276, + "grad_norm": 1.878483461012236, + "learning_rate": 3.99644434293922e-06, + "loss": 1.0463, + "step": 645 + }, + { + "epoch": 0.04854952652938524, + "grad_norm": 1.689439577257819, + "learning_rate": 3.99641526662263e-06, + "loss": 1.0944, + "step": 646 + }, + { + "epoch": 0.0486246805952202, + "grad_norm": 1.6002633384717233, + "learning_rate": 3.99638607201079e-06, + "loss": 0.9983, + "step": 647 + }, + { + "epoch": 0.048699834661055166, + "grad_norm": 1.5534815879739916, + "learning_rate": 3.996356759105431e-06, + "loss": 0.9804, + "step": 648 + }, + { + "epoch": 0.04877498872689012, + "grad_norm": 1.5198043930779361, + "learning_rate": 3.996327327908289e-06, + "loss": 0.9327, + "step": 649 + }, + { + "epoch": 0.048850142792725086, + "grad_norm": 1.5747924213766762, + "learning_rate": 3.996297778421109e-06, + "loss": 0.9581, + "step": 650 + }, + { + "epoch": 0.04892529685856005, + "grad_norm": 1.523651890463497, + "learning_rate": 3.996268110645641e-06, + "loss": 0.9444, + "step": 651 + }, + { + "epoch": 0.04900045092439501, + "grad_norm": 1.5629872897382437, + "learning_rate": 3.996238324583643e-06, + "loss": 0.9646, + "step": 652 + }, + { + "epoch": 0.04907560499022997, + "grad_norm": 1.4956093743572494, + "learning_rate": 3.99620842023688e-06, + "loss": 1.0476, + "step": 653 + }, + { + "epoch": 0.04915075905606493, + "grad_norm": 0.9948432403031114, + "learning_rate": 3.996178397607125e-06, + "loss": 0.8112, + "step": 654 + }, + { + "epoch": 0.049225913121899896, + "grad_norm": 1.6040826833299793, + "learning_rate": 3.996148256696155e-06, + "loss": 1.0891, + "step": 655 + }, + { + "epoch": 0.04930106718773486, + "grad_norm": 1.927205640331073, + "learning_rate": 3.996117997505758e-06, + "loss": 1.1119, + "step": 656 + }, + { + "epoch": 0.049376221253569816, + "grad_norm": 1.7475968618111997, + "learning_rate": 3.996087620037725e-06, + "loss": 0.9993, + "step": 657 + }, + { + "epoch": 0.04945137531940478, + "grad_norm": 0.8423139572748003, + "learning_rate": 3.996057124293857e-06, + "loss": 0.7733, + "step": 658 + }, + { + "epoch": 0.04952652938523974, + "grad_norm": 1.6071352141062332, + "learning_rate": 3.996026510275962e-06, + "loss": 1.0201, + "step": 659 + }, + { + "epoch": 0.049601683451074706, + "grad_norm": 1.9903116407735453, + "learning_rate": 3.995995777985852e-06, + "loss": 0.9604, + "step": 660 + }, + { + "epoch": 0.04967683751690966, + "grad_norm": 1.5366322650630622, + "learning_rate": 3.995964927425349e-06, + "loss": 1.0021, + "step": 661 + }, + { + "epoch": 0.049751991582744626, + "grad_norm": 2.0476927543334966, + "learning_rate": 3.995933958596282e-06, + "loss": 0.9341, + "step": 662 + }, + { + "epoch": 0.04982714564857959, + "grad_norm": 2.101376096265331, + "learning_rate": 3.995902871500485e-06, + "loss": 1.155, + "step": 663 + }, + { + "epoch": 0.04990229971441455, + "grad_norm": 1.6171299283749503, + "learning_rate": 3.995871666139799e-06, + "loss": 1.0478, + "step": 664 + }, + { + "epoch": 0.04997745378024951, + "grad_norm": 1.3618224025196286, + "learning_rate": 3.995840342516074e-06, + "loss": 1.0691, + "step": 665 + }, + { + "epoch": 0.05005260784608447, + "grad_norm": 1.559169882698487, + "learning_rate": 3.995808900631167e-06, + "loss": 1.0423, + "step": 666 + }, + { + "epoch": 0.050127761911919436, + "grad_norm": 2.0074977999208397, + "learning_rate": 3.99577734048694e-06, + "loss": 1.0245, + "step": 667 + }, + { + "epoch": 0.0502029159777544, + "grad_norm": 1.635291084602746, + "learning_rate": 3.9957456620852636e-06, + "loss": 0.9253, + "step": 668 + }, + { + "epoch": 0.050278070043589355, + "grad_norm": 1.7737495047075953, + "learning_rate": 3.995713865428014e-06, + "loss": 0.9867, + "step": 669 + }, + { + "epoch": 0.05035322410942432, + "grad_norm": 1.7370748727064305, + "learning_rate": 3.995681950517075e-06, + "loss": 1.0192, + "step": 670 + }, + { + "epoch": 0.05042837817525928, + "grad_norm": 1.6232318061291935, + "learning_rate": 3.995649917354339e-06, + "loss": 1.0151, + "step": 671 + }, + { + "epoch": 0.050503532241094246, + "grad_norm": 1.8904004554762515, + "learning_rate": 3.9956177659417036e-06, + "loss": 0.9954, + "step": 672 + }, + { + "epoch": 0.0505786863069292, + "grad_norm": 1.6028591210659322, + "learning_rate": 3.995585496281074e-06, + "loss": 1.0687, + "step": 673 + }, + { + "epoch": 0.050653840372764165, + "grad_norm": 1.9727404888031599, + "learning_rate": 3.995553108374362e-06, + "loss": 1.0568, + "step": 674 + }, + { + "epoch": 0.05072899443859913, + "grad_norm": 1.8754649539481, + "learning_rate": 3.995520602223487e-06, + "loss": 0.8993, + "step": 675 + }, + { + "epoch": 0.05080414850443409, + "grad_norm": 2.3205830323204983, + "learning_rate": 3.995487977830375e-06, + "loss": 1.094, + "step": 676 + }, + { + "epoch": 0.05087930257026905, + "grad_norm": 1.666777352790232, + "learning_rate": 3.995455235196959e-06, + "loss": 0.9798, + "step": 677 + }, + { + "epoch": 0.05095445663610401, + "grad_norm": 1.8436140883185745, + "learning_rate": 3.995422374325179e-06, + "loss": 1.0152, + "step": 678 + }, + { + "epoch": 0.051029610701938975, + "grad_norm": 1.8649090388311205, + "learning_rate": 3.995389395216983e-06, + "loss": 1.0046, + "step": 679 + }, + { + "epoch": 0.05110476476777394, + "grad_norm": 1.63124118378694, + "learning_rate": 3.9953562978743244e-06, + "loss": 0.942, + "step": 680 + }, + { + "epoch": 0.051179918833608895, + "grad_norm": 1.575234740793157, + "learning_rate": 3.995323082299164e-06, + "loss": 1.0298, + "step": 681 + }, + { + "epoch": 0.05125507289944386, + "grad_norm": 1.387368493248845, + "learning_rate": 3.9952897484934706e-06, + "loss": 0.9453, + "step": 682 + }, + { + "epoch": 0.05133022696527882, + "grad_norm": 1.637911341595523, + "learning_rate": 3.9952562964592184e-06, + "loss": 1.0316, + "step": 683 + }, + { + "epoch": 0.051405381031113785, + "grad_norm": 1.6664368788939743, + "learning_rate": 3.995222726198391e-06, + "loss": 1.0747, + "step": 684 + }, + { + "epoch": 0.05148053509694874, + "grad_norm": 2.0547582809962717, + "learning_rate": 3.995189037712977e-06, + "loss": 0.9929, + "step": 685 + }, + { + "epoch": 0.051555689162783705, + "grad_norm": 1.4975596285811146, + "learning_rate": 3.9951552310049715e-06, + "loss": 1.051, + "step": 686 + }, + { + "epoch": 0.05163084322861867, + "grad_norm": 1.8157181944048806, + "learning_rate": 3.99512130607638e-06, + "loss": 0.993, + "step": 687 + }, + { + "epoch": 0.05170599729445363, + "grad_norm": 1.683340881192247, + "learning_rate": 3.995087262929209e-06, + "loss": 0.938, + "step": 688 + }, + { + "epoch": 0.051781151360288595, + "grad_norm": 1.6354604905527548, + "learning_rate": 3.99505310156548e-06, + "loss": 1.0173, + "step": 689 + }, + { + "epoch": 0.05185630542612355, + "grad_norm": 1.6478445423931152, + "learning_rate": 3.995018821987215e-06, + "loss": 0.9714, + "step": 690 + }, + { + "epoch": 0.051931459491958515, + "grad_norm": 2.048937740975563, + "learning_rate": 3.994984424196445e-06, + "loss": 1.0708, + "step": 691 + }, + { + "epoch": 0.05200661355779348, + "grad_norm": 1.6651029244534519, + "learning_rate": 3.994949908195208e-06, + "loss": 1.0032, + "step": 692 + }, + { + "epoch": 0.05208176762362844, + "grad_norm": 0.8277970870395088, + "learning_rate": 3.994915273985551e-06, + "loss": 0.7962, + "step": 693 + }, + { + "epoch": 0.0521569216894634, + "grad_norm": 1.5735213862524318, + "learning_rate": 3.994880521569524e-06, + "loss": 0.9443, + "step": 694 + }, + { + "epoch": 0.05223207575529836, + "grad_norm": 4.639297655157976, + "learning_rate": 3.994845650949187e-06, + "loss": 0.9865, + "step": 695 + }, + { + "epoch": 0.052307229821133325, + "grad_norm": 1.4186449356500148, + "learning_rate": 3.994810662126607e-06, + "loss": 0.9004, + "step": 696 + }, + { + "epoch": 0.05238238388696829, + "grad_norm": 2.435526317763584, + "learning_rate": 3.994775555103857e-06, + "loss": 1.0052, + "step": 697 + }, + { + "epoch": 0.052457537952803245, + "grad_norm": 1.361088341676041, + "learning_rate": 3.994740329883016e-06, + "loss": 0.9921, + "step": 698 + }, + { + "epoch": 0.05253269201863821, + "grad_norm": 1.7702487728705623, + "learning_rate": 3.994704986466172e-06, + "loss": 0.9582, + "step": 699 + }, + { + "epoch": 0.05260784608447317, + "grad_norm": 2.4792172616449397, + "learning_rate": 3.99466952485542e-06, + "loss": 1.0099, + "step": 700 + }, + { + "epoch": 0.052683000150308135, + "grad_norm": 1.8819303226605828, + "learning_rate": 3.994633945052861e-06, + "loss": 0.9162, + "step": 701 + }, + { + "epoch": 0.05275815421614309, + "grad_norm": 2.257726331076279, + "learning_rate": 3.994598247060602e-06, + "loss": 1.0134, + "step": 702 + }, + { + "epoch": 0.052833308281978054, + "grad_norm": 1.8744090738535195, + "learning_rate": 3.9945624308807585e-06, + "loss": 1.0547, + "step": 703 + }, + { + "epoch": 0.05290846234781302, + "grad_norm": 1.7616979762711253, + "learning_rate": 3.994526496515454e-06, + "loss": 1.0522, + "step": 704 + }, + { + "epoch": 0.05298361641364798, + "grad_norm": 1.8557439660199446, + "learning_rate": 3.994490443966818e-06, + "loss": 1.0266, + "step": 705 + }, + { + "epoch": 0.05305877047948294, + "grad_norm": 0.910184704249842, + "learning_rate": 3.994454273236984e-06, + "loss": 0.7803, + "step": 706 + }, + { + "epoch": 0.0531339245453179, + "grad_norm": 1.8718567250271965, + "learning_rate": 3.994417984328098e-06, + "loss": 1.0475, + "step": 707 + }, + { + "epoch": 0.053209078611152864, + "grad_norm": 1.9708842864382412, + "learning_rate": 3.994381577242309e-06, + "loss": 1.0786, + "step": 708 + }, + { + "epoch": 0.05328423267698783, + "grad_norm": 1.5511046103708939, + "learning_rate": 3.994345051981774e-06, + "loss": 1.071, + "step": 709 + }, + { + "epoch": 0.053359386742822784, + "grad_norm": 1.9273529139108754, + "learning_rate": 3.994308408548659e-06, + "loss": 1.1116, + "step": 710 + }, + { + "epoch": 0.05343454080865775, + "grad_norm": 1.8511444134837627, + "learning_rate": 3.994271646945133e-06, + "loss": 1.1402, + "step": 711 + }, + { + "epoch": 0.05350969487449271, + "grad_norm": 1.9549410502334068, + "learning_rate": 3.994234767173376e-06, + "loss": 0.9644, + "step": 712 + }, + { + "epoch": 0.053584848940327674, + "grad_norm": 1.6125902076742755, + "learning_rate": 3.994197769235572e-06, + "loss": 1.0416, + "step": 713 + }, + { + "epoch": 0.05366000300616263, + "grad_norm": 1.4459453949755, + "learning_rate": 3.994160653133915e-06, + "loss": 1.0045, + "step": 714 + }, + { + "epoch": 0.053735157071997594, + "grad_norm": 1.2343040156174492, + "learning_rate": 3.994123418870603e-06, + "loss": 1.0641, + "step": 715 + }, + { + "epoch": 0.05381031113783256, + "grad_norm": 1.6410039554899292, + "learning_rate": 3.994086066447841e-06, + "loss": 0.9839, + "step": 716 + }, + { + "epoch": 0.05388546520366752, + "grad_norm": 0.8670136904730057, + "learning_rate": 3.994048595867845e-06, + "loss": 0.8267, + "step": 717 + }, + { + "epoch": 0.05396061926950248, + "grad_norm": 1.5486185285602527, + "learning_rate": 3.994011007132833e-06, + "loss": 1.0262, + "step": 718 + }, + { + "epoch": 0.05403577333533744, + "grad_norm": 0.7692988088209063, + "learning_rate": 3.993973300245034e-06, + "loss": 0.7887, + "step": 719 + }, + { + "epoch": 0.054110927401172404, + "grad_norm": 1.8944982781441406, + "learning_rate": 3.993935475206682e-06, + "loss": 0.993, + "step": 720 + }, + { + "epoch": 0.05418608146700737, + "grad_norm": 2.4393760113099505, + "learning_rate": 3.993897532020017e-06, + "loss": 0.964, + "step": 721 + }, + { + "epoch": 0.054261235532842324, + "grad_norm": 1.9346976779872742, + "learning_rate": 3.993859470687288e-06, + "loss": 1.0209, + "step": 722 + }, + { + "epoch": 0.05433638959867729, + "grad_norm": 1.592525235968732, + "learning_rate": 3.993821291210751e-06, + "loss": 1.0221, + "step": 723 + }, + { + "epoch": 0.05441154366451225, + "grad_norm": 1.4871854978557142, + "learning_rate": 3.993782993592667e-06, + "loss": 0.956, + "step": 724 + }, + { + "epoch": 0.054486697730347214, + "grad_norm": 1.8538272061157393, + "learning_rate": 3.993744577835306e-06, + "loss": 0.9669, + "step": 725 + }, + { + "epoch": 0.05456185179618217, + "grad_norm": 1.7287465086796192, + "learning_rate": 3.993706043940945e-06, + "loss": 1.1071, + "step": 726 + }, + { + "epoch": 0.054637005862017134, + "grad_norm": 1.6176784341492736, + "learning_rate": 3.993667391911866e-06, + "loss": 1.0146, + "step": 727 + }, + { + "epoch": 0.0547121599278521, + "grad_norm": 2.051093493714632, + "learning_rate": 3.993628621750359e-06, + "loss": 1.0122, + "step": 728 + }, + { + "epoch": 0.05478731399368706, + "grad_norm": 2.496076939008335, + "learning_rate": 3.993589733458723e-06, + "loss": 0.9715, + "step": 729 + }, + { + "epoch": 0.05486246805952202, + "grad_norm": 2.013129530353535, + "learning_rate": 3.993550727039261e-06, + "loss": 0.9299, + "step": 730 + }, + { + "epoch": 0.05493762212535698, + "grad_norm": 2.762587608915966, + "learning_rate": 3.993511602494285e-06, + "loss": 0.9332, + "step": 731 + }, + { + "epoch": 0.055012776191191944, + "grad_norm": 1.6354279082412038, + "learning_rate": 3.993472359826112e-06, + "loss": 0.9874, + "step": 732 + }, + { + "epoch": 0.05508793025702691, + "grad_norm": 1.8992928669761566, + "learning_rate": 3.993432999037068e-06, + "loss": 0.9827, + "step": 733 + }, + { + "epoch": 0.05516308432286186, + "grad_norm": 1.5965980812239942, + "learning_rate": 3.993393520129487e-06, + "loss": 1.0364, + "step": 734 + }, + { + "epoch": 0.05523823838869683, + "grad_norm": 1.8730596675177396, + "learning_rate": 3.993353923105705e-06, + "loss": 1.0313, + "step": 735 + }, + { + "epoch": 0.05531339245453179, + "grad_norm": 0.9525959547775878, + "learning_rate": 3.993314207968071e-06, + "loss": 0.8955, + "step": 736 + }, + { + "epoch": 0.05538854652036675, + "grad_norm": 0.7443018861187315, + "learning_rate": 3.993274374718938e-06, + "loss": 0.7667, + "step": 737 + }, + { + "epoch": 0.05546370058620172, + "grad_norm": 1.8356939925365063, + "learning_rate": 3.9932344233606634e-06, + "loss": 1.0277, + "step": 738 + }, + { + "epoch": 0.05553885465203667, + "grad_norm": 1.674828318437515, + "learning_rate": 3.993194353895618e-06, + "loss": 1.0134, + "step": 739 + }, + { + "epoch": 0.05561400871787164, + "grad_norm": 2.3883766122833414, + "learning_rate": 3.9931541663261756e-06, + "loss": 1.0234, + "step": 740 + }, + { + "epoch": 0.0556891627837066, + "grad_norm": 1.643103045073478, + "learning_rate": 3.993113860654715e-06, + "loss": 0.9999, + "step": 741 + }, + { + "epoch": 0.05576431684954156, + "grad_norm": 1.8560657062944064, + "learning_rate": 3.993073436883627e-06, + "loss": 0.9877, + "step": 742 + }, + { + "epoch": 0.05583947091537652, + "grad_norm": 1.3688135559097467, + "learning_rate": 3.993032895015304e-06, + "loss": 0.991, + "step": 743 + }, + { + "epoch": 0.05591462498121148, + "grad_norm": 1.7615054523731213, + "learning_rate": 3.992992235052152e-06, + "loss": 1.059, + "step": 744 + }, + { + "epoch": 0.05598977904704645, + "grad_norm": 1.6735300446324457, + "learning_rate": 3.992951456996578e-06, + "loss": 1.0816, + "step": 745 + }, + { + "epoch": 0.05606493311288141, + "grad_norm": 1.5327754304935564, + "learning_rate": 3.9929105608509984e-06, + "loss": 1.0718, + "step": 746 + }, + { + "epoch": 0.056140087178716366, + "grad_norm": 1.5315014379720635, + "learning_rate": 3.9928695466178375e-06, + "loss": 0.9217, + "step": 747 + }, + { + "epoch": 0.05621524124455133, + "grad_norm": 1.8137328460323845, + "learning_rate": 3.992828414299524e-06, + "loss": 0.949, + "step": 748 + }, + { + "epoch": 0.05629039531038629, + "grad_norm": 1.4129901845635637, + "learning_rate": 3.9927871638984955e-06, + "loss": 1.0452, + "step": 749 + }, + { + "epoch": 0.056365549376221256, + "grad_norm": 1.9957877146604912, + "learning_rate": 3.992745795417198e-06, + "loss": 1.0257, + "step": 750 + }, + { + "epoch": 0.05644070344205621, + "grad_norm": 1.9451121031710712, + "learning_rate": 3.99270430885808e-06, + "loss": 1.0441, + "step": 751 + }, + { + "epoch": 0.056515857507891176, + "grad_norm": 1.5677914856142388, + "learning_rate": 3.992662704223602e-06, + "loss": 1.0268, + "step": 752 + }, + { + "epoch": 0.05659101157372614, + "grad_norm": 1.7132638939962492, + "learning_rate": 3.992620981516228e-06, + "loss": 1.021, + "step": 753 + }, + { + "epoch": 0.0566661656395611, + "grad_norm": 1.530626208864469, + "learning_rate": 3.9925791407384304e-06, + "loss": 1.0671, + "step": 754 + }, + { + "epoch": 0.05674131970539606, + "grad_norm": 1.6129822534766207, + "learning_rate": 3.9925371818926884e-06, + "loss": 1.0546, + "step": 755 + }, + { + "epoch": 0.05681647377123102, + "grad_norm": 0.9466167626581026, + "learning_rate": 3.992495104981489e-06, + "loss": 0.846, + "step": 756 + }, + { + "epoch": 0.056891627837065986, + "grad_norm": 3.071585122107723, + "learning_rate": 3.992452910007325e-06, + "loss": 0.9744, + "step": 757 + }, + { + "epoch": 0.05696678190290095, + "grad_norm": 1.8337108339002857, + "learning_rate": 3.992410596972696e-06, + "loss": 1.1388, + "step": 758 + }, + { + "epoch": 0.057041935968735906, + "grad_norm": 1.6841036208160565, + "learning_rate": 3.99236816588011e-06, + "loss": 1.0488, + "step": 759 + }, + { + "epoch": 0.05711709003457087, + "grad_norm": 2.215763930254462, + "learning_rate": 3.992325616732081e-06, + "loss": 0.9604, + "step": 760 + }, + { + "epoch": 0.05719224410040583, + "grad_norm": 1.6056647633536056, + "learning_rate": 3.992282949531129e-06, + "loss": 0.9953, + "step": 761 + }, + { + "epoch": 0.057267398166240796, + "grad_norm": 1.8735199490969683, + "learning_rate": 3.992240164279785e-06, + "loss": 1.1063, + "step": 762 + }, + { + "epoch": 0.05734255223207575, + "grad_norm": 1.7444793953131108, + "learning_rate": 3.9921972609805815e-06, + "loss": 1.0608, + "step": 763 + }, + { + "epoch": 0.057417706297910716, + "grad_norm": 1.5039408097693043, + "learning_rate": 3.992154239636062e-06, + "loss": 1.0658, + "step": 764 + }, + { + "epoch": 0.05749286036374568, + "grad_norm": 0.9704709917201357, + "learning_rate": 3.992111100248775e-06, + "loss": 0.8245, + "step": 765 + }, + { + "epoch": 0.05756801442958064, + "grad_norm": 1.6651370022514933, + "learning_rate": 3.992067842821277e-06, + "loss": 1.0248, + "step": 766 + }, + { + "epoch": 0.0576431684954156, + "grad_norm": 1.8592031062835574, + "learning_rate": 3.992024467356132e-06, + "loss": 0.9532, + "step": 767 + }, + { + "epoch": 0.05771832256125056, + "grad_norm": 1.758567139446672, + "learning_rate": 3.991980973855908e-06, + "loss": 1.0709, + "step": 768 + }, + { + "epoch": 0.057793476627085526, + "grad_norm": 1.9724504117933213, + "learning_rate": 3.991937362323183e-06, + "loss": 1.1027, + "step": 769 + }, + { + "epoch": 0.05786863069292049, + "grad_norm": 1.8389228735386398, + "learning_rate": 3.991893632760544e-06, + "loss": 1.0235, + "step": 770 + }, + { + "epoch": 0.057943784758755446, + "grad_norm": 1.1097185382031234, + "learning_rate": 3.991849785170578e-06, + "loss": 0.8149, + "step": 771 + }, + { + "epoch": 0.05801893882459041, + "grad_norm": 3.518565664663351, + "learning_rate": 3.991805819555885e-06, + "loss": 0.9419, + "step": 772 + }, + { + "epoch": 0.05809409289042537, + "grad_norm": 1.6601923680583979, + "learning_rate": 3.991761735919071e-06, + "loss": 0.9638, + "step": 773 + }, + { + "epoch": 0.058169246956260336, + "grad_norm": 3.945520376639876, + "learning_rate": 3.991717534262747e-06, + "loss": 1.0719, + "step": 774 + }, + { + "epoch": 0.05824440102209529, + "grad_norm": 1.7425688595622735, + "learning_rate": 3.991673214589532e-06, + "loss": 1.0951, + "step": 775 + }, + { + "epoch": 0.058319555087930255, + "grad_norm": 1.8217982830560902, + "learning_rate": 3.991628776902052e-06, + "loss": 0.9994, + "step": 776 + }, + { + "epoch": 0.05839470915376522, + "grad_norm": 2.004827514420965, + "learning_rate": 3.991584221202942e-06, + "loss": 1.0526, + "step": 777 + }, + { + "epoch": 0.05846986321960018, + "grad_norm": 1.9188010591056721, + "learning_rate": 3.991539547494839e-06, + "loss": 1.0722, + "step": 778 + }, + { + "epoch": 0.05854501728543514, + "grad_norm": 1.9531015013234836, + "learning_rate": 3.991494755780392e-06, + "loss": 0.9438, + "step": 779 + }, + { + "epoch": 0.0586201713512701, + "grad_norm": 1.5886561951755724, + "learning_rate": 3.991449846062255e-06, + "loss": 1.1094, + "step": 780 + }, + { + "epoch": 0.058695325417105065, + "grad_norm": 1.6044965710292753, + "learning_rate": 3.991404818343089e-06, + "loss": 1.0863, + "step": 781 + }, + { + "epoch": 0.05877047948294003, + "grad_norm": 1.7818299403492672, + "learning_rate": 3.991359672625562e-06, + "loss": 0.9875, + "step": 782 + }, + { + "epoch": 0.05884563354877499, + "grad_norm": 1.5366062761191748, + "learning_rate": 3.9913144089123485e-06, + "loss": 1.0505, + "step": 783 + }, + { + "epoch": 0.05892078761460995, + "grad_norm": 2.199854702919886, + "learning_rate": 3.991269027206131e-06, + "loss": 0.9955, + "step": 784 + }, + { + "epoch": 0.05899594168044491, + "grad_norm": 1.967070756497774, + "learning_rate": 3.991223527509599e-06, + "loss": 0.9493, + "step": 785 + }, + { + "epoch": 0.059071095746279875, + "grad_norm": 1.4930134828933457, + "learning_rate": 3.991177909825448e-06, + "loss": 0.9642, + "step": 786 + }, + { + "epoch": 0.05914624981211484, + "grad_norm": 1.8846074962001735, + "learning_rate": 3.991132174156381e-06, + "loss": 1.034, + "step": 787 + }, + { + "epoch": 0.059221403877949795, + "grad_norm": 1.4762220040556835, + "learning_rate": 3.991086320505108e-06, + "loss": 1.0653, + "step": 788 + }, + { + "epoch": 0.05929655794378476, + "grad_norm": 1.4805901711627858, + "learning_rate": 3.991040348874346e-06, + "loss": 1.0731, + "step": 789 + }, + { + "epoch": 0.05937171200961972, + "grad_norm": 1.8080897112929026, + "learning_rate": 3.99099425926682e-06, + "loss": 1.0872, + "step": 790 + }, + { + "epoch": 0.059446866075454685, + "grad_norm": 1.642304351162763, + "learning_rate": 3.990948051685259e-06, + "loss": 1.0132, + "step": 791 + }, + { + "epoch": 0.05952202014128964, + "grad_norm": 1.560588965009477, + "learning_rate": 3.990901726132403e-06, + "loss": 1.0962, + "step": 792 + }, + { + "epoch": 0.059597174207124605, + "grad_norm": 1.9230798805654232, + "learning_rate": 3.990855282610996e-06, + "loss": 0.9261, + "step": 793 + }, + { + "epoch": 0.05967232827295957, + "grad_norm": 1.6729647213442083, + "learning_rate": 3.990808721123789e-06, + "loss": 0.9445, + "step": 794 + }, + { + "epoch": 0.05974748233879453, + "grad_norm": 1.5999753950134474, + "learning_rate": 3.990762041673543e-06, + "loss": 1.0238, + "step": 795 + }, + { + "epoch": 0.05982263640462949, + "grad_norm": 1.9540436072193463, + "learning_rate": 3.990715244263023e-06, + "loss": 0.9673, + "step": 796 + }, + { + "epoch": 0.05989779047046445, + "grad_norm": 1.9442558636120137, + "learning_rate": 3.9906683288950005e-06, + "loss": 1.0295, + "step": 797 + }, + { + "epoch": 0.059972944536299415, + "grad_norm": 1.3751011724530713, + "learning_rate": 3.990621295572258e-06, + "loss": 1.0397, + "step": 798 + }, + { + "epoch": 0.06004809860213438, + "grad_norm": 1.5454633031688751, + "learning_rate": 3.99057414429758e-06, + "loss": 1.0878, + "step": 799 + }, + { + "epoch": 0.060123252667969335, + "grad_norm": 1.4802304682551974, + "learning_rate": 3.9905268750737625e-06, + "loss": 1.0067, + "step": 800 + }, + { + "epoch": 0.0601984067338043, + "grad_norm": 1.6168786090210403, + "learning_rate": 3.990479487903605e-06, + "loss": 0.9711, + "step": 801 + }, + { + "epoch": 0.06027356079963926, + "grad_norm": 1.4731557793167285, + "learning_rate": 3.990431982789917e-06, + "loss": 0.9971, + "step": 802 + }, + { + "epoch": 0.060348714865474225, + "grad_norm": 2.8478330706419905, + "learning_rate": 3.9903843597355105e-06, + "loss": 1.0906, + "step": 803 + }, + { + "epoch": 0.06042386893130918, + "grad_norm": 1.8688402083960582, + "learning_rate": 3.99033661874321e-06, + "loss": 1.1428, + "step": 804 + }, + { + "epoch": 0.060499022997144145, + "grad_norm": 1.8041156341641593, + "learning_rate": 3.990288759815843e-06, + "loss": 0.7991, + "step": 805 + }, + { + "epoch": 0.06057417706297911, + "grad_norm": 1.6347150168111089, + "learning_rate": 3.990240782956245e-06, + "loss": 1.0908, + "step": 806 + }, + { + "epoch": 0.06064933112881407, + "grad_norm": 1.6780220469002851, + "learning_rate": 3.99019268816726e-06, + "loss": 1.0179, + "step": 807 + }, + { + "epoch": 0.06072448519464903, + "grad_norm": 1.664404963683098, + "learning_rate": 3.990144475451738e-06, + "loss": 1.0589, + "step": 808 + }, + { + "epoch": 0.06079963926048399, + "grad_norm": 1.416701998924514, + "learning_rate": 3.990096144812534e-06, + "loss": 1.1014, + "step": 809 + }, + { + "epoch": 0.060874793326318954, + "grad_norm": 1.6258187798820292, + "learning_rate": 3.9900476962525125e-06, + "loss": 1.0538, + "step": 810 + }, + { + "epoch": 0.06094994739215392, + "grad_norm": 1.9892161032078015, + "learning_rate": 3.989999129774546e-06, + "loss": 0.9363, + "step": 811 + }, + { + "epoch": 0.061025101457988874, + "grad_norm": 2.0303217013788943, + "learning_rate": 3.989950445381511e-06, + "loss": 0.9757, + "step": 812 + }, + { + "epoch": 0.06110025552382384, + "grad_norm": 1.3437354952007172, + "learning_rate": 3.98990164307629e-06, + "loss": 1.0111, + "step": 813 + }, + { + "epoch": 0.0611754095896588, + "grad_norm": 1.930253721864272, + "learning_rate": 3.989852722861778e-06, + "loss": 1.1242, + "step": 814 + }, + { + "epoch": 0.061250563655493764, + "grad_norm": 1.7462192294031844, + "learning_rate": 3.989803684740873e-06, + "loss": 1.034, + "step": 815 + }, + { + "epoch": 0.06132571772132872, + "grad_norm": 1.541106208101917, + "learning_rate": 3.9897545287164795e-06, + "loss": 0.9542, + "step": 816 + }, + { + "epoch": 0.061400871787163684, + "grad_norm": 1.4583314713965745, + "learning_rate": 3.9897052547915115e-06, + "loss": 1.0444, + "step": 817 + }, + { + "epoch": 0.06147602585299865, + "grad_norm": 1.4628793794940766, + "learning_rate": 3.989655862968887e-06, + "loss": 1.0222, + "step": 818 + }, + { + "epoch": 0.06155117991883361, + "grad_norm": 1.2066642919450299, + "learning_rate": 3.989606353251535e-06, + "loss": 0.9727, + "step": 819 + }, + { + "epoch": 0.06162633398466857, + "grad_norm": 1.210743008653461, + "learning_rate": 3.989556725642388e-06, + "loss": 0.9556, + "step": 820 + }, + { + "epoch": 0.06170148805050353, + "grad_norm": 1.725796836136789, + "learning_rate": 3.989506980144385e-06, + "loss": 0.9919, + "step": 821 + }, + { + "epoch": 0.061776642116338494, + "grad_norm": 1.4163889425727865, + "learning_rate": 3.989457116760477e-06, + "loss": 0.9781, + "step": 822 + }, + { + "epoch": 0.06185179618217346, + "grad_norm": 0.8037586962761893, + "learning_rate": 3.989407135493615e-06, + "loss": 0.8719, + "step": 823 + }, + { + "epoch": 0.061926950248008414, + "grad_norm": 1.499074730011095, + "learning_rate": 3.9893570363467625e-06, + "loss": 0.8534, + "step": 824 + }, + { + "epoch": 0.06200210431384338, + "grad_norm": 1.8931084707652464, + "learning_rate": 3.9893068193228885e-06, + "loss": 1.0309, + "step": 825 + }, + { + "epoch": 0.06207725837967834, + "grad_norm": 1.6823764303325885, + "learning_rate": 3.989256484424968e-06, + "loss": 0.9822, + "step": 826 + }, + { + "epoch": 0.062152412445513304, + "grad_norm": 1.5449250386394995, + "learning_rate": 3.989206031655982e-06, + "loss": 0.9666, + "step": 827 + }, + { + "epoch": 0.06222756651134827, + "grad_norm": 1.8023270532225704, + "learning_rate": 3.989155461018923e-06, + "loss": 1.0701, + "step": 828 + }, + { + "epoch": 0.062302720577183224, + "grad_norm": 1.5299964756395226, + "learning_rate": 3.989104772516785e-06, + "loss": 1.0267, + "step": 829 + }, + { + "epoch": 0.06237787464301819, + "grad_norm": 1.7471410695434344, + "learning_rate": 3.989053966152573e-06, + "loss": 0.9911, + "step": 830 + }, + { + "epoch": 0.06245302870885315, + "grad_norm": 1.672177438921345, + "learning_rate": 3.9890030419292965e-06, + "loss": 1.0117, + "step": 831 + }, + { + "epoch": 0.0625281827746881, + "grad_norm": 2.141076924310051, + "learning_rate": 3.988951999849974e-06, + "loss": 0.919, + "step": 832 + }, + { + "epoch": 0.06260333684052308, + "grad_norm": 1.6670571187742367, + "learning_rate": 3.988900839917628e-06, + "loss": 1.0213, + "step": 833 + }, + { + "epoch": 0.06267849090635803, + "grad_norm": 1.5257580420882029, + "learning_rate": 3.988849562135293e-06, + "loss": 0.9938, + "step": 834 + }, + { + "epoch": 0.06275364497219299, + "grad_norm": 1.683061838760368, + "learning_rate": 3.988798166506005e-06, + "loss": 1.015, + "step": 835 + }, + { + "epoch": 0.06282879903802796, + "grad_norm": 0.9303433004604151, + "learning_rate": 3.98874665303281e-06, + "loss": 0.8436, + "step": 836 + }, + { + "epoch": 0.06290395310386292, + "grad_norm": 1.6895360319930173, + "learning_rate": 3.98869502171876e-06, + "loss": 1.1332, + "step": 837 + }, + { + "epoch": 0.06297910716969789, + "grad_norm": 1.565132898288062, + "learning_rate": 3.9886432725669146e-06, + "loss": 1.0439, + "step": 838 + }, + { + "epoch": 0.06305426123553284, + "grad_norm": 1.656347451851846, + "learning_rate": 3.988591405580341e-06, + "loss": 1.0484, + "step": 839 + }, + { + "epoch": 0.0631294153013678, + "grad_norm": 1.3823149049823107, + "learning_rate": 3.988539420762111e-06, + "loss": 1.0137, + "step": 840 + }, + { + "epoch": 0.06320456936720277, + "grad_norm": 1.5818867181760663, + "learning_rate": 3.988487318115306e-06, + "loss": 1.0205, + "step": 841 + }, + { + "epoch": 0.06327972343303773, + "grad_norm": 1.6361007069787803, + "learning_rate": 3.9884350976430136e-06, + "loss": 1.051, + "step": 842 + }, + { + "epoch": 0.06335487749887268, + "grad_norm": 2.4750302746395003, + "learning_rate": 3.988382759348327e-06, + "loss": 1.0032, + "step": 843 + }, + { + "epoch": 0.06343003156470765, + "grad_norm": 1.7188339212910342, + "learning_rate": 3.988330303234347e-06, + "loss": 1.1316, + "step": 844 + }, + { + "epoch": 0.06350518563054261, + "grad_norm": 1.4424662840289553, + "learning_rate": 3.988277729304184e-06, + "loss": 1.034, + "step": 845 + }, + { + "epoch": 0.06358033969637758, + "grad_norm": 1.9896238699040623, + "learning_rate": 3.988225037560951e-06, + "loss": 0.9657, + "step": 846 + }, + { + "epoch": 0.06365549376221254, + "grad_norm": 1.8107408997769057, + "learning_rate": 3.988172228007771e-06, + "loss": 1.0389, + "step": 847 + }, + { + "epoch": 0.0637306478280475, + "grad_norm": 1.7503069880840934, + "learning_rate": 3.9881193006477745e-06, + "loss": 0.9807, + "step": 848 + }, + { + "epoch": 0.06380580189388246, + "grad_norm": 1.6289813346847228, + "learning_rate": 3.9880662554840955e-06, + "loss": 0.9926, + "step": 849 + }, + { + "epoch": 0.06388095595971742, + "grad_norm": 1.56224386023116, + "learning_rate": 3.9880130925198786e-06, + "loss": 1.0924, + "step": 850 + }, + { + "epoch": 0.06395611002555238, + "grad_norm": 2.3682380586055496, + "learning_rate": 3.987959811758273e-06, + "loss": 0.951, + "step": 851 + }, + { + "epoch": 0.06403126409138735, + "grad_norm": 1.492284482031251, + "learning_rate": 3.9879064132024365e-06, + "loss": 0.946, + "step": 852 + }, + { + "epoch": 0.0641064181572223, + "grad_norm": 1.52503507917738, + "learning_rate": 3.987852896855532e-06, + "loss": 1.0697, + "step": 853 + }, + { + "epoch": 0.06418157222305727, + "grad_norm": 1.3458957838719576, + "learning_rate": 3.987799262720732e-06, + "loss": 1.0538, + "step": 854 + }, + { + "epoch": 0.06425672628889223, + "grad_norm": 1.9073001907910951, + "learning_rate": 3.987745510801214e-06, + "loss": 1.0164, + "step": 855 + }, + { + "epoch": 0.06433188035472719, + "grad_norm": 1.3757600417660565, + "learning_rate": 3.987691641100162e-06, + "loss": 1.0094, + "step": 856 + }, + { + "epoch": 0.06440703442056216, + "grad_norm": 2.260845219878143, + "learning_rate": 3.98763765362077e-06, + "loss": 1.09, + "step": 857 + }, + { + "epoch": 0.06448218848639711, + "grad_norm": 1.3965635904106002, + "learning_rate": 3.987583548366235e-06, + "loss": 0.8886, + "step": 858 + }, + { + "epoch": 0.06455734255223207, + "grad_norm": 1.583790570878282, + "learning_rate": 3.987529325339764e-06, + "loss": 0.9186, + "step": 859 + }, + { + "epoch": 0.06463249661806704, + "grad_norm": 1.619599832866695, + "learning_rate": 3.98747498454457e-06, + "loss": 0.9258, + "step": 860 + }, + { + "epoch": 0.064707650683902, + "grad_norm": 1.6004794296568259, + "learning_rate": 3.987420525983873e-06, + "loss": 1.0788, + "step": 861 + }, + { + "epoch": 0.06478280474973697, + "grad_norm": 1.3833341745963852, + "learning_rate": 3.9873659496608985e-06, + "loss": 0.9403, + "step": 862 + }, + { + "epoch": 0.06485795881557192, + "grad_norm": 1.726540543774775, + "learning_rate": 3.9873112555788816e-06, + "loss": 1.0606, + "step": 863 + }, + { + "epoch": 0.06493311288140688, + "grad_norm": 1.5448332904636368, + "learning_rate": 3.987256443741063e-06, + "loss": 0.9548, + "step": 864 + }, + { + "epoch": 0.06500826694724185, + "grad_norm": 1.7076139873922749, + "learning_rate": 3.9872015141506905e-06, + "loss": 1.0991, + "step": 865 + }, + { + "epoch": 0.0650834210130768, + "grad_norm": 1.2309809511819634, + "learning_rate": 3.987146466811019e-06, + "loss": 1.0363, + "step": 866 + }, + { + "epoch": 0.06515857507891176, + "grad_norm": 2.006500527970002, + "learning_rate": 3.98709130172531e-06, + "loss": 1.064, + "step": 867 + }, + { + "epoch": 0.06523372914474673, + "grad_norm": 1.849465127897215, + "learning_rate": 3.987036018896832e-06, + "loss": 0.9341, + "step": 868 + }, + { + "epoch": 0.06530888321058169, + "grad_norm": 1.6289900431132578, + "learning_rate": 3.986980618328861e-06, + "loss": 0.9719, + "step": 869 + }, + { + "epoch": 0.06538403727641666, + "grad_norm": 1.5219518998474357, + "learning_rate": 3.98692510002468e-06, + "loss": 0.9651, + "step": 870 + }, + { + "epoch": 0.06545919134225162, + "grad_norm": 1.7974004226076234, + "learning_rate": 3.986869463987578e-06, + "loss": 0.9756, + "step": 871 + }, + { + "epoch": 0.06553434540808657, + "grad_norm": 1.7928581785460658, + "learning_rate": 3.9868137102208525e-06, + "loss": 0.9582, + "step": 872 + }, + { + "epoch": 0.06560949947392154, + "grad_norm": 1.4949318779061733, + "learning_rate": 3.9867578387278065e-06, + "loss": 1.0931, + "step": 873 + }, + { + "epoch": 0.0656846535397565, + "grad_norm": 1.6344068855982712, + "learning_rate": 3.986701849511751e-06, + "loss": 1.0029, + "step": 874 + }, + { + "epoch": 0.06575980760559147, + "grad_norm": 1.6271565872224563, + "learning_rate": 3.986645742576002e-06, + "loss": 1.0035, + "step": 875 + }, + { + "epoch": 0.06583496167142643, + "grad_norm": 1.409289656463649, + "learning_rate": 3.986589517923887e-06, + "loss": 1.0183, + "step": 876 + }, + { + "epoch": 0.06591011573726138, + "grad_norm": 2.43107609457841, + "learning_rate": 3.986533175558735e-06, + "loss": 0.9993, + "step": 877 + }, + { + "epoch": 0.06598526980309635, + "grad_norm": 1.463058075024004, + "learning_rate": 3.9864767154838856e-06, + "loss": 0.987, + "step": 878 + }, + { + "epoch": 0.06606042386893131, + "grad_norm": 1.7577820499749772, + "learning_rate": 3.986420137702684e-06, + "loss": 0.9344, + "step": 879 + }, + { + "epoch": 0.06613557793476627, + "grad_norm": 0.8774065574387041, + "learning_rate": 3.9863634422184835e-06, + "loss": 0.7985, + "step": 880 + }, + { + "epoch": 0.06621073200060124, + "grad_norm": 1.6849900405544176, + "learning_rate": 3.986306629034642e-06, + "loss": 1.0995, + "step": 881 + }, + { + "epoch": 0.06628588606643619, + "grad_norm": 1.235782834489853, + "learning_rate": 3.9862496981545265e-06, + "loss": 0.9355, + "step": 882 + }, + { + "epoch": 0.06636104013227116, + "grad_norm": 1.812086189649618, + "learning_rate": 3.986192649581511e-06, + "loss": 0.9671, + "step": 883 + }, + { + "epoch": 0.06643619419810612, + "grad_norm": 1.7211057990678793, + "learning_rate": 3.986135483318975e-06, + "loss": 1.01, + "step": 884 + }, + { + "epoch": 0.06651134826394108, + "grad_norm": 2.101199093848207, + "learning_rate": 3.986078199370307e-06, + "loss": 0.9512, + "step": 885 + }, + { + "epoch": 0.06658650232977605, + "grad_norm": 2.1104512047710857, + "learning_rate": 3.9860207977388994e-06, + "loss": 0.8966, + "step": 886 + }, + { + "epoch": 0.066661656395611, + "grad_norm": 1.8514369487446776, + "learning_rate": 3.985963278428155e-06, + "loss": 1.0593, + "step": 887 + }, + { + "epoch": 0.06673681046144596, + "grad_norm": 0.8895998961420307, + "learning_rate": 3.985905641441482e-06, + "loss": 0.826, + "step": 888 + }, + { + "epoch": 0.06681196452728093, + "grad_norm": 0.8609291588574782, + "learning_rate": 3.9858478867822945e-06, + "loss": 0.8169, + "step": 889 + }, + { + "epoch": 0.06688711859311589, + "grad_norm": 1.4723980826581198, + "learning_rate": 3.985790014454016e-06, + "loss": 1.0118, + "step": 890 + }, + { + "epoch": 0.06696227265895086, + "grad_norm": 1.8657243813988686, + "learning_rate": 3.985732024460074e-06, + "loss": 1.0724, + "step": 891 + }, + { + "epoch": 0.06703742672478581, + "grad_norm": 1.6164464312242766, + "learning_rate": 3.985673916803907e-06, + "loss": 1.042, + "step": 892 + }, + { + "epoch": 0.06711258079062077, + "grad_norm": 1.6388059431663347, + "learning_rate": 3.9856156914889556e-06, + "loss": 1.0821, + "step": 893 + }, + { + "epoch": 0.06718773485645574, + "grad_norm": 1.6965889761673965, + "learning_rate": 3.985557348518672e-06, + "loss": 0.975, + "step": 894 + }, + { + "epoch": 0.0672628889222907, + "grad_norm": 1.4723855333897935, + "learning_rate": 3.9854988878965125e-06, + "loss": 0.987, + "step": 895 + }, + { + "epoch": 0.06733804298812565, + "grad_norm": 1.473402486892824, + "learning_rate": 3.98544030962594e-06, + "loss": 1.0803, + "step": 896 + }, + { + "epoch": 0.06741319705396062, + "grad_norm": 1.7431412895702583, + "learning_rate": 3.985381613710427e-06, + "loss": 1.0205, + "step": 897 + }, + { + "epoch": 0.06748835111979558, + "grad_norm": 2.0701063521334033, + "learning_rate": 3.98532280015345e-06, + "loss": 0.9791, + "step": 898 + }, + { + "epoch": 0.06756350518563055, + "grad_norm": 1.4910514461460815, + "learning_rate": 3.985263868958496e-06, + "loss": 0.9781, + "step": 899 + }, + { + "epoch": 0.0676386592514655, + "grad_norm": 1.4520066582655822, + "learning_rate": 3.9852048201290545e-06, + "loss": 0.9562, + "step": 900 + }, + { + "epoch": 0.06771381331730046, + "grad_norm": 1.555554938694665, + "learning_rate": 3.985145653668626e-06, + "loss": 1.0217, + "step": 901 + }, + { + "epoch": 0.06778896738313543, + "grad_norm": 1.4967880829043607, + "learning_rate": 3.985086369580716e-06, + "loss": 1.0028, + "step": 902 + }, + { + "epoch": 0.06786412144897039, + "grad_norm": 1.5570760043880845, + "learning_rate": 3.985026967868837e-06, + "loss": 0.9373, + "step": 903 + }, + { + "epoch": 0.06793927551480534, + "grad_norm": 1.8966129781882182, + "learning_rate": 3.9849674485365094e-06, + "loss": 0.9834, + "step": 904 + }, + { + "epoch": 0.06801442958064031, + "grad_norm": 1.6420838908604558, + "learning_rate": 3.98490781158726e-06, + "loss": 1.071, + "step": 905 + }, + { + "epoch": 0.06808958364647527, + "grad_norm": 1.607007108737911, + "learning_rate": 3.98484805702462e-06, + "loss": 1.0476, + "step": 906 + }, + { + "epoch": 0.06816473771231024, + "grad_norm": 1.3622615465559378, + "learning_rate": 3.9847881848521345e-06, + "loss": 0.9651, + "step": 907 + }, + { + "epoch": 0.0682398917781452, + "grad_norm": 1.532802417655862, + "learning_rate": 3.984728195073347e-06, + "loss": 0.9704, + "step": 908 + }, + { + "epoch": 0.06831504584398015, + "grad_norm": 1.492903532622887, + "learning_rate": 3.984668087691815e-06, + "loss": 1.0485, + "step": 909 + }, + { + "epoch": 0.06839019990981512, + "grad_norm": 1.6156392080246726, + "learning_rate": 3.984607862711099e-06, + "loss": 0.884, + "step": 910 + }, + { + "epoch": 0.06846535397565008, + "grad_norm": 1.433265918802315, + "learning_rate": 3.984547520134767e-06, + "loss": 0.9408, + "step": 911 + }, + { + "epoch": 0.06854050804148504, + "grad_norm": 1.5919732306209013, + "learning_rate": 3.9844870599663954e-06, + "loss": 0.9516, + "step": 912 + }, + { + "epoch": 0.06861566210732001, + "grad_norm": 1.92125164560559, + "learning_rate": 3.984426482209567e-06, + "loss": 1.0327, + "step": 913 + }, + { + "epoch": 0.06869081617315496, + "grad_norm": 1.884803447442715, + "learning_rate": 3.98436578686787e-06, + "loss": 1.0079, + "step": 914 + }, + { + "epoch": 0.06876597023898993, + "grad_norm": 1.501182090373127, + "learning_rate": 3.984304973944901e-06, + "loss": 1.046, + "step": 915 + }, + { + "epoch": 0.06884112430482489, + "grad_norm": 1.6797092879900481, + "learning_rate": 3.984244043444264e-06, + "loss": 0.961, + "step": 916 + }, + { + "epoch": 0.06891627837065985, + "grad_norm": 1.7738525308300581, + "learning_rate": 3.98418299536957e-06, + "loss": 1.0548, + "step": 917 + }, + { + "epoch": 0.06899143243649482, + "grad_norm": 1.6132007791248184, + "learning_rate": 3.984121829724435e-06, + "loss": 0.9836, + "step": 918 + }, + { + "epoch": 0.06906658650232977, + "grad_norm": 1.629707706432212, + "learning_rate": 3.984060546512484e-06, + "loss": 1.1162, + "step": 919 + }, + { + "epoch": 0.06914174056816474, + "grad_norm": 1.5002313276570933, + "learning_rate": 3.983999145737348e-06, + "loss": 1.0005, + "step": 920 + }, + { + "epoch": 0.0692168946339997, + "grad_norm": 2.3850376433274825, + "learning_rate": 3.983937627402665e-06, + "loss": 0.9545, + "step": 921 + }, + { + "epoch": 0.06929204869983466, + "grad_norm": 1.6899241237423495, + "learning_rate": 3.983875991512082e-06, + "loss": 0.9847, + "step": 922 + }, + { + "epoch": 0.06936720276566963, + "grad_norm": 1.6483714013610378, + "learning_rate": 3.983814238069249e-06, + "loss": 1.0431, + "step": 923 + }, + { + "epoch": 0.06944235683150458, + "grad_norm": 1.5380575704229353, + "learning_rate": 3.983752367077826e-06, + "loss": 0.9689, + "step": 924 + }, + { + "epoch": 0.06951751089733954, + "grad_norm": 1.6259151962659701, + "learning_rate": 3.983690378541478e-06, + "loss": 0.9514, + "step": 925 + }, + { + "epoch": 0.06959266496317451, + "grad_norm": 1.3965544569663522, + "learning_rate": 3.9836282724638805e-06, + "loss": 0.9921, + "step": 926 + }, + { + "epoch": 0.06966781902900947, + "grad_norm": 1.9126089376701685, + "learning_rate": 3.983566048848711e-06, + "loss": 1.0799, + "step": 927 + }, + { + "epoch": 0.06974297309484444, + "grad_norm": 1.5894117213547465, + "learning_rate": 3.983503707699658e-06, + "loss": 0.8926, + "step": 928 + }, + { + "epoch": 0.0698181271606794, + "grad_norm": 1.6771813437055851, + "learning_rate": 3.983441249020414e-06, + "loss": 1.0683, + "step": 929 + }, + { + "epoch": 0.06989328122651435, + "grad_norm": 1.7826105911167944, + "learning_rate": 3.983378672814682e-06, + "loss": 1.0029, + "step": 930 + }, + { + "epoch": 0.06996843529234932, + "grad_norm": 1.496518306402262, + "learning_rate": 3.983315979086169e-06, + "loss": 1.0698, + "step": 931 + }, + { + "epoch": 0.07004358935818428, + "grad_norm": 1.6060343022370842, + "learning_rate": 3.9832531678385885e-06, + "loss": 1.0213, + "step": 932 + }, + { + "epoch": 0.07011874342401923, + "grad_norm": 4.568242678674662, + "learning_rate": 3.983190239075664e-06, + "loss": 1.0956, + "step": 933 + }, + { + "epoch": 0.0701938974898542, + "grad_norm": 2.1118926670659737, + "learning_rate": 3.983127192801123e-06, + "loss": 0.9531, + "step": 934 + }, + { + "epoch": 0.07026905155568916, + "grad_norm": 1.7985557325061985, + "learning_rate": 3.983064029018703e-06, + "loss": 1.0713, + "step": 935 + }, + { + "epoch": 0.07034420562152413, + "grad_norm": 1.3649592088186708, + "learning_rate": 3.983000747732145e-06, + "loss": 0.9667, + "step": 936 + }, + { + "epoch": 0.07041935968735909, + "grad_norm": 1.7483693427470743, + "learning_rate": 3.9829373489452e-06, + "loss": 1.113, + "step": 937 + }, + { + "epoch": 0.07049451375319404, + "grad_norm": 1.6187254738146355, + "learning_rate": 3.982873832661623e-06, + "loss": 1.033, + "step": 938 + }, + { + "epoch": 0.07056966781902901, + "grad_norm": 1.4505447942739695, + "learning_rate": 3.982810198885179e-06, + "loss": 1.0785, + "step": 939 + }, + { + "epoch": 0.07064482188486397, + "grad_norm": 1.6466949074880775, + "learning_rate": 3.982746447619638e-06, + "loss": 1.0804, + "step": 940 + }, + { + "epoch": 0.07071997595069893, + "grad_norm": 1.7971590563944841, + "learning_rate": 3.982682578868777e-06, + "loss": 1.0323, + "step": 941 + }, + { + "epoch": 0.0707951300165339, + "grad_norm": 1.931940223286406, + "learning_rate": 3.982618592636381e-06, + "loss": 1.0689, + "step": 942 + }, + { + "epoch": 0.07087028408236885, + "grad_norm": 1.6986372731064592, + "learning_rate": 3.982554488926242e-06, + "loss": 1.0942, + "step": 943 + }, + { + "epoch": 0.07094543814820382, + "grad_norm": 1.6723379759718684, + "learning_rate": 3.982490267742158e-06, + "loss": 1.0011, + "step": 944 + }, + { + "epoch": 0.07102059221403878, + "grad_norm": 1.4549122965489756, + "learning_rate": 3.9824259290879336e-06, + "loss": 1.0425, + "step": 945 + }, + { + "epoch": 0.07109574627987374, + "grad_norm": 1.9006050200636795, + "learning_rate": 3.982361472967382e-06, + "loss": 0.9576, + "step": 946 + }, + { + "epoch": 0.0711709003457087, + "grad_norm": 2.0754386346319773, + "learning_rate": 3.982296899384322e-06, + "loss": 1.0106, + "step": 947 + }, + { + "epoch": 0.07124605441154366, + "grad_norm": 1.8749560033086807, + "learning_rate": 3.9822322083425805e-06, + "loss": 1.0222, + "step": 948 + }, + { + "epoch": 0.07132120847737862, + "grad_norm": 1.842543126174789, + "learning_rate": 3.982167399845989e-06, + "loss": 0.9796, + "step": 949 + }, + { + "epoch": 0.07139636254321359, + "grad_norm": 1.3467267797920928, + "learning_rate": 3.982102473898391e-06, + "loss": 0.9888, + "step": 950 + }, + { + "epoch": 0.07147151660904855, + "grad_norm": 1.719521990711703, + "learning_rate": 3.9820374305036295e-06, + "loss": 1.0244, + "step": 951 + }, + { + "epoch": 0.07154667067488352, + "grad_norm": 1.6984467254360047, + "learning_rate": 3.981972269665561e-06, + "loss": 0.9563, + "step": 952 + }, + { + "epoch": 0.07162182474071847, + "grad_norm": 0.9692113762107188, + "learning_rate": 3.981906991388046e-06, + "loss": 0.749, + "step": 953 + }, + { + "epoch": 0.07169697880655343, + "grad_norm": 1.7497435109322974, + "learning_rate": 3.981841595674952e-06, + "loss": 1.0662, + "step": 954 + }, + { + "epoch": 0.0717721328723884, + "grad_norm": 1.6588572008553328, + "learning_rate": 3.981776082530156e-06, + "loss": 1.0139, + "step": 955 + }, + { + "epoch": 0.07184728693822336, + "grad_norm": 1.7474283669215092, + "learning_rate": 3.981710451957537e-06, + "loss": 1.0254, + "step": 956 + }, + { + "epoch": 0.07192244100405831, + "grad_norm": 1.6184646060291952, + "learning_rate": 3.981644703960986e-06, + "loss": 1.065, + "step": 957 + }, + { + "epoch": 0.07199759506989328, + "grad_norm": 1.599216659835247, + "learning_rate": 3.981578838544398e-06, + "loss": 1.0795, + "step": 958 + }, + { + "epoch": 0.07207274913572824, + "grad_norm": 1.8598092477670483, + "learning_rate": 3.981512855711675e-06, + "loss": 0.954, + "step": 959 + }, + { + "epoch": 0.07214790320156321, + "grad_norm": 1.5760435219917883, + "learning_rate": 3.981446755466729e-06, + "loss": 1.0061, + "step": 960 + }, + { + "epoch": 0.07222305726739817, + "grad_norm": 1.5817100669820723, + "learning_rate": 3.981380537813474e-06, + "loss": 1.057, + "step": 961 + }, + { + "epoch": 0.07229821133323312, + "grad_norm": 1.8104645052940997, + "learning_rate": 3.981314202755835e-06, + "loss": 0.9468, + "step": 962 + }, + { + "epoch": 0.0723733653990681, + "grad_norm": 2.0639446863111073, + "learning_rate": 3.981247750297744e-06, + "loss": 1.0037, + "step": 963 + }, + { + "epoch": 0.07244851946490305, + "grad_norm": 1.2133499725392096, + "learning_rate": 3.9811811804431355e-06, + "loss": 0.8268, + "step": 964 + }, + { + "epoch": 0.07252367353073802, + "grad_norm": 2.0900972301736562, + "learning_rate": 3.981114493195956e-06, + "loss": 0.9598, + "step": 965 + }, + { + "epoch": 0.07259882759657298, + "grad_norm": 1.8914664421603804, + "learning_rate": 3.981047688560156e-06, + "loss": 0.978, + "step": 966 + }, + { + "epoch": 0.07267398166240793, + "grad_norm": 1.8715965535162515, + "learning_rate": 3.980980766539696e-06, + "loss": 1.0171, + "step": 967 + }, + { + "epoch": 0.0727491357282429, + "grad_norm": 1.4304855704034796, + "learning_rate": 3.980913727138539e-06, + "loss": 0.9891, + "step": 968 + }, + { + "epoch": 0.07282428979407786, + "grad_norm": 1.6373560718444413, + "learning_rate": 3.980846570360658e-06, + "loss": 0.922, + "step": 969 + }, + { + "epoch": 0.07289944385991282, + "grad_norm": 1.6125081498287568, + "learning_rate": 3.980779296210033e-06, + "loss": 1.03, + "step": 970 + }, + { + "epoch": 0.07297459792574779, + "grad_norm": 1.6037128434072907, + "learning_rate": 3.98071190469065e-06, + "loss": 1.0064, + "step": 971 + }, + { + "epoch": 0.07304975199158274, + "grad_norm": 1.8872983054213344, + "learning_rate": 3.980644395806502e-06, + "loss": 0.9648, + "step": 972 + }, + { + "epoch": 0.07312490605741771, + "grad_norm": 1.6250671226136717, + "learning_rate": 3.980576769561588e-06, + "loss": 0.9358, + "step": 973 + }, + { + "epoch": 0.07320006012325267, + "grad_norm": 1.9389621385888332, + "learning_rate": 3.980509025959918e-06, + "loss": 0.9829, + "step": 974 + }, + { + "epoch": 0.07327521418908763, + "grad_norm": 1.8898955269856936, + "learning_rate": 3.980441165005503e-06, + "loss": 1.0462, + "step": 975 + }, + { + "epoch": 0.0733503682549226, + "grad_norm": 1.6239184668514615, + "learning_rate": 3.9803731867023665e-06, + "loss": 1.0977, + "step": 976 + }, + { + "epoch": 0.07342552232075755, + "grad_norm": 1.6745363960354338, + "learning_rate": 3.980305091054534e-06, + "loss": 0.9927, + "step": 977 + }, + { + "epoch": 0.07350067638659251, + "grad_norm": 1.494144454651955, + "learning_rate": 3.980236878066042e-06, + "loss": 1.0215, + "step": 978 + }, + { + "epoch": 0.07357583045242748, + "grad_norm": 1.798317040353998, + "learning_rate": 3.9801685477409336e-06, + "loss": 0.9956, + "step": 979 + }, + { + "epoch": 0.07365098451826244, + "grad_norm": 1.667796856422815, + "learning_rate": 3.980100100083254e-06, + "loss": 0.8992, + "step": 980 + }, + { + "epoch": 0.0737261385840974, + "grad_norm": 1.4580887543988348, + "learning_rate": 3.980031535097063e-06, + "loss": 1.064, + "step": 981 + }, + { + "epoch": 0.07380129264993236, + "grad_norm": 0.8289726166039093, + "learning_rate": 3.9799628527864205e-06, + "loss": 0.8559, + "step": 982 + }, + { + "epoch": 0.07387644671576732, + "grad_norm": 1.9629117594477898, + "learning_rate": 3.979894053155398e-06, + "loss": 1.028, + "step": 983 + }, + { + "epoch": 0.07395160078160229, + "grad_norm": 1.5808242526251892, + "learning_rate": 3.979825136208071e-06, + "loss": 0.967, + "step": 984 + }, + { + "epoch": 0.07402675484743725, + "grad_norm": 1.717446474465629, + "learning_rate": 3.979756101948523e-06, + "loss": 1.0569, + "step": 985 + }, + { + "epoch": 0.0741019089132722, + "grad_norm": 1.8089295440120339, + "learning_rate": 3.979686950380845e-06, + "loss": 0.9794, + "step": 986 + }, + { + "epoch": 0.07417706297910717, + "grad_norm": 1.3626336688346925, + "learning_rate": 3.979617681509135e-06, + "loss": 0.9661, + "step": 987 + }, + { + "epoch": 0.07425221704494213, + "grad_norm": 0.8168440379642787, + "learning_rate": 3.979548295337496e-06, + "loss": 0.7879, + "step": 988 + }, + { + "epoch": 0.0743273711107771, + "grad_norm": 1.7604471764563512, + "learning_rate": 3.979478791870041e-06, + "loss": 1.0496, + "step": 989 + }, + { + "epoch": 0.07440252517661206, + "grad_norm": 1.465115127856977, + "learning_rate": 3.9794091711108875e-06, + "loss": 0.9467, + "step": 990 + }, + { + "epoch": 0.07447767924244701, + "grad_norm": 1.5156092692079908, + "learning_rate": 3.9793394330641614e-06, + "loss": 0.9858, + "step": 991 + }, + { + "epoch": 0.07455283330828198, + "grad_norm": 1.0601928007477628, + "learning_rate": 3.979269577733994e-06, + "loss": 0.8161, + "step": 992 + }, + { + "epoch": 0.07462798737411694, + "grad_norm": 0.8465804555572234, + "learning_rate": 3.979199605124525e-06, + "loss": 0.9064, + "step": 993 + }, + { + "epoch": 0.0747031414399519, + "grad_norm": 3.502817246243871, + "learning_rate": 3.979129515239901e-06, + "loss": 0.9099, + "step": 994 + }, + { + "epoch": 0.07477829550578687, + "grad_norm": 1.7603466162955448, + "learning_rate": 3.979059308084274e-06, + "loss": 1.0468, + "step": 995 + }, + { + "epoch": 0.07485344957162182, + "grad_norm": 2.0704428577768903, + "learning_rate": 3.9789889836618045e-06, + "loss": 0.9862, + "step": 996 + }, + { + "epoch": 0.07492860363745679, + "grad_norm": 1.649283314158959, + "learning_rate": 3.97891854197666e-06, + "loss": 1.0351, + "step": 997 + }, + { + "epoch": 0.07500375770329175, + "grad_norm": 1.538601336638201, + "learning_rate": 3.978847983033014e-06, + "loss": 0.9477, + "step": 998 + }, + { + "epoch": 0.0750789117691267, + "grad_norm": 0.9940064944932028, + "learning_rate": 3.978777306835048e-06, + "loss": 0.8251, + "step": 999 + }, + { + "epoch": 0.07515406583496168, + "grad_norm": 1.5470262703961353, + "learning_rate": 3.978706513386949e-06, + "loss": 0.9764, + "step": 1000 + }, + { + "epoch": 0.07522921990079663, + "grad_norm": 1.8994957756764672, + "learning_rate": 3.978635602692912e-06, + "loss": 1.0457, + "step": 1001 + }, + { + "epoch": 0.07530437396663159, + "grad_norm": 1.6910536753179648, + "learning_rate": 3.978564574757139e-06, + "loss": 1.0508, + "step": 1002 + }, + { + "epoch": 0.07537952803246656, + "grad_norm": 1.489466077906814, + "learning_rate": 3.9784934295838385e-06, + "loss": 0.9456, + "step": 1003 + }, + { + "epoch": 0.07545468209830151, + "grad_norm": 1.8872732291510779, + "learning_rate": 3.978422167177226e-06, + "loss": 1.0007, + "step": 1004 + }, + { + "epoch": 0.07552983616413649, + "grad_norm": 1.789206753041695, + "learning_rate": 3.9783507875415245e-06, + "loss": 1.1525, + "step": 1005 + }, + { + "epoch": 0.07560499022997144, + "grad_norm": 1.3876148176370324, + "learning_rate": 3.9782792906809625e-06, + "loss": 1.0264, + "step": 1006 + }, + { + "epoch": 0.0756801442958064, + "grad_norm": 1.7912805791828215, + "learning_rate": 3.978207676599778e-06, + "loss": 1.0426, + "step": 1007 + }, + { + "epoch": 0.07575529836164137, + "grad_norm": 1.5113997406209572, + "learning_rate": 3.978135945302213e-06, + "loss": 0.9973, + "step": 1008 + }, + { + "epoch": 0.07583045242747632, + "grad_norm": 1.5998438103257906, + "learning_rate": 3.978064096792519e-06, + "loss": 0.9407, + "step": 1009 + }, + { + "epoch": 0.0759056064933113, + "grad_norm": 1.511916036817678, + "learning_rate": 3.977992131074953e-06, + "loss": 1.016, + "step": 1010 + }, + { + "epoch": 0.07598076055914625, + "grad_norm": 1.4328006556696327, + "learning_rate": 3.9779200481537775e-06, + "loss": 1.0135, + "step": 1011 + }, + { + "epoch": 0.07605591462498121, + "grad_norm": 1.5987177449328749, + "learning_rate": 3.977847848033267e-06, + "loss": 1.0087, + "step": 1012 + }, + { + "epoch": 0.07613106869081618, + "grad_norm": 1.8638888793821504, + "learning_rate": 3.977775530717696e-06, + "loss": 1.0001, + "step": 1013 + }, + { + "epoch": 0.07620622275665113, + "grad_norm": 2.313297031275324, + "learning_rate": 3.977703096211354e-06, + "loss": 0.982, + "step": 1014 + }, + { + "epoch": 0.07628137682248609, + "grad_norm": 1.439049590443249, + "learning_rate": 3.977630544518529e-06, + "loss": 0.9838, + "step": 1015 + }, + { + "epoch": 0.07635653088832106, + "grad_norm": 1.658228933540138, + "learning_rate": 3.97755787564352e-06, + "loss": 1.1041, + "step": 1016 + }, + { + "epoch": 0.07643168495415602, + "grad_norm": 1.7541392963777995, + "learning_rate": 3.977485089590636e-06, + "loss": 0.9845, + "step": 1017 + }, + { + "epoch": 0.07650683901999099, + "grad_norm": 1.5830269399966403, + "learning_rate": 3.977412186364187e-06, + "loss": 1.0305, + "step": 1018 + }, + { + "epoch": 0.07658199308582594, + "grad_norm": 1.7333208609401247, + "learning_rate": 3.977339165968495e-06, + "loss": 1.0959, + "step": 1019 + }, + { + "epoch": 0.0766571471516609, + "grad_norm": 2.0913117288712493, + "learning_rate": 3.977266028407885e-06, + "loss": 0.9912, + "step": 1020 + }, + { + "epoch": 0.07673230121749587, + "grad_norm": 1.6265718487297756, + "learning_rate": 3.977192773686692e-06, + "loss": 1.0162, + "step": 1021 + }, + { + "epoch": 0.07680745528333083, + "grad_norm": 1.2905124254984552, + "learning_rate": 3.977119401809255e-06, + "loss": 0.9657, + "step": 1022 + }, + { + "epoch": 0.07688260934916578, + "grad_norm": 1.2764751035546333, + "learning_rate": 3.977045912779924e-06, + "loss": 0.9938, + "step": 1023 + }, + { + "epoch": 0.07695776341500075, + "grad_norm": 1.2834393009593914, + "learning_rate": 3.9769723066030505e-06, + "loss": 0.9536, + "step": 1024 + }, + { + "epoch": 0.07703291748083571, + "grad_norm": 1.9757139171079472, + "learning_rate": 3.976898583282998e-06, + "loss": 0.8834, + "step": 1025 + }, + { + "epoch": 0.07710807154667068, + "grad_norm": 1.2111903757196727, + "learning_rate": 3.976824742824135e-06, + "loss": 0.9808, + "step": 1026 + }, + { + "epoch": 0.07718322561250564, + "grad_norm": 1.6778311473927814, + "learning_rate": 3.976750785230835e-06, + "loss": 1.0126, + "step": 1027 + }, + { + "epoch": 0.0772583796783406, + "grad_norm": 1.3961093639147462, + "learning_rate": 3.976676710507483e-06, + "loss": 0.9855, + "step": 1028 + }, + { + "epoch": 0.07733353374417556, + "grad_norm": 1.7279996112496327, + "learning_rate": 3.976602518658466e-06, + "loss": 1.0778, + "step": 1029 + }, + { + "epoch": 0.07740868781001052, + "grad_norm": 1.7700404093596256, + "learning_rate": 3.976528209688181e-06, + "loss": 0.9902, + "step": 1030 + }, + { + "epoch": 0.07748384187584548, + "grad_norm": 1.393970750852134, + "learning_rate": 3.976453783601031e-06, + "loss": 0.9983, + "step": 1031 + }, + { + "epoch": 0.07755899594168045, + "grad_norm": 1.6137116899382642, + "learning_rate": 3.976379240401426e-06, + "loss": 0.9931, + "step": 1032 + }, + { + "epoch": 0.0776341500075154, + "grad_norm": 1.6317094060945718, + "learning_rate": 3.976304580093782e-06, + "loss": 0.9163, + "step": 1033 + }, + { + "epoch": 0.07770930407335037, + "grad_norm": 1.62882655106565, + "learning_rate": 3.976229802682524e-06, + "loss": 1.0508, + "step": 1034 + }, + { + "epoch": 0.07778445813918533, + "grad_norm": 1.6390305536900662, + "learning_rate": 3.9761549081720845e-06, + "loss": 0.9415, + "step": 1035 + }, + { + "epoch": 0.07785961220502029, + "grad_norm": 1.6889173418849504, + "learning_rate": 3.976079896566898e-06, + "loss": 0.9244, + "step": 1036 + }, + { + "epoch": 0.07793476627085526, + "grad_norm": 1.6272040605224314, + "learning_rate": 3.976004767871411e-06, + "loss": 0.9837, + "step": 1037 + }, + { + "epoch": 0.07800992033669021, + "grad_norm": 1.8910084485329768, + "learning_rate": 3.975929522090075e-06, + "loss": 0.9497, + "step": 1038 + }, + { + "epoch": 0.07808507440252517, + "grad_norm": 1.5232102444680902, + "learning_rate": 3.9758541592273485e-06, + "loss": 1.0348, + "step": 1039 + }, + { + "epoch": 0.07816022846836014, + "grad_norm": 1.5157207183501078, + "learning_rate": 3.975778679287697e-06, + "loss": 1.0185, + "step": 1040 + }, + { + "epoch": 0.0782353825341951, + "grad_norm": 0.8416093232351876, + "learning_rate": 3.975703082275592e-06, + "loss": 0.812, + "step": 1041 + }, + { + "epoch": 0.07831053660003007, + "grad_norm": 1.7076487526391768, + "learning_rate": 3.975627368195515e-06, + "loss": 0.9153, + "step": 1042 + }, + { + "epoch": 0.07838569066586502, + "grad_norm": 1.7748052221536488, + "learning_rate": 3.9755515370519515e-06, + "loss": 0.9803, + "step": 1043 + }, + { + "epoch": 0.07846084473169998, + "grad_norm": 1.7070332053585768, + "learning_rate": 3.975475588849394e-06, + "loss": 1.051, + "step": 1044 + }, + { + "epoch": 0.07853599879753495, + "grad_norm": 1.8728411007560226, + "learning_rate": 3.975399523592343e-06, + "loss": 1.0506, + "step": 1045 + }, + { + "epoch": 0.0786111528633699, + "grad_norm": 1.3159204599785093, + "learning_rate": 3.975323341285306e-06, + "loss": 1.028, + "step": 1046 + }, + { + "epoch": 0.07868630692920486, + "grad_norm": 1.3867539092428818, + "learning_rate": 3.975247041932797e-06, + "loss": 1.0082, + "step": 1047 + }, + { + "epoch": 0.07876146099503983, + "grad_norm": 1.6263988782686492, + "learning_rate": 3.975170625539338e-06, + "loss": 0.9575, + "step": 1048 + }, + { + "epoch": 0.07883661506087479, + "grad_norm": 1.4260194344339852, + "learning_rate": 3.975094092109455e-06, + "loss": 1.0552, + "step": 1049 + }, + { + "epoch": 0.07891176912670976, + "grad_norm": 1.514526913242237, + "learning_rate": 3.975017441647684e-06, + "loss": 0.9626, + "step": 1050 + }, + { + "epoch": 0.07898692319254472, + "grad_norm": 1.6052087726464517, + "learning_rate": 3.974940674158567e-06, + "loss": 0.9481, + "step": 1051 + }, + { + "epoch": 0.07906207725837967, + "grad_norm": 0.7378287568617866, + "learning_rate": 3.9748637896466526e-06, + "loss": 0.7697, + "step": 1052 + }, + { + "epoch": 0.07913723132421464, + "grad_norm": 1.4430535886863511, + "learning_rate": 3.974786788116496e-06, + "loss": 0.9594, + "step": 1053 + }, + { + "epoch": 0.0792123853900496, + "grad_norm": 1.5382841824125333, + "learning_rate": 3.974709669572661e-06, + "loss": 0.9909, + "step": 1054 + }, + { + "epoch": 0.07928753945588456, + "grad_norm": 1.629503831178247, + "learning_rate": 3.974632434019716e-06, + "loss": 0.9327, + "step": 1055 + }, + { + "epoch": 0.07936269352171953, + "grad_norm": 1.4633486836583411, + "learning_rate": 3.974555081462237e-06, + "loss": 0.9818, + "step": 1056 + }, + { + "epoch": 0.07943784758755448, + "grad_norm": 1.8682756644862002, + "learning_rate": 3.97447761190481e-06, + "loss": 0.9577, + "step": 1057 + }, + { + "epoch": 0.07951300165338945, + "grad_norm": 1.89667520240691, + "learning_rate": 3.974400025352022e-06, + "loss": 1.1083, + "step": 1058 + }, + { + "epoch": 0.07958815571922441, + "grad_norm": 1.4547099012377118, + "learning_rate": 3.974322321808473e-06, + "loss": 1.0414, + "step": 1059 + }, + { + "epoch": 0.07966330978505937, + "grad_norm": 1.733774167025121, + "learning_rate": 3.974244501278766e-06, + "loss": 1.0255, + "step": 1060 + }, + { + "epoch": 0.07973846385089434, + "grad_norm": 1.498665613466506, + "learning_rate": 3.974166563767513e-06, + "loss": 0.9353, + "step": 1061 + }, + { + "epoch": 0.0798136179167293, + "grad_norm": 1.7196590447155569, + "learning_rate": 3.974088509279331e-06, + "loss": 1.0574, + "step": 1062 + }, + { + "epoch": 0.07988877198256426, + "grad_norm": 1.4897664986772439, + "learning_rate": 3.9740103378188455e-06, + "loss": 0.9378, + "step": 1063 + }, + { + "epoch": 0.07996392604839922, + "grad_norm": 2.156091435448677, + "learning_rate": 3.973932049390688e-06, + "loss": 1.1802, + "step": 1064 + }, + { + "epoch": 0.08003908011423418, + "grad_norm": 1.479839906023938, + "learning_rate": 3.973853643999499e-06, + "loss": 0.9874, + "step": 1065 + }, + { + "epoch": 0.08011423418006915, + "grad_norm": 1.765208372400251, + "learning_rate": 3.973775121649922e-06, + "loss": 1.0457, + "step": 1066 + }, + { + "epoch": 0.0801893882459041, + "grad_norm": 1.4551231838136665, + "learning_rate": 3.973696482346611e-06, + "loss": 0.922, + "step": 1067 + }, + { + "epoch": 0.08026454231173906, + "grad_norm": 1.540614473566148, + "learning_rate": 3.973617726094227e-06, + "loss": 1.0406, + "step": 1068 + }, + { + "epoch": 0.08033969637757403, + "grad_norm": 1.3868865435456388, + "learning_rate": 3.973538852897435e-06, + "loss": 1.0434, + "step": 1069 + }, + { + "epoch": 0.08041485044340899, + "grad_norm": 1.7074363803880102, + "learning_rate": 3.973459862760908e-06, + "loss": 1.1124, + "step": 1070 + }, + { + "epoch": 0.08049000450924396, + "grad_norm": 1.7862939193590486, + "learning_rate": 3.973380755689328e-06, + "loss": 0.9894, + "step": 1071 + }, + { + "epoch": 0.08056515857507891, + "grad_norm": 1.5301951453478309, + "learning_rate": 3.97330153168738e-06, + "loss": 1.0552, + "step": 1072 + }, + { + "epoch": 0.08064031264091387, + "grad_norm": 1.7952131464651921, + "learning_rate": 3.973222190759761e-06, + "loss": 1.0612, + "step": 1073 + }, + { + "epoch": 0.08071546670674884, + "grad_norm": 1.6461706258513706, + "learning_rate": 3.973142732911172e-06, + "loss": 1.061, + "step": 1074 + }, + { + "epoch": 0.0807906207725838, + "grad_norm": 1.4838076661753725, + "learning_rate": 3.9730631581463195e-06, + "loss": 1.0073, + "step": 1075 + }, + { + "epoch": 0.08086577483841875, + "grad_norm": 2.139034153462767, + "learning_rate": 3.972983466469919e-06, + "loss": 1.0548, + "step": 1076 + }, + { + "epoch": 0.08094092890425372, + "grad_norm": 2.2711801759844303, + "learning_rate": 3.972903657886695e-06, + "loss": 1.0145, + "step": 1077 + }, + { + "epoch": 0.08101608297008868, + "grad_norm": 2.394362376793924, + "learning_rate": 3.972823732401373e-06, + "loss": 1.0598, + "step": 1078 + }, + { + "epoch": 0.08109123703592365, + "grad_norm": 1.4132755260446508, + "learning_rate": 3.972743690018691e-06, + "loss": 1.0055, + "step": 1079 + }, + { + "epoch": 0.0811663911017586, + "grad_norm": 2.8014717900089705, + "learning_rate": 3.9726635307433906e-06, + "loss": 1.1258, + "step": 1080 + }, + { + "epoch": 0.08124154516759356, + "grad_norm": 1.4873220802136764, + "learning_rate": 3.972583254580223e-06, + "loss": 0.9567, + "step": 1081 + }, + { + "epoch": 0.08131669923342853, + "grad_norm": 1.4719654355137446, + "learning_rate": 3.972502861533943e-06, + "loss": 1.0269, + "step": 1082 + }, + { + "epoch": 0.08139185329926349, + "grad_norm": 1.7980844159364422, + "learning_rate": 3.9724223516093154e-06, + "loss": 1.0432, + "step": 1083 + }, + { + "epoch": 0.08146700736509845, + "grad_norm": 0.9432472178127206, + "learning_rate": 3.972341724811111e-06, + "loss": 0.8194, + "step": 1084 + }, + { + "epoch": 0.08154216143093342, + "grad_norm": 1.4776477357744737, + "learning_rate": 3.972260981144107e-06, + "loss": 1.0623, + "step": 1085 + }, + { + "epoch": 0.08161731549676837, + "grad_norm": 1.8775998211165328, + "learning_rate": 3.972180120613087e-06, + "loss": 1.0729, + "step": 1086 + }, + { + "epoch": 0.08169246956260334, + "grad_norm": 2.370674209305243, + "learning_rate": 3.972099143222844e-06, + "loss": 1.0615, + "step": 1087 + }, + { + "epoch": 0.0817676236284383, + "grad_norm": 1.497497537448495, + "learning_rate": 3.972018048978174e-06, + "loss": 1.0466, + "step": 1088 + }, + { + "epoch": 0.08184277769427326, + "grad_norm": 2.119940625204992, + "learning_rate": 3.971936837883884e-06, + "loss": 1.0487, + "step": 1089 + }, + { + "epoch": 0.08191793176010823, + "grad_norm": 1.6336937675342715, + "learning_rate": 3.971855509944784e-06, + "loss": 1.0179, + "step": 1090 + }, + { + "epoch": 0.08199308582594318, + "grad_norm": 2.195730692173448, + "learning_rate": 3.971774065165696e-06, + "loss": 0.9357, + "step": 1091 + }, + { + "epoch": 0.08206823989177814, + "grad_norm": 1.3281885917249454, + "learning_rate": 3.971692503551443e-06, + "loss": 1.0864, + "step": 1092 + }, + { + "epoch": 0.08214339395761311, + "grad_norm": 1.8311509089301283, + "learning_rate": 3.971610825106859e-06, + "loss": 0.9676, + "step": 1093 + }, + { + "epoch": 0.08221854802344807, + "grad_norm": 1.7814063957220319, + "learning_rate": 3.971529029836785e-06, + "loss": 0.8584, + "step": 1094 + }, + { + "epoch": 0.08229370208928304, + "grad_norm": 1.9878540779302798, + "learning_rate": 3.971447117746065e-06, + "loss": 0.8498, + "step": 1095 + }, + { + "epoch": 0.08236885615511799, + "grad_norm": 1.825592503975098, + "learning_rate": 3.9713650888395555e-06, + "loss": 0.9624, + "step": 1096 + }, + { + "epoch": 0.08244401022095295, + "grad_norm": 1.6761329833056862, + "learning_rate": 3.971282943122115e-06, + "loss": 0.9844, + "step": 1097 + }, + { + "epoch": 0.08251916428678792, + "grad_norm": 1.9153503445101019, + "learning_rate": 3.971200680598611e-06, + "loss": 0.9442, + "step": 1098 + }, + { + "epoch": 0.08259431835262288, + "grad_norm": 1.4581872685262902, + "learning_rate": 3.971118301273919e-06, + "loss": 0.941, + "step": 1099 + }, + { + "epoch": 0.08266947241845783, + "grad_norm": 1.613893037303, + "learning_rate": 3.97103580515292e-06, + "loss": 0.9692, + "step": 1100 + }, + { + "epoch": 0.0827446264842928, + "grad_norm": 1.6036099628014353, + "learning_rate": 3.970953192240502e-06, + "loss": 1.0575, + "step": 1101 + }, + { + "epoch": 0.08281978055012776, + "grad_norm": 1.5738774159112747, + "learning_rate": 3.970870462541559e-06, + "loss": 1.0168, + "step": 1102 + }, + { + "epoch": 0.08289493461596273, + "grad_norm": 1.4057057917110667, + "learning_rate": 3.970787616060995e-06, + "loss": 0.9907, + "step": 1103 + }, + { + "epoch": 0.08297008868179769, + "grad_norm": 1.6050476115880419, + "learning_rate": 3.970704652803718e-06, + "loss": 0.9815, + "step": 1104 + }, + { + "epoch": 0.08304524274763264, + "grad_norm": 1.2964086958177181, + "learning_rate": 3.970621572774644e-06, + "loss": 0.9991, + "step": 1105 + }, + { + "epoch": 0.08312039681346761, + "grad_norm": 1.7182122516368008, + "learning_rate": 3.970538375978695e-06, + "loss": 0.9198, + "step": 1106 + }, + { + "epoch": 0.08319555087930257, + "grad_norm": 1.3881896908116633, + "learning_rate": 3.970455062420803e-06, + "loss": 0.9741, + "step": 1107 + }, + { + "epoch": 0.08327070494513754, + "grad_norm": 1.7661608667689102, + "learning_rate": 3.9703716321059026e-06, + "loss": 1.0662, + "step": 1108 + }, + { + "epoch": 0.0833458590109725, + "grad_norm": 0.9100960642025463, + "learning_rate": 3.970288085038938e-06, + "loss": 0.8397, + "step": 1109 + }, + { + "epoch": 0.08342101307680745, + "grad_norm": 1.5327952626043628, + "learning_rate": 3.970204421224859e-06, + "loss": 0.9851, + "step": 1110 + }, + { + "epoch": 0.08349616714264242, + "grad_norm": 1.3177718598302768, + "learning_rate": 3.970120640668623e-06, + "loss": 1.0983, + "step": 1111 + }, + { + "epoch": 0.08357132120847738, + "grad_norm": 1.4228787913757446, + "learning_rate": 3.970036743375197e-06, + "loss": 1.0034, + "step": 1112 + }, + { + "epoch": 0.08364647527431233, + "grad_norm": 1.358620050317219, + "learning_rate": 3.9699527293495485e-06, + "loss": 0.9176, + "step": 1113 + }, + { + "epoch": 0.0837216293401473, + "grad_norm": 1.6786766381612699, + "learning_rate": 3.969868598596658e-06, + "loss": 1.0413, + "step": 1114 + }, + { + "epoch": 0.08379678340598226, + "grad_norm": 1.5412017361771493, + "learning_rate": 3.9697843511215104e-06, + "loss": 0.9988, + "step": 1115 + }, + { + "epoch": 0.08387193747181723, + "grad_norm": 6.455691513027628, + "learning_rate": 3.969699986929096e-06, + "loss": 0.9789, + "step": 1116 + }, + { + "epoch": 0.08394709153765219, + "grad_norm": 1.6242991183068536, + "learning_rate": 3.969615506024416e-06, + "loss": 0.9537, + "step": 1117 + }, + { + "epoch": 0.08402224560348714, + "grad_norm": 3.7088446246792555, + "learning_rate": 3.969530908412474e-06, + "loss": 1.0181, + "step": 1118 + }, + { + "epoch": 0.08409739966932211, + "grad_norm": 1.9039135356351997, + "learning_rate": 3.969446194098286e-06, + "loss": 1.0704, + "step": 1119 + }, + { + "epoch": 0.08417255373515707, + "grad_norm": 1.5450253458011913, + "learning_rate": 3.969361363086867e-06, + "loss": 1.0581, + "step": 1120 + }, + { + "epoch": 0.08424770780099203, + "grad_norm": 1.7063713980724682, + "learning_rate": 3.969276415383248e-06, + "loss": 1.0258, + "step": 1121 + }, + { + "epoch": 0.084322861866827, + "grad_norm": 1.469844491479313, + "learning_rate": 3.9691913509924586e-06, + "loss": 1.0641, + "step": 1122 + }, + { + "epoch": 0.08439801593266195, + "grad_norm": 1.8010325225486163, + "learning_rate": 3.969106169919542e-06, + "loss": 1.0879, + "step": 1123 + }, + { + "epoch": 0.08447316999849692, + "grad_norm": 1.7055797560000379, + "learning_rate": 3.969020872169545e-06, + "loss": 1.0985, + "step": 1124 + }, + { + "epoch": 0.08454832406433188, + "grad_norm": 2.210880346199767, + "learning_rate": 3.9689354577475206e-06, + "loss": 0.9964, + "step": 1125 + }, + { + "epoch": 0.08462347813016684, + "grad_norm": 2.538360216879631, + "learning_rate": 3.968849926658532e-06, + "loss": 0.9176, + "step": 1126 + }, + { + "epoch": 0.08469863219600181, + "grad_norm": 1.590011250438114, + "learning_rate": 3.968764278907645e-06, + "loss": 0.9465, + "step": 1127 + }, + { + "epoch": 0.08477378626183676, + "grad_norm": 2.1172582185202637, + "learning_rate": 3.968678514499936e-06, + "loss": 1.0701, + "step": 1128 + }, + { + "epoch": 0.08484894032767172, + "grad_norm": 1.5317305764420102, + "learning_rate": 3.968592633440486e-06, + "loss": 0.9488, + "step": 1129 + }, + { + "epoch": 0.08492409439350669, + "grad_norm": 1.5295570512739323, + "learning_rate": 3.968506635734385e-06, + "loss": 0.9853, + "step": 1130 + }, + { + "epoch": 0.08499924845934165, + "grad_norm": 1.6766336047336325, + "learning_rate": 3.9684205213867275e-06, + "loss": 0.9897, + "step": 1131 + }, + { + "epoch": 0.08507440252517662, + "grad_norm": 1.7151230430394468, + "learning_rate": 3.968334290402616e-06, + "loss": 0.9968, + "step": 1132 + }, + { + "epoch": 0.08514955659101157, + "grad_norm": 2.131493911881223, + "learning_rate": 3.968247942787161e-06, + "loss": 1.0482, + "step": 1133 + }, + { + "epoch": 0.08522471065684653, + "grad_norm": 1.7830892737520339, + "learning_rate": 3.968161478545479e-06, + "loss": 1.0356, + "step": 1134 + }, + { + "epoch": 0.0852998647226815, + "grad_norm": 1.4720963848164206, + "learning_rate": 3.968074897682692e-06, + "loss": 1.0011, + "step": 1135 + }, + { + "epoch": 0.08537501878851646, + "grad_norm": 2.0207322347724443, + "learning_rate": 3.967988200203931e-06, + "loss": 0.9806, + "step": 1136 + }, + { + "epoch": 0.08545017285435141, + "grad_norm": 1.5571388586425439, + "learning_rate": 3.967901386114334e-06, + "loss": 0.9212, + "step": 1137 + }, + { + "epoch": 0.08552532692018638, + "grad_norm": 2.5249061680345966, + "learning_rate": 3.967814455419044e-06, + "loss": 0.9969, + "step": 1138 + }, + { + "epoch": 0.08560048098602134, + "grad_norm": 1.7666415773565527, + "learning_rate": 3.9677274081232116e-06, + "loss": 0.9334, + "step": 1139 + }, + { + "epoch": 0.08567563505185631, + "grad_norm": 1.5895398198576103, + "learning_rate": 3.967640244231996e-06, + "loss": 1.1303, + "step": 1140 + }, + { + "epoch": 0.08575078911769127, + "grad_norm": 1.7158895097564943, + "learning_rate": 3.9675529637505615e-06, + "loss": 1.0441, + "step": 1141 + }, + { + "epoch": 0.08582594318352622, + "grad_norm": 1.481426160936277, + "learning_rate": 3.967465566684079e-06, + "loss": 1.0054, + "step": 1142 + }, + { + "epoch": 0.0859010972493612, + "grad_norm": 1.5233161291045478, + "learning_rate": 3.967378053037728e-06, + "loss": 1.0427, + "step": 1143 + }, + { + "epoch": 0.08597625131519615, + "grad_norm": 1.8453936605023575, + "learning_rate": 3.967290422816693e-06, + "loss": 1.0529, + "step": 1144 + }, + { + "epoch": 0.0860514053810311, + "grad_norm": 1.6255686571916224, + "learning_rate": 3.967202676026169e-06, + "loss": 0.9724, + "step": 1145 + }, + { + "epoch": 0.08612655944686608, + "grad_norm": 1.944314969005781, + "learning_rate": 3.967114812671353e-06, + "loss": 1.0459, + "step": 1146 + }, + { + "epoch": 0.08620171351270103, + "grad_norm": 1.6701552200135767, + "learning_rate": 3.967026832757451e-06, + "loss": 1.102, + "step": 1147 + }, + { + "epoch": 0.086276867578536, + "grad_norm": 1.401579794888655, + "learning_rate": 3.966938736289677e-06, + "loss": 0.9965, + "step": 1148 + }, + { + "epoch": 0.08635202164437096, + "grad_norm": 1.5212214074938848, + "learning_rate": 3.966850523273251e-06, + "loss": 1.096, + "step": 1149 + }, + { + "epoch": 0.08642717571020592, + "grad_norm": 1.7754275545784448, + "learning_rate": 3.9667621937134e-06, + "loss": 1.0456, + "step": 1150 + }, + { + "epoch": 0.08650232977604089, + "grad_norm": 1.7491510239268633, + "learning_rate": 3.966673747615358e-06, + "loss": 1.0077, + "step": 1151 + }, + { + "epoch": 0.08657748384187584, + "grad_norm": 1.6154268419943807, + "learning_rate": 3.966585184984366e-06, + "loss": 0.9735, + "step": 1152 + }, + { + "epoch": 0.08665263790771081, + "grad_norm": 1.6127695375842959, + "learning_rate": 3.96649650582567e-06, + "loss": 0.9588, + "step": 1153 + }, + { + "epoch": 0.08672779197354577, + "grad_norm": 1.6029292531812955, + "learning_rate": 3.966407710144527e-06, + "loss": 0.9921, + "step": 1154 + }, + { + "epoch": 0.08680294603938073, + "grad_norm": 0.9127292527970373, + "learning_rate": 3.966318797946196e-06, + "loss": 0.806, + "step": 1155 + }, + { + "epoch": 0.0868781001052157, + "grad_norm": 1.4905727097398638, + "learning_rate": 3.966229769235948e-06, + "loss": 1.093, + "step": 1156 + }, + { + "epoch": 0.08695325417105065, + "grad_norm": 1.6187100085175246, + "learning_rate": 3.966140624019056e-06, + "loss": 0.9691, + "step": 1157 + }, + { + "epoch": 0.08702840823688561, + "grad_norm": 1.4709159626142774, + "learning_rate": 3.966051362300804e-06, + "loss": 0.9564, + "step": 1158 + }, + { + "epoch": 0.08710356230272058, + "grad_norm": 1.5398728614195873, + "learning_rate": 3.9659619840864795e-06, + "loss": 0.9377, + "step": 1159 + }, + { + "epoch": 0.08717871636855554, + "grad_norm": 1.8826993536290906, + "learning_rate": 3.96587248938138e-06, + "loss": 1.0716, + "step": 1160 + }, + { + "epoch": 0.0872538704343905, + "grad_norm": 0.7319453609403711, + "learning_rate": 3.965782878190807e-06, + "loss": 0.7792, + "step": 1161 + }, + { + "epoch": 0.08732902450022546, + "grad_norm": 1.7472775945630843, + "learning_rate": 3.965693150520071e-06, + "loss": 0.9287, + "step": 1162 + }, + { + "epoch": 0.08740417856606042, + "grad_norm": 2.206352845299275, + "learning_rate": 3.965603306374489e-06, + "loss": 0.9378, + "step": 1163 + }, + { + "epoch": 0.08747933263189539, + "grad_norm": 1.2712794078624985, + "learning_rate": 3.9655133457593845e-06, + "loss": 1.0224, + "step": 1164 + }, + { + "epoch": 0.08755448669773035, + "grad_norm": 0.9695414109969411, + "learning_rate": 3.965423268680087e-06, + "loss": 0.8307, + "step": 1165 + }, + { + "epoch": 0.0876296407635653, + "grad_norm": 1.5877971551277918, + "learning_rate": 3.965333075141936e-06, + "loss": 1.0811, + "step": 1166 + }, + { + "epoch": 0.08770479482940027, + "grad_norm": 1.693132044262488, + "learning_rate": 3.965242765150274e-06, + "loss": 1.0096, + "step": 1167 + }, + { + "epoch": 0.08777994889523523, + "grad_norm": 2.3117773628088223, + "learning_rate": 3.9651523387104526e-06, + "loss": 0.9688, + "step": 1168 + }, + { + "epoch": 0.0878551029610702, + "grad_norm": 1.776519651407881, + "learning_rate": 3.96506179582783e-06, + "loss": 0.985, + "step": 1169 + }, + { + "epoch": 0.08793025702690516, + "grad_norm": 2.3491840691094987, + "learning_rate": 3.964971136507771e-06, + "loss": 1.1069, + "step": 1170 + }, + { + "epoch": 0.08800541109274011, + "grad_norm": 2.3398239822698996, + "learning_rate": 3.964880360755648e-06, + "loss": 1.0379, + "step": 1171 + }, + { + "epoch": 0.08808056515857508, + "grad_norm": 1.6220032863560132, + "learning_rate": 3.96478946857684e-06, + "loss": 1.0468, + "step": 1172 + }, + { + "epoch": 0.08815571922441004, + "grad_norm": 2.187537165734778, + "learning_rate": 3.964698459976732e-06, + "loss": 0.9927, + "step": 1173 + }, + { + "epoch": 0.088230873290245, + "grad_norm": 1.8603967375789654, + "learning_rate": 3.9646073349607165e-06, + "loss": 1.1538, + "step": 1174 + }, + { + "epoch": 0.08830602735607997, + "grad_norm": 1.514550526231723, + "learning_rate": 3.964516093534194e-06, + "loss": 1.0275, + "step": 1175 + }, + { + "epoch": 0.08838118142191492, + "grad_norm": 1.6802622676674535, + "learning_rate": 3.964424735702571e-06, + "loss": 0.8999, + "step": 1176 + }, + { + "epoch": 0.0884563354877499, + "grad_norm": 1.4268796269546058, + "learning_rate": 3.964333261471258e-06, + "loss": 0.9649, + "step": 1177 + }, + { + "epoch": 0.08853148955358485, + "grad_norm": 1.8839979436624243, + "learning_rate": 3.964241670845679e-06, + "loss": 0.9672, + "step": 1178 + }, + { + "epoch": 0.0886066436194198, + "grad_norm": 2.0229881700894894, + "learning_rate": 3.9641499638312595e-06, + "loss": 1.0339, + "step": 1179 + }, + { + "epoch": 0.08868179768525478, + "grad_norm": 1.432376771190243, + "learning_rate": 3.964058140433434e-06, + "loss": 0.9881, + "step": 1180 + }, + { + "epoch": 0.08875695175108973, + "grad_norm": 1.6254934148687334, + "learning_rate": 3.9639662006576405e-06, + "loss": 1.0232, + "step": 1181 + }, + { + "epoch": 0.08883210581692469, + "grad_norm": 1.6316037609273029, + "learning_rate": 3.963874144509331e-06, + "loss": 1.0413, + "step": 1182 + }, + { + "epoch": 0.08890725988275966, + "grad_norm": 1.2281877962709675, + "learning_rate": 3.963781971993957e-06, + "loss": 0.8316, + "step": 1183 + }, + { + "epoch": 0.08898241394859462, + "grad_norm": 1.972591285768932, + "learning_rate": 3.963689683116982e-06, + "loss": 0.9325, + "step": 1184 + }, + { + "epoch": 0.08905756801442959, + "grad_norm": 1.690197660410197, + "learning_rate": 3.963597277883874e-06, + "loss": 1.0326, + "step": 1185 + }, + { + "epoch": 0.08913272208026454, + "grad_norm": 1.7282946429187644, + "learning_rate": 3.963504756300107e-06, + "loss": 1.0259, + "step": 1186 + }, + { + "epoch": 0.0892078761460995, + "grad_norm": 1.0092085402852717, + "learning_rate": 3.963412118371166e-06, + "loss": 0.8495, + "step": 1187 + }, + { + "epoch": 0.08928303021193447, + "grad_norm": 2.1395717730731416, + "learning_rate": 3.963319364102538e-06, + "loss": 0.9835, + "step": 1188 + }, + { + "epoch": 0.08935818427776943, + "grad_norm": 1.941390346408232, + "learning_rate": 3.96322649349972e-06, + "loss": 1.017, + "step": 1189 + }, + { + "epoch": 0.08943333834360438, + "grad_norm": 1.6399494051688521, + "learning_rate": 3.963133506568214e-06, + "loss": 1.0995, + "step": 1190 + }, + { + "epoch": 0.08950849240943935, + "grad_norm": 1.9926576545786359, + "learning_rate": 3.96304040331353e-06, + "loss": 0.9823, + "step": 1191 + }, + { + "epoch": 0.08958364647527431, + "grad_norm": 1.8058560374049841, + "learning_rate": 3.9629471837411855e-06, + "loss": 1.071, + "step": 1192 + }, + { + "epoch": 0.08965880054110928, + "grad_norm": 1.944220056150924, + "learning_rate": 3.962853847856704e-06, + "loss": 1.0789, + "step": 1193 + }, + { + "epoch": 0.08973395460694424, + "grad_norm": 1.2590184087884548, + "learning_rate": 3.962760395665616e-06, + "loss": 0.982, + "step": 1194 + }, + { + "epoch": 0.08980910867277919, + "grad_norm": 1.7839331913357328, + "learning_rate": 3.962666827173458e-06, + "loss": 1.0562, + "step": 1195 + }, + { + "epoch": 0.08988426273861416, + "grad_norm": 0.9012997355077379, + "learning_rate": 3.9625731423857745e-06, + "loss": 0.8104, + "step": 1196 + }, + { + "epoch": 0.08995941680444912, + "grad_norm": 1.6959450240536769, + "learning_rate": 3.9624793413081185e-06, + "loss": 1.0315, + "step": 1197 + }, + { + "epoch": 0.09003457087028409, + "grad_norm": 1.4555965447336874, + "learning_rate": 3.962385423946046e-06, + "loss": 1.0631, + "step": 1198 + }, + { + "epoch": 0.09010972493611905, + "grad_norm": 1.898220222503848, + "learning_rate": 3.962291390305123e-06, + "loss": 0.9847, + "step": 1199 + }, + { + "epoch": 0.090184879001954, + "grad_norm": 1.8984519424203088, + "learning_rate": 3.96219724039092e-06, + "loss": 0.9864, + "step": 1200 + }, + { + "epoch": 0.09026003306778897, + "grad_norm": 1.5088251695972859, + "learning_rate": 3.962102974209018e-06, + "loss": 1.0636, + "step": 1201 + }, + { + "epoch": 0.09033518713362393, + "grad_norm": 1.492374588187733, + "learning_rate": 3.962008591765e-06, + "loss": 1.039, + "step": 1202 + }, + { + "epoch": 0.09041034119945889, + "grad_norm": 1.5523157479112872, + "learning_rate": 3.961914093064461e-06, + "loss": 0.9771, + "step": 1203 + }, + { + "epoch": 0.09048549526529386, + "grad_norm": 1.7709792656699805, + "learning_rate": 3.961819478112999e-06, + "loss": 1.0814, + "step": 1204 + }, + { + "epoch": 0.09056064933112881, + "grad_norm": 10.606891953884828, + "learning_rate": 3.961724746916221e-06, + "loss": 0.991, + "step": 1205 + }, + { + "epoch": 0.09063580339696378, + "grad_norm": 1.9045137722031187, + "learning_rate": 3.961629899479739e-06, + "loss": 1.0669, + "step": 1206 + }, + { + "epoch": 0.09071095746279874, + "grad_norm": 1.4892766232162424, + "learning_rate": 3.961534935809174e-06, + "loss": 1.0348, + "step": 1207 + }, + { + "epoch": 0.0907861115286337, + "grad_norm": 1.549667227309179, + "learning_rate": 3.961439855910154e-06, + "loss": 0.967, + "step": 1208 + }, + { + "epoch": 0.09086126559446867, + "grad_norm": 1.7723149851667814, + "learning_rate": 3.961344659788311e-06, + "loss": 1.0384, + "step": 1209 + }, + { + "epoch": 0.09093641966030362, + "grad_norm": 1.4741075078559933, + "learning_rate": 3.961249347449286e-06, + "loss": 1.0612, + "step": 1210 + }, + { + "epoch": 0.09101157372613858, + "grad_norm": 1.5455864057029929, + "learning_rate": 3.961153918898727e-06, + "loss": 0.997, + "step": 1211 + }, + { + "epoch": 0.09108672779197355, + "grad_norm": 1.6232854160597228, + "learning_rate": 3.961058374142289e-06, + "loss": 1.1252, + "step": 1212 + }, + { + "epoch": 0.0911618818578085, + "grad_norm": 1.6561141048317463, + "learning_rate": 3.960962713185633e-06, + "loss": 0.9792, + "step": 1213 + }, + { + "epoch": 0.09123703592364348, + "grad_norm": 1.4438019152798613, + "learning_rate": 3.960866936034426e-06, + "loss": 1.0284, + "step": 1214 + }, + { + "epoch": 0.09131218998947843, + "grad_norm": 1.5014845336615457, + "learning_rate": 3.960771042694346e-06, + "loss": 1.0235, + "step": 1215 + }, + { + "epoch": 0.09138734405531339, + "grad_norm": 1.541803465607549, + "learning_rate": 3.960675033171072e-06, + "loss": 0.956, + "step": 1216 + }, + { + "epoch": 0.09146249812114836, + "grad_norm": 1.8233105769762663, + "learning_rate": 3.960578907470295e-06, + "loss": 1.0498, + "step": 1217 + }, + { + "epoch": 0.09153765218698331, + "grad_norm": 1.6955919167327969, + "learning_rate": 3.960482665597711e-06, + "loss": 0.918, + "step": 1218 + }, + { + "epoch": 0.09161280625281827, + "grad_norm": 1.545270530817331, + "learning_rate": 3.960386307559021e-06, + "loss": 1.0197, + "step": 1219 + }, + { + "epoch": 0.09168796031865324, + "grad_norm": 2.0088479307429163, + "learning_rate": 3.960289833359936e-06, + "loss": 0.9863, + "step": 1220 + }, + { + "epoch": 0.0917631143844882, + "grad_norm": 1.2792384942398187, + "learning_rate": 3.9601932430061714e-06, + "loss": 0.9981, + "step": 1221 + }, + { + "epoch": 0.09183826845032317, + "grad_norm": 1.6743430343828887, + "learning_rate": 3.9600965365034515e-06, + "loss": 1.0783, + "step": 1222 + }, + { + "epoch": 0.09191342251615812, + "grad_norm": 1.540440286392154, + "learning_rate": 3.959999713857505e-06, + "loss": 0.7433, + "step": 1223 + }, + { + "epoch": 0.09198857658199308, + "grad_norm": 1.6455924264224497, + "learning_rate": 3.959902775074072e-06, + "loss": 1.0669, + "step": 1224 + }, + { + "epoch": 0.09206373064782805, + "grad_norm": 1.6205084421984886, + "learning_rate": 3.9598057201588926e-06, + "loss": 1.0555, + "step": 1225 + }, + { + "epoch": 0.09213888471366301, + "grad_norm": 1.4726216036416846, + "learning_rate": 3.959708549117721e-06, + "loss": 0.9913, + "step": 1226 + }, + { + "epoch": 0.09221403877949796, + "grad_norm": 1.6726577210007836, + "learning_rate": 3.959611261956313e-06, + "loss": 1.0241, + "step": 1227 + }, + { + "epoch": 0.09228919284533293, + "grad_norm": 1.7390889209477987, + "learning_rate": 3.959513858680434e-06, + "loss": 0.9663, + "step": 1228 + }, + { + "epoch": 0.09236434691116789, + "grad_norm": 1.6227914033014688, + "learning_rate": 3.9594163392958566e-06, + "loss": 1.0824, + "step": 1229 + }, + { + "epoch": 0.09243950097700286, + "grad_norm": 1.5032894306299278, + "learning_rate": 3.959318703808356e-06, + "loss": 0.9053, + "step": 1230 + }, + { + "epoch": 0.09251465504283782, + "grad_norm": 1.3550195746749862, + "learning_rate": 3.95922095222372e-06, + "loss": 1.0447, + "step": 1231 + }, + { + "epoch": 0.09258980910867277, + "grad_norm": 1.5787214153054887, + "learning_rate": 3.959123084547741e-06, + "loss": 1.0667, + "step": 1232 + }, + { + "epoch": 0.09266496317450774, + "grad_norm": 1.5516776081577652, + "learning_rate": 3.959025100786217e-06, + "loss": 1.059, + "step": 1233 + }, + { + "epoch": 0.0927401172403427, + "grad_norm": 1.5682254415784784, + "learning_rate": 3.958927000944954e-06, + "loss": 1.0442, + "step": 1234 + }, + { + "epoch": 0.09281527130617766, + "grad_norm": 1.689707618905438, + "learning_rate": 3.958828785029765e-06, + "loss": 1.026, + "step": 1235 + }, + { + "epoch": 0.09289042537201263, + "grad_norm": 1.3462717657003311, + "learning_rate": 3.958730453046469e-06, + "loss": 1.0855, + "step": 1236 + }, + { + "epoch": 0.09296557943784758, + "grad_norm": 1.538124999503981, + "learning_rate": 3.958632005000895e-06, + "loss": 1.002, + "step": 1237 + }, + { + "epoch": 0.09304073350368255, + "grad_norm": 1.7613625853638855, + "learning_rate": 3.958533440898873e-06, + "loss": 0.9896, + "step": 1238 + }, + { + "epoch": 0.09311588756951751, + "grad_norm": 1.3693945373088576, + "learning_rate": 3.958434760746245e-06, + "loss": 0.9992, + "step": 1239 + }, + { + "epoch": 0.09319104163535247, + "grad_norm": 1.0259950436812588, + "learning_rate": 3.958335964548859e-06, + "loss": 0.8482, + "step": 1240 + }, + { + "epoch": 0.09326619570118744, + "grad_norm": 1.4632797640392214, + "learning_rate": 3.958237052312568e-06, + "loss": 1.0303, + "step": 1241 + }, + { + "epoch": 0.0933413497670224, + "grad_norm": 1.644396778688869, + "learning_rate": 3.958138024043232e-06, + "loss": 1.0882, + "step": 1242 + }, + { + "epoch": 0.09341650383285736, + "grad_norm": 0.9906638754548207, + "learning_rate": 3.958038879746721e-06, + "loss": 0.7885, + "step": 1243 + }, + { + "epoch": 0.09349165789869232, + "grad_norm": 1.8870723543067431, + "learning_rate": 3.9579396194289075e-06, + "loss": 0.8911, + "step": 1244 + }, + { + "epoch": 0.09356681196452728, + "grad_norm": 1.9582339187170206, + "learning_rate": 3.957840243095675e-06, + "loss": 0.9574, + "step": 1245 + }, + { + "epoch": 0.09364196603036225, + "grad_norm": 1.6343979872186547, + "learning_rate": 3.95774075075291e-06, + "loss": 0.9392, + "step": 1246 + }, + { + "epoch": 0.0937171200961972, + "grad_norm": 1.600746430688108, + "learning_rate": 3.957641142406509e-06, + "loss": 1.0117, + "step": 1247 + }, + { + "epoch": 0.09379227416203216, + "grad_norm": 1.9072241248424322, + "learning_rate": 3.9575414180623746e-06, + "loss": 0.9347, + "step": 1248 + }, + { + "epoch": 0.09386742822786713, + "grad_norm": 1.819283190304035, + "learning_rate": 3.957441577726415e-06, + "loss": 1.0134, + "step": 1249 + }, + { + "epoch": 0.09394258229370209, + "grad_norm": 1.552224129386124, + "learning_rate": 3.957341621404547e-06, + "loss": 1.0199, + "step": 1250 + }, + { + "epoch": 0.09401773635953706, + "grad_norm": 1.3170247421981847, + "learning_rate": 3.957241549102692e-06, + "loss": 1.0493, + "step": 1251 + }, + { + "epoch": 0.09409289042537201, + "grad_norm": 1.8847661309908648, + "learning_rate": 3.957141360826781e-06, + "loss": 1.0148, + "step": 1252 + }, + { + "epoch": 0.09416804449120697, + "grad_norm": 1.45384510204819, + "learning_rate": 3.95704105658275e-06, + "loss": 0.9027, + "step": 1253 + }, + { + "epoch": 0.09424319855704194, + "grad_norm": 1.0263772922239862, + "learning_rate": 3.9569406363765415e-06, + "loss": 0.8433, + "step": 1254 + }, + { + "epoch": 0.0943183526228769, + "grad_norm": 2.0722977242963374, + "learning_rate": 3.956840100214107e-06, + "loss": 0.9461, + "step": 1255 + }, + { + "epoch": 0.09439350668871185, + "grad_norm": 1.4624615232182763, + "learning_rate": 3.956739448101404e-06, + "loss": 1.0109, + "step": 1256 + }, + { + "epoch": 0.09446866075454682, + "grad_norm": 1.8034603025309917, + "learning_rate": 3.956638680044396e-06, + "loss": 1.0524, + "step": 1257 + }, + { + "epoch": 0.09454381482038178, + "grad_norm": 1.925844017383228, + "learning_rate": 3.956537796049052e-06, + "loss": 0.9683, + "step": 1258 + }, + { + "epoch": 0.09461896888621675, + "grad_norm": 1.5652004557190915, + "learning_rate": 3.9564367961213536e-06, + "loss": 0.9571, + "step": 1259 + }, + { + "epoch": 0.0946941229520517, + "grad_norm": 1.576342643175216, + "learning_rate": 3.956335680267282e-06, + "loss": 1.0718, + "step": 1260 + }, + { + "epoch": 0.09476927701788666, + "grad_norm": 2.2547824683646924, + "learning_rate": 3.956234448492831e-06, + "loss": 0.9727, + "step": 1261 + }, + { + "epoch": 0.09484443108372163, + "grad_norm": 1.5872211992442011, + "learning_rate": 3.956133100803996e-06, + "loss": 0.9628, + "step": 1262 + }, + { + "epoch": 0.09491958514955659, + "grad_norm": 4.678528605590659, + "learning_rate": 3.956031637206786e-06, + "loss": 1.025, + "step": 1263 + }, + { + "epoch": 0.09499473921539155, + "grad_norm": 1.7342024021355973, + "learning_rate": 3.955930057707211e-06, + "loss": 0.9187, + "step": 1264 + }, + { + "epoch": 0.09506989328122652, + "grad_norm": 1.7346349985255518, + "learning_rate": 3.95582836231129e-06, + "loss": 1.0003, + "step": 1265 + }, + { + "epoch": 0.09514504734706147, + "grad_norm": 1.8399254523600432, + "learning_rate": 3.9557265510250505e-06, + "loss": 0.9775, + "step": 1266 + }, + { + "epoch": 0.09522020141289644, + "grad_norm": 1.8567090837894267, + "learning_rate": 3.955624623854522e-06, + "loss": 0.9626, + "step": 1267 + }, + { + "epoch": 0.0952953554787314, + "grad_norm": 1.5069351410859384, + "learning_rate": 3.9555225808057475e-06, + "loss": 0.8892, + "step": 1268 + }, + { + "epoch": 0.09537050954456636, + "grad_norm": 1.9340792576231374, + "learning_rate": 3.9554204218847705e-06, + "loss": 0.9868, + "step": 1269 + }, + { + "epoch": 0.09544566361040133, + "grad_norm": 1.4635316389109125, + "learning_rate": 3.955318147097647e-06, + "loss": 1.0227, + "step": 1270 + }, + { + "epoch": 0.09552081767623628, + "grad_norm": 2.2936275646992246, + "learning_rate": 3.955215756450435e-06, + "loss": 0.9865, + "step": 1271 + }, + { + "epoch": 0.09559597174207124, + "grad_norm": 2.025490209325567, + "learning_rate": 3.955113249949203e-06, + "loss": 0.8935, + "step": 1272 + }, + { + "epoch": 0.09567112580790621, + "grad_norm": 5.123788271684512, + "learning_rate": 3.955010627600024e-06, + "loss": 1.0017, + "step": 1273 + }, + { + "epoch": 0.09574627987374117, + "grad_norm": 1.4169292580573198, + "learning_rate": 3.954907889408979e-06, + "loss": 1.0695, + "step": 1274 + }, + { + "epoch": 0.09582143393957614, + "grad_norm": 1.502490598550793, + "learning_rate": 3.954805035382155e-06, + "loss": 1.0837, + "step": 1275 + }, + { + "epoch": 0.0958965880054111, + "grad_norm": 1.898000910285434, + "learning_rate": 3.954702065525649e-06, + "loss": 0.9574, + "step": 1276 + }, + { + "epoch": 0.09597174207124605, + "grad_norm": 1.5632161597294165, + "learning_rate": 3.954598979845559e-06, + "loss": 1.0383, + "step": 1277 + }, + { + "epoch": 0.09604689613708102, + "grad_norm": 1.695087119032597, + "learning_rate": 3.954495778347996e-06, + "loss": 0.8766, + "step": 1278 + }, + { + "epoch": 0.09612205020291598, + "grad_norm": 1.3697875231932812, + "learning_rate": 3.954392461039073e-06, + "loss": 1.0622, + "step": 1279 + }, + { + "epoch": 0.09619720426875093, + "grad_norm": 1.8883585941255794, + "learning_rate": 3.954289027924912e-06, + "loss": 0.9489, + "step": 1280 + }, + { + "epoch": 0.0962723583345859, + "grad_norm": 1.5545671119276454, + "learning_rate": 3.954185479011644e-06, + "loss": 1.0441, + "step": 1281 + }, + { + "epoch": 0.09634751240042086, + "grad_norm": 1.0452466798595954, + "learning_rate": 3.954081814305403e-06, + "loss": 0.7871, + "step": 1282 + }, + { + "epoch": 0.09642266646625583, + "grad_norm": 1.6762110730230466, + "learning_rate": 3.953978033812332e-06, + "loss": 1.0075, + "step": 1283 + }, + { + "epoch": 0.09649782053209079, + "grad_norm": 1.89249730995622, + "learning_rate": 3.95387413753858e-06, + "loss": 0.9325, + "step": 1284 + }, + { + "epoch": 0.09657297459792574, + "grad_norm": 1.5664430837510017, + "learning_rate": 3.9537701254903034e-06, + "loss": 0.8842, + "step": 1285 + }, + { + "epoch": 0.09664812866376071, + "grad_norm": 1.4543355186096836, + "learning_rate": 3.953665997673665e-06, + "loss": 0.993, + "step": 1286 + }, + { + "epoch": 0.09672328272959567, + "grad_norm": 1.7033132746511381, + "learning_rate": 3.953561754094836e-06, + "loss": 0.9862, + "step": 1287 + }, + { + "epoch": 0.09679843679543064, + "grad_norm": 1.3251339610775301, + "learning_rate": 3.953457394759992e-06, + "loss": 1.0783, + "step": 1288 + }, + { + "epoch": 0.0968735908612656, + "grad_norm": 1.6030345571485205, + "learning_rate": 3.953352919675317e-06, + "loss": 1.0596, + "step": 1289 + }, + { + "epoch": 0.09694874492710055, + "grad_norm": 3.2950852584806873, + "learning_rate": 3.953248328847001e-06, + "loss": 0.9351, + "step": 1290 + }, + { + "epoch": 0.09702389899293552, + "grad_norm": 1.5634242566691718, + "learning_rate": 3.953143622281243e-06, + "loss": 1.0558, + "step": 1291 + }, + { + "epoch": 0.09709905305877048, + "grad_norm": 1.5613202791300789, + "learning_rate": 3.953038799984246e-06, + "loss": 1.0026, + "step": 1292 + }, + { + "epoch": 0.09717420712460544, + "grad_norm": 2.1023273423432456, + "learning_rate": 3.952933861962222e-06, + "loss": 1.093, + "step": 1293 + }, + { + "epoch": 0.0972493611904404, + "grad_norm": 1.7332976148055124, + "learning_rate": 3.952828808221387e-06, + "loss": 1.0006, + "step": 1294 + }, + { + "epoch": 0.09732451525627536, + "grad_norm": 1.5743301390056093, + "learning_rate": 3.952723638767968e-06, + "loss": 1.0994, + "step": 1295 + }, + { + "epoch": 0.09739966932211033, + "grad_norm": 1.7398936819063884, + "learning_rate": 3.952618353608196e-06, + "loss": 0.9871, + "step": 1296 + }, + { + "epoch": 0.09747482338794529, + "grad_norm": 1.5661560010902416, + "learning_rate": 3.95251295274831e-06, + "loss": 1.0132, + "step": 1297 + }, + { + "epoch": 0.09754997745378025, + "grad_norm": 1.5768987593824528, + "learning_rate": 3.952407436194554e-06, + "loss": 0.9901, + "step": 1298 + }, + { + "epoch": 0.09762513151961522, + "grad_norm": 1.3102250137435951, + "learning_rate": 3.9523018039531816e-06, + "loss": 0.9198, + "step": 1299 + }, + { + "epoch": 0.09770028558545017, + "grad_norm": 2.991642755467767, + "learning_rate": 3.952196056030451e-06, + "loss": 0.9823, + "step": 1300 + }, + { + "epoch": 0.09777543965128513, + "grad_norm": 0.8655401856223928, + "learning_rate": 3.952090192432629e-06, + "loss": 0.8028, + "step": 1301 + }, + { + "epoch": 0.0978505937171201, + "grad_norm": 1.558680864160672, + "learning_rate": 3.951984213165988e-06, + "loss": 1.0126, + "step": 1302 + }, + { + "epoch": 0.09792574778295506, + "grad_norm": 1.6233609075900233, + "learning_rate": 3.951878118236807e-06, + "loss": 0.8402, + "step": 1303 + }, + { + "epoch": 0.09800090184879003, + "grad_norm": 2.0131232534576866, + "learning_rate": 3.951771907651374e-06, + "loss": 0.9525, + "step": 1304 + }, + { + "epoch": 0.09807605591462498, + "grad_norm": 1.6243508582909938, + "learning_rate": 3.951665581415982e-06, + "loss": 1.0108, + "step": 1305 + }, + { + "epoch": 0.09815120998045994, + "grad_norm": 1.469031367845406, + "learning_rate": 3.9515591395369305e-06, + "loss": 0.9952, + "step": 1306 + }, + { + "epoch": 0.09822636404629491, + "grad_norm": 1.6034058917804894, + "learning_rate": 3.9514525820205265e-06, + "loss": 0.9977, + "step": 1307 + }, + { + "epoch": 0.09830151811212987, + "grad_norm": 1.8851766978046913, + "learning_rate": 3.951345908873085e-06, + "loss": 0.9785, + "step": 1308 + }, + { + "epoch": 0.09837667217796482, + "grad_norm": 1.5702727962807546, + "learning_rate": 3.9512391201009265e-06, + "loss": 0.9808, + "step": 1309 + }, + { + "epoch": 0.09845182624379979, + "grad_norm": 1.6666202856603862, + "learning_rate": 3.9511322157103776e-06, + "loss": 1.0169, + "step": 1310 + }, + { + "epoch": 0.09852698030963475, + "grad_norm": 1.6817350460752591, + "learning_rate": 3.951025195707774e-06, + "loss": 1.0571, + "step": 1311 + }, + { + "epoch": 0.09860213437546972, + "grad_norm": 1.5416558583198596, + "learning_rate": 3.950918060099456e-06, + "loss": 1.091, + "step": 1312 + }, + { + "epoch": 0.09867728844130468, + "grad_norm": 1.6101752208349163, + "learning_rate": 3.950810808891773e-06, + "loss": 0.9474, + "step": 1313 + }, + { + "epoch": 0.09875244250713963, + "grad_norm": 1.645931852587037, + "learning_rate": 3.950703442091079e-06, + "loss": 1.0154, + "step": 1314 + }, + { + "epoch": 0.0988275965729746, + "grad_norm": 1.6610268856768713, + "learning_rate": 3.950595959703736e-06, + "loss": 1.0023, + "step": 1315 + }, + { + "epoch": 0.09890275063880956, + "grad_norm": 1.6311629144216095, + "learning_rate": 3.950488361736114e-06, + "loss": 0.9851, + "step": 1316 + }, + { + "epoch": 0.09897790470464451, + "grad_norm": 1.8546268520707212, + "learning_rate": 3.950380648194587e-06, + "loss": 1.0515, + "step": 1317 + }, + { + "epoch": 0.09905305877047949, + "grad_norm": 1.2533914335122156, + "learning_rate": 3.950272819085538e-06, + "loss": 0.9759, + "step": 1318 + }, + { + "epoch": 0.09912821283631444, + "grad_norm": 1.631943905156184, + "learning_rate": 3.950164874415357e-06, + "loss": 0.9815, + "step": 1319 + }, + { + "epoch": 0.09920336690214941, + "grad_norm": 1.8525797950557892, + "learning_rate": 3.950056814190439e-06, + "loss": 0.9112, + "step": 1320 + }, + { + "epoch": 0.09927852096798437, + "grad_norm": 2.9013553543388784, + "learning_rate": 3.949948638417188e-06, + "loss": 0.9887, + "step": 1321 + }, + { + "epoch": 0.09935367503381932, + "grad_norm": 1.8856807404378926, + "learning_rate": 3.949840347102013e-06, + "loss": 0.9942, + "step": 1322 + }, + { + "epoch": 0.0994288290996543, + "grad_norm": 1.7447701372175288, + "learning_rate": 3.949731940251331e-06, + "loss": 0.9708, + "step": 1323 + }, + { + "epoch": 0.09950398316548925, + "grad_norm": 1.7922413000597242, + "learning_rate": 3.949623417871565e-06, + "loss": 0.9842, + "step": 1324 + }, + { + "epoch": 0.09957913723132421, + "grad_norm": 0.8726270037252243, + "learning_rate": 3.949514779969147e-06, + "loss": 0.7787, + "step": 1325 + }, + { + "epoch": 0.09965429129715918, + "grad_norm": 1.9305847800014992, + "learning_rate": 3.949406026550512e-06, + "loss": 1.0939, + "step": 1326 + }, + { + "epoch": 0.09972944536299413, + "grad_norm": 1.5009128653835562, + "learning_rate": 3.949297157622105e-06, + "loss": 1.0452, + "step": 1327 + }, + { + "epoch": 0.0998045994288291, + "grad_norm": 1.6625526276483151, + "learning_rate": 3.949188173190378e-06, + "loss": 0.9648, + "step": 1328 + }, + { + "epoch": 0.09987975349466406, + "grad_norm": 2.010826727862469, + "learning_rate": 3.949079073261788e-06, + "loss": 0.9692, + "step": 1329 + }, + { + "epoch": 0.09995490756049902, + "grad_norm": 1.889337842388511, + "learning_rate": 3.948969857842799e-06, + "loss": 0.9519, + "step": 1330 + }, + { + "epoch": 0.10003006162633399, + "grad_norm": 1.8405801064889054, + "learning_rate": 3.948860526939882e-06, + "loss": 0.9931, + "step": 1331 + }, + { + "epoch": 0.10010521569216894, + "grad_norm": 2.6368478701023834, + "learning_rate": 3.948751080559517e-06, + "loss": 1.0332, + "step": 1332 + }, + { + "epoch": 0.10018036975800391, + "grad_norm": 1.5002039929598088, + "learning_rate": 3.948641518708188e-06, + "loss": 1.0716, + "step": 1333 + }, + { + "epoch": 0.10025552382383887, + "grad_norm": 1.8796133507161275, + "learning_rate": 3.9485318413923865e-06, + "loss": 0.9512, + "step": 1334 + }, + { + "epoch": 0.10033067788967383, + "grad_norm": 1.501654110276328, + "learning_rate": 3.948422048618612e-06, + "loss": 0.9866, + "step": 1335 + }, + { + "epoch": 0.1004058319555088, + "grad_norm": 1.5094263798595215, + "learning_rate": 3.948312140393372e-06, + "loss": 1.0043, + "step": 1336 + }, + { + "epoch": 0.10048098602134375, + "grad_norm": 0.8025936179650927, + "learning_rate": 3.948202116723176e-06, + "loss": 0.7824, + "step": 1337 + }, + { + "epoch": 0.10055614008717871, + "grad_norm": 1.3832232624905347, + "learning_rate": 3.948091977614544e-06, + "loss": 0.9551, + "step": 1338 + }, + { + "epoch": 0.10063129415301368, + "grad_norm": 1.5238940215170174, + "learning_rate": 3.947981723074003e-06, + "loss": 0.9444, + "step": 1339 + }, + { + "epoch": 0.10070644821884864, + "grad_norm": 1.6760979911044287, + "learning_rate": 3.947871353108085e-06, + "loss": 0.9215, + "step": 1340 + }, + { + "epoch": 0.10078160228468361, + "grad_norm": 1.601125145719106, + "learning_rate": 3.947760867723331e-06, + "loss": 1.0938, + "step": 1341 + }, + { + "epoch": 0.10085675635051856, + "grad_norm": 1.6018992744478926, + "learning_rate": 3.9476502669262866e-06, + "loss": 1.0026, + "step": 1342 + }, + { + "epoch": 0.10093191041635352, + "grad_norm": 1.5572786447582347, + "learning_rate": 3.947539550723506e-06, + "loss": 0.9782, + "step": 1343 + }, + { + "epoch": 0.10100706448218849, + "grad_norm": 0.7857727742719378, + "learning_rate": 3.94742871912155e-06, + "loss": 0.8038, + "step": 1344 + }, + { + "epoch": 0.10108221854802345, + "grad_norm": 2.9145826432884188, + "learning_rate": 3.947317772126985e-06, + "loss": 1.011, + "step": 1345 + }, + { + "epoch": 0.1011573726138584, + "grad_norm": 1.70323775731867, + "learning_rate": 3.947206709746385e-06, + "loss": 1.1099, + "step": 1346 + }, + { + "epoch": 0.10123252667969337, + "grad_norm": 1.5511106050063008, + "learning_rate": 3.947095531986331e-06, + "loss": 1.0809, + "step": 1347 + }, + { + "epoch": 0.10130768074552833, + "grad_norm": 1.413180966928727, + "learning_rate": 3.9469842388534105e-06, + "loss": 1.0161, + "step": 1348 + }, + { + "epoch": 0.1013828348113633, + "grad_norm": 1.3561008969103343, + "learning_rate": 3.946872830354219e-06, + "loss": 0.9846, + "step": 1349 + }, + { + "epoch": 0.10145798887719826, + "grad_norm": 2.412685260676108, + "learning_rate": 3.946761306495357e-06, + "loss": 1.1314, + "step": 1350 + }, + { + "epoch": 0.10153314294303321, + "grad_norm": 2.1994981063206787, + "learning_rate": 3.946649667283433e-06, + "loss": 1.002, + "step": 1351 + }, + { + "epoch": 0.10160829700886818, + "grad_norm": 1.6944299325512648, + "learning_rate": 3.946537912725062e-06, + "loss": 0.9698, + "step": 1352 + }, + { + "epoch": 0.10168345107470314, + "grad_norm": 1.5825524495910988, + "learning_rate": 3.946426042826865e-06, + "loss": 0.9239, + "step": 1353 + }, + { + "epoch": 0.1017586051405381, + "grad_norm": 3.222287058634694, + "learning_rate": 3.946314057595473e-06, + "loss": 0.9737, + "step": 1354 + }, + { + "epoch": 0.10183375920637307, + "grad_norm": 2.051552589317027, + "learning_rate": 3.94620195703752e-06, + "loss": 0.9503, + "step": 1355 + }, + { + "epoch": 0.10190891327220802, + "grad_norm": 1.5449570604795415, + "learning_rate": 3.946089741159647e-06, + "loss": 0.881, + "step": 1356 + }, + { + "epoch": 0.101984067338043, + "grad_norm": 1.0587242017933431, + "learning_rate": 3.9459774099685065e-06, + "loss": 0.8415, + "step": 1357 + }, + { + "epoch": 0.10205922140387795, + "grad_norm": 1.7360553334071023, + "learning_rate": 3.945864963470752e-06, + "loss": 1.0186, + "step": 1358 + }, + { + "epoch": 0.1021343754697129, + "grad_norm": 2.284367978023899, + "learning_rate": 3.945752401673047e-06, + "loss": 0.8934, + "step": 1359 + }, + { + "epoch": 0.10220952953554788, + "grad_norm": 1.6097208780994934, + "learning_rate": 3.945639724582062e-06, + "loss": 1.0933, + "step": 1360 + }, + { + "epoch": 0.10228468360138283, + "grad_norm": 1.5652869737173154, + "learning_rate": 3.9455269322044725e-06, + "loss": 1.0589, + "step": 1361 + }, + { + "epoch": 0.10235983766721779, + "grad_norm": 1.6975363635071494, + "learning_rate": 3.945414024546963e-06, + "loss": 0.9856, + "step": 1362 + }, + { + "epoch": 0.10243499173305276, + "grad_norm": 1.5160363808999704, + "learning_rate": 3.945301001616222e-06, + "loss": 1.0268, + "step": 1363 + }, + { + "epoch": 0.10251014579888772, + "grad_norm": 1.4755958743616029, + "learning_rate": 3.945187863418949e-06, + "loss": 0.9344, + "step": 1364 + }, + { + "epoch": 0.10258529986472269, + "grad_norm": 1.7159317492803952, + "learning_rate": 3.945074609961845e-06, + "loss": 1.0367, + "step": 1365 + }, + { + "epoch": 0.10266045393055764, + "grad_norm": 1.56852445766848, + "learning_rate": 3.944961241251623e-06, + "loss": 1.0223, + "step": 1366 + }, + { + "epoch": 0.1027356079963926, + "grad_norm": 1.6519735295805114, + "learning_rate": 3.944847757295e-06, + "loss": 0.934, + "step": 1367 + }, + { + "epoch": 0.10281076206222757, + "grad_norm": 1.5354064271638073, + "learning_rate": 3.9447341580987e-06, + "loss": 0.9815, + "step": 1368 + }, + { + "epoch": 0.10288591612806253, + "grad_norm": 1.6399710668942438, + "learning_rate": 3.944620443669453e-06, + "loss": 1.0327, + "step": 1369 + }, + { + "epoch": 0.10296107019389748, + "grad_norm": 1.7047602651085343, + "learning_rate": 3.9445066140139995e-06, + "loss": 0.9144, + "step": 1370 + }, + { + "epoch": 0.10303622425973245, + "grad_norm": 1.4258688703813058, + "learning_rate": 3.944392669139083e-06, + "loss": 1.0076, + "step": 1371 + }, + { + "epoch": 0.10311137832556741, + "grad_norm": 1.393699504810001, + "learning_rate": 3.944278609051455e-06, + "loss": 0.8841, + "step": 1372 + }, + { + "epoch": 0.10318653239140238, + "grad_norm": 1.565632921758472, + "learning_rate": 3.944164433757874e-06, + "loss": 1.0101, + "step": 1373 + }, + { + "epoch": 0.10326168645723734, + "grad_norm": 1.6723854497846384, + "learning_rate": 3.944050143265106e-06, + "loss": 0.9297, + "step": 1374 + }, + { + "epoch": 0.1033368405230723, + "grad_norm": 1.7792027403534225, + "learning_rate": 3.943935737579923e-06, + "loss": 1.0394, + "step": 1375 + }, + { + "epoch": 0.10341199458890726, + "grad_norm": 1.4749251071860772, + "learning_rate": 3.943821216709103e-06, + "loss": 0.9286, + "step": 1376 + }, + { + "epoch": 0.10348714865474222, + "grad_norm": 1.809213025579597, + "learning_rate": 3.943706580659433e-06, + "loss": 0.9834, + "step": 1377 + }, + { + "epoch": 0.10356230272057719, + "grad_norm": 1.5859433629721522, + "learning_rate": 3.943591829437705e-06, + "loss": 1.0979, + "step": 1378 + }, + { + "epoch": 0.10363745678641215, + "grad_norm": 2.2262317077125107, + "learning_rate": 3.943476963050719e-06, + "loss": 0.984, + "step": 1379 + }, + { + "epoch": 0.1037126108522471, + "grad_norm": 0.9851679559990798, + "learning_rate": 3.94336198150528e-06, + "loss": 0.8769, + "step": 1380 + }, + { + "epoch": 0.10378776491808207, + "grad_norm": 1.9379324276642116, + "learning_rate": 3.9432468848082024e-06, + "loss": 0.9704, + "step": 1381 + }, + { + "epoch": 0.10386291898391703, + "grad_norm": 1.7181741616829413, + "learning_rate": 3.9431316729663055e-06, + "loss": 1.0096, + "step": 1382 + }, + { + "epoch": 0.10393807304975199, + "grad_norm": 2.4732308586395555, + "learning_rate": 3.943016345986417e-06, + "loss": 0.9272, + "step": 1383 + }, + { + "epoch": 0.10401322711558696, + "grad_norm": 1.4632141000113155, + "learning_rate": 3.942900903875369e-06, + "loss": 1.0285, + "step": 1384 + }, + { + "epoch": 0.10408838118142191, + "grad_norm": 1.5934394833121681, + "learning_rate": 3.942785346640003e-06, + "loss": 1.082, + "step": 1385 + }, + { + "epoch": 0.10416353524725688, + "grad_norm": 2.2019064489062896, + "learning_rate": 3.942669674287166e-06, + "loss": 1.0298, + "step": 1386 + }, + { + "epoch": 0.10423868931309184, + "grad_norm": 1.6591992301306415, + "learning_rate": 3.942553886823711e-06, + "loss": 1.0479, + "step": 1387 + }, + { + "epoch": 0.1043138433789268, + "grad_norm": 1.823688176697545, + "learning_rate": 3.9424379842565005e-06, + "loss": 1.0268, + "step": 1388 + }, + { + "epoch": 0.10438899744476177, + "grad_norm": 1.556309146599749, + "learning_rate": 3.9423219665924e-06, + "loss": 1.0486, + "step": 1389 + }, + { + "epoch": 0.10446415151059672, + "grad_norm": 1.6221374828525315, + "learning_rate": 3.942205833838287e-06, + "loss": 0.9791, + "step": 1390 + }, + { + "epoch": 0.10453930557643168, + "grad_norm": 2.02146114022728, + "learning_rate": 3.942089586001039e-06, + "loss": 1.1275, + "step": 1391 + }, + { + "epoch": 0.10461445964226665, + "grad_norm": 1.668746159581255, + "learning_rate": 3.941973223087548e-06, + "loss": 0.9225, + "step": 1392 + }, + { + "epoch": 0.1046896137081016, + "grad_norm": 1.545324453957572, + "learning_rate": 3.941856745104707e-06, + "loss": 0.9894, + "step": 1393 + }, + { + "epoch": 0.10476476777393658, + "grad_norm": 1.5940680093077395, + "learning_rate": 3.941740152059418e-06, + "loss": 0.994, + "step": 1394 + }, + { + "epoch": 0.10483992183977153, + "grad_norm": 1.4356587269084236, + "learning_rate": 3.94162344395859e-06, + "loss": 0.9772, + "step": 1395 + }, + { + "epoch": 0.10491507590560649, + "grad_norm": 1.5337494672480547, + "learning_rate": 3.941506620809137e-06, + "loss": 1.0236, + "step": 1396 + }, + { + "epoch": 0.10499022997144146, + "grad_norm": 1.4830154617697087, + "learning_rate": 3.941389682617982e-06, + "loss": 0.9469, + "step": 1397 + }, + { + "epoch": 0.10506538403727642, + "grad_norm": 1.5444609946836214, + "learning_rate": 3.9412726293920555e-06, + "loss": 0.9556, + "step": 1398 + }, + { + "epoch": 0.10514053810311137, + "grad_norm": 1.6281634862528145, + "learning_rate": 3.9411554611382914e-06, + "loss": 1.0922, + "step": 1399 + }, + { + "epoch": 0.10521569216894634, + "grad_norm": 1.632573160398625, + "learning_rate": 3.941038177863633e-06, + "loss": 1.0072, + "step": 1400 + }, + { + "epoch": 0.1052908462347813, + "grad_norm": 0.9005565035396196, + "learning_rate": 3.940920779575029e-06, + "loss": 0.8375, + "step": 1401 + }, + { + "epoch": 0.10536600030061627, + "grad_norm": 1.9082970181248398, + "learning_rate": 3.940803266279438e-06, + "loss": 1.0623, + "step": 1402 + }, + { + "epoch": 0.10544115436645123, + "grad_norm": 1.6316332779696778, + "learning_rate": 3.940685637983822e-06, + "loss": 0.9874, + "step": 1403 + }, + { + "epoch": 0.10551630843228618, + "grad_norm": 1.6798097787749784, + "learning_rate": 3.940567894695149e-06, + "loss": 1.0504, + "step": 1404 + }, + { + "epoch": 0.10559146249812115, + "grad_norm": 1.8171055913884386, + "learning_rate": 3.940450036420397e-06, + "loss": 0.9385, + "step": 1405 + }, + { + "epoch": 0.10566661656395611, + "grad_norm": 0.8444053083129163, + "learning_rate": 3.940332063166551e-06, + "loss": 0.7989, + "step": 1406 + }, + { + "epoch": 0.10574177062979107, + "grad_norm": 1.6573893389002095, + "learning_rate": 3.9402139749406e-06, + "loss": 1.059, + "step": 1407 + }, + { + "epoch": 0.10581692469562604, + "grad_norm": 1.7596988305425012, + "learning_rate": 3.940095771749542e-06, + "loss": 1.0221, + "step": 1408 + }, + { + "epoch": 0.10589207876146099, + "grad_norm": 1.8748343712449482, + "learning_rate": 3.939977453600379e-06, + "loss": 0.9676, + "step": 1409 + }, + { + "epoch": 0.10596723282729596, + "grad_norm": 1.601648021368502, + "learning_rate": 3.939859020500124e-06, + "loss": 0.9841, + "step": 1410 + }, + { + "epoch": 0.10604238689313092, + "grad_norm": 1.8449338571327765, + "learning_rate": 3.939740472455794e-06, + "loss": 0.9191, + "step": 1411 + }, + { + "epoch": 0.10611754095896588, + "grad_norm": 1.4325432111470806, + "learning_rate": 3.939621809474413e-06, + "loss": 1.034, + "step": 1412 + }, + { + "epoch": 0.10619269502480085, + "grad_norm": 1.560907238732804, + "learning_rate": 3.9395030315630124e-06, + "loss": 1.0578, + "step": 1413 + }, + { + "epoch": 0.1062678490906358, + "grad_norm": 1.4928327832871342, + "learning_rate": 3.939384138728631e-06, + "loss": 1.0117, + "step": 1414 + }, + { + "epoch": 0.10634300315647076, + "grad_norm": 1.8145568620902093, + "learning_rate": 3.939265130978312e-06, + "loss": 0.9856, + "step": 1415 + }, + { + "epoch": 0.10641815722230573, + "grad_norm": 1.6601930142988082, + "learning_rate": 3.939146008319109e-06, + "loss": 0.9962, + "step": 1416 + }, + { + "epoch": 0.10649331128814069, + "grad_norm": 1.8331073780689007, + "learning_rate": 3.939026770758079e-06, + "loss": 0.9787, + "step": 1417 + }, + { + "epoch": 0.10656846535397566, + "grad_norm": 1.4632245282423504, + "learning_rate": 3.938907418302288e-06, + "loss": 1.0012, + "step": 1418 + }, + { + "epoch": 0.10664361941981061, + "grad_norm": 1.6583105261430813, + "learning_rate": 3.938787950958807e-06, + "loss": 1.0021, + "step": 1419 + }, + { + "epoch": 0.10671877348564557, + "grad_norm": 1.4340555138583, + "learning_rate": 3.938668368734717e-06, + "loss": 0.9541, + "step": 1420 + }, + { + "epoch": 0.10679392755148054, + "grad_norm": 1.7141255675738936, + "learning_rate": 3.938548671637102e-06, + "loss": 0.982, + "step": 1421 + }, + { + "epoch": 0.1068690816173155, + "grad_norm": 1.6857402244170154, + "learning_rate": 3.938428859673055e-06, + "loss": 1.0224, + "step": 1422 + }, + { + "epoch": 0.10694423568315047, + "grad_norm": 1.7389989849875123, + "learning_rate": 3.9383089328496755e-06, + "loss": 0.977, + "step": 1423 + }, + { + "epoch": 0.10701938974898542, + "grad_norm": 1.5029590348519535, + "learning_rate": 3.938188891174069e-06, + "loss": 0.9479, + "step": 1424 + }, + { + "epoch": 0.10709454381482038, + "grad_norm": 1.785541250837169, + "learning_rate": 3.9380687346533495e-06, + "loss": 0.9714, + "step": 1425 + }, + { + "epoch": 0.10716969788065535, + "grad_norm": 2.0663220220753526, + "learning_rate": 3.9379484632946355e-06, + "loss": 1.0233, + "step": 1426 + }, + { + "epoch": 0.1072448519464903, + "grad_norm": 1.6639177142989632, + "learning_rate": 3.937828077105054e-06, + "loss": 0.8595, + "step": 1427 + }, + { + "epoch": 0.10732000601232526, + "grad_norm": 1.4224179247892796, + "learning_rate": 3.9377075760917396e-06, + "loss": 1.0586, + "step": 1428 + }, + { + "epoch": 0.10739516007816023, + "grad_norm": 1.5725587180717442, + "learning_rate": 3.93758696026183e-06, + "loss": 1.046, + "step": 1429 + }, + { + "epoch": 0.10747031414399519, + "grad_norm": 1.42302028935887, + "learning_rate": 3.9374662296224746e-06, + "loss": 1.056, + "step": 1430 + }, + { + "epoch": 0.10754546820983016, + "grad_norm": 1.48570545201767, + "learning_rate": 3.937345384180826e-06, + "loss": 1.0058, + "step": 1431 + }, + { + "epoch": 0.10762062227566511, + "grad_norm": 1.80485058332485, + "learning_rate": 3.937224423944044e-06, + "loss": 1.0227, + "step": 1432 + }, + { + "epoch": 0.10769577634150007, + "grad_norm": 1.3126014034517741, + "learning_rate": 3.937103348919297e-06, + "loss": 0.9125, + "step": 1433 + }, + { + "epoch": 0.10777093040733504, + "grad_norm": 1.8960435631974495, + "learning_rate": 3.936982159113759e-06, + "loss": 0.9368, + "step": 1434 + }, + { + "epoch": 0.10784608447317, + "grad_norm": 1.5705919918503666, + "learning_rate": 3.936860854534611e-06, + "loss": 0.9324, + "step": 1435 + }, + { + "epoch": 0.10792123853900495, + "grad_norm": 1.810914826271756, + "learning_rate": 3.936739435189041e-06, + "loss": 0.9828, + "step": 1436 + }, + { + "epoch": 0.10799639260483992, + "grad_norm": 1.9609548256401252, + "learning_rate": 3.936617901084243e-06, + "loss": 0.9468, + "step": 1437 + }, + { + "epoch": 0.10807154667067488, + "grad_norm": 1.9018774694852314, + "learning_rate": 3.936496252227417e-06, + "loss": 0.9855, + "step": 1438 + }, + { + "epoch": 0.10814670073650985, + "grad_norm": 2.039462346302894, + "learning_rate": 3.936374488625775e-06, + "loss": 1.0111, + "step": 1439 + }, + { + "epoch": 0.10822185480234481, + "grad_norm": 1.5679672720179414, + "learning_rate": 3.936252610286528e-06, + "loss": 1.0396, + "step": 1440 + }, + { + "epoch": 0.10829700886817976, + "grad_norm": 2.0119433231371993, + "learning_rate": 3.9361306172169005e-06, + "loss": 0.9187, + "step": 1441 + }, + { + "epoch": 0.10837216293401473, + "grad_norm": 2.680471233387894, + "learning_rate": 3.93600850942412e-06, + "loss": 0.8565, + "step": 1442 + }, + { + "epoch": 0.10844731699984969, + "grad_norm": 1.6222450672062316, + "learning_rate": 3.935886286915421e-06, + "loss": 0.9067, + "step": 1443 + }, + { + "epoch": 0.10852247106568465, + "grad_norm": 1.441663352139288, + "learning_rate": 3.935763949698047e-06, + "loss": 0.9253, + "step": 1444 + }, + { + "epoch": 0.10859762513151962, + "grad_norm": 1.6328633131671815, + "learning_rate": 3.935641497779247e-06, + "loss": 1.0277, + "step": 1445 + }, + { + "epoch": 0.10867277919735457, + "grad_norm": 1.7968692281879108, + "learning_rate": 3.935518931166275e-06, + "loss": 0.9936, + "step": 1446 + }, + { + "epoch": 0.10874793326318954, + "grad_norm": 1.4583841950765988, + "learning_rate": 3.935396249866396e-06, + "loss": 0.9627, + "step": 1447 + }, + { + "epoch": 0.1088230873290245, + "grad_norm": 2.635899847515841, + "learning_rate": 3.935273453886877e-06, + "loss": 0.9779, + "step": 1448 + }, + { + "epoch": 0.10889824139485946, + "grad_norm": 1.4972559147576388, + "learning_rate": 3.935150543234996e-06, + "loss": 0.9296, + "step": 1449 + }, + { + "epoch": 0.10897339546069443, + "grad_norm": 1.9801109938866968, + "learning_rate": 3.935027517918034e-06, + "loss": 0.9901, + "step": 1450 + }, + { + "epoch": 0.10904854952652938, + "grad_norm": 2.2016925889750496, + "learning_rate": 3.9349043779432825e-06, + "loss": 1.0085, + "step": 1451 + }, + { + "epoch": 0.10912370359236434, + "grad_norm": 2.709505433987513, + "learning_rate": 3.934781123318037e-06, + "loss": 1.0265, + "step": 1452 + }, + { + "epoch": 0.10919885765819931, + "grad_norm": 1.641593501918231, + "learning_rate": 3.934657754049602e-06, + "loss": 0.9795, + "step": 1453 + }, + { + "epoch": 0.10927401172403427, + "grad_norm": 1.446131376315896, + "learning_rate": 3.934534270145287e-06, + "loss": 1.0422, + "step": 1454 + }, + { + "epoch": 0.10934916578986924, + "grad_norm": 2.1698628717155284, + "learning_rate": 3.934410671612408e-06, + "loss": 1.0346, + "step": 1455 + }, + { + "epoch": 0.1094243198557042, + "grad_norm": 1.7939718926650894, + "learning_rate": 3.934286958458289e-06, + "loss": 1.0352, + "step": 1456 + }, + { + "epoch": 0.10949947392153915, + "grad_norm": 1.781417905690567, + "learning_rate": 3.934163130690262e-06, + "loss": 1.0047, + "step": 1457 + }, + { + "epoch": 0.10957462798737412, + "grad_norm": 1.6822031035205411, + "learning_rate": 3.9340391883156614e-06, + "loss": 0.9209, + "step": 1458 + }, + { + "epoch": 0.10964978205320908, + "grad_norm": 1.7590856392091254, + "learning_rate": 3.933915131341834e-06, + "loss": 1.0295, + "step": 1459 + }, + { + "epoch": 0.10972493611904403, + "grad_norm": 1.6920510840268448, + "learning_rate": 3.93379095977613e-06, + "loss": 0.9215, + "step": 1460 + }, + { + "epoch": 0.109800090184879, + "grad_norm": 1.4449369200299371, + "learning_rate": 3.9336666736259055e-06, + "loss": 1.0158, + "step": 1461 + }, + { + "epoch": 0.10987524425071396, + "grad_norm": 1.5165669768271486, + "learning_rate": 3.933542272898527e-06, + "loss": 0.9676, + "step": 1462 + }, + { + "epoch": 0.10995039831654893, + "grad_norm": 1.7862471371328104, + "learning_rate": 3.933417757601365e-06, + "loss": 0.9143, + "step": 1463 + }, + { + "epoch": 0.11002555238238389, + "grad_norm": 1.580361999738248, + "learning_rate": 3.933293127741796e-06, + "loss": 0.9646, + "step": 1464 + }, + { + "epoch": 0.11010070644821884, + "grad_norm": 1.6554962458359146, + "learning_rate": 3.933168383327207e-06, + "loss": 1.0345, + "step": 1465 + }, + { + "epoch": 0.11017586051405381, + "grad_norm": 1.3836617313616493, + "learning_rate": 3.933043524364989e-06, + "loss": 0.9801, + "step": 1466 + }, + { + "epoch": 0.11025101457988877, + "grad_norm": 1.3713625130965026, + "learning_rate": 3.932918550862539e-06, + "loss": 0.9743, + "step": 1467 + }, + { + "epoch": 0.11032616864572373, + "grad_norm": 2.0597740423477973, + "learning_rate": 3.932793462827265e-06, + "loss": 1.0153, + "step": 1468 + }, + { + "epoch": 0.1104013227115587, + "grad_norm": 1.592701585787378, + "learning_rate": 3.932668260266576e-06, + "loss": 0.9688, + "step": 1469 + }, + { + "epoch": 0.11047647677739365, + "grad_norm": 1.582250358573064, + "learning_rate": 3.932542943187892e-06, + "loss": 1.0735, + "step": 1470 + }, + { + "epoch": 0.11055163084322862, + "grad_norm": 1.581066148043701, + "learning_rate": 3.932417511598638e-06, + "loss": 1.0407, + "step": 1471 + }, + { + "epoch": 0.11062678490906358, + "grad_norm": 2.191740013993546, + "learning_rate": 3.932291965506247e-06, + "loss": 0.9401, + "step": 1472 + }, + { + "epoch": 0.11070193897489854, + "grad_norm": 1.5297095666371143, + "learning_rate": 3.932166304918158e-06, + "loss": 0.9742, + "step": 1473 + }, + { + "epoch": 0.1107770930407335, + "grad_norm": 1.6208164943157044, + "learning_rate": 3.9320405298418175e-06, + "loss": 1.0163, + "step": 1474 + }, + { + "epoch": 0.11085224710656846, + "grad_norm": 1.4676272382278082, + "learning_rate": 3.931914640284676e-06, + "loss": 0.9794, + "step": 1475 + }, + { + "epoch": 0.11092740117240343, + "grad_norm": 1.5536419597206161, + "learning_rate": 3.931788636254195e-06, + "loss": 1.0254, + "step": 1476 + }, + { + "epoch": 0.11100255523823839, + "grad_norm": 1.7253631983436792, + "learning_rate": 3.931662517757839e-06, + "loss": 0.9331, + "step": 1477 + }, + { + "epoch": 0.11107770930407335, + "grad_norm": 1.6303860828479515, + "learning_rate": 3.931536284803083e-06, + "loss": 0.9477, + "step": 1478 + }, + { + "epoch": 0.11115286336990832, + "grad_norm": 1.644254103841209, + "learning_rate": 3.931409937397406e-06, + "loss": 1.0734, + "step": 1479 + }, + { + "epoch": 0.11122801743574327, + "grad_norm": 0.8816291845523336, + "learning_rate": 3.931283475548293e-06, + "loss": 0.8701, + "step": 1480 + }, + { + "epoch": 0.11130317150157823, + "grad_norm": 1.5771985497795247, + "learning_rate": 3.93115689926324e-06, + "loss": 0.9491, + "step": 1481 + }, + { + "epoch": 0.1113783255674132, + "grad_norm": 1.5808176165684096, + "learning_rate": 3.931030208549745e-06, + "loss": 1.0337, + "step": 1482 + }, + { + "epoch": 0.11145347963324816, + "grad_norm": 1.8779687267417409, + "learning_rate": 3.930903403415316e-06, + "loss": 0.9912, + "step": 1483 + }, + { + "epoch": 0.11152863369908313, + "grad_norm": 1.7926887509659801, + "learning_rate": 3.930776483867467e-06, + "loss": 1.0357, + "step": 1484 + }, + { + "epoch": 0.11160378776491808, + "grad_norm": 1.5214399495329383, + "learning_rate": 3.9306494499137175e-06, + "loss": 1.0125, + "step": 1485 + }, + { + "epoch": 0.11167894183075304, + "grad_norm": 1.9015895593954373, + "learning_rate": 3.930522301561595e-06, + "loss": 0.982, + "step": 1486 + }, + { + "epoch": 0.11175409589658801, + "grad_norm": 1.5681288230244272, + "learning_rate": 3.930395038818633e-06, + "loss": 0.9918, + "step": 1487 + }, + { + "epoch": 0.11182924996242297, + "grad_norm": 1.913636016042534, + "learning_rate": 3.930267661692374e-06, + "loss": 1.0669, + "step": 1488 + }, + { + "epoch": 0.11190440402825792, + "grad_norm": 4.104273889886981, + "learning_rate": 3.930140170190364e-06, + "loss": 0.9977, + "step": 1489 + }, + { + "epoch": 0.1119795580940929, + "grad_norm": 1.082240831480361, + "learning_rate": 3.930012564320159e-06, + "loss": 0.879, + "step": 1490 + }, + { + "epoch": 0.11205471215992785, + "grad_norm": 1.5126569335482023, + "learning_rate": 3.929884844089318e-06, + "loss": 0.9906, + "step": 1491 + }, + { + "epoch": 0.11212986622576282, + "grad_norm": 1.3321225311236373, + "learning_rate": 3.92975700950541e-06, + "loss": 1.0381, + "step": 1492 + }, + { + "epoch": 0.11220502029159778, + "grad_norm": 1.4147299355828147, + "learning_rate": 3.92962906057601e-06, + "loss": 0.8923, + "step": 1493 + }, + { + "epoch": 0.11228017435743273, + "grad_norm": 2.243105497745778, + "learning_rate": 3.929500997308698e-06, + "loss": 1.0219, + "step": 1494 + }, + { + "epoch": 0.1123553284232677, + "grad_norm": 1.394003503778616, + "learning_rate": 3.929372819711065e-06, + "loss": 0.9731, + "step": 1495 + }, + { + "epoch": 0.11243048248910266, + "grad_norm": 1.9066625054965374, + "learning_rate": 3.929244527790703e-06, + "loss": 0.9607, + "step": 1496 + }, + { + "epoch": 0.11250563655493762, + "grad_norm": 1.870450731617208, + "learning_rate": 3.929116121555216e-06, + "loss": 1.0078, + "step": 1497 + }, + { + "epoch": 0.11258079062077259, + "grad_norm": 1.7314037268518412, + "learning_rate": 3.928987601012212e-06, + "loss": 1.0708, + "step": 1498 + }, + { + "epoch": 0.11265594468660754, + "grad_norm": 1.699746867575763, + "learning_rate": 3.928858966169306e-06, + "loss": 0.9525, + "step": 1499 + }, + { + "epoch": 0.11273109875244251, + "grad_norm": 1.6950183384671293, + "learning_rate": 3.928730217034119e-06, + "loss": 0.9826, + "step": 1500 + }, + { + "epoch": 0.11280625281827747, + "grad_norm": 1.4546463908318852, + "learning_rate": 3.928601353614282e-06, + "loss": 0.9096, + "step": 1501 + }, + { + "epoch": 0.11288140688411243, + "grad_norm": 1.9955490175256674, + "learning_rate": 3.92847237591743e-06, + "loss": 0.9749, + "step": 1502 + }, + { + "epoch": 0.1129565609499474, + "grad_norm": 1.3697612406127921, + "learning_rate": 3.928343283951204e-06, + "loss": 0.9999, + "step": 1503 + }, + { + "epoch": 0.11303171501578235, + "grad_norm": 1.7769527466329766, + "learning_rate": 3.928214077723255e-06, + "loss": 0.9933, + "step": 1504 + }, + { + "epoch": 0.11310686908161731, + "grad_norm": 1.7896502999272408, + "learning_rate": 3.928084757241239e-06, + "loss": 0.99, + "step": 1505 + }, + { + "epoch": 0.11318202314745228, + "grad_norm": 1.6862653524889006, + "learning_rate": 3.9279553225128165e-06, + "loss": 0.8014, + "step": 1506 + }, + { + "epoch": 0.11325717721328724, + "grad_norm": 1.5870597138687415, + "learning_rate": 3.92782577354566e-06, + "loss": 0.9399, + "step": 1507 + }, + { + "epoch": 0.1133323312791222, + "grad_norm": 1.5311041003415684, + "learning_rate": 3.927696110347443e-06, + "loss": 0.9957, + "step": 1508 + }, + { + "epoch": 0.11340748534495716, + "grad_norm": 1.79492240631793, + "learning_rate": 3.92756633292585e-06, + "loss": 0.9942, + "step": 1509 + }, + { + "epoch": 0.11348263941079212, + "grad_norm": 1.6219158248320338, + "learning_rate": 3.927436441288571e-06, + "loss": 1.0538, + "step": 1510 + }, + { + "epoch": 0.11355779347662709, + "grad_norm": 1.6863285519774114, + "learning_rate": 3.9273064354433025e-06, + "loss": 0.9314, + "step": 1511 + }, + { + "epoch": 0.11363294754246205, + "grad_norm": 1.305968883143708, + "learning_rate": 3.927176315397747e-06, + "loss": 1.0282, + "step": 1512 + }, + { + "epoch": 0.113708101608297, + "grad_norm": 0.8073858185696063, + "learning_rate": 3.927046081159615e-06, + "loss": 0.7616, + "step": 1513 + }, + { + "epoch": 0.11378325567413197, + "grad_norm": 2.0222211444185807, + "learning_rate": 3.926915732736624e-06, + "loss": 0.8612, + "step": 1514 + }, + { + "epoch": 0.11385840973996693, + "grad_norm": 1.8721785616036521, + "learning_rate": 3.926785270136497e-06, + "loss": 1.035, + "step": 1515 + }, + { + "epoch": 0.1139335638058019, + "grad_norm": 1.7528584160302136, + "learning_rate": 3.926654693366965e-06, + "loss": 0.9016, + "step": 1516 + }, + { + "epoch": 0.11400871787163686, + "grad_norm": 1.9785820214596388, + "learning_rate": 3.926524002435764e-06, + "loss": 1.0211, + "step": 1517 + }, + { + "epoch": 0.11408387193747181, + "grad_norm": 1.82470703125, + "learning_rate": 3.9263931973506395e-06, + "loss": 0.9051, + "step": 1518 + }, + { + "epoch": 0.11415902600330678, + "grad_norm": 1.4914971796994427, + "learning_rate": 3.926262278119341e-06, + "loss": 1.045, + "step": 1519 + }, + { + "epoch": 0.11423418006914174, + "grad_norm": 1.5235208341840925, + "learning_rate": 3.9261312447496265e-06, + "loss": 1.0725, + "step": 1520 + }, + { + "epoch": 0.11430933413497671, + "grad_norm": 1.7544838139158898, + "learning_rate": 3.92600009724926e-06, + "loss": 1.0013, + "step": 1521 + }, + { + "epoch": 0.11438448820081167, + "grad_norm": 1.6421807315298564, + "learning_rate": 3.925868835626012e-06, + "loss": 1.0446, + "step": 1522 + }, + { + "epoch": 0.11445964226664662, + "grad_norm": 1.489233478284798, + "learning_rate": 3.925737459887662e-06, + "loss": 0.9703, + "step": 1523 + }, + { + "epoch": 0.11453479633248159, + "grad_norm": 1.4572317685989213, + "learning_rate": 3.925605970041992e-06, + "loss": 1.0454, + "step": 1524 + }, + { + "epoch": 0.11460995039831655, + "grad_norm": 1.3632671278957171, + "learning_rate": 3.925474366096796e-06, + "loss": 0.9803, + "step": 1525 + }, + { + "epoch": 0.1146851044641515, + "grad_norm": 1.4175549788804678, + "learning_rate": 3.92534264805987e-06, + "loss": 1.098, + "step": 1526 + }, + { + "epoch": 0.11476025852998648, + "grad_norm": 2.5187740163865, + "learning_rate": 3.92521081593902e-06, + "loss": 1.049, + "step": 1527 + }, + { + "epoch": 0.11483541259582143, + "grad_norm": 1.4830718092095112, + "learning_rate": 3.925078869742056e-06, + "loss": 0.9622, + "step": 1528 + }, + { + "epoch": 0.1149105666616564, + "grad_norm": 1.4014391093123848, + "learning_rate": 3.924946809476798e-06, + "loss": 0.928, + "step": 1529 + }, + { + "epoch": 0.11498572072749136, + "grad_norm": 1.7687717854266867, + "learning_rate": 3.924814635151071e-06, + "loss": 1.0003, + "step": 1530 + }, + { + "epoch": 0.11506087479332631, + "grad_norm": 1.8120824398630055, + "learning_rate": 3.924682346772705e-06, + "loss": 1.0611, + "step": 1531 + }, + { + "epoch": 0.11513602885916129, + "grad_norm": 1.5551657396408542, + "learning_rate": 3.92454994434954e-06, + "loss": 1.0751, + "step": 1532 + }, + { + "epoch": 0.11521118292499624, + "grad_norm": 1.5000217754055019, + "learning_rate": 3.9244174278894226e-06, + "loss": 0.9686, + "step": 1533 + }, + { + "epoch": 0.1152863369908312, + "grad_norm": 1.3941627250580706, + "learning_rate": 3.924284797400202e-06, + "loss": 1.0099, + "step": 1534 + }, + { + "epoch": 0.11536149105666617, + "grad_norm": 1.4970194290979666, + "learning_rate": 3.92415205288974e-06, + "loss": 1.0144, + "step": 1535 + }, + { + "epoch": 0.11543664512250112, + "grad_norm": 1.457531308840606, + "learning_rate": 3.9240191943659e-06, + "loss": 0.9942, + "step": 1536 + }, + { + "epoch": 0.1155117991883361, + "grad_norm": 1.4093953454691188, + "learning_rate": 3.923886221836555e-06, + "loss": 0.963, + "step": 1537 + }, + { + "epoch": 0.11558695325417105, + "grad_norm": 1.303605230382146, + "learning_rate": 3.923753135309584e-06, + "loss": 1.0902, + "step": 1538 + }, + { + "epoch": 0.11566210732000601, + "grad_norm": 1.546378932748989, + "learning_rate": 3.923619934792873e-06, + "loss": 1.0787, + "step": 1539 + }, + { + "epoch": 0.11573726138584098, + "grad_norm": 1.8138205223122972, + "learning_rate": 3.923486620294316e-06, + "loss": 0.9936, + "step": 1540 + }, + { + "epoch": 0.11581241545167593, + "grad_norm": 2.2027785657849477, + "learning_rate": 3.923353191821811e-06, + "loss": 0.9718, + "step": 1541 + }, + { + "epoch": 0.11588756951751089, + "grad_norm": 1.435013154132379, + "learning_rate": 3.923219649383264e-06, + "loss": 0.9931, + "step": 1542 + }, + { + "epoch": 0.11596272358334586, + "grad_norm": 1.7433504063783494, + "learning_rate": 3.923085992986588e-06, + "loss": 1.0081, + "step": 1543 + }, + { + "epoch": 0.11603787764918082, + "grad_norm": 1.6209902810776708, + "learning_rate": 3.922952222639703e-06, + "loss": 1.0151, + "step": 1544 + }, + { + "epoch": 0.11611303171501579, + "grad_norm": 1.6685822048907717, + "learning_rate": 3.922818338350536e-06, + "loss": 1.0801, + "step": 1545 + }, + { + "epoch": 0.11618818578085074, + "grad_norm": 1.5827329233393468, + "learning_rate": 3.9226843401270195e-06, + "loss": 0.9856, + "step": 1546 + }, + { + "epoch": 0.1162633398466857, + "grad_norm": 1.444994917335943, + "learning_rate": 3.922550227977093e-06, + "loss": 1.0202, + "step": 1547 + }, + { + "epoch": 0.11633849391252067, + "grad_norm": 0.8311856409842764, + "learning_rate": 3.9224160019087036e-06, + "loss": 0.7849, + "step": 1548 + }, + { + "epoch": 0.11641364797835563, + "grad_norm": 1.3090076759789415, + "learning_rate": 3.922281661929804e-06, + "loss": 0.9195, + "step": 1549 + }, + { + "epoch": 0.11648880204419058, + "grad_norm": 1.6315681059660114, + "learning_rate": 3.922147208048356e-06, + "loss": 0.9523, + "step": 1550 + }, + { + "epoch": 0.11656395611002555, + "grad_norm": 1.7124088875694756, + "learning_rate": 3.922012640272325e-06, + "loss": 1.0509, + "step": 1551 + }, + { + "epoch": 0.11663911017586051, + "grad_norm": 1.4136755819659448, + "learning_rate": 3.921877958609685e-06, + "loss": 1.0171, + "step": 1552 + }, + { + "epoch": 0.11671426424169548, + "grad_norm": 1.6011322187811483, + "learning_rate": 3.9217431630684174e-06, + "loss": 1.1117, + "step": 1553 + }, + { + "epoch": 0.11678941830753044, + "grad_norm": 1.645542368540162, + "learning_rate": 3.921608253656508e-06, + "loss": 0.9877, + "step": 1554 + }, + { + "epoch": 0.1168645723733654, + "grad_norm": 2.0951342206034798, + "learning_rate": 3.921473230381951e-06, + "loss": 0.9373, + "step": 1555 + }, + { + "epoch": 0.11693972643920036, + "grad_norm": 1.449306187215506, + "learning_rate": 3.921338093252748e-06, + "loss": 0.9447, + "step": 1556 + }, + { + "epoch": 0.11701488050503532, + "grad_norm": 1.6110912171958893, + "learning_rate": 3.921202842276906e-06, + "loss": 0.968, + "step": 1557 + }, + { + "epoch": 0.11709003457087028, + "grad_norm": 1.9597571114470562, + "learning_rate": 3.921067477462437e-06, + "loss": 0.9303, + "step": 1558 + }, + { + "epoch": 0.11716518863670525, + "grad_norm": 0.8586841233707205, + "learning_rate": 3.920931998817365e-06, + "loss": 0.8572, + "step": 1559 + }, + { + "epoch": 0.1172403427025402, + "grad_norm": 1.3032218329954781, + "learning_rate": 3.920796406349717e-06, + "loss": 1.0288, + "step": 1560 + }, + { + "epoch": 0.11731549676837517, + "grad_norm": 1.500717706640734, + "learning_rate": 3.920660700067525e-06, + "loss": 1.0188, + "step": 1561 + }, + { + "epoch": 0.11739065083421013, + "grad_norm": 1.846767348816715, + "learning_rate": 3.920524879978833e-06, + "loss": 0.9864, + "step": 1562 + }, + { + "epoch": 0.11746580490004509, + "grad_norm": 1.5557632941466193, + "learning_rate": 3.920388946091687e-06, + "loss": 0.9065, + "step": 1563 + }, + { + "epoch": 0.11754095896588006, + "grad_norm": 1.361060840084894, + "learning_rate": 3.920252898414143e-06, + "loss": 1.0667, + "step": 1564 + }, + { + "epoch": 0.11761611303171501, + "grad_norm": 1.597134082776424, + "learning_rate": 3.920116736954261e-06, + "loss": 0.9151, + "step": 1565 + }, + { + "epoch": 0.11769126709754998, + "grad_norm": 1.419928675324371, + "learning_rate": 3.91998046172011e-06, + "loss": 0.9291, + "step": 1566 + }, + { + "epoch": 0.11776642116338494, + "grad_norm": 1.824872767572679, + "learning_rate": 3.9198440727197645e-06, + "loss": 1.0431, + "step": 1567 + }, + { + "epoch": 0.1178415752292199, + "grad_norm": 3.234874889424927, + "learning_rate": 3.919707569961306e-06, + "loss": 0.9975, + "step": 1568 + }, + { + "epoch": 0.11791672929505487, + "grad_norm": 1.4217457974598007, + "learning_rate": 3.9195709534528235e-06, + "loss": 1.061, + "step": 1569 + }, + { + "epoch": 0.11799188336088982, + "grad_norm": 1.3551633639233027, + "learning_rate": 3.919434223202411e-06, + "loss": 0.9395, + "step": 1570 + }, + { + "epoch": 0.11806703742672478, + "grad_norm": 1.502641339120488, + "learning_rate": 3.919297379218171e-06, + "loss": 1.0827, + "step": 1571 + }, + { + "epoch": 0.11814219149255975, + "grad_norm": 2.1006462056600013, + "learning_rate": 3.919160421508211e-06, + "loss": 0.9924, + "step": 1572 + }, + { + "epoch": 0.1182173455583947, + "grad_norm": 1.836801398724678, + "learning_rate": 3.919023350080648e-06, + "loss": 1.0494, + "step": 1573 + }, + { + "epoch": 0.11829249962422968, + "grad_norm": 1.3743999645816012, + "learning_rate": 3.918886164943603e-06, + "loss": 1.047, + "step": 1574 + }, + { + "epoch": 0.11836765369006463, + "grad_norm": 1.6465129736682569, + "learning_rate": 3.918748866105204e-06, + "loss": 1.0584, + "step": 1575 + }, + { + "epoch": 0.11844280775589959, + "grad_norm": 1.749630480309186, + "learning_rate": 3.918611453573589e-06, + "loss": 0.935, + "step": 1576 + }, + { + "epoch": 0.11851796182173456, + "grad_norm": 1.3184652554415375, + "learning_rate": 3.918473927356896e-06, + "loss": 0.9493, + "step": 1577 + }, + { + "epoch": 0.11859311588756952, + "grad_norm": 1.5944619925829608, + "learning_rate": 3.918336287463279e-06, + "loss": 0.9364, + "step": 1578 + }, + { + "epoch": 0.11866826995340447, + "grad_norm": 1.5481256380722832, + "learning_rate": 3.9181985339008895e-06, + "loss": 1.0914, + "step": 1579 + }, + { + "epoch": 0.11874342401923944, + "grad_norm": 1.7268503907058954, + "learning_rate": 3.918060666677892e-06, + "loss": 1.0412, + "step": 1580 + }, + { + "epoch": 0.1188185780850744, + "grad_norm": 0.8538164878636479, + "learning_rate": 3.9179226858024555e-06, + "loss": 0.8502, + "step": 1581 + }, + { + "epoch": 0.11889373215090937, + "grad_norm": 1.345617903404109, + "learning_rate": 3.917784591282756e-06, + "loss": 0.9119, + "step": 1582 + }, + { + "epoch": 0.11896888621674433, + "grad_norm": 1.554336201452371, + "learning_rate": 3.917646383126975e-06, + "loss": 0.9682, + "step": 1583 + }, + { + "epoch": 0.11904404028257928, + "grad_norm": 1.3424642754886011, + "learning_rate": 3.917508061343303e-06, + "loss": 1.0852, + "step": 1584 + }, + { + "epoch": 0.11911919434841425, + "grad_norm": 1.403170328746191, + "learning_rate": 3.917369625939936e-06, + "loss": 1.0423, + "step": 1585 + }, + { + "epoch": 0.11919434841424921, + "grad_norm": 1.5863132877214885, + "learning_rate": 3.917231076925076e-06, + "loss": 0.9958, + "step": 1586 + }, + { + "epoch": 0.11926950248008417, + "grad_norm": 1.7069445136270007, + "learning_rate": 3.917092414306933e-06, + "loss": 1.0215, + "step": 1587 + }, + { + "epoch": 0.11934465654591914, + "grad_norm": 1.7163810356555753, + "learning_rate": 3.916953638093725e-06, + "loss": 0.9181, + "step": 1588 + }, + { + "epoch": 0.1194198106117541, + "grad_norm": 1.7124189120954951, + "learning_rate": 3.9168147482936715e-06, + "loss": 0.98, + "step": 1589 + }, + { + "epoch": 0.11949496467758906, + "grad_norm": 1.8054597201548652, + "learning_rate": 3.916675744915005e-06, + "loss": 0.93, + "step": 1590 + }, + { + "epoch": 0.11957011874342402, + "grad_norm": 1.8317493473976985, + "learning_rate": 3.916536627965961e-06, + "loss": 0.8981, + "step": 1591 + }, + { + "epoch": 0.11964527280925898, + "grad_norm": 1.4929812252655295, + "learning_rate": 3.916397397454783e-06, + "loss": 0.9779, + "step": 1592 + }, + { + "epoch": 0.11972042687509395, + "grad_norm": 1.3576687156130431, + "learning_rate": 3.916258053389721e-06, + "loss": 0.9515, + "step": 1593 + }, + { + "epoch": 0.1197955809409289, + "grad_norm": 1.5093876808534183, + "learning_rate": 3.916118595779031e-06, + "loss": 0.9738, + "step": 1594 + }, + { + "epoch": 0.11987073500676386, + "grad_norm": 1.469656481925529, + "learning_rate": 3.915979024630977e-06, + "loss": 0.9934, + "step": 1595 + }, + { + "epoch": 0.11994588907259883, + "grad_norm": 2.261979004112365, + "learning_rate": 3.91583933995383e-06, + "loss": 1.0059, + "step": 1596 + }, + { + "epoch": 0.12002104313843379, + "grad_norm": 1.813086053218918, + "learning_rate": 3.915699541755865e-06, + "loss": 0.9852, + "step": 1597 + }, + { + "epoch": 0.12009619720426876, + "grad_norm": 1.3149606708258565, + "learning_rate": 3.915559630045367e-06, + "loss": 0.9555, + "step": 1598 + }, + { + "epoch": 0.12017135127010371, + "grad_norm": 1.718414135108295, + "learning_rate": 3.9154196048306244e-06, + "loss": 1.0471, + "step": 1599 + }, + { + "epoch": 0.12024650533593867, + "grad_norm": 1.5664223838983635, + "learning_rate": 3.915279466119937e-06, + "loss": 1.0172, + "step": 1600 + }, + { + "epoch": 0.12032165940177364, + "grad_norm": 1.588962845129651, + "learning_rate": 3.915139213921606e-06, + "loss": 0.955, + "step": 1601 + }, + { + "epoch": 0.1203968134676086, + "grad_norm": 2.3814113812566204, + "learning_rate": 3.914998848243944e-06, + "loss": 0.9031, + "step": 1602 + }, + { + "epoch": 0.12047196753344355, + "grad_norm": 1.7857727054849133, + "learning_rate": 3.914858369095267e-06, + "loss": 1.0066, + "step": 1603 + }, + { + "epoch": 0.12054712159927852, + "grad_norm": 1.8059521492230084, + "learning_rate": 3.914717776483899e-06, + "loss": 0.9784, + "step": 1604 + }, + { + "epoch": 0.12062227566511348, + "grad_norm": 1.782625370084177, + "learning_rate": 3.9145770704181715e-06, + "loss": 1.0405, + "step": 1605 + }, + { + "epoch": 0.12069742973094845, + "grad_norm": 1.5787141663266886, + "learning_rate": 3.9144362509064194e-06, + "loss": 0.9698, + "step": 1606 + }, + { + "epoch": 0.1207725837967834, + "grad_norm": 1.2377101411513567, + "learning_rate": 3.91429531795699e-06, + "loss": 0.9201, + "step": 1607 + }, + { + "epoch": 0.12084773786261836, + "grad_norm": 1.5089746453674888, + "learning_rate": 3.9141542715782325e-06, + "loss": 0.8927, + "step": 1608 + }, + { + "epoch": 0.12092289192845333, + "grad_norm": 1.695554232612806, + "learning_rate": 3.9140131117785045e-06, + "loss": 0.9524, + "step": 1609 + }, + { + "epoch": 0.12099804599428829, + "grad_norm": 1.6983794847281652, + "learning_rate": 3.91387183856617e-06, + "loss": 0.9896, + "step": 1610 + }, + { + "epoch": 0.12107320006012326, + "grad_norm": 1.4790320357540825, + "learning_rate": 3.913730451949601e-06, + "loss": 0.9344, + "step": 1611 + }, + { + "epoch": 0.12114835412595822, + "grad_norm": 1.6551842049371905, + "learning_rate": 3.913588951937174e-06, + "loss": 0.9098, + "step": 1612 + }, + { + "epoch": 0.12122350819179317, + "grad_norm": 0.8435300611013389, + "learning_rate": 3.913447338537274e-06, + "loss": 0.9113, + "step": 1613 + }, + { + "epoch": 0.12129866225762814, + "grad_norm": 2.1679246502764165, + "learning_rate": 3.913305611758292e-06, + "loss": 0.9165, + "step": 1614 + }, + { + "epoch": 0.1213738163234631, + "grad_norm": 1.4764534238864035, + "learning_rate": 3.913163771608627e-06, + "loss": 0.9082, + "step": 1615 + }, + { + "epoch": 0.12144897038929806, + "grad_norm": 1.4816579747812206, + "learning_rate": 3.913021818096682e-06, + "loss": 1.0181, + "step": 1616 + }, + { + "epoch": 0.12152412445513303, + "grad_norm": 1.3187460551270695, + "learning_rate": 3.912879751230868e-06, + "loss": 0.9011, + "step": 1617 + }, + { + "epoch": 0.12159927852096798, + "grad_norm": 1.8027148013225016, + "learning_rate": 3.9127375710196044e-06, + "loss": 1.0086, + "step": 1618 + }, + { + "epoch": 0.12167443258680295, + "grad_norm": 1.5069258064176096, + "learning_rate": 3.912595277471316e-06, + "loss": 0.8998, + "step": 1619 + }, + { + "epoch": 0.12174958665263791, + "grad_norm": 1.4666178221950001, + "learning_rate": 3.912452870594433e-06, + "loss": 1.0143, + "step": 1620 + }, + { + "epoch": 0.12182474071847287, + "grad_norm": 1.6102135704976361, + "learning_rate": 3.912310350397394e-06, + "loss": 1.0023, + "step": 1621 + }, + { + "epoch": 0.12189989478430784, + "grad_norm": 1.6410404224333728, + "learning_rate": 3.912167716888644e-06, + "loss": 1.1025, + "step": 1622 + }, + { + "epoch": 0.12197504885014279, + "grad_norm": 1.4646415062208893, + "learning_rate": 3.912024970076636e-06, + "loss": 1.0023, + "step": 1623 + }, + { + "epoch": 0.12205020291597775, + "grad_norm": 1.4504077897005208, + "learning_rate": 3.911882109969825e-06, + "loss": 1.0193, + "step": 1624 + }, + { + "epoch": 0.12212535698181272, + "grad_norm": 1.759604257177227, + "learning_rate": 3.9117391365766785e-06, + "loss": 0.98, + "step": 1625 + }, + { + "epoch": 0.12220051104764768, + "grad_norm": 1.4758385392191804, + "learning_rate": 3.9115960499056674e-06, + "loss": 1.0044, + "step": 1626 + }, + { + "epoch": 0.12227566511348265, + "grad_norm": 1.6504749163709524, + "learning_rate": 3.911452849965271e-06, + "loss": 0.99, + "step": 1627 + }, + { + "epoch": 0.1223508191793176, + "grad_norm": 1.4305860686931648, + "learning_rate": 3.911309536763974e-06, + "loss": 0.9394, + "step": 1628 + }, + { + "epoch": 0.12242597324515256, + "grad_norm": 1.801385939350653, + "learning_rate": 3.911166110310267e-06, + "loss": 0.9973, + "step": 1629 + }, + { + "epoch": 0.12250112731098753, + "grad_norm": 1.4718184377246055, + "learning_rate": 3.91102257061265e-06, + "loss": 1.0669, + "step": 1630 + }, + { + "epoch": 0.12257628137682249, + "grad_norm": 1.5996344893909509, + "learning_rate": 3.9108789176796285e-06, + "loss": 0.9947, + "step": 1631 + }, + { + "epoch": 0.12265143544265744, + "grad_norm": 1.5222910161650198, + "learning_rate": 3.910735151519713e-06, + "loss": 1.0342, + "step": 1632 + }, + { + "epoch": 0.12272658950849241, + "grad_norm": 1.453851825693732, + "learning_rate": 3.910591272141424e-06, + "loss": 0.93, + "step": 1633 + }, + { + "epoch": 0.12280174357432737, + "grad_norm": 1.4826359248770697, + "learning_rate": 3.910447279553285e-06, + "loss": 1.0011, + "step": 1634 + }, + { + "epoch": 0.12287689764016234, + "grad_norm": 1.6241534302069203, + "learning_rate": 3.91030317376383e-06, + "loss": 1.007, + "step": 1635 + }, + { + "epoch": 0.1229520517059973, + "grad_norm": 0.7845461928344953, + "learning_rate": 3.9101589547815965e-06, + "loss": 0.7877, + "step": 1636 + }, + { + "epoch": 0.12302720577183225, + "grad_norm": 1.448260131642458, + "learning_rate": 3.91001462261513e-06, + "loss": 1.0202, + "step": 1637 + }, + { + "epoch": 0.12310235983766722, + "grad_norm": 0.8326250921093853, + "learning_rate": 3.909870177272984e-06, + "loss": 0.81, + "step": 1638 + }, + { + "epoch": 0.12317751390350218, + "grad_norm": 2.108673205014678, + "learning_rate": 3.909725618763716e-06, + "loss": 0.8939, + "step": 1639 + }, + { + "epoch": 0.12325266796933713, + "grad_norm": 1.5351161029409688, + "learning_rate": 3.909580947095892e-06, + "loss": 0.9425, + "step": 1640 + }, + { + "epoch": 0.1233278220351721, + "grad_norm": 1.7370048723125038, + "learning_rate": 3.909436162278085e-06, + "loss": 0.9895, + "step": 1641 + }, + { + "epoch": 0.12340297610100706, + "grad_norm": 1.4092513374835223, + "learning_rate": 3.9092912643188745e-06, + "loss": 1.0004, + "step": 1642 + }, + { + "epoch": 0.12347813016684203, + "grad_norm": 1.699956133220401, + "learning_rate": 3.909146253226844e-06, + "loss": 1.0547, + "step": 1643 + }, + { + "epoch": 0.12355328423267699, + "grad_norm": 1.8078513547633832, + "learning_rate": 3.909001129010588e-06, + "loss": 1.0438, + "step": 1644 + }, + { + "epoch": 0.12362843829851194, + "grad_norm": 1.4847500778945875, + "learning_rate": 3.908855891678706e-06, + "loss": 1.0281, + "step": 1645 + }, + { + "epoch": 0.12370359236434691, + "grad_norm": 1.704975138172548, + "learning_rate": 3.908710541239802e-06, + "loss": 1.0904, + "step": 1646 + }, + { + "epoch": 0.12377874643018187, + "grad_norm": 1.4665760427711707, + "learning_rate": 3.90856507770249e-06, + "loss": 0.9286, + "step": 1647 + }, + { + "epoch": 0.12385390049601683, + "grad_norm": 1.239748401496966, + "learning_rate": 3.908419501075388e-06, + "loss": 0.8908, + "step": 1648 + }, + { + "epoch": 0.1239290545618518, + "grad_norm": 1.8255485572469259, + "learning_rate": 3.908273811367123e-06, + "loss": 0.9426, + "step": 1649 + }, + { + "epoch": 0.12400420862768675, + "grad_norm": 1.5540964354463702, + "learning_rate": 3.908128008586328e-06, + "loss": 0.9037, + "step": 1650 + }, + { + "epoch": 0.12407936269352172, + "grad_norm": 1.8761325594586888, + "learning_rate": 3.90798209274164e-06, + "loss": 0.8583, + "step": 1651 + }, + { + "epoch": 0.12415451675935668, + "grad_norm": 1.496164902870068, + "learning_rate": 3.907836063841709e-06, + "loss": 0.9492, + "step": 1652 + }, + { + "epoch": 0.12422967082519164, + "grad_norm": 1.5641245979229912, + "learning_rate": 3.907689921895184e-06, + "loss": 0.9642, + "step": 1653 + }, + { + "epoch": 0.12430482489102661, + "grad_norm": 2.254474535194418, + "learning_rate": 3.9075436669107265e-06, + "loss": 0.9597, + "step": 1654 + }, + { + "epoch": 0.12437997895686156, + "grad_norm": 1.357635612949973, + "learning_rate": 3.907397298897003e-06, + "loss": 0.9341, + "step": 1655 + }, + { + "epoch": 0.12445513302269653, + "grad_norm": 1.4948535689275388, + "learning_rate": 3.907250817862685e-06, + "loss": 0.9269, + "step": 1656 + }, + { + "epoch": 0.12453028708853149, + "grad_norm": 1.4495877107292998, + "learning_rate": 3.907104223816453e-06, + "loss": 0.9477, + "step": 1657 + }, + { + "epoch": 0.12460544115436645, + "grad_norm": 10.92562448213863, + "learning_rate": 3.906957516766993e-06, + "loss": 0.9633, + "step": 1658 + }, + { + "epoch": 0.12468059522020142, + "grad_norm": 1.3570184560995115, + "learning_rate": 3.906810696722997e-06, + "loss": 1.0133, + "step": 1659 + }, + { + "epoch": 0.12475574928603637, + "grad_norm": 1.5810652432669554, + "learning_rate": 3.906663763693167e-06, + "loss": 0.933, + "step": 1660 + }, + { + "epoch": 0.12483090335187133, + "grad_norm": 1.5152378964364774, + "learning_rate": 3.906516717686207e-06, + "loss": 1.045, + "step": 1661 + }, + { + "epoch": 0.1249060574177063, + "grad_norm": 2.1807983193109073, + "learning_rate": 3.906369558710831e-06, + "loss": 0.9389, + "step": 1662 + }, + { + "epoch": 0.12498121148354126, + "grad_norm": 1.7094840926763266, + "learning_rate": 3.906222286775759e-06, + "loss": 0.9297, + "step": 1663 + }, + { + "epoch": 0.1250563655493762, + "grad_norm": 1.3612687085620148, + "learning_rate": 3.906074901889717e-06, + "loss": 0.9633, + "step": 1664 + }, + { + "epoch": 0.12513151961521118, + "grad_norm": 1.5429665867271227, + "learning_rate": 3.905927404061439e-06, + "loss": 1.0332, + "step": 1665 + }, + { + "epoch": 0.12520667368104615, + "grad_norm": 2.0426705545773536, + "learning_rate": 3.905779793299662e-06, + "loss": 0.9664, + "step": 1666 + }, + { + "epoch": 0.1252818277468811, + "grad_norm": 0.8609382975552113, + "learning_rate": 3.905632069613136e-06, + "loss": 0.8403, + "step": 1667 + }, + { + "epoch": 0.12535698181271607, + "grad_norm": 1.471688192706221, + "learning_rate": 3.9054842330106125e-06, + "loss": 1.0171, + "step": 1668 + }, + { + "epoch": 0.12543213587855104, + "grad_norm": 1.902486368899945, + "learning_rate": 3.9053362835008516e-06, + "loss": 1.0216, + "step": 1669 + }, + { + "epoch": 0.12550728994438598, + "grad_norm": 1.9145885134001825, + "learning_rate": 3.9051882210926195e-06, + "loss": 0.8727, + "step": 1670 + }, + { + "epoch": 0.12558244401022095, + "grad_norm": 1.5089903663241788, + "learning_rate": 3.90504004579469e-06, + "loss": 0.9761, + "step": 1671 + }, + { + "epoch": 0.12565759807605592, + "grad_norm": 1.5254720551348406, + "learning_rate": 3.904891757615843e-06, + "loss": 0.989, + "step": 1672 + }, + { + "epoch": 0.12573275214189086, + "grad_norm": 1.613436490549366, + "learning_rate": 3.904743356564865e-06, + "loss": 0.9771, + "step": 1673 + }, + { + "epoch": 0.12580790620772583, + "grad_norm": 1.7110703647531846, + "learning_rate": 3.90459484265055e-06, + "loss": 1.0137, + "step": 1674 + }, + { + "epoch": 0.1258830602735608, + "grad_norm": 1.7750584256939475, + "learning_rate": 3.904446215881697e-06, + "loss": 0.9877, + "step": 1675 + }, + { + "epoch": 0.12595821433939577, + "grad_norm": 1.2971426905844374, + "learning_rate": 3.9042974762671125e-06, + "loss": 0.9993, + "step": 1676 + }, + { + "epoch": 0.12603336840523072, + "grad_norm": 1.5844439158602976, + "learning_rate": 3.904148623815611e-06, + "loss": 1.0587, + "step": 1677 + }, + { + "epoch": 0.1261085224710657, + "grad_norm": 1.6252665667872626, + "learning_rate": 3.903999658536012e-06, + "loss": 1.0031, + "step": 1678 + }, + { + "epoch": 0.12618367653690066, + "grad_norm": 1.2702846697659669, + "learning_rate": 3.903850580437142e-06, + "loss": 1.0925, + "step": 1679 + }, + { + "epoch": 0.1262588306027356, + "grad_norm": 2.4324275325319125, + "learning_rate": 3.903701389527836e-06, + "loss": 1.0235, + "step": 1680 + }, + { + "epoch": 0.12633398466857057, + "grad_norm": 1.6673295451236496, + "learning_rate": 3.903552085816932e-06, + "loss": 0.9788, + "step": 1681 + }, + { + "epoch": 0.12640913873440554, + "grad_norm": 2.0319808818788845, + "learning_rate": 3.903402669313278e-06, + "loss": 1.0914, + "step": 1682 + }, + { + "epoch": 0.12648429280024048, + "grad_norm": 1.5346925146680839, + "learning_rate": 3.903253140025726e-06, + "loss": 0.9355, + "step": 1683 + }, + { + "epoch": 0.12655944686607545, + "grad_norm": 1.3846387280428223, + "learning_rate": 3.9031034979631385e-06, + "loss": 0.9587, + "step": 1684 + }, + { + "epoch": 0.12663460093191042, + "grad_norm": 1.7201393061047747, + "learning_rate": 3.902953743134381e-06, + "loss": 0.9954, + "step": 1685 + }, + { + "epoch": 0.12670975499774537, + "grad_norm": 1.73875938440419, + "learning_rate": 3.9028038755483275e-06, + "loss": 0.9918, + "step": 1686 + }, + { + "epoch": 0.12678490906358034, + "grad_norm": 1.2338430489774166, + "learning_rate": 3.902653895213858e-06, + "loss": 1.0985, + "step": 1687 + }, + { + "epoch": 0.1268600631294153, + "grad_norm": 1.3968064156919135, + "learning_rate": 3.90250380213986e-06, + "loss": 0.9693, + "step": 1688 + }, + { + "epoch": 0.12693521719525025, + "grad_norm": 1.5475517873156388, + "learning_rate": 3.902353596335225e-06, + "loss": 0.9798, + "step": 1689 + }, + { + "epoch": 0.12701037126108522, + "grad_norm": 1.8110325232944597, + "learning_rate": 3.902203277808856e-06, + "loss": 1.0229, + "step": 1690 + }, + { + "epoch": 0.1270855253269202, + "grad_norm": 1.627830680852558, + "learning_rate": 3.902052846569659e-06, + "loss": 1.0456, + "step": 1691 + }, + { + "epoch": 0.12716067939275516, + "grad_norm": 1.61559715025637, + "learning_rate": 3.901902302626547e-06, + "loss": 0.9382, + "step": 1692 + }, + { + "epoch": 0.1272358334585901, + "grad_norm": 1.6254915080979535, + "learning_rate": 3.901751645988441e-06, + "loss": 0.9688, + "step": 1693 + }, + { + "epoch": 0.12731098752442507, + "grad_norm": 1.615582171523611, + "learning_rate": 3.901600876664267e-06, + "loss": 1.0153, + "step": 1694 + }, + { + "epoch": 0.12738614159026004, + "grad_norm": 2.1249601696993845, + "learning_rate": 3.9014499946629595e-06, + "loss": 1.0369, + "step": 1695 + }, + { + "epoch": 0.127461295656095, + "grad_norm": 1.9916468466353785, + "learning_rate": 3.901298999993459e-06, + "loss": 0.9591, + "step": 1696 + }, + { + "epoch": 0.12753644972192996, + "grad_norm": 1.8834912987799959, + "learning_rate": 3.901147892664713e-06, + "loss": 1.021, + "step": 1697 + }, + { + "epoch": 0.12761160378776493, + "grad_norm": 1.5440402788926906, + "learning_rate": 3.9009966726856725e-06, + "loss": 0.9853, + "step": 1698 + }, + { + "epoch": 0.12768675785359987, + "grad_norm": 1.4185640120748022, + "learning_rate": 3.900845340065301e-06, + "loss": 0.9687, + "step": 1699 + }, + { + "epoch": 0.12776191191943484, + "grad_norm": 1.4793201510177414, + "learning_rate": 3.900693894812564e-06, + "loss": 1.0497, + "step": 1700 + }, + { + "epoch": 0.1278370659852698, + "grad_norm": 1.563584294798041, + "learning_rate": 3.900542336936436e-06, + "loss": 0.9649, + "step": 1701 + }, + { + "epoch": 0.12791222005110475, + "grad_norm": 1.877962696342763, + "learning_rate": 3.900390666445896e-06, + "loss": 1.0023, + "step": 1702 + }, + { + "epoch": 0.12798737411693972, + "grad_norm": 1.910707608709141, + "learning_rate": 3.900238883349932e-06, + "loss": 1.0651, + "step": 1703 + }, + { + "epoch": 0.1280625281827747, + "grad_norm": 1.5209794779890982, + "learning_rate": 3.900086987657539e-06, + "loss": 1.0699, + "step": 1704 + }, + { + "epoch": 0.12813768224860966, + "grad_norm": 1.4888796266738993, + "learning_rate": 3.899934979377714e-06, + "loss": 1.0116, + "step": 1705 + }, + { + "epoch": 0.1282128363144446, + "grad_norm": 1.424098027187749, + "learning_rate": 3.899782858519467e-06, + "loss": 1.0583, + "step": 1706 + }, + { + "epoch": 0.12828799038027958, + "grad_norm": 1.500623891147647, + "learning_rate": 3.899630625091811e-06, + "loss": 1.0436, + "step": 1707 + }, + { + "epoch": 0.12836314444611455, + "grad_norm": 1.6221527684961117, + "learning_rate": 3.899478279103767e-06, + "loss": 0.928, + "step": 1708 + }, + { + "epoch": 0.1284382985119495, + "grad_norm": 1.6945509364671627, + "learning_rate": 3.89932582056436e-06, + "loss": 1.0504, + "step": 1709 + }, + { + "epoch": 0.12851345257778446, + "grad_norm": 1.520581037244124, + "learning_rate": 3.899173249482626e-06, + "loss": 1.0161, + "step": 1710 + }, + { + "epoch": 0.12858860664361943, + "grad_norm": 1.5851651269956515, + "learning_rate": 3.899020565867604e-06, + "loss": 1.0488, + "step": 1711 + }, + { + "epoch": 0.12866376070945437, + "grad_norm": 1.5254192276347647, + "learning_rate": 3.898867769728342e-06, + "loss": 0.9667, + "step": 1712 + }, + { + "epoch": 0.12873891477528934, + "grad_norm": 1.963730726238539, + "learning_rate": 3.8987148610738935e-06, + "loss": 0.9691, + "step": 1713 + }, + { + "epoch": 0.1288140688411243, + "grad_norm": 1.4966381068715522, + "learning_rate": 3.898561839913319e-06, + "loss": 0.9989, + "step": 1714 + }, + { + "epoch": 0.12888922290695926, + "grad_norm": 2.1915672637482984, + "learning_rate": 3.898408706255685e-06, + "loss": 1.0791, + "step": 1715 + }, + { + "epoch": 0.12896437697279423, + "grad_norm": 2.0467257845704054, + "learning_rate": 3.898255460110066e-06, + "loss": 0.9425, + "step": 1716 + }, + { + "epoch": 0.1290395310386292, + "grad_norm": 2.011849467650863, + "learning_rate": 3.898102101485542e-06, + "loss": 0.9813, + "step": 1717 + }, + { + "epoch": 0.12911468510446414, + "grad_norm": 1.7318631102182385, + "learning_rate": 3.8979486303912e-06, + "loss": 0.9214, + "step": 1718 + }, + { + "epoch": 0.1291898391702991, + "grad_norm": 1.3331613081983227, + "learning_rate": 3.8977950468361335e-06, + "loss": 1.0481, + "step": 1719 + }, + { + "epoch": 0.12926499323613408, + "grad_norm": 1.516558015865077, + "learning_rate": 3.897641350829444e-06, + "loss": 0.8916, + "step": 1720 + }, + { + "epoch": 0.12934014730196905, + "grad_norm": 1.40906078447655, + "learning_rate": 3.8974875423802385e-06, + "loss": 1.1501, + "step": 1721 + }, + { + "epoch": 0.129415301367804, + "grad_norm": 1.5666629267032433, + "learning_rate": 3.897333621497629e-06, + "loss": 1.0761, + "step": 1722 + }, + { + "epoch": 0.12949045543363896, + "grad_norm": 2.253593013160012, + "learning_rate": 3.897179588190737e-06, + "loss": 1.0024, + "step": 1723 + }, + { + "epoch": 0.12956560949947393, + "grad_norm": 1.5501355573382178, + "learning_rate": 3.89702544246869e-06, + "loss": 0.9307, + "step": 1724 + }, + { + "epoch": 0.12964076356530888, + "grad_norm": 1.3848976750222315, + "learning_rate": 3.896871184340622e-06, + "loss": 1.0132, + "step": 1725 + }, + { + "epoch": 0.12971591763114385, + "grad_norm": 1.7412660496965435, + "learning_rate": 3.896716813815672e-06, + "loss": 1.0513, + "step": 1726 + }, + { + "epoch": 0.12979107169697882, + "grad_norm": 1.5628789060840247, + "learning_rate": 3.8965623309029876e-06, + "loss": 1.0524, + "step": 1727 + }, + { + "epoch": 0.12986622576281376, + "grad_norm": 1.5326715215932694, + "learning_rate": 3.896407735611722e-06, + "loss": 1.0055, + "step": 1728 + }, + { + "epoch": 0.12994137982864873, + "grad_norm": 1.4380628064569856, + "learning_rate": 3.896253027951038e-06, + "loss": 0.9858, + "step": 1729 + }, + { + "epoch": 0.1300165338944837, + "grad_norm": 2.0978678733305958, + "learning_rate": 3.8960982079301e-06, + "loss": 0.9519, + "step": 1730 + }, + { + "epoch": 0.13009168796031864, + "grad_norm": 1.6827126164311184, + "learning_rate": 3.895943275558083e-06, + "loss": 1.0337, + "step": 1731 + }, + { + "epoch": 0.1301668420261536, + "grad_norm": 1.851197283172592, + "learning_rate": 3.895788230844166e-06, + "loss": 0.9262, + "step": 1732 + }, + { + "epoch": 0.13024199609198858, + "grad_norm": 4.271468746399163, + "learning_rate": 3.895633073797537e-06, + "loss": 0.9406, + "step": 1733 + }, + { + "epoch": 0.13031715015782352, + "grad_norm": 1.8005642377458426, + "learning_rate": 3.89547780442739e-06, + "loss": 1.0567, + "step": 1734 + }, + { + "epoch": 0.1303923042236585, + "grad_norm": 1.6714810771869488, + "learning_rate": 3.895322422742924e-06, + "loss": 0.9339, + "step": 1735 + }, + { + "epoch": 0.13046745828949347, + "grad_norm": 1.9877670612488192, + "learning_rate": 3.895166928753348e-06, + "loss": 1.0308, + "step": 1736 + }, + { + "epoch": 0.13054261235532844, + "grad_norm": 1.8776506126327435, + "learning_rate": 3.895011322467874e-06, + "loss": 1.0168, + "step": 1737 + }, + { + "epoch": 0.13061776642116338, + "grad_norm": 1.4637254031726525, + "learning_rate": 3.894855603895723e-06, + "loss": 0.9558, + "step": 1738 + }, + { + "epoch": 0.13069292048699835, + "grad_norm": 1.6801873461395433, + "learning_rate": 3.89469977304612e-06, + "loss": 0.9933, + "step": 1739 + }, + { + "epoch": 0.13076807455283332, + "grad_norm": 1.9517975225105124, + "learning_rate": 3.894543829928302e-06, + "loss": 0.9475, + "step": 1740 + }, + { + "epoch": 0.13084322861866826, + "grad_norm": 1.5698304456525023, + "learning_rate": 3.894387774551506e-06, + "loss": 0.9905, + "step": 1741 + }, + { + "epoch": 0.13091838268450323, + "grad_norm": 1.6386678011346858, + "learning_rate": 3.894231606924981e-06, + "loss": 0.9852, + "step": 1742 + }, + { + "epoch": 0.1309935367503382, + "grad_norm": 1.9422767575684785, + "learning_rate": 3.89407532705798e-06, + "loss": 0.9124, + "step": 1743 + }, + { + "epoch": 0.13106869081617314, + "grad_norm": 0.8420560574978432, + "learning_rate": 3.893918934959762e-06, + "loss": 0.8455, + "step": 1744 + }, + { + "epoch": 0.13114384488200811, + "grad_norm": 1.6852521406611978, + "learning_rate": 3.893762430639596e-06, + "loss": 0.9323, + "step": 1745 + }, + { + "epoch": 0.13121899894784309, + "grad_norm": 1.7024563517976021, + "learning_rate": 3.893605814106753e-06, + "loss": 1.0081, + "step": 1746 + }, + { + "epoch": 0.13129415301367803, + "grad_norm": 1.75865539686763, + "learning_rate": 3.893449085370515e-06, + "loss": 1.026, + "step": 1747 + }, + { + "epoch": 0.131369307079513, + "grad_norm": 1.6810599464289038, + "learning_rate": 3.893292244440168e-06, + "loss": 1.0445, + "step": 1748 + }, + { + "epoch": 0.13144446114534797, + "grad_norm": 1.4962505527326413, + "learning_rate": 3.893135291325006e-06, + "loss": 1.0514, + "step": 1749 + }, + { + "epoch": 0.13151961521118294, + "grad_norm": 1.6758654024206836, + "learning_rate": 3.892978226034329e-06, + "loss": 0.9994, + "step": 1750 + }, + { + "epoch": 0.13159476927701788, + "grad_norm": 1.571300897737435, + "learning_rate": 3.892821048577443e-06, + "loss": 0.957, + "step": 1751 + }, + { + "epoch": 0.13166992334285285, + "grad_norm": 1.929154666124538, + "learning_rate": 3.892663758963661e-06, + "loss": 1.0415, + "step": 1752 + }, + { + "epoch": 0.13174507740868782, + "grad_norm": 1.5757334500024245, + "learning_rate": 3.892506357202305e-06, + "loss": 0.8873, + "step": 1753 + }, + { + "epoch": 0.13182023147452276, + "grad_norm": 1.6945660613491127, + "learning_rate": 3.8923488433027e-06, + "loss": 0.9956, + "step": 1754 + }, + { + "epoch": 0.13189538554035773, + "grad_norm": 1.3627092095839628, + "learning_rate": 3.89219121727418e-06, + "loss": 1.0301, + "step": 1755 + }, + { + "epoch": 0.1319705396061927, + "grad_norm": 1.589257659480798, + "learning_rate": 3.892033479126084e-06, + "loss": 0.9778, + "step": 1756 + }, + { + "epoch": 0.13204569367202765, + "grad_norm": 1.6030922631681983, + "learning_rate": 3.89187562886776e-06, + "loss": 1.0333, + "step": 1757 + }, + { + "epoch": 0.13212084773786262, + "grad_norm": 1.4496401767855631, + "learning_rate": 3.89171766650856e-06, + "loss": 1.0334, + "step": 1758 + }, + { + "epoch": 0.1321960018036976, + "grad_norm": 2.136917913942638, + "learning_rate": 3.891559592057845e-06, + "loss": 1.0912, + "step": 1759 + }, + { + "epoch": 0.13227115586953253, + "grad_norm": 2.0109984061881208, + "learning_rate": 3.8914014055249805e-06, + "loss": 0.9405, + "step": 1760 + }, + { + "epoch": 0.1323463099353675, + "grad_norm": 1.4843012741250965, + "learning_rate": 3.89124310691934e-06, + "loss": 0.8954, + "step": 1761 + }, + { + "epoch": 0.13242146400120247, + "grad_norm": 2.003294258286691, + "learning_rate": 3.891084696250304e-06, + "loss": 1.0559, + "step": 1762 + }, + { + "epoch": 0.1324966180670374, + "grad_norm": 1.600835793470286, + "learning_rate": 3.890926173527258e-06, + "loss": 0.9923, + "step": 1763 + }, + { + "epoch": 0.13257177213287238, + "grad_norm": 1.690100961299104, + "learning_rate": 3.8907675387595944e-06, + "loss": 0.9334, + "step": 1764 + }, + { + "epoch": 0.13264692619870735, + "grad_norm": 1.6420131078979818, + "learning_rate": 3.890608791956714e-06, + "loss": 0.9359, + "step": 1765 + }, + { + "epoch": 0.13272208026454232, + "grad_norm": 1.8427295770565293, + "learning_rate": 3.890449933128025e-06, + "loss": 1.0107, + "step": 1766 + }, + { + "epoch": 0.13279723433037727, + "grad_norm": 1.66639417169113, + "learning_rate": 3.890290962282937e-06, + "loss": 0.9674, + "step": 1767 + }, + { + "epoch": 0.13287238839621224, + "grad_norm": 1.4240786066540314, + "learning_rate": 3.890131879430871e-06, + "loss": 0.9421, + "step": 1768 + }, + { + "epoch": 0.1329475424620472, + "grad_norm": 1.9816678421855463, + "learning_rate": 3.889972684581253e-06, + "loss": 0.9616, + "step": 1769 + }, + { + "epoch": 0.13302269652788215, + "grad_norm": 1.6101664106482467, + "learning_rate": 3.889813377743517e-06, + "loss": 1.0211, + "step": 1770 + }, + { + "epoch": 0.13309785059371712, + "grad_norm": 1.8566723583871565, + "learning_rate": 3.8896539589271016e-06, + "loss": 1.0419, + "step": 1771 + }, + { + "epoch": 0.1331730046595521, + "grad_norm": 1.6009481600887343, + "learning_rate": 3.889494428141453e-06, + "loss": 1.0, + "step": 1772 + }, + { + "epoch": 0.13324815872538703, + "grad_norm": 1.3315134244522133, + "learning_rate": 3.889334785396024e-06, + "loss": 0.9422, + "step": 1773 + }, + { + "epoch": 0.133323312791222, + "grad_norm": 1.600489261446974, + "learning_rate": 3.8891750307002746e-06, + "loss": 1.0333, + "step": 1774 + }, + { + "epoch": 0.13339846685705697, + "grad_norm": 1.376472161886918, + "learning_rate": 3.889015164063671e-06, + "loss": 1.0031, + "step": 1775 + }, + { + "epoch": 0.13347362092289192, + "grad_norm": 1.7143167368034626, + "learning_rate": 3.888855185495685e-06, + "loss": 1.0079, + "step": 1776 + }, + { + "epoch": 0.1335487749887269, + "grad_norm": 1.9877072088322825, + "learning_rate": 3.8886950950057965e-06, + "loss": 0.883, + "step": 1777 + }, + { + "epoch": 0.13362392905456186, + "grad_norm": 1.6113864217674008, + "learning_rate": 3.888534892603491e-06, + "loss": 0.8582, + "step": 1778 + }, + { + "epoch": 0.1336990831203968, + "grad_norm": 1.6089302855918441, + "learning_rate": 3.888374578298261e-06, + "loss": 1.0348, + "step": 1779 + }, + { + "epoch": 0.13377423718623177, + "grad_norm": 1.5837190643396346, + "learning_rate": 3.888214152099607e-06, + "loss": 1.0295, + "step": 1780 + }, + { + "epoch": 0.13384939125206674, + "grad_norm": 1.3430565220065465, + "learning_rate": 3.888053614017034e-06, + "loss": 0.977, + "step": 1781 + }, + { + "epoch": 0.1339245453179017, + "grad_norm": 1.3908493203961776, + "learning_rate": 3.887892964060054e-06, + "loss": 0.9916, + "step": 1782 + }, + { + "epoch": 0.13399969938373665, + "grad_norm": 1.7161588383723576, + "learning_rate": 3.887732202238186e-06, + "loss": 0.9335, + "step": 1783 + }, + { + "epoch": 0.13407485344957162, + "grad_norm": 4.45457568715723, + "learning_rate": 3.887571328560958e-06, + "loss": 0.9572, + "step": 1784 + }, + { + "epoch": 0.1341500075154066, + "grad_norm": 1.633417117960672, + "learning_rate": 3.8874103430379e-06, + "loss": 0.9872, + "step": 1785 + }, + { + "epoch": 0.13422516158124154, + "grad_norm": 1.6831068852891486, + "learning_rate": 3.887249245678552e-06, + "loss": 1.0987, + "step": 1786 + }, + { + "epoch": 0.1343003156470765, + "grad_norm": 1.397695163208556, + "learning_rate": 3.887088036492459e-06, + "loss": 0.9667, + "step": 1787 + }, + { + "epoch": 0.13437546971291148, + "grad_norm": 1.3834423635021171, + "learning_rate": 3.886926715489173e-06, + "loss": 0.9952, + "step": 1788 + }, + { + "epoch": 0.13445062377874642, + "grad_norm": 2.029940608616843, + "learning_rate": 3.8867652826782555e-06, + "loss": 1.0262, + "step": 1789 + }, + { + "epoch": 0.1345257778445814, + "grad_norm": 4.1819430245924565, + "learning_rate": 3.886603738069269e-06, + "loss": 0.9872, + "step": 1790 + }, + { + "epoch": 0.13460093191041636, + "grad_norm": 9.0450005310632, + "learning_rate": 3.886442081671787e-06, + "loss": 0.9748, + "step": 1791 + }, + { + "epoch": 0.1346760859762513, + "grad_norm": 2.1225073162981696, + "learning_rate": 3.886280313495388e-06, + "loss": 0.9258, + "step": 1792 + }, + { + "epoch": 0.13475124004208627, + "grad_norm": 1.632679692028067, + "learning_rate": 3.886118433549657e-06, + "loss": 1.0294, + "step": 1793 + }, + { + "epoch": 0.13482639410792124, + "grad_norm": 1.3093251020806627, + "learning_rate": 3.8859564418441865e-06, + "loss": 0.9648, + "step": 1794 + }, + { + "epoch": 0.1349015481737562, + "grad_norm": 1.4928864444899912, + "learning_rate": 3.885794338388575e-06, + "loss": 0.9754, + "step": 1795 + }, + { + "epoch": 0.13497670223959116, + "grad_norm": 1.3971452763029704, + "learning_rate": 3.8856321231924275e-06, + "loss": 1.0535, + "step": 1796 + }, + { + "epoch": 0.13505185630542613, + "grad_norm": 1.546640551746712, + "learning_rate": 3.885469796265357e-06, + "loss": 1.0415, + "step": 1797 + }, + { + "epoch": 0.1351270103712611, + "grad_norm": 1.6773893736563084, + "learning_rate": 3.885307357616981e-06, + "loss": 1.051, + "step": 1798 + }, + { + "epoch": 0.13520216443709604, + "grad_norm": 1.836750905458438, + "learning_rate": 3.8851448072569245e-06, + "loss": 0.9747, + "step": 1799 + }, + { + "epoch": 0.135277318502931, + "grad_norm": 6.224851136170283, + "learning_rate": 3.884982145194819e-06, + "loss": 0.9943, + "step": 1800 + }, + { + "epoch": 0.13535247256876598, + "grad_norm": 1.2689566382897484, + "learning_rate": 3.8848193714403035e-06, + "loss": 1.0395, + "step": 1801 + }, + { + "epoch": 0.13542762663460092, + "grad_norm": 1.2952193757048154, + "learning_rate": 3.884656486003023e-06, + "loss": 0.9912, + "step": 1802 + }, + { + "epoch": 0.1355027807004359, + "grad_norm": 1.7780112553638234, + "learning_rate": 3.8844934888926295e-06, + "loss": 0.9896, + "step": 1803 + }, + { + "epoch": 0.13557793476627086, + "grad_norm": 1.3075640370604393, + "learning_rate": 3.884330380118779e-06, + "loss": 0.9517, + "step": 1804 + }, + { + "epoch": 0.1356530888321058, + "grad_norm": 1.7488105001066305, + "learning_rate": 3.884167159691139e-06, + "loss": 0.9752, + "step": 1805 + }, + { + "epoch": 0.13572824289794078, + "grad_norm": 1.618820618963267, + "learning_rate": 3.88400382761938e-06, + "loss": 0.9002, + "step": 1806 + }, + { + "epoch": 0.13580339696377575, + "grad_norm": 1.7052323489011736, + "learning_rate": 3.883840383913179e-06, + "loss": 0.9933, + "step": 1807 + }, + { + "epoch": 0.1358785510296107, + "grad_norm": 1.758341662973257, + "learning_rate": 3.8836768285822225e-06, + "loss": 0.8884, + "step": 1808 + }, + { + "epoch": 0.13595370509544566, + "grad_norm": 1.694456314955581, + "learning_rate": 3.8835131616362005e-06, + "loss": 1.0136, + "step": 1809 + }, + { + "epoch": 0.13602885916128063, + "grad_norm": 1.4202615995277559, + "learning_rate": 3.883349383084811e-06, + "loss": 1.0817, + "step": 1810 + }, + { + "epoch": 0.1361040132271156, + "grad_norm": 1.3955181060048332, + "learning_rate": 3.883185492937759e-06, + "loss": 1.047, + "step": 1811 + }, + { + "epoch": 0.13617916729295054, + "grad_norm": 1.5029351604280226, + "learning_rate": 3.883021491204755e-06, + "loss": 1.0364, + "step": 1812 + }, + { + "epoch": 0.1362543213587855, + "grad_norm": 1.9944395612776733, + "learning_rate": 3.8828573778955175e-06, + "loss": 1.0845, + "step": 1813 + }, + { + "epoch": 0.13632947542462048, + "grad_norm": 0.9973782503787684, + "learning_rate": 3.88269315301977e-06, + "loss": 0.804, + "step": 1814 + }, + { + "epoch": 0.13640462949045543, + "grad_norm": 1.761372263599198, + "learning_rate": 3.882528816587244e-06, + "loss": 0.946, + "step": 1815 + }, + { + "epoch": 0.1364797835562904, + "grad_norm": 1.6791017442494376, + "learning_rate": 3.882364368607677e-06, + "loss": 0.9472, + "step": 1816 + }, + { + "epoch": 0.13655493762212537, + "grad_norm": 1.5794604099549283, + "learning_rate": 3.882199809090813e-06, + "loss": 0.947, + "step": 1817 + }, + { + "epoch": 0.1366300916879603, + "grad_norm": 1.8674334120720206, + "learning_rate": 3.8820351380464035e-06, + "loss": 1.1076, + "step": 1818 + }, + { + "epoch": 0.13670524575379528, + "grad_norm": 1.8346227605278027, + "learning_rate": 3.881870355484204e-06, + "loss": 1.0162, + "step": 1819 + }, + { + "epoch": 0.13678039981963025, + "grad_norm": 1.4284033506154907, + "learning_rate": 3.88170546141398e-06, + "loss": 0.992, + "step": 1820 + }, + { + "epoch": 0.1368555538854652, + "grad_norm": 1.679900186956897, + "learning_rate": 3.881540455845503e-06, + "loss": 0.9789, + "step": 1821 + }, + { + "epoch": 0.13693070795130016, + "grad_norm": 1.617573493926939, + "learning_rate": 3.881375338788549e-06, + "loss": 0.9298, + "step": 1822 + }, + { + "epoch": 0.13700586201713513, + "grad_norm": 1.6886613170003406, + "learning_rate": 3.881210110252901e-06, + "loss": 0.9688, + "step": 1823 + }, + { + "epoch": 0.13708101608297008, + "grad_norm": 1.455213706849248, + "learning_rate": 3.881044770248351e-06, + "loss": 1.0164, + "step": 1824 + }, + { + "epoch": 0.13715617014880505, + "grad_norm": 1.4935967147259779, + "learning_rate": 3.880879318784695e-06, + "loss": 1.0387, + "step": 1825 + }, + { + "epoch": 0.13723132421464002, + "grad_norm": 1.8119140203931763, + "learning_rate": 3.8807137558717375e-06, + "loss": 1.061, + "step": 1826 + }, + { + "epoch": 0.137306478280475, + "grad_norm": 1.7993695400058474, + "learning_rate": 3.880548081519287e-06, + "loss": 0.9183, + "step": 1827 + }, + { + "epoch": 0.13738163234630993, + "grad_norm": 1.6749624817475288, + "learning_rate": 3.880382295737163e-06, + "loss": 0.8823, + "step": 1828 + }, + { + "epoch": 0.1374567864121449, + "grad_norm": 1.3469579684376913, + "learning_rate": 3.880216398535187e-06, + "loss": 0.9545, + "step": 1829 + }, + { + "epoch": 0.13753194047797987, + "grad_norm": 1.5568191253669375, + "learning_rate": 3.8800503899231895e-06, + "loss": 0.9982, + "step": 1830 + }, + { + "epoch": 0.1376070945438148, + "grad_norm": 1.7174858299237201, + "learning_rate": 3.879884269911007e-06, + "loss": 0.8733, + "step": 1831 + }, + { + "epoch": 0.13768224860964978, + "grad_norm": 1.9624553140788887, + "learning_rate": 3.879718038508483e-06, + "loss": 0.8885, + "step": 1832 + }, + { + "epoch": 0.13775740267548475, + "grad_norm": 2.0144873668873866, + "learning_rate": 3.8795516957254675e-06, + "loss": 1.0268, + "step": 1833 + }, + { + "epoch": 0.1378325567413197, + "grad_norm": 1.7242625526080817, + "learning_rate": 3.8793852415718165e-06, + "loss": 0.9659, + "step": 1834 + }, + { + "epoch": 0.13790771080715467, + "grad_norm": 1.5576218560127921, + "learning_rate": 3.879218676057394e-06, + "loss": 1.0012, + "step": 1835 + }, + { + "epoch": 0.13798286487298964, + "grad_norm": 1.7375983450521664, + "learning_rate": 3.879051999192068e-06, + "loss": 1.0342, + "step": 1836 + }, + { + "epoch": 0.13805801893882458, + "grad_norm": 1.6666271522924079, + "learning_rate": 3.8788852109857166e-06, + "loss": 0.9985, + "step": 1837 + }, + { + "epoch": 0.13813317300465955, + "grad_norm": 1.5988139316162773, + "learning_rate": 3.878718311448221e-06, + "loss": 1.0093, + "step": 1838 + }, + { + "epoch": 0.13820832707049452, + "grad_norm": 1.582263467958909, + "learning_rate": 3.878551300589471e-06, + "loss": 0.927, + "step": 1839 + }, + { + "epoch": 0.1382834811363295, + "grad_norm": 1.9314704719792657, + "learning_rate": 3.8783841784193635e-06, + "loss": 1.0106, + "step": 1840 + }, + { + "epoch": 0.13835863520216443, + "grad_norm": 1.6295027338258437, + "learning_rate": 3.878216944947801e-06, + "loss": 0.9011, + "step": 1841 + }, + { + "epoch": 0.1384337892679994, + "grad_norm": 1.5541376262879976, + "learning_rate": 3.878049600184692e-06, + "loss": 0.9711, + "step": 1842 + }, + { + "epoch": 0.13850894333383437, + "grad_norm": 1.8798801491458406, + "learning_rate": 3.877882144139952e-06, + "loss": 0.8916, + "step": 1843 + }, + { + "epoch": 0.13858409739966931, + "grad_norm": 2.064051420310559, + "learning_rate": 3.8777145768235054e-06, + "loss": 0.9716, + "step": 1844 + }, + { + "epoch": 0.13865925146550429, + "grad_norm": 1.8668298099387965, + "learning_rate": 3.877546898245279e-06, + "loss": 1.0059, + "step": 1845 + }, + { + "epoch": 0.13873440553133926, + "grad_norm": 1.6543167635500153, + "learning_rate": 3.877379108415209e-06, + "loss": 0.927, + "step": 1846 + }, + { + "epoch": 0.1388095595971742, + "grad_norm": 1.4721997917113994, + "learning_rate": 3.8772112073432385e-06, + "loss": 0.9949, + "step": 1847 + }, + { + "epoch": 0.13888471366300917, + "grad_norm": 1.6576535018335747, + "learning_rate": 3.8770431950393154e-06, + "loss": 1.0042, + "step": 1848 + }, + { + "epoch": 0.13895986772884414, + "grad_norm": 1.4791579268649517, + "learning_rate": 3.876875071513395e-06, + "loss": 0.988, + "step": 1849 + }, + { + "epoch": 0.13903502179467908, + "grad_norm": 1.7263135773748026, + "learning_rate": 3.87670683677544e-06, + "loss": 0.9005, + "step": 1850 + }, + { + "epoch": 0.13911017586051405, + "grad_norm": 1.609574462974149, + "learning_rate": 3.876538490835419e-06, + "loss": 0.9683, + "step": 1851 + }, + { + "epoch": 0.13918532992634902, + "grad_norm": 1.4270581523966201, + "learning_rate": 3.876370033703307e-06, + "loss": 1.0609, + "step": 1852 + }, + { + "epoch": 0.13926048399218396, + "grad_norm": 0.897190854482515, + "learning_rate": 3.876201465389084e-06, + "loss": 0.8284, + "step": 1853 + }, + { + "epoch": 0.13933563805801893, + "grad_norm": 1.3443948396134813, + "learning_rate": 3.87603278590274e-06, + "loss": 1.0579, + "step": 1854 + }, + { + "epoch": 0.1394107921238539, + "grad_norm": 1.3547887938049226, + "learning_rate": 3.8758639952542695e-06, + "loss": 0.9982, + "step": 1855 + }, + { + "epoch": 0.13948594618968888, + "grad_norm": 1.518700693845022, + "learning_rate": 3.875695093453675e-06, + "loss": 0.9624, + "step": 1856 + }, + { + "epoch": 0.13956110025552382, + "grad_norm": 0.6928645342676972, + "learning_rate": 3.875526080510963e-06, + "loss": 0.7743, + "step": 1857 + }, + { + "epoch": 0.1396362543213588, + "grad_norm": 1.8813223105692782, + "learning_rate": 3.8753569564361495e-06, + "loss": 1.1269, + "step": 1858 + }, + { + "epoch": 0.13971140838719376, + "grad_norm": 1.4429346267473973, + "learning_rate": 3.875187721239254e-06, + "loss": 0.8984, + "step": 1859 + }, + { + "epoch": 0.1397865624530287, + "grad_norm": 1.460932767319791, + "learning_rate": 3.8750183749303066e-06, + "loss": 0.9652, + "step": 1860 + }, + { + "epoch": 0.13986171651886367, + "grad_norm": 1.59235972183388, + "learning_rate": 3.87484891751934e-06, + "loss": 1.0035, + "step": 1861 + }, + { + "epoch": 0.13993687058469864, + "grad_norm": 2.2025682543983067, + "learning_rate": 3.874679349016396e-06, + "loss": 0.889, + "step": 1862 + }, + { + "epoch": 0.14001202465053358, + "grad_norm": 1.4557341255371945, + "learning_rate": 3.874509669431521e-06, + "loss": 1.0655, + "step": 1863 + }, + { + "epoch": 0.14008717871636855, + "grad_norm": 1.873745180496624, + "learning_rate": 3.874339878774771e-06, + "loss": 1.0288, + "step": 1864 + }, + { + "epoch": 0.14016233278220352, + "grad_norm": 1.6522765089561553, + "learning_rate": 3.8741699770562065e-06, + "loss": 1.0102, + "step": 1865 + }, + { + "epoch": 0.14023748684803847, + "grad_norm": 1.8149506017304717, + "learning_rate": 3.873999964285893e-06, + "loss": 0.799, + "step": 1866 + }, + { + "epoch": 0.14031264091387344, + "grad_norm": 1.8589589270517006, + "learning_rate": 3.873829840473906e-06, + "loss": 0.9931, + "step": 1867 + }, + { + "epoch": 0.1403877949797084, + "grad_norm": 1.5094906654031381, + "learning_rate": 3.873659605630325e-06, + "loss": 1.029, + "step": 1868 + }, + { + "epoch": 0.14046294904554335, + "grad_norm": 1.462587292014215, + "learning_rate": 3.873489259765239e-06, + "loss": 1.0505, + "step": 1869 + }, + { + "epoch": 0.14053810311137832, + "grad_norm": 1.6501467552811306, + "learning_rate": 3.873318802888739e-06, + "loss": 1.0283, + "step": 1870 + }, + { + "epoch": 0.1406132571772133, + "grad_norm": 1.8865736753553903, + "learning_rate": 3.8731482350109276e-06, + "loss": 0.9367, + "step": 1871 + }, + { + "epoch": 0.14068841124304826, + "grad_norm": 1.631558242263309, + "learning_rate": 3.87297755614191e-06, + "loss": 1.0542, + "step": 1872 + }, + { + "epoch": 0.1407635653088832, + "grad_norm": 1.807178050344169, + "learning_rate": 3.8728067662918e-06, + "loss": 1.0548, + "step": 1873 + }, + { + "epoch": 0.14083871937471817, + "grad_norm": 1.6177003204022797, + "learning_rate": 3.872635865470718e-06, + "loss": 1.0678, + "step": 1874 + }, + { + "epoch": 0.14091387344055314, + "grad_norm": 1.4066329434647764, + "learning_rate": 3.8724648536887895e-06, + "loss": 1.0371, + "step": 1875 + }, + { + "epoch": 0.1409890275063881, + "grad_norm": 2.428526475233874, + "learning_rate": 3.872293730956149e-06, + "loss": 0.9422, + "step": 1876 + }, + { + "epoch": 0.14106418157222306, + "grad_norm": 1.4576866986945114, + "learning_rate": 3.872122497282935e-06, + "loss": 0.9616, + "step": 1877 + }, + { + "epoch": 0.14113933563805803, + "grad_norm": 1.5410826925905767, + "learning_rate": 3.871951152679294e-06, + "loss": 1.0039, + "step": 1878 + }, + { + "epoch": 0.14121448970389297, + "grad_norm": 1.6199591285353467, + "learning_rate": 3.871779697155379e-06, + "loss": 1.0433, + "step": 1879 + }, + { + "epoch": 0.14128964376972794, + "grad_norm": 1.3823902757088498, + "learning_rate": 3.87160813072135e-06, + "loss": 0.9473, + "step": 1880 + }, + { + "epoch": 0.1413647978355629, + "grad_norm": 1.6347346331394348, + "learning_rate": 3.871436453387372e-06, + "loss": 1.0413, + "step": 1881 + }, + { + "epoch": 0.14143995190139785, + "grad_norm": 1.9362825444874496, + "learning_rate": 3.8712646651636185e-06, + "loss": 1.0217, + "step": 1882 + }, + { + "epoch": 0.14151510596723282, + "grad_norm": 1.4831406932568005, + "learning_rate": 3.8710927660602676e-06, + "loss": 0.8931, + "step": 1883 + }, + { + "epoch": 0.1415902600330678, + "grad_norm": 1.6918757890376044, + "learning_rate": 3.870920756087505e-06, + "loss": 0.9994, + "step": 1884 + }, + { + "epoch": 0.14166541409890276, + "grad_norm": 1.60474314384226, + "learning_rate": 3.870748635255524e-06, + "loss": 0.9684, + "step": 1885 + }, + { + "epoch": 0.1417405681647377, + "grad_norm": 1.4619805187748645, + "learning_rate": 3.870576403574523e-06, + "loss": 1.0023, + "step": 1886 + }, + { + "epoch": 0.14181572223057268, + "grad_norm": 1.8951122533519131, + "learning_rate": 3.870404061054706e-06, + "loss": 0.9809, + "step": 1887 + }, + { + "epoch": 0.14189087629640765, + "grad_norm": 1.6233092094890045, + "learning_rate": 3.870231607706287e-06, + "loss": 0.9413, + "step": 1888 + }, + { + "epoch": 0.1419660303622426, + "grad_norm": 1.8725456068242612, + "learning_rate": 3.870059043539484e-06, + "loss": 0.9379, + "step": 1889 + }, + { + "epoch": 0.14204118442807756, + "grad_norm": 1.9880603000506734, + "learning_rate": 3.869886368564521e-06, + "loss": 1.181, + "step": 1890 + }, + { + "epoch": 0.14211633849391253, + "grad_norm": 1.537180335456048, + "learning_rate": 3.869713582791631e-06, + "loss": 1.0326, + "step": 1891 + }, + { + "epoch": 0.14219149255974747, + "grad_norm": 1.4862407804031532, + "learning_rate": 3.869540686231051e-06, + "loss": 0.993, + "step": 1892 + }, + { + "epoch": 0.14226664662558244, + "grad_norm": 1.6926943343146088, + "learning_rate": 3.8693676788930264e-06, + "loss": 0.9783, + "step": 1893 + }, + { + "epoch": 0.1423418006914174, + "grad_norm": 1.830795619693656, + "learning_rate": 3.869194560787808e-06, + "loss": 0.9819, + "step": 1894 + }, + { + "epoch": 0.14241695475725236, + "grad_norm": 1.541808413956971, + "learning_rate": 3.8690213319256555e-06, + "loss": 0.949, + "step": 1895 + }, + { + "epoch": 0.14249210882308733, + "grad_norm": 0.8672516601471834, + "learning_rate": 3.8688479923168316e-06, + "loss": 0.8917, + "step": 1896 + }, + { + "epoch": 0.1425672628889223, + "grad_norm": 1.5582043919112267, + "learning_rate": 3.868674541971608e-06, + "loss": 0.9168, + "step": 1897 + }, + { + "epoch": 0.14264241695475724, + "grad_norm": 1.5966652032504347, + "learning_rate": 3.868500980900262e-06, + "loss": 0.9495, + "step": 1898 + }, + { + "epoch": 0.1427175710205922, + "grad_norm": 1.516310310768172, + "learning_rate": 3.868327309113079e-06, + "loss": 1.0695, + "step": 1899 + }, + { + "epoch": 0.14279272508642718, + "grad_norm": 2.25495978579643, + "learning_rate": 3.8681535266203464e-06, + "loss": 0.9902, + "step": 1900 + }, + { + "epoch": 0.14286787915226215, + "grad_norm": 1.8800062104386361, + "learning_rate": 3.867979633432365e-06, + "loss": 0.9754, + "step": 1901 + }, + { + "epoch": 0.1429430332180971, + "grad_norm": 0.8326534756388574, + "learning_rate": 3.867805629559438e-06, + "loss": 0.8052, + "step": 1902 + }, + { + "epoch": 0.14301818728393206, + "grad_norm": 1.6527925788081312, + "learning_rate": 3.867631515011874e-06, + "loss": 1.0183, + "step": 1903 + }, + { + "epoch": 0.14309334134976703, + "grad_norm": 1.7510236062550601, + "learning_rate": 3.8674572897999915e-06, + "loss": 1.03, + "step": 1904 + }, + { + "epoch": 0.14316849541560198, + "grad_norm": 1.5413517071779357, + "learning_rate": 3.8672829539341136e-06, + "loss": 1.0127, + "step": 1905 + }, + { + "epoch": 0.14324364948143695, + "grad_norm": 1.4506367534932125, + "learning_rate": 3.8671085074245704e-06, + "loss": 1.0413, + "step": 1906 + }, + { + "epoch": 0.14331880354727192, + "grad_norm": 1.353151875288484, + "learning_rate": 3.8669339502816985e-06, + "loss": 0.8727, + "step": 1907 + }, + { + "epoch": 0.14339395761310686, + "grad_norm": 2.1493529432828344, + "learning_rate": 3.866759282515841e-06, + "loss": 0.9562, + "step": 1908 + }, + { + "epoch": 0.14346911167894183, + "grad_norm": 1.4565561455729397, + "learning_rate": 3.866584504137347e-06, + "loss": 0.9414, + "step": 1909 + }, + { + "epoch": 0.1435442657447768, + "grad_norm": 1.5606162350629607, + "learning_rate": 3.8664096151565755e-06, + "loss": 0.9011, + "step": 1910 + }, + { + "epoch": 0.14361941981061174, + "grad_norm": 2.0813434889132436, + "learning_rate": 3.8662346155838855e-06, + "loss": 1.0766, + "step": 1911 + }, + { + "epoch": 0.1436945738764467, + "grad_norm": 1.6208435600893007, + "learning_rate": 3.866059505429649e-06, + "loss": 1.0051, + "step": 1912 + }, + { + "epoch": 0.14376972794228168, + "grad_norm": 1.5951031942677436, + "learning_rate": 3.865884284704241e-06, + "loss": 1.0609, + "step": 1913 + }, + { + "epoch": 0.14384488200811663, + "grad_norm": 1.4290340577078593, + "learning_rate": 3.8657089534180445e-06, + "loss": 0.9777, + "step": 1914 + }, + { + "epoch": 0.1439200360739516, + "grad_norm": 1.628469138649753, + "learning_rate": 3.865533511581448e-06, + "loss": 0.9927, + "step": 1915 + }, + { + "epoch": 0.14399519013978657, + "grad_norm": 1.6363688165409285, + "learning_rate": 3.865357959204847e-06, + "loss": 0.98, + "step": 1916 + }, + { + "epoch": 0.14407034420562154, + "grad_norm": 1.5322860793740767, + "learning_rate": 3.865182296298644e-06, + "loss": 0.9996, + "step": 1917 + }, + { + "epoch": 0.14414549827145648, + "grad_norm": 1.5388001349034361, + "learning_rate": 3.865006522873249e-06, + "loss": 1.0668, + "step": 1918 + }, + { + "epoch": 0.14422065233729145, + "grad_norm": 1.3975566455368207, + "learning_rate": 3.864830638939074e-06, + "loss": 1.0006, + "step": 1919 + }, + { + "epoch": 0.14429580640312642, + "grad_norm": 0.8933978834305203, + "learning_rate": 3.864654644506544e-06, + "loss": 0.8733, + "step": 1920 + }, + { + "epoch": 0.14437096046896136, + "grad_norm": 1.5600296328248857, + "learning_rate": 3.864478539586085e-06, + "loss": 1.0655, + "step": 1921 + }, + { + "epoch": 0.14444611453479633, + "grad_norm": 6.78275517929592, + "learning_rate": 3.8643023241881344e-06, + "loss": 1.1156, + "step": 1922 + }, + { + "epoch": 0.1445212686006313, + "grad_norm": 1.8611811993182064, + "learning_rate": 3.864125998323131e-06, + "loss": 1.004, + "step": 1923 + }, + { + "epoch": 0.14459642266646625, + "grad_norm": 1.5169101261178903, + "learning_rate": 3.863949562001524e-06, + "loss": 0.9362, + "step": 1924 + }, + { + "epoch": 0.14467157673230122, + "grad_norm": 1.4982500199665314, + "learning_rate": 3.863773015233769e-06, + "loss": 1.0337, + "step": 1925 + }, + { + "epoch": 0.1447467307981362, + "grad_norm": 1.4847297646061626, + "learning_rate": 3.863596358030326e-06, + "loss": 0.8921, + "step": 1926 + }, + { + "epoch": 0.14482188486397113, + "grad_norm": 1.6363384378544903, + "learning_rate": 3.863419590401661e-06, + "loss": 1.0044, + "step": 1927 + }, + { + "epoch": 0.1448970389298061, + "grad_norm": 1.6065512022234463, + "learning_rate": 3.8632427123582505e-06, + "loss": 0.9568, + "step": 1928 + }, + { + "epoch": 0.14497219299564107, + "grad_norm": 1.687709865942162, + "learning_rate": 3.8630657239105754e-06, + "loss": 1.0023, + "step": 1929 + }, + { + "epoch": 0.14504734706147604, + "grad_norm": 3.3194547375493273, + "learning_rate": 3.862888625069121e-06, + "loss": 0.9157, + "step": 1930 + }, + { + "epoch": 0.14512250112731098, + "grad_norm": 1.7232966540999917, + "learning_rate": 3.8627114158443825e-06, + "loss": 1.0859, + "step": 1931 + }, + { + "epoch": 0.14519765519314595, + "grad_norm": 2.09107070984269, + "learning_rate": 3.862534096246859e-06, + "loss": 1.0289, + "step": 1932 + }, + { + "epoch": 0.14527280925898092, + "grad_norm": 1.8988365099387678, + "learning_rate": 3.862356666287059e-06, + "loss": 0.9368, + "step": 1933 + }, + { + "epoch": 0.14534796332481587, + "grad_norm": 0.9304255913526919, + "learning_rate": 3.862179125975495e-06, + "loss": 0.804, + "step": 1934 + }, + { + "epoch": 0.14542311739065084, + "grad_norm": 1.3589353617804099, + "learning_rate": 3.862001475322687e-06, + "loss": 1.0613, + "step": 1935 + }, + { + "epoch": 0.1454982714564858, + "grad_norm": 1.6885882506709429, + "learning_rate": 3.861823714339162e-06, + "loss": 0.9132, + "step": 1936 + }, + { + "epoch": 0.14557342552232075, + "grad_norm": 1.8160235914554912, + "learning_rate": 3.861645843035452e-06, + "loss": 0.9432, + "step": 1937 + }, + { + "epoch": 0.14564857958815572, + "grad_norm": 2.5672517774246812, + "learning_rate": 3.861467861422096e-06, + "loss": 1.0334, + "step": 1938 + }, + { + "epoch": 0.1457237336539907, + "grad_norm": 2.024687038774257, + "learning_rate": 3.861289769509643e-06, + "loss": 0.9281, + "step": 1939 + }, + { + "epoch": 0.14579888771982563, + "grad_norm": 1.5013347885747015, + "learning_rate": 3.861111567308643e-06, + "loss": 0.9788, + "step": 1940 + }, + { + "epoch": 0.1458740417856606, + "grad_norm": 1.8993420741700333, + "learning_rate": 3.860933254829656e-06, + "loss": 1.0842, + "step": 1941 + }, + { + "epoch": 0.14594919585149557, + "grad_norm": 1.4707117169092487, + "learning_rate": 3.860754832083247e-06, + "loss": 0.8932, + "step": 1942 + }, + { + "epoch": 0.14602434991733051, + "grad_norm": 2.3962252973498086, + "learning_rate": 3.86057629907999e-06, + "loss": 0.9009, + "step": 1943 + }, + { + "epoch": 0.14609950398316549, + "grad_norm": 1.6883906380636708, + "learning_rate": 3.8603976558304624e-06, + "loss": 0.8884, + "step": 1944 + }, + { + "epoch": 0.14617465804900046, + "grad_norm": 1.8030667644332488, + "learning_rate": 3.86021890234525e-06, + "loss": 0.9508, + "step": 1945 + }, + { + "epoch": 0.14624981211483543, + "grad_norm": 2.0041746439414827, + "learning_rate": 3.860040038634944e-06, + "loss": 0.8844, + "step": 1946 + }, + { + "epoch": 0.14632496618067037, + "grad_norm": 3.9418335334238317, + "learning_rate": 3.8598610647101426e-06, + "loss": 1.037, + "step": 1947 + }, + { + "epoch": 0.14640012024650534, + "grad_norm": 1.7810682906777435, + "learning_rate": 3.859681980581452e-06, + "loss": 0.9132, + "step": 1948 + }, + { + "epoch": 0.1464752743123403, + "grad_norm": 1.4801310050822565, + "learning_rate": 3.859502786259482e-06, + "loss": 0.9807, + "step": 1949 + }, + { + "epoch": 0.14655042837817525, + "grad_norm": 1.3407258447096118, + "learning_rate": 3.8593234817548525e-06, + "loss": 0.9872, + "step": 1950 + }, + { + "epoch": 0.14662558244401022, + "grad_norm": 1.5930354442544588, + "learning_rate": 3.859144067078186e-06, + "loss": 0.9677, + "step": 1951 + }, + { + "epoch": 0.1467007365098452, + "grad_norm": 0.8907265187010256, + "learning_rate": 3.858964542240115e-06, + "loss": 0.8735, + "step": 1952 + }, + { + "epoch": 0.14677589057568013, + "grad_norm": 1.7684032895664836, + "learning_rate": 3.8587849072512755e-06, + "loss": 1.0058, + "step": 1953 + }, + { + "epoch": 0.1468510446415151, + "grad_norm": 1.7058081533377816, + "learning_rate": 3.858605162122314e-06, + "loss": 0.9978, + "step": 1954 + }, + { + "epoch": 0.14692619870735008, + "grad_norm": 1.922843812972417, + "learning_rate": 3.858425306863878e-06, + "loss": 0.8587, + "step": 1955 + }, + { + "epoch": 0.14700135277318502, + "grad_norm": 2.163584411913898, + "learning_rate": 3.858245341486627e-06, + "loss": 1.009, + "step": 1956 + }, + { + "epoch": 0.14707650683902, + "grad_norm": 1.6788328613600718, + "learning_rate": 3.858065266001224e-06, + "loss": 0.9791, + "step": 1957 + }, + { + "epoch": 0.14715166090485496, + "grad_norm": 1.648020904112149, + "learning_rate": 3.857885080418339e-06, + "loss": 0.9718, + "step": 1958 + }, + { + "epoch": 0.1472268149706899, + "grad_norm": 1.7718654840876527, + "learning_rate": 3.857704784748648e-06, + "loss": 1.0165, + "step": 1959 + }, + { + "epoch": 0.14730196903652487, + "grad_norm": 1.6588328393576233, + "learning_rate": 3.857524379002835e-06, + "loss": 0.914, + "step": 1960 + }, + { + "epoch": 0.14737712310235984, + "grad_norm": 1.6843240309100485, + "learning_rate": 3.85734386319159e-06, + "loss": 1.0587, + "step": 1961 + }, + { + "epoch": 0.1474522771681948, + "grad_norm": 1.7053040030395146, + "learning_rate": 3.857163237325608e-06, + "loss": 0.9667, + "step": 1962 + }, + { + "epoch": 0.14752743123402975, + "grad_norm": 1.654847468927653, + "learning_rate": 3.856982501415595e-06, + "loss": 1.0356, + "step": 1963 + }, + { + "epoch": 0.14760258529986472, + "grad_norm": 1.8802377023983274, + "learning_rate": 3.8568016554722554e-06, + "loss": 1.044, + "step": 1964 + }, + { + "epoch": 0.1476777393656997, + "grad_norm": 1.5338138991695722, + "learning_rate": 3.856620699506308e-06, + "loss": 1.0264, + "step": 1965 + }, + { + "epoch": 0.14775289343153464, + "grad_norm": 1.929575681342423, + "learning_rate": 3.856439633528476e-06, + "loss": 1.0333, + "step": 1966 + }, + { + "epoch": 0.1478280474973696, + "grad_norm": 1.4139372206352392, + "learning_rate": 3.856258457549486e-06, + "loss": 0.8984, + "step": 1967 + }, + { + "epoch": 0.14790320156320458, + "grad_norm": 1.852041319408832, + "learning_rate": 3.856077171580074e-06, + "loss": 1.0164, + "step": 1968 + }, + { + "epoch": 0.14797835562903952, + "grad_norm": 1.7487023856038935, + "learning_rate": 3.855895775630983e-06, + "loss": 0.8824, + "step": 1969 + }, + { + "epoch": 0.1480535096948745, + "grad_norm": 1.5747915129954395, + "learning_rate": 3.85571426971296e-06, + "loss": 1.0641, + "step": 1970 + }, + { + "epoch": 0.14812866376070946, + "grad_norm": 1.905330936692432, + "learning_rate": 3.8555326538367605e-06, + "loss": 1.0759, + "step": 1971 + }, + { + "epoch": 0.1482038178265444, + "grad_norm": 1.3709386319394261, + "learning_rate": 3.855350928013145e-06, + "loss": 1.0514, + "step": 1972 + }, + { + "epoch": 0.14827897189237937, + "grad_norm": 2.5156567494539073, + "learning_rate": 3.855169092252884e-06, + "loss": 0.9965, + "step": 1973 + }, + { + "epoch": 0.14835412595821434, + "grad_norm": 1.4190958550991095, + "learning_rate": 3.85498714656675e-06, + "loss": 0.9642, + "step": 1974 + }, + { + "epoch": 0.14842928002404931, + "grad_norm": 1.6967715124674978, + "learning_rate": 3.854805090965525e-06, + "loss": 0.971, + "step": 1975 + }, + { + "epoch": 0.14850443408988426, + "grad_norm": 1.6214548467136018, + "learning_rate": 3.854622925459994e-06, + "loss": 1.0057, + "step": 1976 + }, + { + "epoch": 0.14857958815571923, + "grad_norm": 1.9697793358696407, + "learning_rate": 3.854440650060955e-06, + "loss": 0.9662, + "step": 1977 + }, + { + "epoch": 0.1486547422215542, + "grad_norm": 1.3452324784744747, + "learning_rate": 3.854258264779204e-06, + "loss": 1.006, + "step": 1978 + }, + { + "epoch": 0.14872989628738914, + "grad_norm": 1.663975386510889, + "learning_rate": 3.854075769625552e-06, + "loss": 1.0, + "step": 1979 + }, + { + "epoch": 0.1488050503532241, + "grad_norm": 1.5969252277992738, + "learning_rate": 3.8538931646108105e-06, + "loss": 0.9377, + "step": 1980 + }, + { + "epoch": 0.14888020441905908, + "grad_norm": 1.378983103741987, + "learning_rate": 3.853710449745801e-06, + "loss": 0.9829, + "step": 1981 + }, + { + "epoch": 0.14895535848489402, + "grad_norm": 1.3615777595440643, + "learning_rate": 3.853527625041347e-06, + "loss": 0.9768, + "step": 1982 + }, + { + "epoch": 0.149030512550729, + "grad_norm": 1.3377070614263578, + "learning_rate": 3.853344690508285e-06, + "loss": 0.8984, + "step": 1983 + }, + { + "epoch": 0.14910566661656396, + "grad_norm": 1.515730706687475, + "learning_rate": 3.853161646157453e-06, + "loss": 1.0303, + "step": 1984 + }, + { + "epoch": 0.1491808206823989, + "grad_norm": 1.665237298614348, + "learning_rate": 3.852978491999697e-06, + "loss": 1.0498, + "step": 1985 + }, + { + "epoch": 0.14925597474823388, + "grad_norm": 1.3117388834495898, + "learning_rate": 3.852795228045869e-06, + "loss": 1.0475, + "step": 1986 + }, + { + "epoch": 0.14933112881406885, + "grad_norm": 1.9356833524348849, + "learning_rate": 3.85261185430683e-06, + "loss": 1.0016, + "step": 1987 + }, + { + "epoch": 0.1494062828799038, + "grad_norm": 1.497086556602184, + "learning_rate": 3.8524283707934445e-06, + "loss": 1.1056, + "step": 1988 + }, + { + "epoch": 0.14948143694573876, + "grad_norm": 1.4497691135083925, + "learning_rate": 3.8522447775165845e-06, + "loss": 0.9546, + "step": 1989 + }, + { + "epoch": 0.14955659101157373, + "grad_norm": 1.4421875033063465, + "learning_rate": 3.852061074487129e-06, + "loss": 0.9913, + "step": 1990 + }, + { + "epoch": 0.1496317450774087, + "grad_norm": 1.443903963541054, + "learning_rate": 3.851877261715961e-06, + "loss": 1.036, + "step": 1991 + }, + { + "epoch": 0.14970689914324364, + "grad_norm": 1.5054137761798936, + "learning_rate": 3.851693339213976e-06, + "loss": 1.0594, + "step": 1992 + }, + { + "epoch": 0.1497820532090786, + "grad_norm": 1.4306550634315949, + "learning_rate": 3.8515093069920695e-06, + "loss": 0.9796, + "step": 1993 + }, + { + "epoch": 0.14985720727491358, + "grad_norm": 1.479819605890868, + "learning_rate": 3.851325165061147e-06, + "loss": 1.0327, + "step": 1994 + }, + { + "epoch": 0.14993236134074853, + "grad_norm": 1.5324667164360384, + "learning_rate": 3.851140913432118e-06, + "loss": 0.9978, + "step": 1995 + }, + { + "epoch": 0.1500075154065835, + "grad_norm": 1.5936636059850076, + "learning_rate": 3.850956552115903e-06, + "loss": 0.9487, + "step": 1996 + }, + { + "epoch": 0.15008266947241847, + "grad_norm": 1.7552748609965576, + "learning_rate": 3.850772081123423e-06, + "loss": 1.069, + "step": 1997 + }, + { + "epoch": 0.1501578235382534, + "grad_norm": 1.388720461911535, + "learning_rate": 3.850587500465611e-06, + "loss": 0.9805, + "step": 1998 + }, + { + "epoch": 0.15023297760408838, + "grad_norm": 1.4766897318282246, + "learning_rate": 3.850402810153403e-06, + "loss": 0.9379, + "step": 1999 + }, + { + "epoch": 0.15030813166992335, + "grad_norm": 1.7361914942464163, + "learning_rate": 3.850218010197743e-06, + "loss": 0.8979, + "step": 2000 + }, + { + "epoch": 0.1503832857357583, + "grad_norm": 1.4710773939134054, + "learning_rate": 3.850033100609581e-06, + "loss": 1.0725, + "step": 2001 + }, + { + "epoch": 0.15045843980159326, + "grad_norm": 1.6544570575074193, + "learning_rate": 3.8498480813998735e-06, + "loss": 0.9678, + "step": 2002 + }, + { + "epoch": 0.15053359386742823, + "grad_norm": 1.793169596274716, + "learning_rate": 3.849662952579583e-06, + "loss": 0.9908, + "step": 2003 + }, + { + "epoch": 0.15060874793326318, + "grad_norm": 1.8307181982486889, + "learning_rate": 3.8494777141596805e-06, + "loss": 1.0106, + "step": 2004 + }, + { + "epoch": 0.15068390199909815, + "grad_norm": 0.9627220826500702, + "learning_rate": 3.8492923661511405e-06, + "loss": 0.8497, + "step": 2005 + }, + { + "epoch": 0.15075905606493312, + "grad_norm": 1.3615297800282176, + "learning_rate": 3.8491069085649475e-06, + "loss": 1.0452, + "step": 2006 + }, + { + "epoch": 0.1508342101307681, + "grad_norm": 1.502518209007481, + "learning_rate": 3.848921341412088e-06, + "loss": 1.011, + "step": 2007 + }, + { + "epoch": 0.15090936419660303, + "grad_norm": 2.093096503003679, + "learning_rate": 3.848735664703561e-06, + "loss": 1.0443, + "step": 2008 + }, + { + "epoch": 0.150984518262438, + "grad_norm": 1.453616315577745, + "learning_rate": 3.848549878450365e-06, + "loss": 1.0484, + "step": 2009 + }, + { + "epoch": 0.15105967232827297, + "grad_norm": 1.9903730319822615, + "learning_rate": 3.84836398266351e-06, + "loss": 1.0459, + "step": 2010 + }, + { + "epoch": 0.1511348263941079, + "grad_norm": 2.0821818284769464, + "learning_rate": 3.848177977354012e-06, + "loss": 1.0614, + "step": 2011 + }, + { + "epoch": 0.15120998045994288, + "grad_norm": 0.9399962887284895, + "learning_rate": 3.847991862532892e-06, + "loss": 0.8616, + "step": 2012 + }, + { + "epoch": 0.15128513452577785, + "grad_norm": 1.6605122981019524, + "learning_rate": 3.847805638211177e-06, + "loss": 1.042, + "step": 2013 + }, + { + "epoch": 0.1513602885916128, + "grad_norm": 1.7533699694070646, + "learning_rate": 3.847619304399902e-06, + "loss": 1.0005, + "step": 2014 + }, + { + "epoch": 0.15143544265744777, + "grad_norm": 1.4138918610354005, + "learning_rate": 3.847432861110109e-06, + "loss": 0.9794, + "step": 2015 + }, + { + "epoch": 0.15151059672328274, + "grad_norm": 2.186870048279674, + "learning_rate": 3.847246308352844e-06, + "loss": 0.9184, + "step": 2016 + }, + { + "epoch": 0.15158575078911768, + "grad_norm": 1.8351820020301968, + "learning_rate": 3.847059646139162e-06, + "loss": 0.9938, + "step": 2017 + }, + { + "epoch": 0.15166090485495265, + "grad_norm": 1.8206443647827373, + "learning_rate": 3.846872874480123e-06, + "loss": 1.0223, + "step": 2018 + }, + { + "epoch": 0.15173605892078762, + "grad_norm": 1.3591105324438182, + "learning_rate": 3.8466859933867945e-06, + "loss": 0.9878, + "step": 2019 + }, + { + "epoch": 0.1518112129866226, + "grad_norm": 1.0513283252000105, + "learning_rate": 3.846499002870249e-06, + "loss": 0.9686, + "step": 2020 + }, + { + "epoch": 0.15188636705245753, + "grad_norm": 1.5591709530946634, + "learning_rate": 3.846311902941567e-06, + "loss": 0.8806, + "step": 2021 + }, + { + "epoch": 0.1519615211182925, + "grad_norm": 1.4939607798727836, + "learning_rate": 3.846124693611835e-06, + "loss": 0.9521, + "step": 2022 + }, + { + "epoch": 0.15203667518412747, + "grad_norm": 1.5239538638122614, + "learning_rate": 3.845937374892145e-06, + "loss": 0.8047, + "step": 2023 + }, + { + "epoch": 0.15211182924996242, + "grad_norm": 1.837200493309973, + "learning_rate": 3.845749946793597e-06, + "loss": 0.7865, + "step": 2024 + }, + { + "epoch": 0.1521869833157974, + "grad_norm": 1.6722168617472413, + "learning_rate": 3.845562409327297e-06, + "loss": 1.0115, + "step": 2025 + }, + { + "epoch": 0.15226213738163236, + "grad_norm": 1.5249215746866138, + "learning_rate": 3.8453747625043575e-06, + "loss": 0.9746, + "step": 2026 + }, + { + "epoch": 0.1523372914474673, + "grad_norm": 1.6142146109946196, + "learning_rate": 3.8451870063358966e-06, + "loss": 1.1026, + "step": 2027 + }, + { + "epoch": 0.15241244551330227, + "grad_norm": 1.6866717248521632, + "learning_rate": 3.844999140833039e-06, + "loss": 0.9174, + "step": 2028 + }, + { + "epoch": 0.15248759957913724, + "grad_norm": 1.4049263552713223, + "learning_rate": 3.844811166006919e-06, + "loss": 1.0058, + "step": 2029 + }, + { + "epoch": 0.15256275364497218, + "grad_norm": 1.5453644900934604, + "learning_rate": 3.844623081868672e-06, + "loss": 1.0014, + "step": 2030 + }, + { + "epoch": 0.15263790771080715, + "grad_norm": 1.42503550635825, + "learning_rate": 3.844434888429444e-06, + "loss": 0.9508, + "step": 2031 + }, + { + "epoch": 0.15271306177664212, + "grad_norm": 1.6815283335766584, + "learning_rate": 3.8442465857003864e-06, + "loss": 1.0009, + "step": 2032 + }, + { + "epoch": 0.15278821584247707, + "grad_norm": 1.3434874699543864, + "learning_rate": 3.844058173692657e-06, + "loss": 1.0499, + "step": 2033 + }, + { + "epoch": 0.15286336990831204, + "grad_norm": 1.7575760745952085, + "learning_rate": 3.843869652417418e-06, + "loss": 1.0167, + "step": 2034 + }, + { + "epoch": 0.152938523974147, + "grad_norm": 1.3258100192939153, + "learning_rate": 3.843681021885842e-06, + "loss": 1.0106, + "step": 2035 + }, + { + "epoch": 0.15301367803998198, + "grad_norm": 1.5623377906524665, + "learning_rate": 3.843492282109107e-06, + "loss": 0.8198, + "step": 2036 + }, + { + "epoch": 0.15308883210581692, + "grad_norm": 1.453698158028606, + "learning_rate": 3.843303433098393e-06, + "loss": 0.9835, + "step": 2037 + }, + { + "epoch": 0.1531639861716519, + "grad_norm": 1.5979746009374145, + "learning_rate": 3.843114474864894e-06, + "loss": 1.1007, + "step": 2038 + }, + { + "epoch": 0.15323914023748686, + "grad_norm": 0.8479438329057946, + "learning_rate": 3.842925407419803e-06, + "loss": 0.7948, + "step": 2039 + }, + { + "epoch": 0.1533142943033218, + "grad_norm": 1.6782683298856498, + "learning_rate": 3.842736230774325e-06, + "loss": 0.8965, + "step": 2040 + }, + { + "epoch": 0.15338944836915677, + "grad_norm": 1.4535674375276415, + "learning_rate": 3.842546944939669e-06, + "loss": 0.9113, + "step": 2041 + }, + { + "epoch": 0.15346460243499174, + "grad_norm": 1.2821508357630502, + "learning_rate": 3.842357549927051e-06, + "loss": 0.9908, + "step": 2042 + }, + { + "epoch": 0.15353975650082669, + "grad_norm": 1.7532943643776904, + "learning_rate": 3.842168045747693e-06, + "loss": 1.0071, + "step": 2043 + }, + { + "epoch": 0.15361491056666166, + "grad_norm": 1.5114488779723505, + "learning_rate": 3.8419784324128256e-06, + "loss": 1.013, + "step": 2044 + }, + { + "epoch": 0.15369006463249663, + "grad_norm": 1.4873534186452009, + "learning_rate": 3.841788709933682e-06, + "loss": 0.8302, + "step": 2045 + }, + { + "epoch": 0.15376521869833157, + "grad_norm": 1.6138284784997965, + "learning_rate": 3.841598878321503e-06, + "loss": 1.0315, + "step": 2046 + }, + { + "epoch": 0.15384037276416654, + "grad_norm": 1.5390164905777532, + "learning_rate": 3.84140893758754e-06, + "loss": 1.1595, + "step": 2047 + }, + { + "epoch": 0.1539155268300015, + "grad_norm": 2.528064748755729, + "learning_rate": 3.841218887743046e-06, + "loss": 1.0164, + "step": 2048 + }, + { + "epoch": 0.15399068089583645, + "grad_norm": 1.593818812661205, + "learning_rate": 3.8410287287992825e-06, + "loss": 0.9727, + "step": 2049 + }, + { + "epoch": 0.15406583496167142, + "grad_norm": 1.4023672542886845, + "learning_rate": 3.840838460767517e-06, + "loss": 0.886, + "step": 2050 + }, + { + "epoch": 0.1541409890275064, + "grad_norm": 1.6766127010910536, + "learning_rate": 3.840648083659024e-06, + "loss": 0.982, + "step": 2051 + }, + { + "epoch": 0.15421614309334136, + "grad_norm": 2.0166798278547273, + "learning_rate": 3.840457597485083e-06, + "loss": 0.993, + "step": 2052 + }, + { + "epoch": 0.1542912971591763, + "grad_norm": 1.16208244741321, + "learning_rate": 3.840267002256983e-06, + "loss": 0.8611, + "step": 2053 + }, + { + "epoch": 0.15436645122501128, + "grad_norm": 1.4782126757565894, + "learning_rate": 3.840076297986015e-06, + "loss": 0.9724, + "step": 2054 + }, + { + "epoch": 0.15444160529084625, + "grad_norm": 1.9093517645438207, + "learning_rate": 3.839885484683481e-06, + "loss": 0.9727, + "step": 2055 + }, + { + "epoch": 0.1545167593566812, + "grad_norm": 1.6749587096633602, + "learning_rate": 3.839694562360686e-06, + "loss": 0.9643, + "step": 2056 + }, + { + "epoch": 0.15459191342251616, + "grad_norm": 1.8288534946320552, + "learning_rate": 3.839503531028944e-06, + "loss": 0.9425, + "step": 2057 + }, + { + "epoch": 0.15466706748835113, + "grad_norm": 1.526711101190734, + "learning_rate": 3.839312390699573e-06, + "loss": 1.0104, + "step": 2058 + }, + { + "epoch": 0.15474222155418607, + "grad_norm": 1.3266986592826644, + "learning_rate": 3.8391211413839005e-06, + "loss": 1.0141, + "step": 2059 + }, + { + "epoch": 0.15481737562002104, + "grad_norm": 1.6962274288736334, + "learning_rate": 3.838929783093258e-06, + "loss": 0.9544, + "step": 2060 + }, + { + "epoch": 0.154892529685856, + "grad_norm": 1.7825742115981416, + "learning_rate": 3.838738315838983e-06, + "loss": 1.0059, + "step": 2061 + }, + { + "epoch": 0.15496768375169095, + "grad_norm": 1.5575385094745149, + "learning_rate": 3.838546739632423e-06, + "loss": 1.0124, + "step": 2062 + }, + { + "epoch": 0.15504283781752592, + "grad_norm": 1.7352246102268754, + "learning_rate": 3.838355054484928e-06, + "loss": 1.0589, + "step": 2063 + }, + { + "epoch": 0.1551179918833609, + "grad_norm": 1.4422950382166626, + "learning_rate": 3.838163260407857e-06, + "loss": 0.9791, + "step": 2064 + }, + { + "epoch": 0.15519314594919587, + "grad_norm": 1.7309071005007648, + "learning_rate": 3.837971357412573e-06, + "loss": 0.9023, + "step": 2065 + }, + { + "epoch": 0.1552683000150308, + "grad_norm": 1.6093093710545265, + "learning_rate": 3.837779345510449e-06, + "loss": 0.9787, + "step": 2066 + }, + { + "epoch": 0.15534345408086578, + "grad_norm": 1.5244811566853238, + "learning_rate": 3.837587224712861e-06, + "loss": 0.9295, + "step": 2067 + }, + { + "epoch": 0.15541860814670075, + "grad_norm": 1.6376421014993323, + "learning_rate": 3.837394995031193e-06, + "loss": 0.8232, + "step": 2068 + }, + { + "epoch": 0.1554937622125357, + "grad_norm": 1.9285062811711875, + "learning_rate": 3.837202656476836e-06, + "loss": 1.032, + "step": 2069 + }, + { + "epoch": 0.15556891627837066, + "grad_norm": 1.5509708778652829, + "learning_rate": 3.837010209061187e-06, + "loss": 0.9078, + "step": 2070 + }, + { + "epoch": 0.15564407034420563, + "grad_norm": 1.4028253604930805, + "learning_rate": 3.836817652795648e-06, + "loss": 0.9571, + "step": 2071 + }, + { + "epoch": 0.15571922441004057, + "grad_norm": 2.0322084953095247, + "learning_rate": 3.8366249876916294e-06, + "loss": 0.9986, + "step": 2072 + }, + { + "epoch": 0.15579437847587554, + "grad_norm": 1.4738048657791876, + "learning_rate": 3.8364322137605484e-06, + "loss": 0.9274, + "step": 2073 + }, + { + "epoch": 0.15586953254171051, + "grad_norm": 1.6925696056934225, + "learning_rate": 3.836239331013825e-06, + "loss": 1.0418, + "step": 2074 + }, + { + "epoch": 0.15594468660754546, + "grad_norm": 0.8831635173974942, + "learning_rate": 3.836046339462891e-06, + "loss": 0.8153, + "step": 2075 + }, + { + "epoch": 0.15601984067338043, + "grad_norm": 1.6235079150675138, + "learning_rate": 3.83585323911918e-06, + "loss": 1.0368, + "step": 2076 + }, + { + "epoch": 0.1560949947392154, + "grad_norm": 1.6421863937011245, + "learning_rate": 3.835660029994135e-06, + "loss": 0.9992, + "step": 2077 + }, + { + "epoch": 0.15617014880505034, + "grad_norm": 1.7373051678215679, + "learning_rate": 3.835466712099204e-06, + "loss": 0.937, + "step": 2078 + }, + { + "epoch": 0.1562453028708853, + "grad_norm": 1.6145329457285835, + "learning_rate": 3.835273285445842e-06, + "loss": 1.0037, + "step": 2079 + }, + { + "epoch": 0.15632045693672028, + "grad_norm": 1.320209657173683, + "learning_rate": 3.83507975004551e-06, + "loss": 0.9886, + "step": 2080 + }, + { + "epoch": 0.15639561100255525, + "grad_norm": 1.7452541440601161, + "learning_rate": 3.8348861059096755e-06, + "loss": 0.9974, + "step": 2081 + }, + { + "epoch": 0.1564707650683902, + "grad_norm": 2.3065793306248055, + "learning_rate": 3.834692353049814e-06, + "loss": 0.9869, + "step": 2082 + }, + { + "epoch": 0.15654591913422516, + "grad_norm": 1.5743146919187676, + "learning_rate": 3.834498491477403e-06, + "loss": 1.0342, + "step": 2083 + }, + { + "epoch": 0.15662107320006013, + "grad_norm": 1.6224633739081638, + "learning_rate": 3.834304521203934e-06, + "loss": 1.0249, + "step": 2084 + }, + { + "epoch": 0.15669622726589508, + "grad_norm": 1.6005144365130908, + "learning_rate": 3.834110442240896e-06, + "loss": 1.0006, + "step": 2085 + }, + { + "epoch": 0.15677138133173005, + "grad_norm": 1.7305244109628954, + "learning_rate": 3.833916254599792e-06, + "loss": 0.957, + "step": 2086 + }, + { + "epoch": 0.15684653539756502, + "grad_norm": 1.4258613459415828, + "learning_rate": 3.833721958292128e-06, + "loss": 0.98, + "step": 2087 + }, + { + "epoch": 0.15692168946339996, + "grad_norm": 1.8388041091934437, + "learning_rate": 3.8335275533294155e-06, + "loss": 0.9986, + "step": 2088 + }, + { + "epoch": 0.15699684352923493, + "grad_norm": 1.6715355644402174, + "learning_rate": 3.833333039723174e-06, + "loss": 0.9407, + "step": 2089 + }, + { + "epoch": 0.1570719975950699, + "grad_norm": 1.6205685419678175, + "learning_rate": 3.83313841748493e-06, + "loss": 1.0545, + "step": 2090 + }, + { + "epoch": 0.15714715166090484, + "grad_norm": 1.5646064292669317, + "learning_rate": 3.832943686626215e-06, + "loss": 1.0596, + "step": 2091 + }, + { + "epoch": 0.1572223057267398, + "grad_norm": 1.5736462330562526, + "learning_rate": 3.832748847158568e-06, + "loss": 0.9671, + "step": 2092 + }, + { + "epoch": 0.15729745979257478, + "grad_norm": 1.8003421670157485, + "learning_rate": 3.8325538990935346e-06, + "loss": 0.9742, + "step": 2093 + }, + { + "epoch": 0.15737261385840973, + "grad_norm": 1.505969250105454, + "learning_rate": 3.832358842442665e-06, + "loss": 1.0503, + "step": 2094 + }, + { + "epoch": 0.1574477679242447, + "grad_norm": 1.3922478282799875, + "learning_rate": 3.832163677217516e-06, + "loss": 0.9815, + "step": 2095 + }, + { + "epoch": 0.15752292199007967, + "grad_norm": 1.3122301732587842, + "learning_rate": 3.831968403429655e-06, + "loss": 0.9298, + "step": 2096 + }, + { + "epoch": 0.15759807605591464, + "grad_norm": 1.8351401687638522, + "learning_rate": 3.83177302109065e-06, + "loss": 0.9754, + "step": 2097 + }, + { + "epoch": 0.15767323012174958, + "grad_norm": 1.662335067828181, + "learning_rate": 3.8315775302120796e-06, + "loss": 1.0131, + "step": 2098 + }, + { + "epoch": 0.15774838418758455, + "grad_norm": 1.6681756420602303, + "learning_rate": 3.831381930805526e-06, + "loss": 1.0096, + "step": 2099 + }, + { + "epoch": 0.15782353825341952, + "grad_norm": 1.61346619217914, + "learning_rate": 3.831186222882582e-06, + "loss": 0.9893, + "step": 2100 + }, + { + "epoch": 0.15789869231925446, + "grad_norm": 1.7391167507980867, + "learning_rate": 3.830990406454841e-06, + "loss": 0.9507, + "step": 2101 + }, + { + "epoch": 0.15797384638508943, + "grad_norm": 1.4983621238230984, + "learning_rate": 3.8307944815339065e-06, + "loss": 0.9717, + "step": 2102 + }, + { + "epoch": 0.1580490004509244, + "grad_norm": 1.5993521719969654, + "learning_rate": 3.83059844813139e-06, + "loss": 1.0177, + "step": 2103 + }, + { + "epoch": 0.15812415451675935, + "grad_norm": 1.682530892982671, + "learning_rate": 3.830402306258904e-06, + "loss": 0.9564, + "step": 2104 + }, + { + "epoch": 0.15819930858259432, + "grad_norm": 2.0398930871286574, + "learning_rate": 3.8302060559280735e-06, + "loss": 0.9534, + "step": 2105 + }, + { + "epoch": 0.1582744626484293, + "grad_norm": 1.5944693194954795, + "learning_rate": 3.830009697150526e-06, + "loss": 0.9077, + "step": 2106 + }, + { + "epoch": 0.15834961671426423, + "grad_norm": 1.3733645162622334, + "learning_rate": 3.829813229937896e-06, + "loss": 0.9051, + "step": 2107 + }, + { + "epoch": 0.1584247707800992, + "grad_norm": 1.5301873548648084, + "learning_rate": 3.829616654301824e-06, + "loss": 1.0185, + "step": 2108 + }, + { + "epoch": 0.15849992484593417, + "grad_norm": 1.7008603639649285, + "learning_rate": 3.829419970253961e-06, + "loss": 0.9255, + "step": 2109 + }, + { + "epoch": 0.1585750789117691, + "grad_norm": 1.5312479758735824, + "learning_rate": 3.829223177805959e-06, + "loss": 1.0561, + "step": 2110 + }, + { + "epoch": 0.15865023297760408, + "grad_norm": 1.936118894981282, + "learning_rate": 3.8290262769694785e-06, + "loss": 0.9269, + "step": 2111 + }, + { + "epoch": 0.15872538704343905, + "grad_norm": 1.5422843864022093, + "learning_rate": 3.828829267756188e-06, + "loss": 1.0279, + "step": 2112 + }, + { + "epoch": 0.15880054110927402, + "grad_norm": 1.7278488683572835, + "learning_rate": 3.82863215017776e-06, + "loss": 1.003, + "step": 2113 + }, + { + "epoch": 0.15887569517510897, + "grad_norm": 1.8813232610384676, + "learning_rate": 3.828434924245874e-06, + "loss": 0.9661, + "step": 2114 + }, + { + "epoch": 0.15895084924094394, + "grad_norm": 1.2713749570395545, + "learning_rate": 3.828237589972218e-06, + "loss": 0.9793, + "step": 2115 + }, + { + "epoch": 0.1590260033067789, + "grad_norm": 1.6752571762562012, + "learning_rate": 3.828040147368484e-06, + "loss": 1.0022, + "step": 2116 + }, + { + "epoch": 0.15910115737261385, + "grad_norm": 1.4033556929065751, + "learning_rate": 3.827842596446372e-06, + "loss": 0.9751, + "step": 2117 + }, + { + "epoch": 0.15917631143844882, + "grad_norm": 1.4520428637753662, + "learning_rate": 3.827644937217585e-06, + "loss": 0.9626, + "step": 2118 + }, + { + "epoch": 0.1592514655042838, + "grad_norm": 1.831296598603204, + "learning_rate": 3.827447169693839e-06, + "loss": 1.066, + "step": 2119 + }, + { + "epoch": 0.15932661957011873, + "grad_norm": 2.3565526590823462, + "learning_rate": 3.827249293886849e-06, + "loss": 1.0349, + "step": 2120 + }, + { + "epoch": 0.1594017736359537, + "grad_norm": 1.7590364391289197, + "learning_rate": 3.827051309808342e-06, + "loss": 1.013, + "step": 2121 + }, + { + "epoch": 0.15947692770178867, + "grad_norm": 1.2895576219763656, + "learning_rate": 3.826853217470048e-06, + "loss": 0.9639, + "step": 2122 + }, + { + "epoch": 0.15955208176762362, + "grad_norm": 1.2920069041318634, + "learning_rate": 3.8266550168837065e-06, + "loss": 0.9756, + "step": 2123 + }, + { + "epoch": 0.1596272358334586, + "grad_norm": 1.522137757477104, + "learning_rate": 3.82645670806106e-06, + "loss": 0.9233, + "step": 2124 + }, + { + "epoch": 0.15970238989929356, + "grad_norm": 1.4990930199394, + "learning_rate": 3.826258291013859e-06, + "loss": 1.0378, + "step": 2125 + }, + { + "epoch": 0.15977754396512853, + "grad_norm": 1.4387428052839986, + "learning_rate": 3.826059765753861e-06, + "loss": 0.9555, + "step": 2126 + }, + { + "epoch": 0.15985269803096347, + "grad_norm": 1.5636829475889427, + "learning_rate": 3.82586113229283e-06, + "loss": 1.0291, + "step": 2127 + }, + { + "epoch": 0.15992785209679844, + "grad_norm": 1.5864085733379212, + "learning_rate": 3.825662390642535e-06, + "loss": 0.9171, + "step": 2128 + }, + { + "epoch": 0.1600030061626334, + "grad_norm": 1.3482614236493065, + "learning_rate": 3.825463540814753e-06, + "loss": 0.9623, + "step": 2129 + }, + { + "epoch": 0.16007816022846835, + "grad_norm": 1.6986635199871492, + "learning_rate": 3.8252645828212655e-06, + "loss": 0.9803, + "step": 2130 + }, + { + "epoch": 0.16015331429430332, + "grad_norm": 1.6292400063920431, + "learning_rate": 3.825065516673862e-06, + "loss": 0.9628, + "step": 2131 + }, + { + "epoch": 0.1602284683601383, + "grad_norm": 1.31090498691838, + "learning_rate": 3.824866342384338e-06, + "loss": 1.054, + "step": 2132 + }, + { + "epoch": 0.16030362242597324, + "grad_norm": 2.3827613700008667, + "learning_rate": 3.824667059964496e-06, + "loss": 0.916, + "step": 2133 + }, + { + "epoch": 0.1603787764918082, + "grad_norm": 2.150575330183404, + "learning_rate": 3.824467669426143e-06, + "loss": 0.9688, + "step": 2134 + }, + { + "epoch": 0.16045393055764318, + "grad_norm": 2.7490983698712936, + "learning_rate": 3.824268170781094e-06, + "loss": 0.9084, + "step": 2135 + }, + { + "epoch": 0.16052908462347812, + "grad_norm": 1.0431946927623081, + "learning_rate": 3.82406856404117e-06, + "loss": 0.8602, + "step": 2136 + }, + { + "epoch": 0.1606042386893131, + "grad_norm": 1.6813163488973693, + "learning_rate": 3.8238688492182e-06, + "loss": 1.0567, + "step": 2137 + }, + { + "epoch": 0.16067939275514806, + "grad_norm": 2.1442268308384267, + "learning_rate": 3.823669026324016e-06, + "loss": 1.0024, + "step": 2138 + }, + { + "epoch": 0.160754546820983, + "grad_norm": 1.9018012493274263, + "learning_rate": 3.823469095370459e-06, + "loss": 1.053, + "step": 2139 + }, + { + "epoch": 0.16082970088681797, + "grad_norm": 1.487420501507988, + "learning_rate": 3.823269056369376e-06, + "loss": 0.9878, + "step": 2140 + }, + { + "epoch": 0.16090485495265294, + "grad_norm": 1.4846061928047563, + "learning_rate": 3.8230689093326185e-06, + "loss": 1.0781, + "step": 2141 + }, + { + "epoch": 0.1609800090184879, + "grad_norm": 1.3712349408695519, + "learning_rate": 3.822868654272048e-06, + "loss": 0.9168, + "step": 2142 + }, + { + "epoch": 0.16105516308432286, + "grad_norm": 1.8359864573342835, + "learning_rate": 3.822668291199529e-06, + "loss": 1.0798, + "step": 2143 + }, + { + "epoch": 0.16113031715015783, + "grad_norm": 2.0880986708642713, + "learning_rate": 3.822467820126935e-06, + "loss": 0.9206, + "step": 2144 + }, + { + "epoch": 0.1612054712159928, + "grad_norm": 1.6287268670196622, + "learning_rate": 3.822267241066143e-06, + "loss": 0.933, + "step": 2145 + }, + { + "epoch": 0.16128062528182774, + "grad_norm": 2.0222305763454003, + "learning_rate": 3.8220665540290395e-06, + "loss": 1.009, + "step": 2146 + }, + { + "epoch": 0.1613557793476627, + "grad_norm": 1.2136666338622952, + "learning_rate": 3.821865759027515e-06, + "loss": 0.9559, + "step": 2147 + }, + { + "epoch": 0.16143093341349768, + "grad_norm": 1.7372107477720378, + "learning_rate": 3.821664856073469e-06, + "loss": 0.9504, + "step": 2148 + }, + { + "epoch": 0.16150608747933262, + "grad_norm": 2.085439329630762, + "learning_rate": 3.821463845178803e-06, + "loss": 1.0179, + "step": 2149 + }, + { + "epoch": 0.1615812415451676, + "grad_norm": 1.965966876308776, + "learning_rate": 3.821262726355431e-06, + "loss": 1.0295, + "step": 2150 + }, + { + "epoch": 0.16165639561100256, + "grad_norm": 1.5690768870180543, + "learning_rate": 3.821061499615268e-06, + "loss": 0.9326, + "step": 2151 + }, + { + "epoch": 0.1617315496768375, + "grad_norm": 0.954031622518504, + "learning_rate": 3.820860164970237e-06, + "loss": 0.7963, + "step": 2152 + }, + { + "epoch": 0.16180670374267248, + "grad_norm": 1.9890616034110926, + "learning_rate": 3.820658722432269e-06, + "loss": 1.0183, + "step": 2153 + }, + { + "epoch": 0.16188185780850745, + "grad_norm": 1.8836205992017236, + "learning_rate": 3.820457172013301e-06, + "loss": 1.0163, + "step": 2154 + }, + { + "epoch": 0.1619570118743424, + "grad_norm": 1.3144299985486465, + "learning_rate": 3.820255513725274e-06, + "loss": 1.0245, + "step": 2155 + }, + { + "epoch": 0.16203216594017736, + "grad_norm": 1.4203955530574524, + "learning_rate": 3.820053747580137e-06, + "loss": 1.0241, + "step": 2156 + }, + { + "epoch": 0.16210732000601233, + "grad_norm": 1.5471762932094348, + "learning_rate": 3.8198518735898465e-06, + "loss": 0.9975, + "step": 2157 + }, + { + "epoch": 0.1621824740718473, + "grad_norm": 1.4238038446896666, + "learning_rate": 3.819649891766364e-06, + "loss": 0.9356, + "step": 2158 + }, + { + "epoch": 0.16225762813768224, + "grad_norm": 1.6624050572181637, + "learning_rate": 3.8194478021216566e-06, + "loss": 0.962, + "step": 2159 + }, + { + "epoch": 0.1623327822035172, + "grad_norm": 1.626472906014242, + "learning_rate": 3.8192456046677004e-06, + "loss": 1.1398, + "step": 2160 + }, + { + "epoch": 0.16240793626935218, + "grad_norm": 2.151263123227865, + "learning_rate": 3.819043299416476e-06, + "loss": 1.0455, + "step": 2161 + }, + { + "epoch": 0.16248309033518712, + "grad_norm": 1.7333302831011612, + "learning_rate": 3.8188408863799706e-06, + "loss": 0.9742, + "step": 2162 + }, + { + "epoch": 0.1625582444010221, + "grad_norm": 2.087536488288361, + "learning_rate": 3.818638365570177e-06, + "loss": 0.9172, + "step": 2163 + }, + { + "epoch": 0.16263339846685707, + "grad_norm": 1.443895790038268, + "learning_rate": 3.818435736999097e-06, + "loss": 1.0278, + "step": 2164 + }, + { + "epoch": 0.162708552532692, + "grad_norm": 1.6030711441958125, + "learning_rate": 3.818233000678736e-06, + "loss": 0.954, + "step": 2165 + }, + { + "epoch": 0.16278370659852698, + "grad_norm": 1.4501048050197551, + "learning_rate": 3.8180301566211075e-06, + "loss": 0.9861, + "step": 2166 + }, + { + "epoch": 0.16285886066436195, + "grad_norm": 1.5641207871832672, + "learning_rate": 3.81782720483823e-06, + "loss": 1.0636, + "step": 2167 + }, + { + "epoch": 0.1629340147301969, + "grad_norm": 1.5740671392466719, + "learning_rate": 3.8176241453421305e-06, + "loss": 1.0327, + "step": 2168 + }, + { + "epoch": 0.16300916879603186, + "grad_norm": 1.6214649924358362, + "learning_rate": 3.81742097814484e-06, + "loss": 0.9414, + "step": 2169 + }, + { + "epoch": 0.16308432286186683, + "grad_norm": 1.0349841856554547, + "learning_rate": 3.817217703258397e-06, + "loss": 0.8728, + "step": 2170 + }, + { + "epoch": 0.1631594769277018, + "grad_norm": 0.7741583345400773, + "learning_rate": 3.817014320694846e-06, + "loss": 0.7971, + "step": 2171 + }, + { + "epoch": 0.16323463099353674, + "grad_norm": 1.632332471164265, + "learning_rate": 3.816810830466239e-06, + "loss": 1.0334, + "step": 2172 + }, + { + "epoch": 0.16330978505937171, + "grad_norm": 1.5486411598760954, + "learning_rate": 3.816607232584633e-06, + "loss": 0.9228, + "step": 2173 + }, + { + "epoch": 0.16338493912520669, + "grad_norm": 1.2821263828350091, + "learning_rate": 3.816403527062093e-06, + "loss": 0.9836, + "step": 2174 + }, + { + "epoch": 0.16346009319104163, + "grad_norm": 1.6833141829405376, + "learning_rate": 3.816199713910688e-06, + "loss": 1.057, + "step": 2175 + }, + { + "epoch": 0.1635352472568766, + "grad_norm": 1.2908293666167086, + "learning_rate": 3.815995793142495e-06, + "loss": 0.9741, + "step": 2176 + }, + { + "epoch": 0.16361040132271157, + "grad_norm": 1.8990851659333927, + "learning_rate": 3.815791764769598e-06, + "loss": 0.9106, + "step": 2177 + }, + { + "epoch": 0.1636855553885465, + "grad_norm": 1.0910515330982202, + "learning_rate": 3.815587628804086e-06, + "loss": 0.8603, + "step": 2178 + }, + { + "epoch": 0.16376070945438148, + "grad_norm": 1.725098554587758, + "learning_rate": 3.815383385258054e-06, + "loss": 0.9804, + "step": 2179 + }, + { + "epoch": 0.16383586352021645, + "grad_norm": 2.0592048215811123, + "learning_rate": 3.8151790341436046e-06, + "loss": 0.9784, + "step": 2180 + }, + { + "epoch": 0.1639110175860514, + "grad_norm": 1.4772811836053996, + "learning_rate": 3.814974575472847e-06, + "loss": 1.0348, + "step": 2181 + }, + { + "epoch": 0.16398617165188636, + "grad_norm": 2.02277932085477, + "learning_rate": 3.814770009257896e-06, + "loss": 0.985, + "step": 2182 + }, + { + "epoch": 0.16406132571772133, + "grad_norm": 1.619980689628329, + "learning_rate": 3.814565335510873e-06, + "loss": 1.0279, + "step": 2183 + }, + { + "epoch": 0.16413647978355628, + "grad_norm": 0.9539748920859381, + "learning_rate": 3.814360554243905e-06, + "loss": 0.7949, + "step": 2184 + }, + { + "epoch": 0.16421163384939125, + "grad_norm": 2.1521739304248944, + "learning_rate": 3.814155665469126e-06, + "loss": 0.9136, + "step": 2185 + }, + { + "epoch": 0.16428678791522622, + "grad_norm": 2.1149240479234965, + "learning_rate": 3.813950669198678e-06, + "loss": 1.0903, + "step": 2186 + }, + { + "epoch": 0.1643619419810612, + "grad_norm": 1.2602073190821328, + "learning_rate": 3.8137455654447063e-06, + "loss": 0.8216, + "step": 2187 + }, + { + "epoch": 0.16443709604689613, + "grad_norm": 1.7257687224127272, + "learning_rate": 3.8135403542193646e-06, + "loss": 0.9382, + "step": 2188 + }, + { + "epoch": 0.1645122501127311, + "grad_norm": 1.3975936645346747, + "learning_rate": 3.8133350355348125e-06, + "loss": 1.0225, + "step": 2189 + }, + { + "epoch": 0.16458740417856607, + "grad_norm": 1.4730744514322012, + "learning_rate": 3.8131296094032158e-06, + "loss": 0.9287, + "step": 2190 + }, + { + "epoch": 0.164662558244401, + "grad_norm": 1.6310605958619862, + "learning_rate": 3.8129240758367463e-06, + "loss": 0.9973, + "step": 2191 + }, + { + "epoch": 0.16473771231023598, + "grad_norm": 1.8587427426717928, + "learning_rate": 3.8127184348475836e-06, + "loss": 0.9853, + "step": 2192 + }, + { + "epoch": 0.16481286637607095, + "grad_norm": 1.5569191256364547, + "learning_rate": 3.8125126864479123e-06, + "loss": 1.0604, + "step": 2193 + }, + { + "epoch": 0.1648880204419059, + "grad_norm": 1.334381491223165, + "learning_rate": 3.8123068306499236e-06, + "loss": 0.9485, + "step": 2194 + }, + { + "epoch": 0.16496317450774087, + "grad_norm": 1.4591898537272596, + "learning_rate": 3.8121008674658154e-06, + "loss": 1.0133, + "step": 2195 + }, + { + "epoch": 0.16503832857357584, + "grad_norm": 1.4409556938152874, + "learning_rate": 3.8118947969077915e-06, + "loss": 1.0166, + "step": 2196 + }, + { + "epoch": 0.16511348263941078, + "grad_norm": 1.756297429214748, + "learning_rate": 3.8116886189880634e-06, + "loss": 1.0053, + "step": 2197 + }, + { + "epoch": 0.16518863670524575, + "grad_norm": 1.785922030217545, + "learning_rate": 3.811482333718847e-06, + "loss": 0.9035, + "step": 2198 + }, + { + "epoch": 0.16526379077108072, + "grad_norm": 1.5778024173806542, + "learning_rate": 3.811275941112366e-06, + "loss": 1.0195, + "step": 2199 + }, + { + "epoch": 0.16533894483691566, + "grad_norm": 1.7324875880425687, + "learning_rate": 3.811069441180849e-06, + "loss": 1.0452, + "step": 2200 + }, + { + "epoch": 0.16541409890275063, + "grad_norm": 1.9525903809322702, + "learning_rate": 3.810862833936532e-06, + "loss": 1.0028, + "step": 2201 + }, + { + "epoch": 0.1654892529685856, + "grad_norm": 1.8403379906294464, + "learning_rate": 3.8106561193916587e-06, + "loss": 0.9304, + "step": 2202 + }, + { + "epoch": 0.16556440703442057, + "grad_norm": 1.5485988990899002, + "learning_rate": 3.810449297558477e-06, + "loss": 0.9104, + "step": 2203 + }, + { + "epoch": 0.16563956110025552, + "grad_norm": 1.8636557404637861, + "learning_rate": 3.810242368449241e-06, + "loss": 0.8923, + "step": 2204 + }, + { + "epoch": 0.1657147151660905, + "grad_norm": 2.559793950818736, + "learning_rate": 3.810035332076214e-06, + "loss": 0.9786, + "step": 2205 + }, + { + "epoch": 0.16578986923192546, + "grad_norm": 1.1441709713022525, + "learning_rate": 3.809828188451662e-06, + "loss": 0.8457, + "step": 2206 + }, + { + "epoch": 0.1658650232977604, + "grad_norm": 1.5979082799473403, + "learning_rate": 3.809620937587859e-06, + "loss": 1.0017, + "step": 2207 + }, + { + "epoch": 0.16594017736359537, + "grad_norm": 1.7962626865180524, + "learning_rate": 3.8094135794970857e-06, + "loss": 0.9838, + "step": 2208 + }, + { + "epoch": 0.16601533142943034, + "grad_norm": 2.7348410073047384, + "learning_rate": 3.80920611419163e-06, + "loss": 1.0288, + "step": 2209 + }, + { + "epoch": 0.16609048549526528, + "grad_norm": 1.6331201290929531, + "learning_rate": 3.808998541683784e-06, + "loss": 1.0178, + "step": 2210 + }, + { + "epoch": 0.16616563956110025, + "grad_norm": 1.7198020489581667, + "learning_rate": 3.8087908619858473e-06, + "loss": 0.9928, + "step": 2211 + }, + { + "epoch": 0.16624079362693522, + "grad_norm": 1.4973369642102137, + "learning_rate": 3.8085830751101253e-06, + "loss": 1.006, + "step": 2212 + }, + { + "epoch": 0.16631594769277017, + "grad_norm": 1.626776091433654, + "learning_rate": 3.8083751810689306e-06, + "loss": 0.9605, + "step": 2213 + }, + { + "epoch": 0.16639110175860514, + "grad_norm": 1.3679406407734538, + "learning_rate": 3.8081671798745817e-06, + "loss": 1.0621, + "step": 2214 + }, + { + "epoch": 0.1664662558244401, + "grad_norm": 1.5390769841630867, + "learning_rate": 3.807959071539404e-06, + "loss": 1.0338, + "step": 2215 + }, + { + "epoch": 0.16654140989027508, + "grad_norm": 1.585218294676971, + "learning_rate": 3.8077508560757275e-06, + "loss": 1.1018, + "step": 2216 + }, + { + "epoch": 0.16661656395611002, + "grad_norm": 1.0708676311675318, + "learning_rate": 3.8075425334958908e-06, + "loss": 0.8806, + "step": 2217 + }, + { + "epoch": 0.166691718021945, + "grad_norm": 1.7615896376434088, + "learning_rate": 3.8073341038122374e-06, + "loss": 0.9415, + "step": 2218 + }, + { + "epoch": 0.16676687208777996, + "grad_norm": 1.4803283141670327, + "learning_rate": 3.8071255670371174e-06, + "loss": 0.9302, + "step": 2219 + }, + { + "epoch": 0.1668420261536149, + "grad_norm": 1.6048176505583365, + "learning_rate": 3.8069169231828875e-06, + "loss": 0.9955, + "step": 2220 + }, + { + "epoch": 0.16691718021944987, + "grad_norm": 1.6225013596539317, + "learning_rate": 3.8067081722619114e-06, + "loss": 0.9888, + "step": 2221 + }, + { + "epoch": 0.16699233428528484, + "grad_norm": 1.349104879463391, + "learning_rate": 3.8064993142865573e-06, + "loss": 0.9503, + "step": 2222 + }, + { + "epoch": 0.1670674883511198, + "grad_norm": 1.0660729079408553, + "learning_rate": 3.8062903492692014e-06, + "loss": 0.7975, + "step": 2223 + }, + { + "epoch": 0.16714264241695476, + "grad_norm": 1.6521977930558254, + "learning_rate": 3.8060812772222255e-06, + "loss": 0.9077, + "step": 2224 + }, + { + "epoch": 0.16721779648278973, + "grad_norm": 1.4517713414997135, + "learning_rate": 3.805872098158018e-06, + "loss": 1.0658, + "step": 2225 + }, + { + "epoch": 0.16729295054862467, + "grad_norm": 2.0185563410884617, + "learning_rate": 3.8056628120889736e-06, + "loss": 1.0342, + "step": 2226 + }, + { + "epoch": 0.16736810461445964, + "grad_norm": 1.6512887170195818, + "learning_rate": 3.805453419027493e-06, + "loss": 1.0115, + "step": 2227 + }, + { + "epoch": 0.1674432586802946, + "grad_norm": 1.4817540369495321, + "learning_rate": 3.805243918985984e-06, + "loss": 1.0235, + "step": 2228 + }, + { + "epoch": 0.16751841274612955, + "grad_norm": 1.749960830795083, + "learning_rate": 3.80503431197686e-06, + "loss": 1.011, + "step": 2229 + }, + { + "epoch": 0.16759356681196452, + "grad_norm": 1.6628097331700267, + "learning_rate": 3.804824598012541e-06, + "loss": 0.9712, + "step": 2230 + }, + { + "epoch": 0.1676687208777995, + "grad_norm": 3.4409723950148736, + "learning_rate": 3.8046147771054536e-06, + "loss": 1.0744, + "step": 2231 + }, + { + "epoch": 0.16774387494363446, + "grad_norm": 5.141413387611859, + "learning_rate": 3.8044048492680297e-06, + "loss": 0.9298, + "step": 2232 + }, + { + "epoch": 0.1678190290094694, + "grad_norm": 2.007153948533598, + "learning_rate": 3.80419481451271e-06, + "loss": 1.0712, + "step": 2233 + }, + { + "epoch": 0.16789418307530438, + "grad_norm": 1.662385767403808, + "learning_rate": 3.8039846728519383e-06, + "loss": 0.9161, + "step": 2234 + }, + { + "epoch": 0.16796933714113935, + "grad_norm": 1.5926417068839895, + "learning_rate": 3.803774424298167e-06, + "loss": 0.9904, + "step": 2235 + }, + { + "epoch": 0.1680444912069743, + "grad_norm": 0.9485656626549914, + "learning_rate": 3.8035640688638537e-06, + "loss": 0.8047, + "step": 2236 + }, + { + "epoch": 0.16811964527280926, + "grad_norm": 1.5559465687018792, + "learning_rate": 3.8033536065614625e-06, + "loss": 0.966, + "step": 2237 + }, + { + "epoch": 0.16819479933864423, + "grad_norm": 1.4774462764674092, + "learning_rate": 3.8031430374034653e-06, + "loss": 1.0246, + "step": 2238 + }, + { + "epoch": 0.16826995340447917, + "grad_norm": 2.0582725665225596, + "learning_rate": 3.802932361402338e-06, + "loss": 1.1196, + "step": 2239 + }, + { + "epoch": 0.16834510747031414, + "grad_norm": 2.2938300469590756, + "learning_rate": 3.8027215785705654e-06, + "loss": 0.8658, + "step": 2240 + }, + { + "epoch": 0.1684202615361491, + "grad_norm": 1.4887336584006245, + "learning_rate": 3.8025106889206353e-06, + "loss": 0.9779, + "step": 2241 + }, + { + "epoch": 0.16849541560198406, + "grad_norm": 0.9713369714000734, + "learning_rate": 3.802299692465045e-06, + "loss": 0.8267, + "step": 2242 + }, + { + "epoch": 0.16857056966781903, + "grad_norm": 1.8169764782438587, + "learning_rate": 3.802088589216296e-06, + "loss": 0.9036, + "step": 2243 + }, + { + "epoch": 0.168645723733654, + "grad_norm": 1.6651170281645433, + "learning_rate": 3.801877379186898e-06, + "loss": 0.9179, + "step": 2244 + }, + { + "epoch": 0.16872087779948894, + "grad_norm": 1.5522676738188106, + "learning_rate": 3.8016660623893653e-06, + "loss": 1.0226, + "step": 2245 + }, + { + "epoch": 0.1687960318653239, + "grad_norm": 3.4848860977364495, + "learning_rate": 3.801454638836219e-06, + "loss": 0.9247, + "step": 2246 + }, + { + "epoch": 0.16887118593115888, + "grad_norm": 1.4331664402223625, + "learning_rate": 3.801243108539987e-06, + "loss": 0.9214, + "step": 2247 + }, + { + "epoch": 0.16894633999699385, + "grad_norm": 2.019502444034164, + "learning_rate": 3.8010314715132037e-06, + "loss": 0.967, + "step": 2248 + }, + { + "epoch": 0.1690214940628288, + "grad_norm": 1.8253506859153545, + "learning_rate": 3.8008197277684094e-06, + "loss": 1.0007, + "step": 2249 + }, + { + "epoch": 0.16909664812866376, + "grad_norm": 2.6224399071746034, + "learning_rate": 3.80060787731815e-06, + "loss": 1.0185, + "step": 2250 + }, + { + "epoch": 0.16917180219449873, + "grad_norm": 1.706375170577403, + "learning_rate": 3.8003959201749793e-06, + "loss": 0.9853, + "step": 2251 + }, + { + "epoch": 0.16924695626033368, + "grad_norm": 1.4840681059958982, + "learning_rate": 3.800183856351456e-06, + "loss": 1.047, + "step": 2252 + }, + { + "epoch": 0.16932211032616865, + "grad_norm": 1.6531724227131785, + "learning_rate": 3.7999716858601456e-06, + "loss": 0.9906, + "step": 2253 + }, + { + "epoch": 0.16939726439200362, + "grad_norm": 1.4319147157260423, + "learning_rate": 3.79975940871362e-06, + "loss": 0.8902, + "step": 2254 + }, + { + "epoch": 0.16947241845783856, + "grad_norm": 1.4043733154205282, + "learning_rate": 3.7995470249244582e-06, + "loss": 1.0072, + "step": 2255 + }, + { + "epoch": 0.16954757252367353, + "grad_norm": 1.8819852412432585, + "learning_rate": 3.7993345345052445e-06, + "loss": 1.0346, + "step": 2256 + }, + { + "epoch": 0.1696227265895085, + "grad_norm": 1.8506074526701155, + "learning_rate": 3.799121937468569e-06, + "loss": 1.0299, + "step": 2257 + }, + { + "epoch": 0.16969788065534344, + "grad_norm": 1.5232246445849982, + "learning_rate": 3.7989092338270295e-06, + "loss": 0.9096, + "step": 2258 + }, + { + "epoch": 0.1697730347211784, + "grad_norm": 1.8517068412307052, + "learning_rate": 3.7986964235932293e-06, + "loss": 1.0278, + "step": 2259 + }, + { + "epoch": 0.16984818878701338, + "grad_norm": 1.7444510359885914, + "learning_rate": 3.7984835067797788e-06, + "loss": 0.9753, + "step": 2260 + }, + { + "epoch": 0.16992334285284835, + "grad_norm": 1.6522039981117171, + "learning_rate": 3.7982704833992933e-06, + "loss": 1.0123, + "step": 2261 + }, + { + "epoch": 0.1699984969186833, + "grad_norm": 1.5628113245754631, + "learning_rate": 3.7980573534643954e-06, + "loss": 1.0343, + "step": 2262 + }, + { + "epoch": 0.17007365098451827, + "grad_norm": 1.770805859820104, + "learning_rate": 3.7978441169877143e-06, + "loss": 0.8751, + "step": 2263 + }, + { + "epoch": 0.17014880505035324, + "grad_norm": 1.9748088309563643, + "learning_rate": 3.7976307739818852e-06, + "loss": 0.9458, + "step": 2264 + }, + { + "epoch": 0.17022395911618818, + "grad_norm": 1.4410816023446031, + "learning_rate": 3.7974173244595493e-06, + "loss": 1.0448, + "step": 2265 + }, + { + "epoch": 0.17029911318202315, + "grad_norm": 1.6890993663000498, + "learning_rate": 3.7972037684333534e-06, + "loss": 0.9932, + "step": 2266 + }, + { + "epoch": 0.17037426724785812, + "grad_norm": 1.462866015157233, + "learning_rate": 3.7969901059159524e-06, + "loss": 0.8773, + "step": 2267 + }, + { + "epoch": 0.17044942131369306, + "grad_norm": 1.4831581348008471, + "learning_rate": 3.796776336920007e-06, + "loss": 0.8796, + "step": 2268 + }, + { + "epoch": 0.17052457537952803, + "grad_norm": 1.70166856279308, + "learning_rate": 3.796562461458183e-06, + "loss": 0.9914, + "step": 2269 + }, + { + "epoch": 0.170599729445363, + "grad_norm": 2.1267483194913375, + "learning_rate": 3.7963484795431537e-06, + "loss": 0.9573, + "step": 2270 + }, + { + "epoch": 0.17067488351119794, + "grad_norm": 0.8895906163676127, + "learning_rate": 3.796134391187598e-06, + "loss": 0.7857, + "step": 2271 + }, + { + "epoch": 0.17075003757703291, + "grad_norm": 1.6005091482924114, + "learning_rate": 3.7959201964042024e-06, + "loss": 0.8549, + "step": 2272 + }, + { + "epoch": 0.17082519164286789, + "grad_norm": 1.473544068189135, + "learning_rate": 3.7957058952056577e-06, + "loss": 0.9119, + "step": 2273 + }, + { + "epoch": 0.17090034570870283, + "grad_norm": 1.7338870625131664, + "learning_rate": 3.7954914876046626e-06, + "loss": 0.9448, + "step": 2274 + }, + { + "epoch": 0.1709754997745378, + "grad_norm": 1.637332264120662, + "learning_rate": 3.795276973613921e-06, + "loss": 0.7988, + "step": 2275 + }, + { + "epoch": 0.17105065384037277, + "grad_norm": 1.4485886018653549, + "learning_rate": 3.795062353246145e-06, + "loss": 0.9849, + "step": 2276 + }, + { + "epoch": 0.17112580790620774, + "grad_norm": 1.7794630222235408, + "learning_rate": 3.79484762651405e-06, + "loss": 0.9657, + "step": 2277 + }, + { + "epoch": 0.17120096197204268, + "grad_norm": 3.87118428544956, + "learning_rate": 3.7946327934303612e-06, + "loss": 0.9932, + "step": 2278 + }, + { + "epoch": 0.17127611603787765, + "grad_norm": 1.8097838076677049, + "learning_rate": 3.7944178540078065e-06, + "loss": 1.0393, + "step": 2279 + }, + { + "epoch": 0.17135127010371262, + "grad_norm": 1.6509353009768355, + "learning_rate": 3.7942028082591227e-06, + "loss": 0.9443, + "step": 2280 + }, + { + "epoch": 0.17142642416954756, + "grad_norm": 1.67539393788653, + "learning_rate": 3.7939876561970526e-06, + "loss": 1.079, + "step": 2281 + }, + { + "epoch": 0.17150157823538253, + "grad_norm": 3.7559167596695224, + "learning_rate": 3.7937723978343437e-06, + "loss": 0.9902, + "step": 2282 + }, + { + "epoch": 0.1715767323012175, + "grad_norm": 1.3844636445264433, + "learning_rate": 3.7935570331837514e-06, + "loss": 0.9816, + "step": 2283 + }, + { + "epoch": 0.17165188636705245, + "grad_norm": 1.4621810924033354, + "learning_rate": 3.793341562258037e-06, + "loss": 1.044, + "step": 2284 + }, + { + "epoch": 0.17172704043288742, + "grad_norm": 1.861458004305881, + "learning_rate": 3.7931259850699678e-06, + "loss": 1.0126, + "step": 2285 + }, + { + "epoch": 0.1718021944987224, + "grad_norm": 0.9311738194278029, + "learning_rate": 3.7929103016323183e-06, + "loss": 0.8239, + "step": 2286 + }, + { + "epoch": 0.17187734856455733, + "grad_norm": 1.520592875174634, + "learning_rate": 3.792694511957867e-06, + "loss": 1.0113, + "step": 2287 + }, + { + "epoch": 0.1719525026303923, + "grad_norm": 1.4586960840925949, + "learning_rate": 3.7924786160594016e-06, + "loss": 1.0748, + "step": 2288 + }, + { + "epoch": 0.17202765669622727, + "grad_norm": 1.626325946650091, + "learning_rate": 3.792262613949714e-06, + "loss": 1.0179, + "step": 2289 + }, + { + "epoch": 0.1721028107620622, + "grad_norm": 1.5210557365663675, + "learning_rate": 3.792046505641604e-06, + "loss": 1.0013, + "step": 2290 + }, + { + "epoch": 0.17217796482789718, + "grad_norm": 1.7807370836056469, + "learning_rate": 3.7918302911478764e-06, + "loss": 1.0026, + "step": 2291 + }, + { + "epoch": 0.17225311889373215, + "grad_norm": 1.5179049460312584, + "learning_rate": 3.791613970481342e-06, + "loss": 0.9623, + "step": 2292 + }, + { + "epoch": 0.17232827295956712, + "grad_norm": 1.4851213686888787, + "learning_rate": 3.7913975436548195e-06, + "loss": 0.9058, + "step": 2293 + }, + { + "epoch": 0.17240342702540207, + "grad_norm": 1.336482154512983, + "learning_rate": 3.7911810106811332e-06, + "loss": 1.0088, + "step": 2294 + }, + { + "epoch": 0.17247858109123704, + "grad_norm": 1.749541154380128, + "learning_rate": 3.7909643715731133e-06, + "loss": 1.0052, + "step": 2295 + }, + { + "epoch": 0.172553735157072, + "grad_norm": 2.0097384585369342, + "learning_rate": 3.790747626343596e-06, + "loss": 1.0204, + "step": 2296 + }, + { + "epoch": 0.17262888922290695, + "grad_norm": 1.5626397642569172, + "learning_rate": 3.7905307750054247e-06, + "loss": 1.0462, + "step": 2297 + }, + { + "epoch": 0.17270404328874192, + "grad_norm": 1.621091543908514, + "learning_rate": 3.790313817571448e-06, + "loss": 1.048, + "step": 2298 + }, + { + "epoch": 0.1727791973545769, + "grad_norm": 1.5462334583726798, + "learning_rate": 3.790096754054523e-06, + "loss": 0.9209, + "step": 2299 + }, + { + "epoch": 0.17285435142041183, + "grad_norm": 1.4488752014460162, + "learning_rate": 3.7898795844675096e-06, + "loss": 0.9941, + "step": 2300 + }, + { + "epoch": 0.1729295054862468, + "grad_norm": 0.952392077451964, + "learning_rate": 3.789662308823278e-06, + "loss": 0.8447, + "step": 2301 + }, + { + "epoch": 0.17300465955208177, + "grad_norm": 1.780066984073548, + "learning_rate": 3.7894449271347e-06, + "loss": 0.9826, + "step": 2302 + }, + { + "epoch": 0.17307981361791672, + "grad_norm": 2.1776414244106688, + "learning_rate": 3.7892274394146592e-06, + "loss": 1.1288, + "step": 2303 + }, + { + "epoch": 0.1731549676837517, + "grad_norm": 1.3223592479924045, + "learning_rate": 3.789009845676041e-06, + "loss": 0.9607, + "step": 2304 + }, + { + "epoch": 0.17323012174958666, + "grad_norm": 1.8467518566923204, + "learning_rate": 3.7887921459317386e-06, + "loss": 1.014, + "step": 2305 + }, + { + "epoch": 0.17330527581542163, + "grad_norm": 1.7052207441427878, + "learning_rate": 3.7885743401946517e-06, + "loss": 0.8854, + "step": 2306 + }, + { + "epoch": 0.17338042988125657, + "grad_norm": 1.5984765817109872, + "learning_rate": 3.7883564284776863e-06, + "loss": 1.0317, + "step": 2307 + }, + { + "epoch": 0.17345558394709154, + "grad_norm": 1.844914085021202, + "learning_rate": 3.7881384107937546e-06, + "loss": 0.9626, + "step": 2308 + }, + { + "epoch": 0.1735307380129265, + "grad_norm": 1.6437274481673179, + "learning_rate": 3.7879202871557742e-06, + "loss": 0.9885, + "step": 2309 + }, + { + "epoch": 0.17360589207876145, + "grad_norm": 0.9048834725173134, + "learning_rate": 3.7877020575766714e-06, + "loss": 0.7832, + "step": 2310 + }, + { + "epoch": 0.17368104614459642, + "grad_norm": 1.2587692697180832, + "learning_rate": 3.7874837220693756e-06, + "loss": 0.962, + "step": 2311 + }, + { + "epoch": 0.1737562002104314, + "grad_norm": 0.8542157058284783, + "learning_rate": 3.7872652806468244e-06, + "loss": 0.825, + "step": 2312 + }, + { + "epoch": 0.17383135427626634, + "grad_norm": 4.1615053380317555, + "learning_rate": 3.7870467333219614e-06, + "loss": 0.9611, + "step": 2313 + }, + { + "epoch": 0.1739065083421013, + "grad_norm": 1.5733165948322234, + "learning_rate": 3.7868280801077368e-06, + "loss": 0.9969, + "step": 2314 + }, + { + "epoch": 0.17398166240793628, + "grad_norm": 1.3987895160237438, + "learning_rate": 3.786609321017106e-06, + "loss": 1.0471, + "step": 2315 + }, + { + "epoch": 0.17405681647377122, + "grad_norm": 1.847091556252246, + "learning_rate": 3.7863904560630315e-06, + "loss": 1.0322, + "step": 2316 + }, + { + "epoch": 0.1741319705396062, + "grad_norm": 0.7974469992984258, + "learning_rate": 3.786171485258482e-06, + "loss": 0.8397, + "step": 2317 + }, + { + "epoch": 0.17420712460544116, + "grad_norm": 1.5956241583457726, + "learning_rate": 3.785952408616432e-06, + "loss": 1.0682, + "step": 2318 + }, + { + "epoch": 0.1742822786712761, + "grad_norm": 1.6868585851039548, + "learning_rate": 3.7857332261498635e-06, + "loss": 1.0107, + "step": 2319 + }, + { + "epoch": 0.17435743273711107, + "grad_norm": 1.2996042786370148, + "learning_rate": 3.785513937871763e-06, + "loss": 0.9977, + "step": 2320 + }, + { + "epoch": 0.17443258680294604, + "grad_norm": 1.4484779130056944, + "learning_rate": 3.785294543795125e-06, + "loss": 1.0437, + "step": 2321 + }, + { + "epoch": 0.174507740868781, + "grad_norm": 1.465365792393687, + "learning_rate": 3.7850750439329477e-06, + "loss": 0.973, + "step": 2322 + }, + { + "epoch": 0.17458289493461596, + "grad_norm": 1.6778941672349421, + "learning_rate": 3.7848554382982398e-06, + "loss": 0.965, + "step": 2323 + }, + { + "epoch": 0.17465804900045093, + "grad_norm": 1.6001643334873574, + "learning_rate": 3.7846357269040115e-06, + "loss": 0.9948, + "step": 2324 + }, + { + "epoch": 0.1747332030662859, + "grad_norm": 1.7863202129683462, + "learning_rate": 3.784415909763283e-06, + "loss": 0.9164, + "step": 2325 + }, + { + "epoch": 0.17480835713212084, + "grad_norm": 1.6524372480195968, + "learning_rate": 3.784195986889079e-06, + "loss": 1.0393, + "step": 2326 + }, + { + "epoch": 0.1748835111979558, + "grad_norm": 1.5012256859836093, + "learning_rate": 3.7839759582944307e-06, + "loss": 1.0601, + "step": 2327 + }, + { + "epoch": 0.17495866526379078, + "grad_norm": 1.6400332882456854, + "learning_rate": 3.783755823992376e-06, + "loss": 0.8905, + "step": 2328 + }, + { + "epoch": 0.17503381932962572, + "grad_norm": 1.498585510899456, + "learning_rate": 3.783535583995957e-06, + "loss": 1.0235, + "step": 2329 + }, + { + "epoch": 0.1751089733954607, + "grad_norm": 2.2441043967036034, + "learning_rate": 3.783315238318226e-06, + "loss": 1.0087, + "step": 2330 + }, + { + "epoch": 0.17518412746129566, + "grad_norm": 1.4220866789817108, + "learning_rate": 3.7830947869722377e-06, + "loss": 0.9329, + "step": 2331 + }, + { + "epoch": 0.1752592815271306, + "grad_norm": 0.9409683918767454, + "learning_rate": 3.7828742299710558e-06, + "loss": 0.8215, + "step": 2332 + }, + { + "epoch": 0.17533443559296558, + "grad_norm": 1.758338612131207, + "learning_rate": 3.782653567327749e-06, + "loss": 1.0801, + "step": 2333 + }, + { + "epoch": 0.17540958965880055, + "grad_norm": 1.5235094884834237, + "learning_rate": 3.7824327990553914e-06, + "loss": 1.0027, + "step": 2334 + }, + { + "epoch": 0.1754847437246355, + "grad_norm": 2.2374276602362007, + "learning_rate": 3.7822119251670657e-06, + "loss": 0.9564, + "step": 2335 + }, + { + "epoch": 0.17555989779047046, + "grad_norm": 1.4476219901957876, + "learning_rate": 3.7819909456758582e-06, + "loss": 0.9759, + "step": 2336 + }, + { + "epoch": 0.17563505185630543, + "grad_norm": 1.4349373905466463, + "learning_rate": 3.7817698605948643e-06, + "loss": 1.0742, + "step": 2337 + }, + { + "epoch": 0.1757102059221404, + "grad_norm": 1.7269448249719255, + "learning_rate": 3.7815486699371826e-06, + "loss": 1.033, + "step": 2338 + }, + { + "epoch": 0.17578535998797534, + "grad_norm": 1.6550683178587478, + "learning_rate": 3.7813273737159205e-06, + "loss": 0.9962, + "step": 2339 + }, + { + "epoch": 0.1758605140538103, + "grad_norm": 1.5055668369838684, + "learning_rate": 3.78110597194419e-06, + "loss": 1.0231, + "step": 2340 + }, + { + "epoch": 0.17593566811964528, + "grad_norm": 1.4451304553057391, + "learning_rate": 3.780884464635111e-06, + "loss": 0.9667, + "step": 2341 + }, + { + "epoch": 0.17601082218548023, + "grad_norm": 1.3779419290822057, + "learning_rate": 3.7806628518018074e-06, + "loss": 0.9884, + "step": 2342 + }, + { + "epoch": 0.1760859762513152, + "grad_norm": 1.8768704303859858, + "learning_rate": 3.7804411334574116e-06, + "loss": 0.9177, + "step": 2343 + }, + { + "epoch": 0.17616113031715017, + "grad_norm": 1.7670856611756247, + "learning_rate": 3.7802193096150606e-06, + "loss": 0.9785, + "step": 2344 + }, + { + "epoch": 0.1762362843829851, + "grad_norm": 1.8539714353268264, + "learning_rate": 3.7799973802878985e-06, + "loss": 0.951, + "step": 2345 + }, + { + "epoch": 0.17631143844882008, + "grad_norm": 1.4469195530007308, + "learning_rate": 3.779775345489076e-06, + "loss": 0.9425, + "step": 2346 + }, + { + "epoch": 0.17638659251465505, + "grad_norm": 1.713284350460672, + "learning_rate": 3.779553205231749e-06, + "loss": 0.972, + "step": 2347 + }, + { + "epoch": 0.17646174658049, + "grad_norm": 1.3444111993758465, + "learning_rate": 3.77933095952908e-06, + "loss": 0.9994, + "step": 2348 + }, + { + "epoch": 0.17653690064632496, + "grad_norm": 1.5548528842084617, + "learning_rate": 3.779108608394238e-06, + "loss": 0.9342, + "step": 2349 + }, + { + "epoch": 0.17661205471215993, + "grad_norm": 1.9572193413229975, + "learning_rate": 3.7788861518403988e-06, + "loss": 0.9135, + "step": 2350 + }, + { + "epoch": 0.1766872087779949, + "grad_norm": 1.5981972671418592, + "learning_rate": 3.778663589880743e-06, + "loss": 1.0576, + "step": 2351 + }, + { + "epoch": 0.17676236284382985, + "grad_norm": 1.5865278225877706, + "learning_rate": 3.7784409225284585e-06, + "loss": 0.9337, + "step": 2352 + }, + { + "epoch": 0.17683751690966482, + "grad_norm": 1.5257764809373273, + "learning_rate": 3.7782181497967393e-06, + "loss": 0.9446, + "step": 2353 + }, + { + "epoch": 0.1769126709754998, + "grad_norm": 4.544051404608514, + "learning_rate": 3.7779952716987856e-06, + "loss": 0.9415, + "step": 2354 + }, + { + "epoch": 0.17698782504133473, + "grad_norm": 2.1806370570777998, + "learning_rate": 3.7777722882478032e-06, + "loss": 0.9393, + "step": 2355 + }, + { + "epoch": 0.1770629791071697, + "grad_norm": 1.6016306792843058, + "learning_rate": 3.7775491994570057e-06, + "loss": 0.9886, + "step": 2356 + }, + { + "epoch": 0.17713813317300467, + "grad_norm": 1.4676858821295158, + "learning_rate": 3.777326005339611e-06, + "loss": 0.9326, + "step": 2357 + }, + { + "epoch": 0.1772132872388396, + "grad_norm": 0.8407650884022466, + "learning_rate": 3.7771027059088454e-06, + "loss": 0.8194, + "step": 2358 + }, + { + "epoch": 0.17728844130467458, + "grad_norm": 2.4399822018709023, + "learning_rate": 3.7768793011779383e-06, + "loss": 1.0886, + "step": 2359 + }, + { + "epoch": 0.17736359537050955, + "grad_norm": 1.5186288698972747, + "learning_rate": 3.7766557911601295e-06, + "loss": 0.9965, + "step": 2360 + }, + { + "epoch": 0.1774387494363445, + "grad_norm": 1.457663718396728, + "learning_rate": 3.7764321758686614e-06, + "loss": 1.0284, + "step": 2361 + }, + { + "epoch": 0.17751390350217947, + "grad_norm": 1.8184136567992133, + "learning_rate": 3.7762084553167846e-06, + "loss": 0.9848, + "step": 2362 + }, + { + "epoch": 0.17758905756801444, + "grad_norm": 1.489500572555345, + "learning_rate": 3.7759846295177552e-06, + "loss": 1.0064, + "step": 2363 + }, + { + "epoch": 0.17766421163384938, + "grad_norm": 1.9755418879094946, + "learning_rate": 3.775760698484836e-06, + "loss": 1.0214, + "step": 2364 + }, + { + "epoch": 0.17773936569968435, + "grad_norm": 1.4994580561415132, + "learning_rate": 3.7755366622312954e-06, + "loss": 1.0081, + "step": 2365 + }, + { + "epoch": 0.17781451976551932, + "grad_norm": 1.5105625827506286, + "learning_rate": 3.7753125207704084e-06, + "loss": 1.0205, + "step": 2366 + }, + { + "epoch": 0.1778896738313543, + "grad_norm": 1.6388150358352487, + "learning_rate": 3.7750882741154566e-06, + "loss": 1.0273, + "step": 2367 + }, + { + "epoch": 0.17796482789718923, + "grad_norm": 1.6376619011434153, + "learning_rate": 3.774863922279727e-06, + "loss": 1.0605, + "step": 2368 + }, + { + "epoch": 0.1780399819630242, + "grad_norm": 1.547916947677539, + "learning_rate": 3.7746394652765136e-06, + "loss": 0.869, + "step": 2369 + }, + { + "epoch": 0.17811513602885917, + "grad_norm": 0.9269938193550357, + "learning_rate": 3.774414903119117e-06, + "loss": 0.8533, + "step": 2370 + }, + { + "epoch": 0.17819029009469411, + "grad_norm": 1.8253977720943968, + "learning_rate": 3.7741902358208427e-06, + "loss": 0.9722, + "step": 2371 + }, + { + "epoch": 0.17826544416052909, + "grad_norm": 1.750410985370664, + "learning_rate": 3.7739654633950023e-06, + "loss": 1.0728, + "step": 2372 + }, + { + "epoch": 0.17834059822636406, + "grad_norm": 1.358936677618181, + "learning_rate": 3.7737405858549156e-06, + "loss": 0.9857, + "step": 2373 + }, + { + "epoch": 0.178415752292199, + "grad_norm": 1.4025830248025368, + "learning_rate": 3.7735156032139066e-06, + "loss": 1.0471, + "step": 2374 + }, + { + "epoch": 0.17849090635803397, + "grad_norm": 1.9402961857816863, + "learning_rate": 3.773290515485308e-06, + "loss": 0.877, + "step": 2375 + }, + { + "epoch": 0.17856606042386894, + "grad_norm": 2.2256657450930533, + "learning_rate": 3.773065322682455e-06, + "loss": 0.9784, + "step": 2376 + }, + { + "epoch": 0.17864121448970388, + "grad_norm": 1.4267124427302555, + "learning_rate": 3.772840024818692e-06, + "loss": 0.8893, + "step": 2377 + }, + { + "epoch": 0.17871636855553885, + "grad_norm": 1.352052632802768, + "learning_rate": 3.7726146219073697e-06, + "loss": 1.0481, + "step": 2378 + }, + { + "epoch": 0.17879152262137382, + "grad_norm": 1.6438000428447261, + "learning_rate": 3.772389113961843e-06, + "loss": 0.9976, + "step": 2379 + }, + { + "epoch": 0.17886667668720876, + "grad_norm": 2.3920301907653077, + "learning_rate": 3.772163500995474e-06, + "loss": 1.0033, + "step": 2380 + }, + { + "epoch": 0.17894183075304373, + "grad_norm": 1.4861056227696563, + "learning_rate": 3.771937783021632e-06, + "loss": 1.0123, + "step": 2381 + }, + { + "epoch": 0.1790169848188787, + "grad_norm": 3.0004620196292757, + "learning_rate": 3.771711960053691e-06, + "loss": 0.9953, + "step": 2382 + }, + { + "epoch": 0.17909213888471368, + "grad_norm": 1.9579853465629784, + "learning_rate": 3.7714860321050316e-06, + "loss": 0.8735, + "step": 2383 + }, + { + "epoch": 0.17916729295054862, + "grad_norm": 1.7016649900173455, + "learning_rate": 3.771259999189042e-06, + "loss": 0.9749, + "step": 2384 + }, + { + "epoch": 0.1792424470163836, + "grad_norm": 1.3984079624231949, + "learning_rate": 3.7710338613191145e-06, + "loss": 0.9796, + "step": 2385 + }, + { + "epoch": 0.17931760108221856, + "grad_norm": 2.174193647145835, + "learning_rate": 3.770807618508649e-06, + "loss": 0.9733, + "step": 2386 + }, + { + "epoch": 0.1793927551480535, + "grad_norm": 1.5416444914314116, + "learning_rate": 3.770581270771051e-06, + "loss": 1.0242, + "step": 2387 + }, + { + "epoch": 0.17946790921388847, + "grad_norm": 1.84369387783692, + "learning_rate": 3.770354818119733e-06, + "loss": 1.0675, + "step": 2388 + }, + { + "epoch": 0.17954306327972344, + "grad_norm": 0.8574267411148221, + "learning_rate": 3.7701282605681123e-06, + "loss": 0.8179, + "step": 2389 + }, + { + "epoch": 0.17961821734555838, + "grad_norm": 4.517571791905418, + "learning_rate": 3.769901598129615e-06, + "loss": 1.0622, + "step": 2390 + }, + { + "epoch": 0.17969337141139335, + "grad_norm": 1.3431418617436237, + "learning_rate": 3.7696748308176698e-06, + "loss": 1.0143, + "step": 2391 + }, + { + "epoch": 0.17976852547722832, + "grad_norm": 1.6622228348374442, + "learning_rate": 3.7694479586457144e-06, + "loss": 1.0869, + "step": 2392 + }, + { + "epoch": 0.17984367954306327, + "grad_norm": 1.4690765667119956, + "learning_rate": 3.7692209816271915e-06, + "loss": 0.9993, + "step": 2393 + }, + { + "epoch": 0.17991883360889824, + "grad_norm": 2.230660932983095, + "learning_rate": 3.7689938997755512e-06, + "loss": 1.0428, + "step": 2394 + }, + { + "epoch": 0.1799939876747332, + "grad_norm": 1.5036096850195677, + "learning_rate": 3.7687667131042487e-06, + "loss": 0.9349, + "step": 2395 + }, + { + "epoch": 0.18006914174056818, + "grad_norm": 1.7488077053013746, + "learning_rate": 3.7685394216267444e-06, + "loss": 1.0288, + "step": 2396 + }, + { + "epoch": 0.18014429580640312, + "grad_norm": 1.818588619347847, + "learning_rate": 3.7683120253565076e-06, + "loss": 0.9643, + "step": 2397 + }, + { + "epoch": 0.1802194498722381, + "grad_norm": 1.6397607616054555, + "learning_rate": 3.7680845243070128e-06, + "loss": 0.9735, + "step": 2398 + }, + { + "epoch": 0.18029460393807306, + "grad_norm": 2.3226564429803043, + "learning_rate": 3.767856918491739e-06, + "loss": 0.974, + "step": 2399 + }, + { + "epoch": 0.180369758003908, + "grad_norm": 1.486509695504574, + "learning_rate": 3.767629207924172e-06, + "loss": 1.0843, + "step": 2400 + }, + { + "epoch": 0.18044491206974297, + "grad_norm": 1.4742731986935076, + "learning_rate": 3.767401392617807e-06, + "loss": 0.9686, + "step": 2401 + }, + { + "epoch": 0.18052006613557794, + "grad_norm": 1.4266140948321415, + "learning_rate": 3.7671734725861413e-06, + "loss": 0.9675, + "step": 2402 + }, + { + "epoch": 0.1805952202014129, + "grad_norm": 1.5994862566134211, + "learning_rate": 3.7669454478426806e-06, + "loss": 0.9748, + "step": 2403 + }, + { + "epoch": 0.18067037426724786, + "grad_norm": 1.9133248912589593, + "learning_rate": 3.7667173184009356e-06, + "loss": 0.9288, + "step": 2404 + }, + { + "epoch": 0.18074552833308283, + "grad_norm": 1.6205469886961072, + "learning_rate": 3.7664890842744248e-06, + "loss": 0.9304, + "step": 2405 + }, + { + "epoch": 0.18082068239891777, + "grad_norm": 2.7804639755683933, + "learning_rate": 3.7662607454766712e-06, + "loss": 0.8896, + "step": 2406 + }, + { + "epoch": 0.18089583646475274, + "grad_norm": 1.4574160642975829, + "learning_rate": 3.7660323020212047e-06, + "loss": 1.0929, + "step": 2407 + }, + { + "epoch": 0.1809709905305877, + "grad_norm": 1.5215280940345068, + "learning_rate": 3.765803753921562e-06, + "loss": 1.0108, + "step": 2408 + }, + { + "epoch": 0.18104614459642265, + "grad_norm": 1.6410792130733796, + "learning_rate": 3.7655751011912852e-06, + "loss": 0.9544, + "step": 2409 + }, + { + "epoch": 0.18112129866225762, + "grad_norm": 1.496386865333152, + "learning_rate": 3.7653463438439225e-06, + "loss": 0.932, + "step": 2410 + }, + { + "epoch": 0.1811964527280926, + "grad_norm": 1.7823252193370158, + "learning_rate": 3.7651174818930293e-06, + "loss": 1.0938, + "step": 2411 + }, + { + "epoch": 0.18127160679392756, + "grad_norm": 1.5453395736991826, + "learning_rate": 3.764888515352166e-06, + "loss": 0.9348, + "step": 2412 + }, + { + "epoch": 0.1813467608597625, + "grad_norm": 1.5042150403360692, + "learning_rate": 3.7646594442349004e-06, + "loss": 1.0155, + "step": 2413 + }, + { + "epoch": 0.18142191492559748, + "grad_norm": 0.9879178672042067, + "learning_rate": 3.764430268554805e-06, + "loss": 0.8457, + "step": 2414 + }, + { + "epoch": 0.18149706899143245, + "grad_norm": 0.8836057694188482, + "learning_rate": 3.7642009883254594e-06, + "loss": 0.8277, + "step": 2415 + }, + { + "epoch": 0.1815722230572674, + "grad_norm": 1.751031503490914, + "learning_rate": 3.7639716035604502e-06, + "loss": 1.0349, + "step": 2416 + }, + { + "epoch": 0.18164737712310236, + "grad_norm": 1.7986047131765401, + "learning_rate": 3.763742114273369e-06, + "loss": 0.9848, + "step": 2417 + }, + { + "epoch": 0.18172253118893733, + "grad_norm": 1.3160060647836325, + "learning_rate": 3.763512520477813e-06, + "loss": 0.9991, + "step": 2418 + }, + { + "epoch": 0.18179768525477227, + "grad_norm": 1.3498073493402483, + "learning_rate": 3.7632828221873876e-06, + "loss": 0.9594, + "step": 2419 + }, + { + "epoch": 0.18187283932060724, + "grad_norm": 1.6156237131847813, + "learning_rate": 3.763053019415703e-06, + "loss": 0.951, + "step": 2420 + }, + { + "epoch": 0.1819479933864422, + "grad_norm": 1.4782941241621534, + "learning_rate": 3.7628231121763757e-06, + "loss": 1.0155, + "step": 2421 + }, + { + "epoch": 0.18202314745227716, + "grad_norm": 1.4657882697093163, + "learning_rate": 3.7625931004830287e-06, + "loss": 1.0755, + "step": 2422 + }, + { + "epoch": 0.18209830151811213, + "grad_norm": 1.8082861752081651, + "learning_rate": 3.762362984349291e-06, + "loss": 0.9143, + "step": 2423 + }, + { + "epoch": 0.1821734555839471, + "grad_norm": 1.802915355322496, + "learning_rate": 3.762132763788798e-06, + "loss": 0.9944, + "step": 2424 + }, + { + "epoch": 0.18224860964978204, + "grad_norm": 1.880488310562799, + "learning_rate": 3.7619024388151914e-06, + "loss": 0.9585, + "step": 2425 + }, + { + "epoch": 0.182323763715617, + "grad_norm": 1.8021414495663195, + "learning_rate": 3.761672009442118e-06, + "loss": 0.9804, + "step": 2426 + }, + { + "epoch": 0.18239891778145198, + "grad_norm": 1.8749156932950641, + "learning_rate": 3.7614414756832328e-06, + "loss": 0.9694, + "step": 2427 + }, + { + "epoch": 0.18247407184728695, + "grad_norm": 1.7447080981230785, + "learning_rate": 3.7612108375521942e-06, + "loss": 0.9937, + "step": 2428 + }, + { + "epoch": 0.1825492259131219, + "grad_norm": 1.4342865062666927, + "learning_rate": 3.76098009506267e-06, + "loss": 0.9549, + "step": 2429 + }, + { + "epoch": 0.18262437997895686, + "grad_norm": 1.50351303706909, + "learning_rate": 3.7607492482283315e-06, + "loss": 0.8921, + "step": 2430 + }, + { + "epoch": 0.18269953404479183, + "grad_norm": 1.7993625836806901, + "learning_rate": 3.7605182970628583e-06, + "loss": 0.9164, + "step": 2431 + }, + { + "epoch": 0.18277468811062678, + "grad_norm": 1.4398204484897275, + "learning_rate": 3.7602872415799347e-06, + "loss": 0.9654, + "step": 2432 + }, + { + "epoch": 0.18284984217646175, + "grad_norm": 1.9564892899775543, + "learning_rate": 3.7600560817932506e-06, + "loss": 0.995, + "step": 2433 + }, + { + "epoch": 0.18292499624229672, + "grad_norm": 1.875144825746319, + "learning_rate": 3.759824817716504e-06, + "loss": 0.914, + "step": 2434 + }, + { + "epoch": 0.18300015030813166, + "grad_norm": 2.531364249959476, + "learning_rate": 3.7595934493633986e-06, + "loss": 0.9836, + "step": 2435 + }, + { + "epoch": 0.18307530437396663, + "grad_norm": 1.8495347391636878, + "learning_rate": 3.7593619767476435e-06, + "loss": 1.0944, + "step": 2436 + }, + { + "epoch": 0.1831504584398016, + "grad_norm": 1.0045858969564068, + "learning_rate": 3.759130399882954e-06, + "loss": 0.7896, + "step": 2437 + }, + { + "epoch": 0.18322561250563654, + "grad_norm": 1.6355559741255206, + "learning_rate": 3.758898718783052e-06, + "loss": 0.8984, + "step": 2438 + }, + { + "epoch": 0.1833007665714715, + "grad_norm": 1.541570024640185, + "learning_rate": 3.758666933461666e-06, + "loss": 1.0424, + "step": 2439 + }, + { + "epoch": 0.18337592063730648, + "grad_norm": 1.4585715326416184, + "learning_rate": 3.7584350439325295e-06, + "loss": 0.9165, + "step": 2440 + }, + { + "epoch": 0.18345107470314145, + "grad_norm": 1.5421454055745893, + "learning_rate": 3.7582030502093833e-06, + "loss": 0.8439, + "step": 2441 + }, + { + "epoch": 0.1835262287689764, + "grad_norm": 1.881979983824318, + "learning_rate": 3.7579709523059736e-06, + "loss": 0.9172, + "step": 2442 + }, + { + "epoch": 0.18360138283481137, + "grad_norm": 1.7338896063584797, + "learning_rate": 3.7577387502360535e-06, + "loss": 0.9728, + "step": 2443 + }, + { + "epoch": 0.18367653690064634, + "grad_norm": 1.5966675177527425, + "learning_rate": 3.757506444013381e-06, + "loss": 1.0361, + "step": 2444 + }, + { + "epoch": 0.18375169096648128, + "grad_norm": 1.381735689550497, + "learning_rate": 3.7572740336517225e-06, + "loss": 0.9696, + "step": 2445 + }, + { + "epoch": 0.18382684503231625, + "grad_norm": 1.6865733392751299, + "learning_rate": 3.757041519164848e-06, + "loss": 1.079, + "step": 2446 + }, + { + "epoch": 0.18390199909815122, + "grad_norm": 0.9368493365334956, + "learning_rate": 3.7568089005665353e-06, + "loss": 0.8095, + "step": 2447 + }, + { + "epoch": 0.18397715316398616, + "grad_norm": 1.736032192555797, + "learning_rate": 3.7565761778705682e-06, + "loss": 1.0577, + "step": 2448 + }, + { + "epoch": 0.18405230722982113, + "grad_norm": 1.4584630726823535, + "learning_rate": 3.756343351090736e-06, + "loss": 0.9725, + "step": 2449 + }, + { + "epoch": 0.1841274612956561, + "grad_norm": 1.9123965740287268, + "learning_rate": 3.756110420240835e-06, + "loss": 0.9622, + "step": 2450 + }, + { + "epoch": 0.18420261536149105, + "grad_norm": 1.653552756105508, + "learning_rate": 3.755877385334667e-06, + "loss": 1.0189, + "step": 2451 + }, + { + "epoch": 0.18427776942732602, + "grad_norm": 1.352357531322512, + "learning_rate": 3.7556442463860406e-06, + "loss": 1.0065, + "step": 2452 + }, + { + "epoch": 0.184352923493161, + "grad_norm": 3.5186560555582718, + "learning_rate": 3.7554110034087686e-06, + "loss": 0.9871, + "step": 2453 + }, + { + "epoch": 0.18442807755899593, + "grad_norm": 3.2048332311414756, + "learning_rate": 3.7551776564166736e-06, + "loss": 0.9695, + "step": 2454 + }, + { + "epoch": 0.1845032316248309, + "grad_norm": 2.0374964989006057, + "learning_rate": 3.7549442054235813e-06, + "loss": 0.9952, + "step": 2455 + }, + { + "epoch": 0.18457838569066587, + "grad_norm": 1.5156432475141872, + "learning_rate": 3.754710650443325e-06, + "loss": 0.8755, + "step": 2456 + }, + { + "epoch": 0.18465353975650084, + "grad_norm": 1.8335818497962788, + "learning_rate": 3.754476991489743e-06, + "loss": 1.0143, + "step": 2457 + }, + { + "epoch": 0.18472869382233578, + "grad_norm": 1.7143559555013905, + "learning_rate": 3.754243228576681e-06, + "loss": 1.0325, + "step": 2458 + }, + { + "epoch": 0.18480384788817075, + "grad_norm": 1.8375245073202267, + "learning_rate": 3.7540093617179904e-06, + "loss": 1.0471, + "step": 2459 + }, + { + "epoch": 0.18487900195400572, + "grad_norm": 1.5493148673861725, + "learning_rate": 3.7537753909275284e-06, + "loss": 1.0191, + "step": 2460 + }, + { + "epoch": 0.18495415601984067, + "grad_norm": 1.4868022147567792, + "learning_rate": 3.7535413162191584e-06, + "loss": 1.0356, + "step": 2461 + }, + { + "epoch": 0.18502931008567564, + "grad_norm": 1.6597829601004743, + "learning_rate": 3.7533071376067514e-06, + "loss": 0.9512, + "step": 2462 + }, + { + "epoch": 0.1851044641515106, + "grad_norm": 1.5761594712763183, + "learning_rate": 3.7530728551041825e-06, + "loss": 0.9817, + "step": 2463 + }, + { + "epoch": 0.18517961821734555, + "grad_norm": 1.5676102802962548, + "learning_rate": 3.7528384687253335e-06, + "loss": 0.9868, + "step": 2464 + }, + { + "epoch": 0.18525477228318052, + "grad_norm": 1.638551109692556, + "learning_rate": 3.752603978484094e-06, + "loss": 0.9466, + "step": 2465 + }, + { + "epoch": 0.1853299263490155, + "grad_norm": 1.7204216197484488, + "learning_rate": 3.752369384394357e-06, + "loss": 1.0131, + "step": 2466 + }, + { + "epoch": 0.18540508041485043, + "grad_norm": 1.6291499332957267, + "learning_rate": 3.7521346864700235e-06, + "loss": 1.0126, + "step": 2467 + }, + { + "epoch": 0.1854802344806854, + "grad_norm": 1.8055621913233701, + "learning_rate": 3.751899884725001e-06, + "loss": 0.9577, + "step": 2468 + }, + { + "epoch": 0.18555538854652037, + "grad_norm": 1.0108458534926654, + "learning_rate": 3.751664979173202e-06, + "loss": 0.8495, + "step": 2469 + }, + { + "epoch": 0.18563054261235531, + "grad_norm": 1.3553467621451525, + "learning_rate": 3.7514299698285447e-06, + "loss": 0.9343, + "step": 2470 + }, + { + "epoch": 0.18570569667819029, + "grad_norm": 0.7688789895242091, + "learning_rate": 3.751194856704955e-06, + "loss": 0.7901, + "step": 2471 + }, + { + "epoch": 0.18578085074402526, + "grad_norm": 1.3600366670209592, + "learning_rate": 3.750959639816365e-06, + "loss": 1.0072, + "step": 2472 + }, + { + "epoch": 0.18585600480986023, + "grad_norm": 1.7342698348869978, + "learning_rate": 3.750724319176711e-06, + "loss": 0.8508, + "step": 2473 + }, + { + "epoch": 0.18593115887569517, + "grad_norm": 0.9293820496466858, + "learning_rate": 3.7504888947999367e-06, + "loss": 0.8758, + "step": 2474 + }, + { + "epoch": 0.18600631294153014, + "grad_norm": 1.4254941886481007, + "learning_rate": 3.7502533666999935e-06, + "loss": 0.9735, + "step": 2475 + }, + { + "epoch": 0.1860814670073651, + "grad_norm": 1.641711138915219, + "learning_rate": 3.7500177348908354e-06, + "loss": 0.9316, + "step": 2476 + }, + { + "epoch": 0.18615662107320005, + "grad_norm": 1.3751142194297774, + "learning_rate": 3.749781999386425e-06, + "loss": 0.9844, + "step": 2477 + }, + { + "epoch": 0.18623177513903502, + "grad_norm": 1.5909369831612623, + "learning_rate": 3.749546160200731e-06, + "loss": 0.9124, + "step": 2478 + }, + { + "epoch": 0.18630692920487, + "grad_norm": 1.4961935064857412, + "learning_rate": 3.7493102173477277e-06, + "loss": 1.1028, + "step": 2479 + }, + { + "epoch": 0.18638208327070493, + "grad_norm": 2.3748319466010996, + "learning_rate": 3.7490741708413954e-06, + "loss": 1.1042, + "step": 2480 + }, + { + "epoch": 0.1864572373365399, + "grad_norm": 1.5441821771382438, + "learning_rate": 3.748838020695721e-06, + "loss": 0.9616, + "step": 2481 + }, + { + "epoch": 0.18653239140237488, + "grad_norm": 1.418297007163152, + "learning_rate": 3.748601766924697e-06, + "loss": 0.982, + "step": 2482 + }, + { + "epoch": 0.18660754546820982, + "grad_norm": 0.8765615425852614, + "learning_rate": 3.7483654095423223e-06, + "loss": 0.7852, + "step": 2483 + }, + { + "epoch": 0.1866826995340448, + "grad_norm": 1.3425665678715002, + "learning_rate": 3.7481289485626024e-06, + "loss": 0.9758, + "step": 2484 + }, + { + "epoch": 0.18675785359987976, + "grad_norm": 2.155410713833073, + "learning_rate": 3.7478923839995477e-06, + "loss": 0.9734, + "step": 2485 + }, + { + "epoch": 0.18683300766571473, + "grad_norm": 1.6341226970172285, + "learning_rate": 3.7476557158671768e-06, + "loss": 0.9522, + "step": 2486 + }, + { + "epoch": 0.18690816173154967, + "grad_norm": 2.1000865873015373, + "learning_rate": 3.747418944179512e-06, + "loss": 1.017, + "step": 2487 + }, + { + "epoch": 0.18698331579738464, + "grad_norm": 2.0550885987499803, + "learning_rate": 3.747182068950584e-06, + "loss": 0.9736, + "step": 2488 + }, + { + "epoch": 0.1870584698632196, + "grad_norm": 1.5067646556120613, + "learning_rate": 3.746945090194428e-06, + "loss": 1.0551, + "step": 2489 + }, + { + "epoch": 0.18713362392905455, + "grad_norm": 1.3576160758536262, + "learning_rate": 3.7467080079250853e-06, + "loss": 1.0712, + "step": 2490 + }, + { + "epoch": 0.18720877799488952, + "grad_norm": 0.8423264822289422, + "learning_rate": 3.7464708221566052e-06, + "loss": 0.8501, + "step": 2491 + }, + { + "epoch": 0.1872839320607245, + "grad_norm": 2.047226387904205, + "learning_rate": 3.7462335329030408e-06, + "loss": 1.1281, + "step": 2492 + }, + { + "epoch": 0.18735908612655944, + "grad_norm": 1.910663997512962, + "learning_rate": 3.7459961401784527e-06, + "loss": 1.0339, + "step": 2493 + }, + { + "epoch": 0.1874342401923944, + "grad_norm": 1.8943144684084419, + "learning_rate": 3.7457586439969076e-06, + "loss": 1.0214, + "step": 2494 + }, + { + "epoch": 0.18750939425822938, + "grad_norm": 1.7840636935474803, + "learning_rate": 3.745521044372478e-06, + "loss": 1.0877, + "step": 2495 + }, + { + "epoch": 0.18758454832406432, + "grad_norm": 1.9123705178104862, + "learning_rate": 3.745283341319242e-06, + "loss": 1.0509, + "step": 2496 + }, + { + "epoch": 0.1876597023898993, + "grad_norm": 2.200819391767126, + "learning_rate": 3.7450455348512854e-06, + "loss": 0.8233, + "step": 2497 + }, + { + "epoch": 0.18773485645573426, + "grad_norm": 1.6766535127223643, + "learning_rate": 3.7448076249826987e-06, + "loss": 0.9279, + "step": 2498 + }, + { + "epoch": 0.1878100105215692, + "grad_norm": 2.053478625078324, + "learning_rate": 3.7445696117275785e-06, + "loss": 0.9564, + "step": 2499 + }, + { + "epoch": 0.18788516458740417, + "grad_norm": 1.5548024350721616, + "learning_rate": 3.7443314951000285e-06, + "loss": 0.9017, + "step": 2500 + }, + { + "epoch": 0.18796031865323914, + "grad_norm": 1.7118710971103952, + "learning_rate": 3.744093275114158e-06, + "loss": 0.9828, + "step": 2501 + }, + { + "epoch": 0.18803547271907411, + "grad_norm": 1.0259026693054867, + "learning_rate": 3.7438549517840823e-06, + "loss": 0.8459, + "step": 2502 + }, + { + "epoch": 0.18811062678490906, + "grad_norm": 14.04245315899812, + "learning_rate": 3.743616525123923e-06, + "loss": 0.8646, + "step": 2503 + }, + { + "epoch": 0.18818578085074403, + "grad_norm": 1.4396751367175988, + "learning_rate": 3.743377995147808e-06, + "loss": 0.9509, + "step": 2504 + }, + { + "epoch": 0.188260934916579, + "grad_norm": 1.6829094082300462, + "learning_rate": 3.743139361869871e-06, + "loss": 0.9514, + "step": 2505 + }, + { + "epoch": 0.18833608898241394, + "grad_norm": 1.8440440315235842, + "learning_rate": 3.7429006253042524e-06, + "loss": 1.0126, + "step": 2506 + }, + { + "epoch": 0.1884112430482489, + "grad_norm": 2.0772991403737238, + "learning_rate": 3.742661785465097e-06, + "loss": 0.9007, + "step": 2507 + }, + { + "epoch": 0.18848639711408388, + "grad_norm": 1.6820828413503912, + "learning_rate": 3.7424228423665578e-06, + "loss": 0.9941, + "step": 2508 + }, + { + "epoch": 0.18856155117991882, + "grad_norm": 1.5946479960218276, + "learning_rate": 3.7421837960227933e-06, + "loss": 0.9588, + "step": 2509 + }, + { + "epoch": 0.1886367052457538, + "grad_norm": 1.5413022083062464, + "learning_rate": 3.741944646447967e-06, + "loss": 0.935, + "step": 2510 + }, + { + "epoch": 0.18871185931158876, + "grad_norm": 2.2258073563549385, + "learning_rate": 3.7417053936562503e-06, + "loss": 0.9602, + "step": 2511 + }, + { + "epoch": 0.1887870133774237, + "grad_norm": 1.6142399412482467, + "learning_rate": 3.7414660376618195e-06, + "loss": 0.8829, + "step": 2512 + }, + { + "epoch": 0.18886216744325868, + "grad_norm": 1.3555663235858568, + "learning_rate": 3.7412265784788577e-06, + "loss": 1.0798, + "step": 2513 + }, + { + "epoch": 0.18893732150909365, + "grad_norm": 1.5123292936983417, + "learning_rate": 3.7409870161215532e-06, + "loss": 0.9115, + "step": 2514 + }, + { + "epoch": 0.1890124755749286, + "grad_norm": 1.4987815835788978, + "learning_rate": 3.740747350604102e-06, + "loss": 1.0416, + "step": 2515 + }, + { + "epoch": 0.18908762964076356, + "grad_norm": 1.6000687495004868, + "learning_rate": 3.7405075819407045e-06, + "loss": 0.973, + "step": 2516 + }, + { + "epoch": 0.18916278370659853, + "grad_norm": 1.1558303071312228, + "learning_rate": 3.7402677101455672e-06, + "loss": 0.8138, + "step": 2517 + }, + { + "epoch": 0.1892379377724335, + "grad_norm": 1.701668913064807, + "learning_rate": 3.740027735232904e-06, + "loss": 1.0444, + "step": 2518 + }, + { + "epoch": 0.18931309183826844, + "grad_norm": 1.3771023716767163, + "learning_rate": 3.7397876572169355e-06, + "loss": 1.0696, + "step": 2519 + }, + { + "epoch": 0.1893882459041034, + "grad_norm": 1.5662235143907908, + "learning_rate": 3.7395474761118856e-06, + "loss": 0.8751, + "step": 2520 + }, + { + "epoch": 0.18946339996993838, + "grad_norm": 1.52296596956963, + "learning_rate": 3.7393071919319864e-06, + "loss": 1.0544, + "step": 2521 + }, + { + "epoch": 0.18953855403577333, + "grad_norm": 1.6125638165524563, + "learning_rate": 3.739066804691476e-06, + "loss": 1.0451, + "step": 2522 + }, + { + "epoch": 0.1896137081016083, + "grad_norm": 1.6576378244028063, + "learning_rate": 3.738826314404598e-06, + "loss": 0.979, + "step": 2523 + }, + { + "epoch": 0.18968886216744327, + "grad_norm": 1.4536281247929985, + "learning_rate": 3.738585721085603e-06, + "loss": 1.1093, + "step": 2524 + }, + { + "epoch": 0.1897640162332782, + "grad_norm": 1.8758412699091598, + "learning_rate": 3.738345024748746e-06, + "loss": 0.9514, + "step": 2525 + }, + { + "epoch": 0.18983917029911318, + "grad_norm": 2.106912489483672, + "learning_rate": 3.73810422540829e-06, + "loss": 1.0119, + "step": 2526 + }, + { + "epoch": 0.18991432436494815, + "grad_norm": 2.19500437030053, + "learning_rate": 3.7378633230785025e-06, + "loss": 1.0488, + "step": 2527 + }, + { + "epoch": 0.1899894784307831, + "grad_norm": 1.5266101372898861, + "learning_rate": 3.7376223177736587e-06, + "loss": 0.9918, + "step": 2528 + }, + { + "epoch": 0.19006463249661806, + "grad_norm": 1.3969804216240393, + "learning_rate": 3.737381209508039e-06, + "loss": 0.8175, + "step": 2529 + }, + { + "epoch": 0.19013978656245303, + "grad_norm": 1.4209880368013033, + "learning_rate": 3.7371399982959294e-06, + "loss": 0.9499, + "step": 2530 + }, + { + "epoch": 0.190214940628288, + "grad_norm": 1.576585453044053, + "learning_rate": 3.736898684151623e-06, + "loss": 1.017, + "step": 2531 + }, + { + "epoch": 0.19029009469412295, + "grad_norm": 1.649901115459923, + "learning_rate": 3.736657267089419e-06, + "loss": 0.9657, + "step": 2532 + }, + { + "epoch": 0.19036524875995792, + "grad_norm": 1.316445913198377, + "learning_rate": 3.7364157471236215e-06, + "loss": 0.9657, + "step": 2533 + }, + { + "epoch": 0.1904404028257929, + "grad_norm": 2.3423465850326384, + "learning_rate": 3.7361741242685417e-06, + "loss": 1.0345, + "step": 2534 + }, + { + "epoch": 0.19051555689162783, + "grad_norm": 15.393127176042317, + "learning_rate": 3.7359323985384966e-06, + "loss": 0.9155, + "step": 2535 + }, + { + "epoch": 0.1905907109574628, + "grad_norm": 1.6824703141380928, + "learning_rate": 3.7356905699478096e-06, + "loss": 0.9784, + "step": 2536 + }, + { + "epoch": 0.19066586502329777, + "grad_norm": 2.49051976382446, + "learning_rate": 3.7354486385108103e-06, + "loss": 0.9335, + "step": 2537 + }, + { + "epoch": 0.1907410190891327, + "grad_norm": 1.4442945039801596, + "learning_rate": 3.735206604241834e-06, + "loss": 1.035, + "step": 2538 + }, + { + "epoch": 0.19081617315496768, + "grad_norm": 1.7128880499686674, + "learning_rate": 3.734964467155221e-06, + "loss": 0.9839, + "step": 2539 + }, + { + "epoch": 0.19089132722080265, + "grad_norm": 2.3049029750635466, + "learning_rate": 3.73472222726532e-06, + "loss": 0.8897, + "step": 2540 + }, + { + "epoch": 0.1909664812866376, + "grad_norm": 1.553396024025585, + "learning_rate": 3.7344798845864846e-06, + "loss": 0.909, + "step": 2541 + }, + { + "epoch": 0.19104163535247257, + "grad_norm": 1.3032647330107925, + "learning_rate": 3.734237439133074e-06, + "loss": 1.0247, + "step": 2542 + }, + { + "epoch": 0.19111678941830754, + "grad_norm": 1.5180809336863184, + "learning_rate": 3.7339948909194543e-06, + "loss": 1.0133, + "step": 2543 + }, + { + "epoch": 0.19119194348414248, + "grad_norm": 1.7513677156953174, + "learning_rate": 3.7337522399599973e-06, + "loss": 0.9252, + "step": 2544 + }, + { + "epoch": 0.19126709754997745, + "grad_norm": 1.5678508689523951, + "learning_rate": 3.7335094862690814e-06, + "loss": 0.9309, + "step": 2545 + }, + { + "epoch": 0.19134225161581242, + "grad_norm": 1.5208314956584261, + "learning_rate": 3.7332666298610906e-06, + "loss": 1.017, + "step": 2546 + }, + { + "epoch": 0.1914174056816474, + "grad_norm": 0.9164907691270107, + "learning_rate": 3.733023670750414e-06, + "loss": 0.7937, + "step": 2547 + }, + { + "epoch": 0.19149255974748233, + "grad_norm": 1.82312939810394, + "learning_rate": 3.7327806089514497e-06, + "loss": 0.9825, + "step": 2548 + }, + { + "epoch": 0.1915677138133173, + "grad_norm": 1.3944298202428327, + "learning_rate": 3.7325374444785983e-06, + "loss": 0.9665, + "step": 2549 + }, + { + "epoch": 0.19164286787915227, + "grad_norm": 1.4800331461261949, + "learning_rate": 3.7322941773462694e-06, + "loss": 1.0711, + "step": 2550 + }, + { + "epoch": 0.19171802194498722, + "grad_norm": 1.4890784984261567, + "learning_rate": 3.732050807568877e-06, + "loss": 1.0346, + "step": 2551 + }, + { + "epoch": 0.1917931760108222, + "grad_norm": 1.7669652392644284, + "learning_rate": 3.731807335160842e-06, + "loss": 1.0028, + "step": 2552 + }, + { + "epoch": 0.19186833007665716, + "grad_norm": 2.628499468638813, + "learning_rate": 3.7315637601365902e-06, + "loss": 1.0061, + "step": 2553 + }, + { + "epoch": 0.1919434841424921, + "grad_norm": 1.4862376522645324, + "learning_rate": 3.731320082510556e-06, + "loss": 0.9904, + "step": 2554 + }, + { + "epoch": 0.19201863820832707, + "grad_norm": 2.8738615020850435, + "learning_rate": 3.7310763022971764e-06, + "loss": 1.0057, + "step": 2555 + }, + { + "epoch": 0.19209379227416204, + "grad_norm": 1.529488854763281, + "learning_rate": 3.730832419510897e-06, + "loss": 1.0761, + "step": 2556 + }, + { + "epoch": 0.19216894633999698, + "grad_norm": 1.1285689750993075, + "learning_rate": 3.73058843416617e-06, + "loss": 0.8837, + "step": 2557 + }, + { + "epoch": 0.19224410040583195, + "grad_norm": 1.4934859293787945, + "learning_rate": 3.7303443462774505e-06, + "loss": 1.0238, + "step": 2558 + }, + { + "epoch": 0.19231925447166692, + "grad_norm": 1.1861211903362459, + "learning_rate": 3.730100155859203e-06, + "loss": 0.8809, + "step": 2559 + }, + { + "epoch": 0.19239440853750187, + "grad_norm": 1.73982269863622, + "learning_rate": 3.7298558629258966e-06, + "loss": 0.9915, + "step": 2560 + }, + { + "epoch": 0.19246956260333684, + "grad_norm": 1.0060937700954387, + "learning_rate": 3.729611467492005e-06, + "loss": 0.7611, + "step": 2561 + }, + { + "epoch": 0.1925447166691718, + "grad_norm": 2.026158215881398, + "learning_rate": 3.7293669695720117e-06, + "loss": 0.9601, + "step": 2562 + }, + { + "epoch": 0.19261987073500678, + "grad_norm": 2.4772301384682747, + "learning_rate": 3.7291223691804038e-06, + "loss": 0.9916, + "step": 2563 + }, + { + "epoch": 0.19269502480084172, + "grad_norm": 1.390464216216978, + "learning_rate": 3.728877666331673e-06, + "loss": 0.9982, + "step": 2564 + }, + { + "epoch": 0.1927701788666767, + "grad_norm": 1.3430564332468469, + "learning_rate": 3.7286328610403207e-06, + "loss": 1.0123, + "step": 2565 + }, + { + "epoch": 0.19284533293251166, + "grad_norm": 1.8908649008637506, + "learning_rate": 3.7283879533208523e-06, + "loss": 0.842, + "step": 2566 + }, + { + "epoch": 0.1929204869983466, + "grad_norm": 1.7035609396068647, + "learning_rate": 3.7281429431877795e-06, + "loss": 1.0336, + "step": 2567 + }, + { + "epoch": 0.19299564106418157, + "grad_norm": 1.622398053979553, + "learning_rate": 3.727897830655619e-06, + "loss": 0.9647, + "step": 2568 + }, + { + "epoch": 0.19307079513001654, + "grad_norm": 1.62433030927473, + "learning_rate": 3.727652615738896e-06, + "loss": 0.9256, + "step": 2569 + }, + { + "epoch": 0.19314594919585149, + "grad_norm": 1.691041338807261, + "learning_rate": 3.7274072984521395e-06, + "loss": 1.0036, + "step": 2570 + }, + { + "epoch": 0.19322110326168646, + "grad_norm": 1.5486450856829372, + "learning_rate": 3.7271618788098864e-06, + "loss": 0.9613, + "step": 2571 + }, + { + "epoch": 0.19329625732752143, + "grad_norm": 1.6240930226936832, + "learning_rate": 3.7269163568266774e-06, + "loss": 1.0229, + "step": 2572 + }, + { + "epoch": 0.19337141139335637, + "grad_norm": 1.5322751097710638, + "learning_rate": 3.7266707325170623e-06, + "loss": 1.0781, + "step": 2573 + }, + { + "epoch": 0.19344656545919134, + "grad_norm": 1.6129334746877537, + "learning_rate": 3.7264250058955938e-06, + "loss": 0.9353, + "step": 2574 + }, + { + "epoch": 0.1935217195250263, + "grad_norm": 2.6667709230864736, + "learning_rate": 3.726179176976833e-06, + "loss": 1.0173, + "step": 2575 + }, + { + "epoch": 0.19359687359086128, + "grad_norm": 2.2691632233405725, + "learning_rate": 3.7259332457753464e-06, + "loss": 0.9768, + "step": 2576 + }, + { + "epoch": 0.19367202765669622, + "grad_norm": 0.8351124051666972, + "learning_rate": 3.725687212305706e-06, + "loss": 0.7975, + "step": 2577 + }, + { + "epoch": 0.1937471817225312, + "grad_norm": 1.613647493258106, + "learning_rate": 3.7254410765824896e-06, + "loss": 1.0349, + "step": 2578 + }, + { + "epoch": 0.19382233578836616, + "grad_norm": 1.5762571857721195, + "learning_rate": 3.7251948386202827e-06, + "loss": 1.0141, + "step": 2579 + }, + { + "epoch": 0.1938974898542011, + "grad_norm": 1.7451072141835187, + "learning_rate": 3.724948498433675e-06, + "loss": 0.9414, + "step": 2580 + }, + { + "epoch": 0.19397264392003608, + "grad_norm": 1.7458347751487053, + "learning_rate": 3.7247020560372635e-06, + "loss": 1.0306, + "step": 2581 + }, + { + "epoch": 0.19404779798587105, + "grad_norm": 2.237387167360607, + "learning_rate": 3.724455511445651e-06, + "loss": 0.8862, + "step": 2582 + }, + { + "epoch": 0.194122952051706, + "grad_norm": 1.5138565443046474, + "learning_rate": 3.724208864673446e-06, + "loss": 0.9097, + "step": 2583 + }, + { + "epoch": 0.19419810611754096, + "grad_norm": 1.5352428305091546, + "learning_rate": 3.7239621157352633e-06, + "loss": 0.9591, + "step": 2584 + }, + { + "epoch": 0.19427326018337593, + "grad_norm": 2.211708284957635, + "learning_rate": 3.723715264645724e-06, + "loss": 0.9903, + "step": 2585 + }, + { + "epoch": 0.19434841424921087, + "grad_norm": 2.649177920387687, + "learning_rate": 3.723468311419455e-06, + "loss": 1.0004, + "step": 2586 + }, + { + "epoch": 0.19442356831504584, + "grad_norm": 1.479175093004466, + "learning_rate": 3.7232212560710883e-06, + "loss": 0.982, + "step": 2587 + }, + { + "epoch": 0.1944987223808808, + "grad_norm": 1.5934440468718927, + "learning_rate": 3.7229740986152636e-06, + "loss": 0.9118, + "step": 2588 + }, + { + "epoch": 0.19457387644671575, + "grad_norm": 2.127573026307038, + "learning_rate": 3.722726839066626e-06, + "loss": 0.8829, + "step": 2589 + }, + { + "epoch": 0.19464903051255072, + "grad_norm": 1.425008820623571, + "learning_rate": 3.722479477439826e-06, + "loss": 1.0566, + "step": 2590 + }, + { + "epoch": 0.1947241845783857, + "grad_norm": 1.6007371962495622, + "learning_rate": 3.722232013749522e-06, + "loss": 0.9611, + "step": 2591 + }, + { + "epoch": 0.19479933864422067, + "grad_norm": 1.4514283150038692, + "learning_rate": 3.721984448010376e-06, + "loss": 1.0057, + "step": 2592 + }, + { + "epoch": 0.1948744927100556, + "grad_norm": 1.3524911148992247, + "learning_rate": 3.7217367802370573e-06, + "loss": 0.9801, + "step": 2593 + }, + { + "epoch": 0.19494964677589058, + "grad_norm": 1.7004830403434212, + "learning_rate": 3.7214890104442413e-06, + "loss": 0.9697, + "step": 2594 + }, + { + "epoch": 0.19502480084172555, + "grad_norm": 2.0382430377505822, + "learning_rate": 3.7212411386466097e-06, + "loss": 1.0564, + "step": 2595 + }, + { + "epoch": 0.1950999549075605, + "grad_norm": 1.8123403018445354, + "learning_rate": 3.72099316485885e-06, + "loss": 0.9292, + "step": 2596 + }, + { + "epoch": 0.19517510897339546, + "grad_norm": 4.3081995212181186, + "learning_rate": 3.7207450890956544e-06, + "loss": 1.0219, + "step": 2597 + }, + { + "epoch": 0.19525026303923043, + "grad_norm": 1.3784730571207273, + "learning_rate": 3.720496911371723e-06, + "loss": 0.9268, + "step": 2598 + }, + { + "epoch": 0.19532541710506537, + "grad_norm": 1.6927989837630815, + "learning_rate": 3.720248631701762e-06, + "loss": 0.862, + "step": 2599 + }, + { + "epoch": 0.19540057117090034, + "grad_norm": 1.6116551662667247, + "learning_rate": 3.720000250100482e-06, + "loss": 0.9584, + "step": 2600 + }, + { + "epoch": 0.19547572523673531, + "grad_norm": 1.6564333832181573, + "learning_rate": 3.719751766582601e-06, + "loss": 0.8971, + "step": 2601 + }, + { + "epoch": 0.19555087930257026, + "grad_norm": 1.555478608786562, + "learning_rate": 3.7195031811628422e-06, + "loss": 0.9546, + "step": 2602 + }, + { + "epoch": 0.19562603336840523, + "grad_norm": 1.4070749194513328, + "learning_rate": 3.719254493855936e-06, + "loss": 1.0439, + "step": 2603 + }, + { + "epoch": 0.1957011874342402, + "grad_norm": 0.9716117490087901, + "learning_rate": 3.719005704676617e-06, + "loss": 0.909, + "step": 2604 + }, + { + "epoch": 0.19577634150007514, + "grad_norm": 1.5314737273281374, + "learning_rate": 3.7187568136396274e-06, + "loss": 1.0219, + "step": 2605 + }, + { + "epoch": 0.1958514955659101, + "grad_norm": 1.6952204525477523, + "learning_rate": 3.7185078207597158e-06, + "loss": 0.9271, + "step": 2606 + }, + { + "epoch": 0.19592664963174508, + "grad_norm": 2.117697003420725, + "learning_rate": 3.7182587260516343e-06, + "loss": 0.9472, + "step": 2607 + }, + { + "epoch": 0.19600180369758005, + "grad_norm": 1.2935542598944259, + "learning_rate": 3.7180095295301443e-06, + "loss": 0.9622, + "step": 2608 + }, + { + "epoch": 0.196076957763415, + "grad_norm": 1.9246170207933044, + "learning_rate": 3.717760231210011e-06, + "loss": 1.0565, + "step": 2609 + }, + { + "epoch": 0.19615211182924996, + "grad_norm": 1.423328704528667, + "learning_rate": 3.7175108311060057e-06, + "loss": 0.8977, + "step": 2610 + }, + { + "epoch": 0.19622726589508493, + "grad_norm": 1.804793053384696, + "learning_rate": 3.717261329232907e-06, + "loss": 0.9976, + "step": 2611 + }, + { + "epoch": 0.19630241996091988, + "grad_norm": 0.7791630144764649, + "learning_rate": 3.717011725605499e-06, + "loss": 0.7249, + "step": 2612 + }, + { + "epoch": 0.19637757402675485, + "grad_norm": 1.7670559781053035, + "learning_rate": 3.7167620202385715e-06, + "loss": 0.9679, + "step": 2613 + }, + { + "epoch": 0.19645272809258982, + "grad_norm": 1.7302070230306417, + "learning_rate": 3.71651221314692e-06, + "loss": 0.879, + "step": 2614 + }, + { + "epoch": 0.19652788215842476, + "grad_norm": 1.935147179715919, + "learning_rate": 3.7162623043453476e-06, + "loss": 1.0842, + "step": 2615 + }, + { + "epoch": 0.19660303622425973, + "grad_norm": 1.7133001449106644, + "learning_rate": 3.716012293848661e-06, + "loss": 1.0109, + "step": 2616 + }, + { + "epoch": 0.1966781902900947, + "grad_norm": 1.9698854230102338, + "learning_rate": 3.7157621816716747e-06, + "loss": 1.0475, + "step": 2617 + }, + { + "epoch": 0.19675334435592964, + "grad_norm": 1.5202801190306423, + "learning_rate": 3.71551196782921e-06, + "loss": 1.0494, + "step": 2618 + }, + { + "epoch": 0.1968284984217646, + "grad_norm": 1.5637713028808347, + "learning_rate": 3.7152616523360913e-06, + "loss": 1.0405, + "step": 2619 + }, + { + "epoch": 0.19690365248759958, + "grad_norm": 1.4338146735906057, + "learning_rate": 3.7150112352071514e-06, + "loss": 1.0383, + "step": 2620 + }, + { + "epoch": 0.19697880655343455, + "grad_norm": 0.889798399830462, + "learning_rate": 3.714760716457229e-06, + "loss": 0.8257, + "step": 2621 + }, + { + "epoch": 0.1970539606192695, + "grad_norm": 1.7041515791146629, + "learning_rate": 3.7145100961011675e-06, + "loss": 0.9487, + "step": 2622 + }, + { + "epoch": 0.19712911468510447, + "grad_norm": 1.6234513753053883, + "learning_rate": 3.714259374153818e-06, + "loss": 0.9607, + "step": 2623 + }, + { + "epoch": 0.19720426875093944, + "grad_norm": 1.396912366465429, + "learning_rate": 3.714008550630036e-06, + "loss": 0.7915, + "step": 2624 + }, + { + "epoch": 0.19727942281677438, + "grad_norm": 1.5838917282987706, + "learning_rate": 3.713757625544684e-06, + "loss": 0.9853, + "step": 2625 + }, + { + "epoch": 0.19735457688260935, + "grad_norm": 1.6057019586409929, + "learning_rate": 3.7135065989126303e-06, + "loss": 0.9209, + "step": 2626 + }, + { + "epoch": 0.19742973094844432, + "grad_norm": 1.4995457438239204, + "learning_rate": 3.7132554707487493e-06, + "loss": 0.9842, + "step": 2627 + }, + { + "epoch": 0.19750488501427926, + "grad_norm": 3.995490393111378, + "learning_rate": 3.713004241067921e-06, + "loss": 0.9168, + "step": 2628 + }, + { + "epoch": 0.19758003908011423, + "grad_norm": 1.625210014857147, + "learning_rate": 3.712752909885032e-06, + "loss": 0.953, + "step": 2629 + }, + { + "epoch": 0.1976551931459492, + "grad_norm": 1.444876280243385, + "learning_rate": 3.7125014772149746e-06, + "loss": 1.047, + "step": 2630 + }, + { + "epoch": 0.19773034721178415, + "grad_norm": 1.7776206000040036, + "learning_rate": 3.712249943072647e-06, + "loss": 0.9811, + "step": 2631 + }, + { + "epoch": 0.19780550127761912, + "grad_norm": 1.5842496412293972, + "learning_rate": 3.7119983074729532e-06, + "loss": 1.0247, + "step": 2632 + }, + { + "epoch": 0.1978806553434541, + "grad_norm": 1.734608883595862, + "learning_rate": 3.7117465704308045e-06, + "loss": 1.042, + "step": 2633 + }, + { + "epoch": 0.19795580940928903, + "grad_norm": 1.5322371434465765, + "learning_rate": 3.7114947319611164e-06, + "loss": 0.865, + "step": 2634 + }, + { + "epoch": 0.198030963475124, + "grad_norm": 1.7259991447384728, + "learning_rate": 3.711242792078812e-06, + "loss": 0.9043, + "step": 2635 + }, + { + "epoch": 0.19810611754095897, + "grad_norm": 1.5092542801436601, + "learning_rate": 3.7109907507988192e-06, + "loss": 0.9716, + "step": 2636 + }, + { + "epoch": 0.19818127160679394, + "grad_norm": 1.8331396983221036, + "learning_rate": 3.710738608136073e-06, + "loss": 0.9668, + "step": 2637 + }, + { + "epoch": 0.19825642567262888, + "grad_norm": 1.6113535746096281, + "learning_rate": 3.710486364105513e-06, + "loss": 0.9554, + "step": 2638 + }, + { + "epoch": 0.19833157973846385, + "grad_norm": 2.4104658574058333, + "learning_rate": 3.7102340187220863e-06, + "loss": 0.9818, + "step": 2639 + }, + { + "epoch": 0.19840673380429882, + "grad_norm": 1.8688833284818223, + "learning_rate": 3.7099815720007447e-06, + "loss": 0.9663, + "step": 2640 + }, + { + "epoch": 0.19848188787013377, + "grad_norm": 1.5682712021836762, + "learning_rate": 3.7097290239564478e-06, + "loss": 1.0624, + "step": 2641 + }, + { + "epoch": 0.19855704193596874, + "grad_norm": 5.168231337515867, + "learning_rate": 3.7094763746041584e-06, + "loss": 0.9568, + "step": 2642 + }, + { + "epoch": 0.1986321960018037, + "grad_norm": 1.664381829368455, + "learning_rate": 3.709223623958848e-06, + "loss": 0.903, + "step": 2643 + }, + { + "epoch": 0.19870735006763865, + "grad_norm": 1.9027225814325404, + "learning_rate": 3.708970772035493e-06, + "loss": 0.9362, + "step": 2644 + }, + { + "epoch": 0.19878250413347362, + "grad_norm": 1.3782529066592462, + "learning_rate": 3.7087178188490754e-06, + "loss": 0.9378, + "step": 2645 + }, + { + "epoch": 0.1988576581993086, + "grad_norm": 1.8840538776128906, + "learning_rate": 3.708464764414584e-06, + "loss": 0.8733, + "step": 2646 + }, + { + "epoch": 0.19893281226514353, + "grad_norm": 1.6627877237402229, + "learning_rate": 3.708211608747013e-06, + "loss": 1.0053, + "step": 2647 + }, + { + "epoch": 0.1990079663309785, + "grad_norm": 1.7580021734690765, + "learning_rate": 3.7079583518613636e-06, + "loss": 0.9807, + "step": 2648 + }, + { + "epoch": 0.19908312039681347, + "grad_norm": 1.3486236196283357, + "learning_rate": 3.707704993772641e-06, + "loss": 1.0006, + "step": 2649 + }, + { + "epoch": 0.19915827446264842, + "grad_norm": 1.3878620001957036, + "learning_rate": 3.7074515344958584e-06, + "loss": 0.9853, + "step": 2650 + }, + { + "epoch": 0.1992334285284834, + "grad_norm": 1.4955216152933706, + "learning_rate": 3.7071979740460345e-06, + "loss": 0.9375, + "step": 2651 + }, + { + "epoch": 0.19930858259431836, + "grad_norm": 1.50396165621862, + "learning_rate": 3.706944312438193e-06, + "loss": 1.024, + "step": 2652 + }, + { + "epoch": 0.19938373666015333, + "grad_norm": 2.168802187069107, + "learning_rate": 3.7066905496873646e-06, + "loss": 0.9536, + "step": 2653 + }, + { + "epoch": 0.19945889072598827, + "grad_norm": 1.6265233309057805, + "learning_rate": 3.706436685808586e-06, + "loss": 0.9495, + "step": 2654 + }, + { + "epoch": 0.19953404479182324, + "grad_norm": 1.9180411579580914, + "learning_rate": 3.7061827208168995e-06, + "loss": 1.003, + "step": 2655 + }, + { + "epoch": 0.1996091988576582, + "grad_norm": 1.6701395171814828, + "learning_rate": 3.705928654727353e-06, + "loss": 0.9279, + "step": 2656 + }, + { + "epoch": 0.19968435292349315, + "grad_norm": 0.9469313828202318, + "learning_rate": 3.7056744875550016e-06, + "loss": 0.8793, + "step": 2657 + }, + { + "epoch": 0.19975950698932812, + "grad_norm": 2.466001501697043, + "learning_rate": 3.7054202193149047e-06, + "loss": 0.9975, + "step": 2658 + }, + { + "epoch": 0.1998346610551631, + "grad_norm": 1.6291890070457482, + "learning_rate": 3.7051658500221297e-06, + "loss": 0.9613, + "step": 2659 + }, + { + "epoch": 0.19990981512099804, + "grad_norm": 1.8114211719365247, + "learning_rate": 3.704911379691749e-06, + "loss": 0.9567, + "step": 2660 + }, + { + "epoch": 0.199984969186833, + "grad_norm": 1.5407785056747108, + "learning_rate": 3.70465680833884e-06, + "loss": 1.0084, + "step": 2661 + }, + { + "epoch": 0.20006012325266798, + "grad_norm": 1.6144163558099305, + "learning_rate": 3.704402135978488e-06, + "loss": 0.9754, + "step": 2662 + }, + { + "epoch": 0.20013527731850292, + "grad_norm": 1.1877505389489147, + "learning_rate": 3.7041473626257823e-06, + "loss": 0.9443, + "step": 2663 + }, + { + "epoch": 0.2002104313843379, + "grad_norm": 1.550432064995201, + "learning_rate": 3.7038924882958204e-06, + "loss": 0.8933, + "step": 2664 + }, + { + "epoch": 0.20028558545017286, + "grad_norm": 1.656410569378853, + "learning_rate": 3.7036375130037037e-06, + "loss": 0.9706, + "step": 2665 + }, + { + "epoch": 0.20036073951600783, + "grad_norm": 1.4617543104567177, + "learning_rate": 3.70338243676454e-06, + "loss": 1.0511, + "step": 2666 + }, + { + "epoch": 0.20043589358184277, + "grad_norm": 2.277712967061707, + "learning_rate": 3.7031272595934453e-06, + "loss": 0.958, + "step": 2667 + }, + { + "epoch": 0.20051104764767774, + "grad_norm": 1.5861446452042935, + "learning_rate": 3.702871981505538e-06, + "loss": 1.0208, + "step": 2668 + }, + { + "epoch": 0.2005862017135127, + "grad_norm": 1.56997525213867, + "learning_rate": 3.7026166025159454e-06, + "loss": 0.9502, + "step": 2669 + }, + { + "epoch": 0.20066135577934766, + "grad_norm": 2.307379647006253, + "learning_rate": 3.7023611226397993e-06, + "loss": 0.8614, + "step": 2670 + }, + { + "epoch": 0.20073650984518263, + "grad_norm": 1.6540931477131648, + "learning_rate": 3.702105541892238e-06, + "loss": 0.9916, + "step": 2671 + }, + { + "epoch": 0.2008116639110176, + "grad_norm": 1.8241288941040592, + "learning_rate": 3.7018498602884053e-06, + "loss": 0.9182, + "step": 2672 + }, + { + "epoch": 0.20088681797685254, + "grad_norm": 1.6264040090131535, + "learning_rate": 3.701594077843452e-06, + "loss": 1.0444, + "step": 2673 + }, + { + "epoch": 0.2009619720426875, + "grad_norm": 2.069984381243619, + "learning_rate": 3.701338194572533e-06, + "loss": 0.9894, + "step": 2674 + }, + { + "epoch": 0.20103712610852248, + "grad_norm": 2.1269968411915814, + "learning_rate": 3.7010822104908116e-06, + "loss": 1.0203, + "step": 2675 + }, + { + "epoch": 0.20111228017435742, + "grad_norm": 1.773606279718916, + "learning_rate": 3.7008261256134556e-06, + "loss": 1.009, + "step": 2676 + }, + { + "epoch": 0.2011874342401924, + "grad_norm": 1.5499121394943578, + "learning_rate": 3.7005699399556383e-06, + "loss": 1.0533, + "step": 2677 + }, + { + "epoch": 0.20126258830602736, + "grad_norm": 1.687882909777011, + "learning_rate": 3.7003136535325405e-06, + "loss": 0.9981, + "step": 2678 + }, + { + "epoch": 0.2013377423718623, + "grad_norm": 1.609564686679625, + "learning_rate": 3.7000572663593475e-06, + "loss": 0.9143, + "step": 2679 + }, + { + "epoch": 0.20141289643769728, + "grad_norm": 0.8338528245720119, + "learning_rate": 3.6998007784512515e-06, + "loss": 0.7752, + "step": 2680 + }, + { + "epoch": 0.20148805050353225, + "grad_norm": 1.4879910243645549, + "learning_rate": 3.6995441898234507e-06, + "loss": 0.941, + "step": 2681 + }, + { + "epoch": 0.20156320456936722, + "grad_norm": 1.634447219449219, + "learning_rate": 3.6992875004911485e-06, + "loss": 0.9401, + "step": 2682 + }, + { + "epoch": 0.20163835863520216, + "grad_norm": 1.58013176561743, + "learning_rate": 3.6990307104695547e-06, + "loss": 0.996, + "step": 2683 + }, + { + "epoch": 0.20171351270103713, + "grad_norm": 1.5002904451827372, + "learning_rate": 3.6987738197738858e-06, + "loss": 0.9698, + "step": 2684 + }, + { + "epoch": 0.2017886667668721, + "grad_norm": 1.537041901364134, + "learning_rate": 3.698516828419362e-06, + "loss": 1.0184, + "step": 2685 + }, + { + "epoch": 0.20186382083270704, + "grad_norm": 1.4415983573444844, + "learning_rate": 3.698259736421213e-06, + "loss": 1.0106, + "step": 2686 + }, + { + "epoch": 0.201938974898542, + "grad_norm": 1.3501376099564024, + "learning_rate": 3.698002543794671e-06, + "loss": 0.9615, + "step": 2687 + }, + { + "epoch": 0.20201412896437698, + "grad_norm": 1.3921121510538421, + "learning_rate": 3.697745250554977e-06, + "loss": 0.9324, + "step": 2688 + }, + { + "epoch": 0.20208928303021192, + "grad_norm": 1.8413587678169636, + "learning_rate": 3.697487856717375e-06, + "loss": 0.9276, + "step": 2689 + }, + { + "epoch": 0.2021644370960469, + "grad_norm": 1.3755707423226529, + "learning_rate": 3.6972303622971177e-06, + "loss": 1.0821, + "step": 2690 + }, + { + "epoch": 0.20223959116188187, + "grad_norm": 1.7254919815606724, + "learning_rate": 3.6969727673094626e-06, + "loss": 0.9692, + "step": 2691 + }, + { + "epoch": 0.2023147452277168, + "grad_norm": 1.6633156546657866, + "learning_rate": 3.696715071769672e-06, + "loss": 0.998, + "step": 2692 + }, + { + "epoch": 0.20238989929355178, + "grad_norm": 1.5792178911474937, + "learning_rate": 3.696457275693017e-06, + "loss": 0.9589, + "step": 2693 + }, + { + "epoch": 0.20246505335938675, + "grad_norm": 1.6175931707371989, + "learning_rate": 3.6961993790947722e-06, + "loss": 1.0242, + "step": 2694 + }, + { + "epoch": 0.2025402074252217, + "grad_norm": 1.5000702523628437, + "learning_rate": 3.695941381990219e-06, + "loss": 0.9624, + "step": 2695 + }, + { + "epoch": 0.20261536149105666, + "grad_norm": 2.646942617009844, + "learning_rate": 3.6956832843946445e-06, + "loss": 0.9626, + "step": 2696 + }, + { + "epoch": 0.20269051555689163, + "grad_norm": 1.334595837510185, + "learning_rate": 3.695425086323342e-06, + "loss": 0.9747, + "step": 2697 + }, + { + "epoch": 0.2027656696227266, + "grad_norm": 1.5724863975433887, + "learning_rate": 3.6951667877916113e-06, + "loss": 0.975, + "step": 2698 + }, + { + "epoch": 0.20284082368856154, + "grad_norm": 1.2248772871095568, + "learning_rate": 3.694908388814757e-06, + "loss": 0.9717, + "step": 2699 + }, + { + "epoch": 0.20291597775439651, + "grad_norm": 1.5284001046213385, + "learning_rate": 3.6946498894080905e-06, + "loss": 0.9405, + "step": 2700 + }, + { + "epoch": 0.20299113182023149, + "grad_norm": 1.344074742310892, + "learning_rate": 3.694391289586929e-06, + "loss": 1.0504, + "step": 2701 + }, + { + "epoch": 0.20306628588606643, + "grad_norm": 1.5492550043970463, + "learning_rate": 3.6941325893665953e-06, + "loss": 0.9965, + "step": 2702 + }, + { + "epoch": 0.2031414399519014, + "grad_norm": 1.7948694065797728, + "learning_rate": 3.693873788762418e-06, + "loss": 0.896, + "step": 2703 + }, + { + "epoch": 0.20321659401773637, + "grad_norm": 1.8856686657283643, + "learning_rate": 3.6936148877897324e-06, + "loss": 1.0704, + "step": 2704 + }, + { + "epoch": 0.2032917480835713, + "grad_norm": 2.1140379053505143, + "learning_rate": 3.6933558864638805e-06, + "loss": 0.9931, + "step": 2705 + }, + { + "epoch": 0.20336690214940628, + "grad_norm": 1.947429684506897, + "learning_rate": 3.6930967848002065e-06, + "loss": 1.0482, + "step": 2706 + }, + { + "epoch": 0.20344205621524125, + "grad_norm": 1.899908806469508, + "learning_rate": 3.6928375828140658e-06, + "loss": 0.8998, + "step": 2707 + }, + { + "epoch": 0.2035172102810762, + "grad_norm": 1.4841445744002022, + "learning_rate": 3.6925782805208156e-06, + "loss": 0.9167, + "step": 2708 + }, + { + "epoch": 0.20359236434691116, + "grad_norm": 1.8743438844301838, + "learning_rate": 3.692318877935821e-06, + "loss": 0.9883, + "step": 2709 + }, + { + "epoch": 0.20366751841274613, + "grad_norm": 2.029569899372877, + "learning_rate": 3.692059375074453e-06, + "loss": 0.9962, + "step": 2710 + }, + { + "epoch": 0.2037426724785811, + "grad_norm": 1.626206317053965, + "learning_rate": 3.6917997719520867e-06, + "loss": 0.8963, + "step": 2711 + }, + { + "epoch": 0.20381782654441605, + "grad_norm": 1.7563710724174029, + "learning_rate": 3.691540068584106e-06, + "loss": 0.9569, + "step": 2712 + }, + { + "epoch": 0.20389298061025102, + "grad_norm": 0.8421077994885647, + "learning_rate": 3.6912802649858995e-06, + "loss": 0.8346, + "step": 2713 + }, + { + "epoch": 0.203968134676086, + "grad_norm": 1.6129451521684823, + "learning_rate": 3.6910203611728603e-06, + "loss": 0.9831, + "step": 2714 + }, + { + "epoch": 0.20404328874192093, + "grad_norm": 1.6495633732467443, + "learning_rate": 3.6907603571603895e-06, + "loss": 0.9594, + "step": 2715 + }, + { + "epoch": 0.2041184428077559, + "grad_norm": 1.5684444266662447, + "learning_rate": 3.690500252963893e-06, + "loss": 0.9957, + "step": 2716 + }, + { + "epoch": 0.20419359687359087, + "grad_norm": 1.4470100125623486, + "learning_rate": 3.6902400485987835e-06, + "loss": 0.9549, + "step": 2717 + }, + { + "epoch": 0.2042687509394258, + "grad_norm": 1.5268999708696933, + "learning_rate": 3.6899797440804788e-06, + "loss": 1.0751, + "step": 2718 + }, + { + "epoch": 0.20434390500526078, + "grad_norm": 1.890097047847216, + "learning_rate": 3.689719339424403e-06, + "loss": 0.9497, + "step": 2719 + }, + { + "epoch": 0.20441905907109575, + "grad_norm": 1.5586280914577022, + "learning_rate": 3.689458834645986e-06, + "loss": 0.8878, + "step": 2720 + }, + { + "epoch": 0.2044942131369307, + "grad_norm": 0.9238276411059284, + "learning_rate": 3.689198229760663e-06, + "loss": 0.8354, + "step": 2721 + }, + { + "epoch": 0.20456936720276567, + "grad_norm": 4.037158037472802, + "learning_rate": 3.6889375247838766e-06, + "loss": 0.9252, + "step": 2722 + }, + { + "epoch": 0.20464452126860064, + "grad_norm": 1.742793572213448, + "learning_rate": 3.6886767197310757e-06, + "loss": 1.029, + "step": 2723 + }, + { + "epoch": 0.20471967533443558, + "grad_norm": 1.5299229223495832, + "learning_rate": 3.688415814617711e-06, + "loss": 0.9372, + "step": 2724 + }, + { + "epoch": 0.20479482940027055, + "grad_norm": 0.7882524619313566, + "learning_rate": 3.688154809459245e-06, + "loss": 0.7924, + "step": 2725 + }, + { + "epoch": 0.20486998346610552, + "grad_norm": 1.447758682011516, + "learning_rate": 3.6878937042711424e-06, + "loss": 1.0359, + "step": 2726 + }, + { + "epoch": 0.2049451375319405, + "grad_norm": 1.5755853140590839, + "learning_rate": 3.687632499068874e-06, + "loss": 0.8608, + "step": 2727 + }, + { + "epoch": 0.20502029159777543, + "grad_norm": 1.53628320073008, + "learning_rate": 3.6873711938679174e-06, + "loss": 1.009, + "step": 2728 + }, + { + "epoch": 0.2050954456636104, + "grad_norm": 1.464169685405528, + "learning_rate": 3.6871097886837565e-06, + "loss": 0.9079, + "step": 2729 + }, + { + "epoch": 0.20517059972944537, + "grad_norm": 1.5654903026513172, + "learning_rate": 3.68684828353188e-06, + "loss": 0.9487, + "step": 2730 + }, + { + "epoch": 0.20524575379528032, + "grad_norm": 1.8752099237547228, + "learning_rate": 3.6865866784277836e-06, + "loss": 1.085, + "step": 2731 + }, + { + "epoch": 0.2053209078611153, + "grad_norm": 1.6302712847790388, + "learning_rate": 3.6863249733869683e-06, + "loss": 1.0358, + "step": 2732 + }, + { + "epoch": 0.20539606192695026, + "grad_norm": 1.7697551661412083, + "learning_rate": 3.6860631684249403e-06, + "loss": 0.9339, + "step": 2733 + }, + { + "epoch": 0.2054712159927852, + "grad_norm": 1.7632231453046263, + "learning_rate": 3.685801263557214e-06, + "loss": 0.9249, + "step": 2734 + }, + { + "epoch": 0.20554637005862017, + "grad_norm": 1.7232265090791221, + "learning_rate": 3.6855392587993065e-06, + "loss": 1.0084, + "step": 2735 + }, + { + "epoch": 0.20562152412445514, + "grad_norm": 1.6749280344747817, + "learning_rate": 3.6852771541667444e-06, + "loss": 0.9373, + "step": 2736 + }, + { + "epoch": 0.20569667819029008, + "grad_norm": 1.4941892129814294, + "learning_rate": 3.6850149496750575e-06, + "loss": 1.0576, + "step": 2737 + }, + { + "epoch": 0.20577183225612505, + "grad_norm": 0.8320298172486197, + "learning_rate": 3.684752645339782e-06, + "loss": 0.8426, + "step": 2738 + }, + { + "epoch": 0.20584698632196002, + "grad_norm": 1.7731785522162034, + "learning_rate": 3.6844902411764612e-06, + "loss": 0.9419, + "step": 2739 + }, + { + "epoch": 0.20592214038779497, + "grad_norm": 1.602839263377221, + "learning_rate": 3.6842277372006434e-06, + "loss": 1.059, + "step": 2740 + }, + { + "epoch": 0.20599729445362994, + "grad_norm": 1.6654495404494989, + "learning_rate": 3.6839651334278823e-06, + "loss": 1.0382, + "step": 2741 + }, + { + "epoch": 0.2060724485194649, + "grad_norm": 1.6680936029594542, + "learning_rate": 3.6837024298737393e-06, + "loss": 0.9938, + "step": 2742 + }, + { + "epoch": 0.20614760258529988, + "grad_norm": 2.2793833790555285, + "learning_rate": 3.68343962655378e-06, + "loss": 0.9917, + "step": 2743 + }, + { + "epoch": 0.20622275665113482, + "grad_norm": 1.2897836690827387, + "learning_rate": 3.6831767234835763e-06, + "loss": 1.0304, + "step": 2744 + }, + { + "epoch": 0.2062979107169698, + "grad_norm": 2.0339125569094785, + "learning_rate": 3.6829137206787065e-06, + "loss": 0.9907, + "step": 2745 + }, + { + "epoch": 0.20637306478280476, + "grad_norm": 1.4244852943451336, + "learning_rate": 3.6826506181547543e-06, + "loss": 0.9812, + "step": 2746 + }, + { + "epoch": 0.2064482188486397, + "grad_norm": 1.4146605012245785, + "learning_rate": 3.6823874159273095e-06, + "loss": 0.9067, + "step": 2747 + }, + { + "epoch": 0.20652337291447467, + "grad_norm": 1.4934578326530006, + "learning_rate": 3.6821241140119685e-06, + "loss": 0.9408, + "step": 2748 + }, + { + "epoch": 0.20659852698030964, + "grad_norm": 2.07947628287099, + "learning_rate": 3.6818607124243322e-06, + "loss": 1.0016, + "step": 2749 + }, + { + "epoch": 0.2066736810461446, + "grad_norm": 1.2262822213639937, + "learning_rate": 3.6815972111800082e-06, + "loss": 0.9583, + "step": 2750 + }, + { + "epoch": 0.20674883511197956, + "grad_norm": 2.1382535621152226, + "learning_rate": 3.6813336102946107e-06, + "loss": 0.93, + "step": 2751 + }, + { + "epoch": 0.20682398917781453, + "grad_norm": 1.7643763760471147, + "learning_rate": 3.681069909783758e-06, + "loss": 0.9291, + "step": 2752 + }, + { + "epoch": 0.20689914324364947, + "grad_norm": 0.8598637664810882, + "learning_rate": 3.6808061096630765e-06, + "loss": 0.815, + "step": 2753 + }, + { + "epoch": 0.20697429730948444, + "grad_norm": 0.8626513707666059, + "learning_rate": 3.6805422099481965e-06, + "loss": 0.827, + "step": 2754 + }, + { + "epoch": 0.2070494513753194, + "grad_norm": 3.0950364029673962, + "learning_rate": 3.6802782106547553e-06, + "loss": 0.976, + "step": 2755 + }, + { + "epoch": 0.20712460544115438, + "grad_norm": 1.56423838233383, + "learning_rate": 3.6800141117983954e-06, + "loss": 0.8792, + "step": 2756 + }, + { + "epoch": 0.20719975950698932, + "grad_norm": 1.6810434944929882, + "learning_rate": 3.679749913394767e-06, + "loss": 0.9142, + "step": 2757 + }, + { + "epoch": 0.2072749135728243, + "grad_norm": 1.5821588912146467, + "learning_rate": 3.6794856154595235e-06, + "loss": 0.9752, + "step": 2758 + }, + { + "epoch": 0.20735006763865926, + "grad_norm": 0.9269241167714666, + "learning_rate": 3.679221218008326e-06, + "loss": 0.835, + "step": 2759 + }, + { + "epoch": 0.2074252217044942, + "grad_norm": 1.7288316685603498, + "learning_rate": 3.6789567210568417e-06, + "loss": 0.9858, + "step": 2760 + }, + { + "epoch": 0.20750037577032918, + "grad_norm": 1.8888667103768075, + "learning_rate": 3.678692124620742e-06, + "loss": 1.0517, + "step": 2761 + }, + { + "epoch": 0.20757552983616415, + "grad_norm": 1.6318865619251302, + "learning_rate": 3.6784274287157066e-06, + "loss": 0.905, + "step": 2762 + }, + { + "epoch": 0.2076506839019991, + "grad_norm": 1.818678683426293, + "learning_rate": 3.678162633357418e-06, + "loss": 1.0048, + "step": 2763 + }, + { + "epoch": 0.20772583796783406, + "grad_norm": 1.6430341198768788, + "learning_rate": 3.6778977385615676e-06, + "loss": 0.9747, + "step": 2764 + }, + { + "epoch": 0.20780099203366903, + "grad_norm": 1.6089985231211388, + "learning_rate": 3.677632744343851e-06, + "loss": 0.9895, + "step": 2765 + }, + { + "epoch": 0.20787614609950397, + "grad_norm": 1.4369592064022756, + "learning_rate": 3.6773676507199703e-06, + "loss": 0.9058, + "step": 2766 + }, + { + "epoch": 0.20795130016533894, + "grad_norm": 1.5123216476583121, + "learning_rate": 3.6771024577056333e-06, + "loss": 0.991, + "step": 2767 + }, + { + "epoch": 0.2080264542311739, + "grad_norm": 2.1527015125616136, + "learning_rate": 3.6768371653165537e-06, + "loss": 0.9471, + "step": 2768 + }, + { + "epoch": 0.20810160829700886, + "grad_norm": 1.7929543222416173, + "learning_rate": 3.676571773568451e-06, + "loss": 0.9295, + "step": 2769 + }, + { + "epoch": 0.20817676236284383, + "grad_norm": 2.042430799565656, + "learning_rate": 3.67630628247705e-06, + "loss": 1.0129, + "step": 2770 + }, + { + "epoch": 0.2082519164286788, + "grad_norm": 2.222625785846806, + "learning_rate": 3.6760406920580834e-06, + "loss": 1.1409, + "step": 2771 + }, + { + "epoch": 0.20832707049451377, + "grad_norm": 1.789197158743165, + "learning_rate": 3.6757750023272882e-06, + "loss": 0.994, + "step": 2772 + }, + { + "epoch": 0.2084022245603487, + "grad_norm": 1.5265457136779383, + "learning_rate": 3.6755092133004062e-06, + "loss": 1.0519, + "step": 2773 + }, + { + "epoch": 0.20847737862618368, + "grad_norm": 1.847541531314092, + "learning_rate": 3.6752433249931876e-06, + "loss": 0.9012, + "step": 2774 + }, + { + "epoch": 0.20855253269201865, + "grad_norm": 1.6508581097569879, + "learning_rate": 3.6749773374213877e-06, + "loss": 0.9156, + "step": 2775 + }, + { + "epoch": 0.2086276867578536, + "grad_norm": 1.6349094195458136, + "learning_rate": 3.674711250600766e-06, + "loss": 0.9291, + "step": 2776 + }, + { + "epoch": 0.20870284082368856, + "grad_norm": 2.160973613268999, + "learning_rate": 3.6744450645470904e-06, + "loss": 0.9672, + "step": 2777 + }, + { + "epoch": 0.20877799488952353, + "grad_norm": 2.609599212333425, + "learning_rate": 3.6741787792761324e-06, + "loss": 0.9311, + "step": 2778 + }, + { + "epoch": 0.20885314895535848, + "grad_norm": 1.8758352326708747, + "learning_rate": 3.673912394803671e-06, + "loss": 0.9442, + "step": 2779 + }, + { + "epoch": 0.20892830302119345, + "grad_norm": 1.3146058400560499, + "learning_rate": 3.6736459111454903e-06, + "loss": 0.9791, + "step": 2780 + }, + { + "epoch": 0.20900345708702842, + "grad_norm": 10.338418837169172, + "learning_rate": 3.6733793283173805e-06, + "loss": 0.9974, + "step": 2781 + }, + { + "epoch": 0.20907861115286336, + "grad_norm": 1.381200119774971, + "learning_rate": 3.673112646335138e-06, + "loss": 1.0534, + "step": 2782 + }, + { + "epoch": 0.20915376521869833, + "grad_norm": 1.6166990712775529, + "learning_rate": 3.672845865214564e-06, + "loss": 1.0138, + "step": 2783 + }, + { + "epoch": 0.2092289192845333, + "grad_norm": 1.6498701911496076, + "learning_rate": 3.6725789849714665e-06, + "loss": 0.9212, + "step": 2784 + }, + { + "epoch": 0.20930407335036824, + "grad_norm": 1.44255371734192, + "learning_rate": 3.67231200562166e-06, + "loss": 0.9477, + "step": 2785 + }, + { + "epoch": 0.2093792274162032, + "grad_norm": 1.9348767888469796, + "learning_rate": 3.6720449271809633e-06, + "loss": 1.0246, + "step": 2786 + }, + { + "epoch": 0.20945438148203818, + "grad_norm": 1.721193553300966, + "learning_rate": 3.6717777496652023e-06, + "loss": 0.9348, + "step": 2787 + }, + { + "epoch": 0.20952953554787315, + "grad_norm": 2.1080037498947464, + "learning_rate": 3.6715104730902074e-06, + "loss": 1.0436, + "step": 2788 + }, + { + "epoch": 0.2096046896137081, + "grad_norm": 1.2682327907544417, + "learning_rate": 3.671243097471817e-06, + "loss": 0.956, + "step": 2789 + }, + { + "epoch": 0.20967984367954307, + "grad_norm": 1.9601260630296562, + "learning_rate": 3.6709756228258728e-06, + "loss": 1.0614, + "step": 2790 + }, + { + "epoch": 0.20975499774537804, + "grad_norm": 1.8527994008147304, + "learning_rate": 3.6707080491682243e-06, + "loss": 0.9957, + "step": 2791 + }, + { + "epoch": 0.20983015181121298, + "grad_norm": 1.7904470491779516, + "learning_rate": 3.670440376514727e-06, + "loss": 0.9091, + "step": 2792 + }, + { + "epoch": 0.20990530587704795, + "grad_norm": 1.6022146990442623, + "learning_rate": 3.67017260488124e-06, + "loss": 0.9482, + "step": 2793 + }, + { + "epoch": 0.20998045994288292, + "grad_norm": 0.8651140185274976, + "learning_rate": 3.6699047342836313e-06, + "loss": 0.8618, + "step": 2794 + }, + { + "epoch": 0.21005561400871786, + "grad_norm": 1.5457201413289725, + "learning_rate": 3.669636764737772e-06, + "loss": 0.9422, + "step": 2795 + }, + { + "epoch": 0.21013076807455283, + "grad_norm": 1.5864558382905947, + "learning_rate": 3.669368696259542e-06, + "loss": 1.077, + "step": 2796 + }, + { + "epoch": 0.2102059221403878, + "grad_norm": 1.5454835124029551, + "learning_rate": 3.669100528864823e-06, + "loss": 1.0331, + "step": 2797 + }, + { + "epoch": 0.21028107620622274, + "grad_norm": 2.2238915200363087, + "learning_rate": 3.6688322625695075e-06, + "loss": 0.9095, + "step": 2798 + }, + { + "epoch": 0.21035623027205771, + "grad_norm": 0.8731233704100511, + "learning_rate": 3.6685638973894896e-06, + "loss": 0.8859, + "step": 2799 + }, + { + "epoch": 0.21043138433789269, + "grad_norm": 2.224754804323742, + "learning_rate": 3.6682954333406707e-06, + "loss": 1.0849, + "step": 2800 + }, + { + "epoch": 0.21050653840372766, + "grad_norm": 2.0495199339195733, + "learning_rate": 3.6680268704389597e-06, + "loss": 0.9689, + "step": 2801 + }, + { + "epoch": 0.2105816924695626, + "grad_norm": 1.4272131849182406, + "learning_rate": 3.6677582087002695e-06, + "loss": 0.9876, + "step": 2802 + }, + { + "epoch": 0.21065684653539757, + "grad_norm": 1.8518020539299245, + "learning_rate": 3.6674894481405184e-06, + "loss": 1.0297, + "step": 2803 + }, + { + "epoch": 0.21073200060123254, + "grad_norm": 1.9816381850360127, + "learning_rate": 3.667220588775633e-06, + "loss": 0.8362, + "step": 2804 + }, + { + "epoch": 0.21080715466706748, + "grad_norm": 1.4543208462248067, + "learning_rate": 3.6669516306215433e-06, + "loss": 0.8958, + "step": 2805 + }, + { + "epoch": 0.21088230873290245, + "grad_norm": 1.5186262009638132, + "learning_rate": 3.666682573694186e-06, + "loss": 0.9311, + "step": 2806 + }, + { + "epoch": 0.21095746279873742, + "grad_norm": 1.655152226964166, + "learning_rate": 3.6664134180095045e-06, + "loss": 1.0027, + "step": 2807 + }, + { + "epoch": 0.21103261686457236, + "grad_norm": 1.659874099989747, + "learning_rate": 3.666144163583446e-06, + "loss": 1.0045, + "step": 2808 + }, + { + "epoch": 0.21110777093040733, + "grad_norm": 1.337913880528951, + "learning_rate": 3.6658748104319667e-06, + "loss": 0.9859, + "step": 2809 + }, + { + "epoch": 0.2111829249962423, + "grad_norm": 1.6633388038091712, + "learning_rate": 3.665605358571026e-06, + "loss": 0.9828, + "step": 2810 + }, + { + "epoch": 0.21125807906207725, + "grad_norm": 1.908960073186419, + "learning_rate": 3.6653358080165893e-06, + "loss": 0.9079, + "step": 2811 + }, + { + "epoch": 0.21133323312791222, + "grad_norm": 0.9111344479528798, + "learning_rate": 3.6650661587846283e-06, + "loss": 0.8845, + "step": 2812 + }, + { + "epoch": 0.2114083871937472, + "grad_norm": 1.5696479567890225, + "learning_rate": 3.6647964108911226e-06, + "loss": 1.0517, + "step": 2813 + }, + { + "epoch": 0.21148354125958213, + "grad_norm": 1.4909467881632388, + "learning_rate": 3.6645265643520536e-06, + "loss": 0.8753, + "step": 2814 + }, + { + "epoch": 0.2115586953254171, + "grad_norm": 1.5035793196798257, + "learning_rate": 3.664256619183413e-06, + "loss": 1.0046, + "step": 2815 + }, + { + "epoch": 0.21163384939125207, + "grad_norm": 0.8418328793077057, + "learning_rate": 3.6639865754011934e-06, + "loss": 0.775, + "step": 2816 + }, + { + "epoch": 0.21170900345708704, + "grad_norm": 1.4209322476379087, + "learning_rate": 3.663716433021398e-06, + "loss": 0.9587, + "step": 2817 + }, + { + "epoch": 0.21178415752292198, + "grad_norm": 2.554396959867977, + "learning_rate": 3.6634461920600337e-06, + "loss": 0.9807, + "step": 2818 + }, + { + "epoch": 0.21185931158875695, + "grad_norm": 1.944378302978143, + "learning_rate": 3.6631758525331124e-06, + "loss": 0.9487, + "step": 2819 + }, + { + "epoch": 0.21193446565459192, + "grad_norm": 1.5171473614679958, + "learning_rate": 3.662905414456653e-06, + "loss": 0.9371, + "step": 2820 + }, + { + "epoch": 0.21200961972042687, + "grad_norm": 1.663630613060311, + "learning_rate": 3.66263487784668e-06, + "loss": 1.0015, + "step": 2821 + }, + { + "epoch": 0.21208477378626184, + "grad_norm": 2.105099562512619, + "learning_rate": 3.6623642427192237e-06, + "loss": 0.8925, + "step": 2822 + }, + { + "epoch": 0.2121599278520968, + "grad_norm": 1.7159890627612506, + "learning_rate": 3.6620935090903205e-06, + "loss": 1.0108, + "step": 2823 + }, + { + "epoch": 0.21223508191793175, + "grad_norm": 2.056396817702432, + "learning_rate": 3.6618226769760127e-06, + "loss": 0.994, + "step": 2824 + }, + { + "epoch": 0.21231023598376672, + "grad_norm": 1.6405322820712998, + "learning_rate": 3.6615517463923477e-06, + "loss": 0.9753, + "step": 2825 + }, + { + "epoch": 0.2123853900496017, + "grad_norm": 1.5908059997102064, + "learning_rate": 3.661280717355379e-06, + "loss": 1.0061, + "step": 2826 + }, + { + "epoch": 0.21246054411543663, + "grad_norm": 1.6334389392940587, + "learning_rate": 3.661009589881166e-06, + "loss": 0.9821, + "step": 2827 + }, + { + "epoch": 0.2125356981812716, + "grad_norm": 1.7015547206620263, + "learning_rate": 3.660738363985775e-06, + "loss": 0.9278, + "step": 2828 + }, + { + "epoch": 0.21261085224710657, + "grad_norm": 0.805450374123559, + "learning_rate": 3.660467039685276e-06, + "loss": 0.7735, + "step": 2829 + }, + { + "epoch": 0.21268600631294152, + "grad_norm": 1.4837531142294516, + "learning_rate": 3.660195616995747e-06, + "loss": 1.0388, + "step": 2830 + }, + { + "epoch": 0.2127611603787765, + "grad_norm": 1.5090978016440013, + "learning_rate": 3.6599240959332704e-06, + "loss": 1.0568, + "step": 2831 + }, + { + "epoch": 0.21283631444461146, + "grad_norm": 1.5613120331363248, + "learning_rate": 3.659652476513934e-06, + "loss": 1.0245, + "step": 2832 + }, + { + "epoch": 0.21291146851044643, + "grad_norm": 1.4779532205717765, + "learning_rate": 3.6593807587538343e-06, + "loss": 0.9946, + "step": 2833 + }, + { + "epoch": 0.21298662257628137, + "grad_norm": 0.9753483357884688, + "learning_rate": 3.6591089426690695e-06, + "loss": 0.9241, + "step": 2834 + }, + { + "epoch": 0.21306177664211634, + "grad_norm": 1.7303715456835997, + "learning_rate": 3.658837028275747e-06, + "loss": 1.0184, + "step": 2835 + }, + { + "epoch": 0.2131369307079513, + "grad_norm": 1.6507192431248878, + "learning_rate": 3.6585650155899786e-06, + "loss": 1.0139, + "step": 2836 + }, + { + "epoch": 0.21321208477378625, + "grad_norm": 1.616800971531753, + "learning_rate": 3.658292904627882e-06, + "loss": 0.96, + "step": 2837 + }, + { + "epoch": 0.21328723883962122, + "grad_norm": 1.4346421900868769, + "learning_rate": 3.6580206954055807e-06, + "loss": 1.0777, + "step": 2838 + }, + { + "epoch": 0.2133623929054562, + "grad_norm": 1.4158308237437878, + "learning_rate": 3.657748387939204e-06, + "loss": 0.943, + "step": 2839 + }, + { + "epoch": 0.21343754697129114, + "grad_norm": 1.583293646599488, + "learning_rate": 3.657475982244888e-06, + "loss": 0.8327, + "step": 2840 + }, + { + "epoch": 0.2135127010371261, + "grad_norm": 1.868600478298547, + "learning_rate": 3.6572034783387725e-06, + "loss": 0.8888, + "step": 2841 + }, + { + "epoch": 0.21358785510296108, + "grad_norm": 1.0004403813570204, + "learning_rate": 3.6569308762370056e-06, + "loss": 0.7681, + "step": 2842 + }, + { + "epoch": 0.21366300916879602, + "grad_norm": 1.1561128689664322, + "learning_rate": 3.6566581759557387e-06, + "loss": 1.0419, + "step": 2843 + }, + { + "epoch": 0.213738163234631, + "grad_norm": 1.5649666961197575, + "learning_rate": 3.656385377511132e-06, + "loss": 0.9603, + "step": 2844 + }, + { + "epoch": 0.21381331730046596, + "grad_norm": 1.4431007576843773, + "learning_rate": 3.656112480919348e-06, + "loss": 0.9431, + "step": 2845 + }, + { + "epoch": 0.21388847136630093, + "grad_norm": 1.5663867673351983, + "learning_rate": 3.6558394861965587e-06, + "loss": 0.924, + "step": 2846 + }, + { + "epoch": 0.21396362543213587, + "grad_norm": 1.6708280190679017, + "learning_rate": 3.6555663933589384e-06, + "loss": 0.9156, + "step": 2847 + }, + { + "epoch": 0.21403877949797084, + "grad_norm": 0.8476491831120129, + "learning_rate": 3.655293202422671e-06, + "loss": 0.8057, + "step": 2848 + }, + { + "epoch": 0.2141139335638058, + "grad_norm": 1.400755171099569, + "learning_rate": 3.6550199134039414e-06, + "loss": 0.9967, + "step": 2849 + }, + { + "epoch": 0.21418908762964076, + "grad_norm": 1.9689223198938355, + "learning_rate": 3.654746526318945e-06, + "loss": 0.943, + "step": 2850 + }, + { + "epoch": 0.21426424169547573, + "grad_norm": 1.6515160329411784, + "learning_rate": 3.6544730411838805e-06, + "loss": 0.9228, + "step": 2851 + }, + { + "epoch": 0.2143393957613107, + "grad_norm": 1.4989452627467954, + "learning_rate": 3.654199458014953e-06, + "loss": 0.9888, + "step": 2852 + }, + { + "epoch": 0.21441454982714564, + "grad_norm": 1.6089574772234299, + "learning_rate": 3.653925776828373e-06, + "loss": 1.0044, + "step": 2853 + }, + { + "epoch": 0.2144897038929806, + "grad_norm": 1.9305978704826665, + "learning_rate": 3.653651997640358e-06, + "loss": 0.8575, + "step": 2854 + }, + { + "epoch": 0.21456485795881558, + "grad_norm": 0.8384219176912806, + "learning_rate": 3.6533781204671296e-06, + "loss": 0.8315, + "step": 2855 + }, + { + "epoch": 0.21464001202465052, + "grad_norm": 1.6126627255917654, + "learning_rate": 3.6531041453249154e-06, + "loss": 1.0385, + "step": 2856 + }, + { + "epoch": 0.2147151660904855, + "grad_norm": 1.6210683062291649, + "learning_rate": 3.6528300722299515e-06, + "loss": 0.9867, + "step": 2857 + }, + { + "epoch": 0.21479032015632046, + "grad_norm": 1.5634664979571435, + "learning_rate": 3.6525559011984768e-06, + "loss": 0.9132, + "step": 2858 + }, + { + "epoch": 0.2148654742221554, + "grad_norm": 1.062439524107544, + "learning_rate": 3.652281632246736e-06, + "loss": 0.8207, + "step": 2859 + }, + { + "epoch": 0.21494062828799038, + "grad_norm": 1.672333698985624, + "learning_rate": 3.6520072653909823e-06, + "loss": 1.0072, + "step": 2860 + }, + { + "epoch": 0.21501578235382535, + "grad_norm": 1.8523504455780047, + "learning_rate": 3.6517328006474717e-06, + "loss": 1.0461, + "step": 2861 + }, + { + "epoch": 0.21509093641966032, + "grad_norm": 1.6896523771591743, + "learning_rate": 3.651458238032468e-06, + "loss": 1.0131, + "step": 2862 + }, + { + "epoch": 0.21516609048549526, + "grad_norm": 1.3755820516271164, + "learning_rate": 3.65118357756224e-06, + "loss": 1.025, + "step": 2863 + }, + { + "epoch": 0.21524124455133023, + "grad_norm": 2.1252094614353583, + "learning_rate": 3.650908819253062e-06, + "loss": 0.9509, + "step": 2864 + }, + { + "epoch": 0.2153163986171652, + "grad_norm": 1.83377772782886, + "learning_rate": 3.6506339631212145e-06, + "loss": 1.0607, + "step": 2865 + }, + { + "epoch": 0.21539155268300014, + "grad_norm": 2.012600307452673, + "learning_rate": 3.650359009182984e-06, + "loss": 0.857, + "step": 2866 + }, + { + "epoch": 0.2154667067488351, + "grad_norm": 1.850474879170853, + "learning_rate": 3.650083957454663e-06, + "loss": 1.0305, + "step": 2867 + }, + { + "epoch": 0.21554186081467008, + "grad_norm": 1.3971538939473318, + "learning_rate": 3.6498088079525487e-06, + "loss": 0.9382, + "step": 2868 + }, + { + "epoch": 0.21561701488050503, + "grad_norm": 2.1041833354428205, + "learning_rate": 3.649533560692945e-06, + "loss": 0.986, + "step": 2869 + }, + { + "epoch": 0.21569216894634, + "grad_norm": 1.5069546804163472, + "learning_rate": 3.6492582156921615e-06, + "loss": 0.9926, + "step": 2870 + }, + { + "epoch": 0.21576732301217497, + "grad_norm": 1.627681134247853, + "learning_rate": 3.648982772966513e-06, + "loss": 1.0112, + "step": 2871 + }, + { + "epoch": 0.2158424770780099, + "grad_norm": 1.603984508658395, + "learning_rate": 3.648707232532321e-06, + "loss": 1.0739, + "step": 2872 + }, + { + "epoch": 0.21591763114384488, + "grad_norm": 1.8186759304419444, + "learning_rate": 3.648431594405912e-06, + "loss": 0.9513, + "step": 2873 + }, + { + "epoch": 0.21599278520967985, + "grad_norm": 1.6626818306518714, + "learning_rate": 3.648155858603619e-06, + "loss": 0.9384, + "step": 2874 + }, + { + "epoch": 0.2160679392755148, + "grad_norm": 1.8384859618770666, + "learning_rate": 3.64788002514178e-06, + "loss": 1.0124, + "step": 2875 + }, + { + "epoch": 0.21614309334134976, + "grad_norm": 1.2711632670294588, + "learning_rate": 3.6476040940367395e-06, + "loss": 0.9721, + "step": 2876 + }, + { + "epoch": 0.21621824740718473, + "grad_norm": 1.5982103949010058, + "learning_rate": 3.647328065304847e-06, + "loss": 0.8559, + "step": 2877 + }, + { + "epoch": 0.2162934014730197, + "grad_norm": 1.319619262000622, + "learning_rate": 3.6470519389624587e-06, + "loss": 0.9953, + "step": 2878 + }, + { + "epoch": 0.21636855553885465, + "grad_norm": 1.8250883263939321, + "learning_rate": 3.646775715025936e-06, + "loss": 1.0194, + "step": 2879 + }, + { + "epoch": 0.21644370960468962, + "grad_norm": 1.5321392669025928, + "learning_rate": 3.6464993935116464e-06, + "loss": 0.9704, + "step": 2880 + }, + { + "epoch": 0.2165188636705246, + "grad_norm": 1.3821548794772356, + "learning_rate": 3.646222974435963e-06, + "loss": 1.0101, + "step": 2881 + }, + { + "epoch": 0.21659401773635953, + "grad_norm": 0.9921147177126036, + "learning_rate": 3.645946457815264e-06, + "loss": 0.8618, + "step": 2882 + }, + { + "epoch": 0.2166691718021945, + "grad_norm": 1.785644397730375, + "learning_rate": 3.6456698436659353e-06, + "loss": 0.934, + "step": 2883 + }, + { + "epoch": 0.21674432586802947, + "grad_norm": 1.2561178697779813, + "learning_rate": 3.645393132004367e-06, + "loss": 0.9419, + "step": 2884 + }, + { + "epoch": 0.2168194799338644, + "grad_norm": 3.221810080672486, + "learning_rate": 3.6451163228469543e-06, + "loss": 0.8607, + "step": 2885 + }, + { + "epoch": 0.21689463399969938, + "grad_norm": 1.467229502506306, + "learning_rate": 3.6448394162100994e-06, + "loss": 0.9867, + "step": 2886 + }, + { + "epoch": 0.21696978806553435, + "grad_norm": 1.7099099753479876, + "learning_rate": 3.644562412110211e-06, + "loss": 0.972, + "step": 2887 + }, + { + "epoch": 0.2170449421313693, + "grad_norm": 1.4271959784756867, + "learning_rate": 3.6442853105637024e-06, + "loss": 0.9362, + "step": 2888 + }, + { + "epoch": 0.21712009619720427, + "grad_norm": 1.554471715012743, + "learning_rate": 3.644008111586993e-06, + "loss": 1.031, + "step": 2889 + }, + { + "epoch": 0.21719525026303924, + "grad_norm": 1.440732721037465, + "learning_rate": 3.6437308151965074e-06, + "loss": 0.8936, + "step": 2890 + }, + { + "epoch": 0.2172704043288742, + "grad_norm": 1.3535434120045235, + "learning_rate": 3.6434534214086767e-06, + "loss": 1.0133, + "step": 2891 + }, + { + "epoch": 0.21734555839470915, + "grad_norm": 1.8148696799033115, + "learning_rate": 3.643175930239938e-06, + "loss": 1.1097, + "step": 2892 + }, + { + "epoch": 0.21742071246054412, + "grad_norm": 1.3664573245484148, + "learning_rate": 3.6428983417067326e-06, + "loss": 0.9331, + "step": 2893 + }, + { + "epoch": 0.2174958665263791, + "grad_norm": 1.5769881128602135, + "learning_rate": 3.64262065582551e-06, + "loss": 0.9658, + "step": 2894 + }, + { + "epoch": 0.21757102059221403, + "grad_norm": 1.5590481585685811, + "learning_rate": 3.6423428726127232e-06, + "loss": 0.9385, + "step": 2895 + }, + { + "epoch": 0.217646174658049, + "grad_norm": 1.5787074458895263, + "learning_rate": 3.6420649920848324e-06, + "loss": 0.9868, + "step": 2896 + }, + { + "epoch": 0.21772132872388397, + "grad_norm": 1.646656538717241, + "learning_rate": 3.641787014258303e-06, + "loss": 0.9265, + "step": 2897 + }, + { + "epoch": 0.21779648278971891, + "grad_norm": 1.8462110355426324, + "learning_rate": 3.641508939149606e-06, + "loss": 0.9256, + "step": 2898 + }, + { + "epoch": 0.21787163685555389, + "grad_norm": 1.3334305201317156, + "learning_rate": 3.6412307667752185e-06, + "loss": 0.8879, + "step": 2899 + }, + { + "epoch": 0.21794679092138886, + "grad_norm": 1.5563677425348585, + "learning_rate": 3.640952497151623e-06, + "loss": 0.9735, + "step": 2900 + }, + { + "epoch": 0.2180219449872238, + "grad_norm": 1.9741506222864245, + "learning_rate": 3.6406741302953093e-06, + "loss": 1.0413, + "step": 2901 + }, + { + "epoch": 0.21809709905305877, + "grad_norm": 1.6849306113787395, + "learning_rate": 3.6403956662227706e-06, + "loss": 0.9865, + "step": 2902 + }, + { + "epoch": 0.21817225311889374, + "grad_norm": 1.5314415792393679, + "learning_rate": 3.640117104950507e-06, + "loss": 0.8932, + "step": 2903 + }, + { + "epoch": 0.21824740718472868, + "grad_norm": 1.3881459373602363, + "learning_rate": 3.639838446495024e-06, + "loss": 1.0075, + "step": 2904 + }, + { + "epoch": 0.21832256125056365, + "grad_norm": 1.6757919204478902, + "learning_rate": 3.6395596908728344e-06, + "loss": 0.9702, + "step": 2905 + }, + { + "epoch": 0.21839771531639862, + "grad_norm": 1.625136809825568, + "learning_rate": 3.639280838100455e-06, + "loss": 0.9452, + "step": 2906 + }, + { + "epoch": 0.2184728693822336, + "grad_norm": 1.5651171319290178, + "learning_rate": 3.639001888194408e-06, + "loss": 1.0038, + "step": 2907 + }, + { + "epoch": 0.21854802344806853, + "grad_norm": 1.7127903351083007, + "learning_rate": 3.638722841171223e-06, + "loss": 0.9038, + "step": 2908 + }, + { + "epoch": 0.2186231775139035, + "grad_norm": 1.0029308166794701, + "learning_rate": 3.6384436970474353e-06, + "loss": 0.7864, + "step": 2909 + }, + { + "epoch": 0.21869833157973848, + "grad_norm": 1.371455826729764, + "learning_rate": 3.638164455839584e-06, + "loss": 1.0439, + "step": 2910 + }, + { + "epoch": 0.21877348564557342, + "grad_norm": 1.3117545599533746, + "learning_rate": 3.637885117564216e-06, + "loss": 0.96, + "step": 2911 + }, + { + "epoch": 0.2188486397114084, + "grad_norm": 1.7256978489283397, + "learning_rate": 3.6376056822378826e-06, + "loss": 1.0775, + "step": 2912 + }, + { + "epoch": 0.21892379377724336, + "grad_norm": 4.956847708259843, + "learning_rate": 3.6373261498771418e-06, + "loss": 0.8917, + "step": 2913 + }, + { + "epoch": 0.2189989478430783, + "grad_norm": 1.5819466273905172, + "learning_rate": 3.6370465204985567e-06, + "loss": 0.9904, + "step": 2914 + }, + { + "epoch": 0.21907410190891327, + "grad_norm": 2.570519992604444, + "learning_rate": 3.636766794118697e-06, + "loss": 1.0518, + "step": 2915 + }, + { + "epoch": 0.21914925597474824, + "grad_norm": 1.560237925554426, + "learning_rate": 3.636486970754137e-06, + "loss": 0.9074, + "step": 2916 + }, + { + "epoch": 0.21922441004058318, + "grad_norm": 2.1933060025577173, + "learning_rate": 3.6362070504214577e-06, + "loss": 0.9348, + "step": 2917 + }, + { + "epoch": 0.21929956410641815, + "grad_norm": 1.8542171714276456, + "learning_rate": 3.6359270331372447e-06, + "loss": 1.0034, + "step": 2918 + }, + { + "epoch": 0.21937471817225312, + "grad_norm": 1.6284217754359553, + "learning_rate": 3.6356469189180907e-06, + "loss": 0.8479, + "step": 2919 + }, + { + "epoch": 0.21944987223808807, + "grad_norm": 2.2314701003820163, + "learning_rate": 3.6353667077805934e-06, + "loss": 1.007, + "step": 2920 + }, + { + "epoch": 0.21952502630392304, + "grad_norm": 1.335408736317972, + "learning_rate": 3.635086399741357e-06, + "loss": 0.9673, + "step": 2921 + }, + { + "epoch": 0.219600180369758, + "grad_norm": 2.5187968285232745, + "learning_rate": 3.6348059948169894e-06, + "loss": 0.9757, + "step": 2922 + }, + { + "epoch": 0.21967533443559298, + "grad_norm": 1.4097304612760935, + "learning_rate": 3.6345254930241075e-06, + "loss": 0.991, + "step": 2923 + }, + { + "epoch": 0.21975048850142792, + "grad_norm": 1.4652632862758523, + "learning_rate": 3.63424489437933e-06, + "loss": 1.0036, + "step": 2924 + }, + { + "epoch": 0.2198256425672629, + "grad_norm": 1.5855792914190168, + "learning_rate": 3.6339641988992853e-06, + "loss": 1.0084, + "step": 2925 + }, + { + "epoch": 0.21990079663309786, + "grad_norm": 1.6162570359031905, + "learning_rate": 3.633683406600605e-06, + "loss": 0.9547, + "step": 2926 + }, + { + "epoch": 0.2199759506989328, + "grad_norm": 1.5330693083272953, + "learning_rate": 3.633402517499927e-06, + "loss": 0.9477, + "step": 2927 + }, + { + "epoch": 0.22005110476476777, + "grad_norm": 1.9722169896923494, + "learning_rate": 3.633121531613895e-06, + "loss": 1.0365, + "step": 2928 + }, + { + "epoch": 0.22012625883060274, + "grad_norm": 1.8765655339682619, + "learning_rate": 3.6328404489591585e-06, + "loss": 1.0621, + "step": 2929 + }, + { + "epoch": 0.2202014128964377, + "grad_norm": 1.3643982254546878, + "learning_rate": 3.6325592695523727e-06, + "loss": 1.0593, + "step": 2930 + }, + { + "epoch": 0.22027656696227266, + "grad_norm": 1.3425735380323565, + "learning_rate": 3.6322779934101995e-06, + "loss": 1.0398, + "step": 2931 + }, + { + "epoch": 0.22035172102810763, + "grad_norm": 1.819079067179029, + "learning_rate": 3.6319966205493044e-06, + "loss": 1.0828, + "step": 2932 + }, + { + "epoch": 0.22042687509394257, + "grad_norm": 1.6703200194686478, + "learning_rate": 3.63171515098636e-06, + "loss": 1.0142, + "step": 2933 + }, + { + "epoch": 0.22050202915977754, + "grad_norm": 1.7032366374813181, + "learning_rate": 3.6314335847380443e-06, + "loss": 0.9489, + "step": 2934 + }, + { + "epoch": 0.2205771832256125, + "grad_norm": 1.5501179465671808, + "learning_rate": 3.631151921821042e-06, + "loss": 0.9909, + "step": 2935 + }, + { + "epoch": 0.22065233729144745, + "grad_norm": 1.6755451311700096, + "learning_rate": 3.630870162252042e-06, + "loss": 0.9918, + "step": 2936 + }, + { + "epoch": 0.22072749135728242, + "grad_norm": 0.9090586908548489, + "learning_rate": 3.63058830604774e-06, + "loss": 0.8574, + "step": 2937 + }, + { + "epoch": 0.2208026454231174, + "grad_norm": 1.733011482997564, + "learning_rate": 3.6303063532248367e-06, + "loss": 0.997, + "step": 2938 + }, + { + "epoch": 0.22087779948895236, + "grad_norm": 1.784251044088269, + "learning_rate": 3.6300243038000397e-06, + "loss": 0.9449, + "step": 2939 + }, + { + "epoch": 0.2209529535547873, + "grad_norm": 2.384085593402364, + "learning_rate": 3.6297421577900608e-06, + "loss": 0.9311, + "step": 2940 + }, + { + "epoch": 0.22102810762062228, + "grad_norm": 1.6323213705619408, + "learning_rate": 3.629459915211618e-06, + "loss": 0.9631, + "step": 2941 + }, + { + "epoch": 0.22110326168645725, + "grad_norm": 1.8481378875169554, + "learning_rate": 3.6291775760814358e-06, + "loss": 0.9867, + "step": 2942 + }, + { + "epoch": 0.2211784157522922, + "grad_norm": 1.5568719593877514, + "learning_rate": 3.6288951404162433e-06, + "loss": 1.129, + "step": 2943 + }, + { + "epoch": 0.22125356981812716, + "grad_norm": 1.6917631904738357, + "learning_rate": 3.6286126082327764e-06, + "loss": 0.9857, + "step": 2944 + }, + { + "epoch": 0.22132872388396213, + "grad_norm": 1.3840790029589658, + "learning_rate": 3.6283299795477767e-06, + "loss": 0.9934, + "step": 2945 + }, + { + "epoch": 0.22140387794979707, + "grad_norm": 1.7240243612900408, + "learning_rate": 3.62804725437799e-06, + "loss": 0.9937, + "step": 2946 + }, + { + "epoch": 0.22147903201563204, + "grad_norm": 1.7626117210961985, + "learning_rate": 3.6277644327401687e-06, + "loss": 1.0333, + "step": 2947 + }, + { + "epoch": 0.221554186081467, + "grad_norm": 1.404758043395486, + "learning_rate": 3.627481514651073e-06, + "loss": 0.9639, + "step": 2948 + }, + { + "epoch": 0.22162934014730196, + "grad_norm": 1.451443180891966, + "learning_rate": 3.6271985001274647e-06, + "loss": 0.8754, + "step": 2949 + }, + { + "epoch": 0.22170449421313693, + "grad_norm": 1.8533204823101013, + "learning_rate": 3.6269153891861137e-06, + "loss": 1.0002, + "step": 2950 + }, + { + "epoch": 0.2217796482789719, + "grad_norm": 1.7130658572264723, + "learning_rate": 3.6266321818437967e-06, + "loss": 0.9388, + "step": 2951 + }, + { + "epoch": 0.22185480234480687, + "grad_norm": 1.5440931641483626, + "learning_rate": 3.6263488781172946e-06, + "loss": 0.8873, + "step": 2952 + }, + { + "epoch": 0.2219299564106418, + "grad_norm": 1.5838743423269077, + "learning_rate": 3.6260654780233927e-06, + "loss": 1.0449, + "step": 2953 + }, + { + "epoch": 0.22200511047647678, + "grad_norm": 1.4916320086383021, + "learning_rate": 3.6257819815788854e-06, + "loss": 0.9538, + "step": 2954 + }, + { + "epoch": 0.22208026454231175, + "grad_norm": 0.832397308774756, + "learning_rate": 3.6254983888005697e-06, + "loss": 0.8769, + "step": 2955 + }, + { + "epoch": 0.2221554186081467, + "grad_norm": 1.4895810036096382, + "learning_rate": 3.6252146997052507e-06, + "loss": 1.026, + "step": 2956 + }, + { + "epoch": 0.22223057267398166, + "grad_norm": 1.9057418896491147, + "learning_rate": 3.624930914309736e-06, + "loss": 1.0077, + "step": 2957 + }, + { + "epoch": 0.22230572673981663, + "grad_norm": 2.991583621290133, + "learning_rate": 3.624647032630844e-06, + "loss": 1.0162, + "step": 2958 + }, + { + "epoch": 0.22238088080565158, + "grad_norm": 1.5365624430239226, + "learning_rate": 3.6243630546853932e-06, + "loss": 0.8853, + "step": 2959 + }, + { + "epoch": 0.22245603487148655, + "grad_norm": 1.6395470847453342, + "learning_rate": 3.6240789804902116e-06, + "loss": 1.0158, + "step": 2960 + }, + { + "epoch": 0.22253118893732152, + "grad_norm": 3.051287776302169, + "learning_rate": 3.623794810062131e-06, + "loss": 1.0775, + "step": 2961 + }, + { + "epoch": 0.22260634300315646, + "grad_norm": 1.6758604942383126, + "learning_rate": 3.62351054341799e-06, + "loss": 0.9015, + "step": 2962 + }, + { + "epoch": 0.22268149706899143, + "grad_norm": 0.743775846529091, + "learning_rate": 3.623226180574633e-06, + "loss": 0.7697, + "step": 2963 + }, + { + "epoch": 0.2227566511348264, + "grad_norm": 1.4773688158383047, + "learning_rate": 3.6229417215489084e-06, + "loss": 1.0081, + "step": 2964 + }, + { + "epoch": 0.22283180520066134, + "grad_norm": 1.3941271968589266, + "learning_rate": 3.6226571663576727e-06, + "loss": 0.9584, + "step": 2965 + }, + { + "epoch": 0.2229069592664963, + "grad_norm": 1.7360078840296451, + "learning_rate": 3.6223725150177858e-06, + "loss": 0.8721, + "step": 2966 + }, + { + "epoch": 0.22298211333233128, + "grad_norm": 1.3358158703140695, + "learning_rate": 3.622087767546116e-06, + "loss": 1.0458, + "step": 2967 + }, + { + "epoch": 0.22305726739816625, + "grad_norm": 1.740638899961682, + "learning_rate": 3.6218029239595332e-06, + "loss": 0.8713, + "step": 2968 + }, + { + "epoch": 0.2231324214640012, + "grad_norm": 1.3581828886317806, + "learning_rate": 3.6215179842749172e-06, + "loss": 1.0383, + "step": 2969 + }, + { + "epoch": 0.22320757552983617, + "grad_norm": 1.7978613758389155, + "learning_rate": 3.6212329485091518e-06, + "loss": 0.9312, + "step": 2970 + }, + { + "epoch": 0.22328272959567114, + "grad_norm": 1.981522018579803, + "learning_rate": 3.620947816679126e-06, + "loss": 1.1147, + "step": 2971 + }, + { + "epoch": 0.22335788366150608, + "grad_norm": 1.3554435969500684, + "learning_rate": 3.6206625888017355e-06, + "loss": 0.9617, + "step": 2972 + }, + { + "epoch": 0.22343303772734105, + "grad_norm": 1.4036887262003463, + "learning_rate": 3.62037726489388e-06, + "loss": 0.8994, + "step": 2973 + }, + { + "epoch": 0.22350819179317602, + "grad_norm": 1.3740540198072453, + "learning_rate": 3.620091844972467e-06, + "loss": 0.9942, + "step": 2974 + }, + { + "epoch": 0.22358334585901096, + "grad_norm": 1.667898136404142, + "learning_rate": 3.619806329054408e-06, + "loss": 1.0284, + "step": 2975 + }, + { + "epoch": 0.22365849992484593, + "grad_norm": 9.019299900264874, + "learning_rate": 3.619520717156622e-06, + "loss": 0.9919, + "step": 2976 + }, + { + "epoch": 0.2237336539906809, + "grad_norm": 1.538288444630085, + "learning_rate": 3.6192350092960315e-06, + "loss": 1.0508, + "step": 2977 + }, + { + "epoch": 0.22380880805651585, + "grad_norm": 1.4864031137405074, + "learning_rate": 3.6189492054895667e-06, + "loss": 0.9625, + "step": 2978 + }, + { + "epoch": 0.22388396212235082, + "grad_norm": 1.4081576230214778, + "learning_rate": 3.6186633057541617e-06, + "loss": 0.9702, + "step": 2979 + }, + { + "epoch": 0.2239591161881858, + "grad_norm": 1.5983594918256823, + "learning_rate": 3.6183773101067575e-06, + "loss": 1.0522, + "step": 2980 + }, + { + "epoch": 0.22403427025402073, + "grad_norm": 0.8056787108043348, + "learning_rate": 3.618091218564301e-06, + "loss": 0.8321, + "step": 2981 + }, + { + "epoch": 0.2241094243198557, + "grad_norm": 1.6397372068966645, + "learning_rate": 3.6178050311437432e-06, + "loss": 0.9642, + "step": 2982 + }, + { + "epoch": 0.22418457838569067, + "grad_norm": 1.7365566656794833, + "learning_rate": 3.6175187478620424e-06, + "loss": 1.0469, + "step": 2983 + }, + { + "epoch": 0.22425973245152564, + "grad_norm": 1.4432373821236262, + "learning_rate": 3.617232368736162e-06, + "loss": 0.9571, + "step": 2984 + }, + { + "epoch": 0.22433488651736058, + "grad_norm": 1.3964753264020657, + "learning_rate": 3.616945893783071e-06, + "loss": 0.9185, + "step": 2985 + }, + { + "epoch": 0.22441004058319555, + "grad_norm": 1.8664751170288323, + "learning_rate": 3.616659323019744e-06, + "loss": 0.9733, + "step": 2986 + }, + { + "epoch": 0.22448519464903052, + "grad_norm": 2.182126338647376, + "learning_rate": 3.616372656463161e-06, + "loss": 1.0105, + "step": 2987 + }, + { + "epoch": 0.22456034871486547, + "grad_norm": 1.6253103546732786, + "learning_rate": 3.6160858941303095e-06, + "loss": 1.053, + "step": 2988 + }, + { + "epoch": 0.22463550278070044, + "grad_norm": 1.1772248287171072, + "learning_rate": 3.61579903603818e-06, + "loss": 0.9809, + "step": 2989 + }, + { + "epoch": 0.2247106568465354, + "grad_norm": 1.4484519883699685, + "learning_rate": 3.6155120822037707e-06, + "loss": 0.9818, + "step": 2990 + }, + { + "epoch": 0.22478581091237035, + "grad_norm": 1.628652868529145, + "learning_rate": 3.6152250326440833e-06, + "loss": 1.0236, + "step": 2991 + }, + { + "epoch": 0.22486096497820532, + "grad_norm": 1.5093685679170339, + "learning_rate": 3.614937887376128e-06, + "loss": 0.8533, + "step": 2992 + }, + { + "epoch": 0.2249361190440403, + "grad_norm": 1.6121330701317362, + "learning_rate": 3.61465064641692e-06, + "loss": 0.9495, + "step": 2993 + }, + { + "epoch": 0.22501127310987523, + "grad_norm": 1.9255834636523375, + "learning_rate": 3.614363309783477e-06, + "loss": 0.9997, + "step": 2994 + }, + { + "epoch": 0.2250864271757102, + "grad_norm": 1.6684044440882266, + "learning_rate": 3.6140758774928265e-06, + "loss": 0.9899, + "step": 2995 + }, + { + "epoch": 0.22516158124154517, + "grad_norm": 1.1972545729060133, + "learning_rate": 3.613788349561999e-06, + "loss": 0.9206, + "step": 2996 + }, + { + "epoch": 0.22523673530738014, + "grad_norm": 1.45447797239428, + "learning_rate": 3.6135007260080334e-06, + "loss": 0.9659, + "step": 2997 + }, + { + "epoch": 0.22531188937321509, + "grad_norm": 1.8015908099263642, + "learning_rate": 3.61321300684797e-06, + "loss": 0.9871, + "step": 2998 + }, + { + "epoch": 0.22538704343905006, + "grad_norm": 1.4073760292787394, + "learning_rate": 3.6129251920988594e-06, + "loss": 1.0191, + "step": 2999 + }, + { + "epoch": 0.22546219750488503, + "grad_norm": 1.5682909654873356, + "learning_rate": 3.612637281777755e-06, + "loss": 1.0181, + "step": 3000 + }, + { + "epoch": 0.22553735157071997, + "grad_norm": 1.6739312477349704, + "learning_rate": 3.6123492759017155e-06, + "loss": 0.9361, + "step": 3001 + }, + { + "epoch": 0.22561250563655494, + "grad_norm": 1.4774973498480406, + "learning_rate": 3.6120611744878076e-06, + "loss": 1.052, + "step": 3002 + }, + { + "epoch": 0.2256876597023899, + "grad_norm": 1.802014439612824, + "learning_rate": 3.6117729775531028e-06, + "loss": 0.9874, + "step": 3003 + }, + { + "epoch": 0.22576281376822485, + "grad_norm": 1.6903203072687618, + "learning_rate": 3.6114846851146767e-06, + "loss": 0.8688, + "step": 3004 + }, + { + "epoch": 0.22583796783405982, + "grad_norm": 1.4724952346646931, + "learning_rate": 3.611196297189612e-06, + "loss": 1.0817, + "step": 3005 + }, + { + "epoch": 0.2259131218998948, + "grad_norm": 1.4883086434794415, + "learning_rate": 3.6109078137949975e-06, + "loss": 0.9024, + "step": 3006 + }, + { + "epoch": 0.22598827596572973, + "grad_norm": 2.0052696185634353, + "learning_rate": 3.6106192349479263e-06, + "loss": 1.095, + "step": 3007 + }, + { + "epoch": 0.2260634300315647, + "grad_norm": 1.8734116502367215, + "learning_rate": 3.610330560665498e-06, + "loss": 1.0106, + "step": 3008 + }, + { + "epoch": 0.22613858409739968, + "grad_norm": 1.896776516665789, + "learning_rate": 3.6100417909648175e-06, + "loss": 0.8562, + "step": 3009 + }, + { + "epoch": 0.22621373816323462, + "grad_norm": 1.5463728426812704, + "learning_rate": 3.6097529258629952e-06, + "loss": 1.0162, + "step": 3010 + }, + { + "epoch": 0.2262888922290696, + "grad_norm": 1.7919233345295955, + "learning_rate": 3.6094639653771486e-06, + "loss": 0.9105, + "step": 3011 + }, + { + "epoch": 0.22636404629490456, + "grad_norm": 1.6496498025271251, + "learning_rate": 3.6091749095243986e-06, + "loss": 0.9734, + "step": 3012 + }, + { + "epoch": 0.22643920036073953, + "grad_norm": 1.6289142269972918, + "learning_rate": 3.6088857583218735e-06, + "loss": 1.0374, + "step": 3013 + }, + { + "epoch": 0.22651435442657447, + "grad_norm": 1.362016369789887, + "learning_rate": 3.6085965117867066e-06, + "loss": 0.917, + "step": 3014 + }, + { + "epoch": 0.22658950849240944, + "grad_norm": 2.340651434836955, + "learning_rate": 3.608307169936036e-06, + "loss": 0.9236, + "step": 3015 + }, + { + "epoch": 0.2266646625582444, + "grad_norm": 1.6146433993415863, + "learning_rate": 3.608017732787007e-06, + "loss": 0.8925, + "step": 3016 + }, + { + "epoch": 0.22673981662407935, + "grad_norm": 2.1171405406966444, + "learning_rate": 3.60772820035677e-06, + "loss": 0.9435, + "step": 3017 + }, + { + "epoch": 0.22681497068991432, + "grad_norm": 1.7728562263678738, + "learning_rate": 3.607438572662481e-06, + "loss": 1.0239, + "step": 3018 + }, + { + "epoch": 0.2268901247557493, + "grad_norm": 1.4272308923027266, + "learning_rate": 3.6071488497213017e-06, + "loss": 0.91, + "step": 3019 + }, + { + "epoch": 0.22696527882158424, + "grad_norm": 1.6389258895978909, + "learning_rate": 3.6068590315503976e-06, + "loss": 0.8809, + "step": 3020 + }, + { + "epoch": 0.2270404328874192, + "grad_norm": 1.5921450743109398, + "learning_rate": 3.606569118166944e-06, + "loss": 0.9634, + "step": 3021 + }, + { + "epoch": 0.22711558695325418, + "grad_norm": 1.870887251192384, + "learning_rate": 3.6062791095881174e-06, + "loss": 0.9922, + "step": 3022 + }, + { + "epoch": 0.22719074101908912, + "grad_norm": 1.3031502993389659, + "learning_rate": 3.6059890058311025e-06, + "loss": 0.9724, + "step": 3023 + }, + { + "epoch": 0.2272658950849241, + "grad_norm": 1.4728766562258213, + "learning_rate": 3.6056988069130903e-06, + "loss": 0.8964, + "step": 3024 + }, + { + "epoch": 0.22734104915075906, + "grad_norm": 1.7533724849859933, + "learning_rate": 3.6054085128512747e-06, + "loss": 0.9561, + "step": 3025 + }, + { + "epoch": 0.227416203216594, + "grad_norm": 1.531882953226275, + "learning_rate": 3.605118123662857e-06, + "loss": 0.9277, + "step": 3026 + }, + { + "epoch": 0.22749135728242897, + "grad_norm": 2.0976546587662726, + "learning_rate": 3.6048276393650434e-06, + "loss": 0.9826, + "step": 3027 + }, + { + "epoch": 0.22756651134826394, + "grad_norm": 1.4526753704154014, + "learning_rate": 3.6045370599750482e-06, + "loss": 0.9643, + "step": 3028 + }, + { + "epoch": 0.22764166541409891, + "grad_norm": 1.63630805860406, + "learning_rate": 3.6042463855100876e-06, + "loss": 0.939, + "step": 3029 + }, + { + "epoch": 0.22771681947993386, + "grad_norm": 1.571727058623049, + "learning_rate": 3.603955615987385e-06, + "loss": 0.9418, + "step": 3030 + }, + { + "epoch": 0.22779197354576883, + "grad_norm": 1.7778624731326698, + "learning_rate": 3.603664751424171e-06, + "loss": 0.9685, + "step": 3031 + }, + { + "epoch": 0.2278671276116038, + "grad_norm": 1.670407373001737, + "learning_rate": 3.603373791837679e-06, + "loss": 1.0299, + "step": 3032 + }, + { + "epoch": 0.22794228167743874, + "grad_norm": 1.9486804287611885, + "learning_rate": 3.6030827372451506e-06, + "loss": 0.8633, + "step": 3033 + }, + { + "epoch": 0.2280174357432737, + "grad_norm": 1.8648880721963097, + "learning_rate": 3.602791587663831e-06, + "loss": 0.8809, + "step": 3034 + }, + { + "epoch": 0.22809258980910868, + "grad_norm": 1.5247253452052134, + "learning_rate": 3.6025003431109722e-06, + "loss": 0.9744, + "step": 3035 + }, + { + "epoch": 0.22816774387494362, + "grad_norm": 1.4564469625845553, + "learning_rate": 3.6022090036038326e-06, + "loss": 1.0879, + "step": 3036 + }, + { + "epoch": 0.2282428979407786, + "grad_norm": 1.2878327023153404, + "learning_rate": 3.601917569159673e-06, + "loss": 1.0784, + "step": 3037 + }, + { + "epoch": 0.22831805200661356, + "grad_norm": 1.8903477993616067, + "learning_rate": 3.6016260397957642e-06, + "loss": 0.9983, + "step": 3038 + }, + { + "epoch": 0.2283932060724485, + "grad_norm": 0.7622354173475839, + "learning_rate": 3.6013344155293792e-06, + "loss": 0.7742, + "step": 3039 + }, + { + "epoch": 0.22846836013828348, + "grad_norm": 1.5198140408300693, + "learning_rate": 3.6010426963777985e-06, + "loss": 0.9508, + "step": 3040 + }, + { + "epoch": 0.22854351420411845, + "grad_norm": 1.4660778492895976, + "learning_rate": 3.600750882358307e-06, + "loss": 1.0606, + "step": 3041 + }, + { + "epoch": 0.22861866826995342, + "grad_norm": 1.955853806615925, + "learning_rate": 3.6004589734881953e-06, + "loss": 0.9864, + "step": 3042 + }, + { + "epoch": 0.22869382233578836, + "grad_norm": 1.6947484638293246, + "learning_rate": 3.600166969784762e-06, + "loss": 0.9994, + "step": 3043 + }, + { + "epoch": 0.22876897640162333, + "grad_norm": 1.8092377342113013, + "learning_rate": 3.5998748712653077e-06, + "loss": 0.9751, + "step": 3044 + }, + { + "epoch": 0.2288441304674583, + "grad_norm": 1.8020630616681426, + "learning_rate": 3.5995826779471408e-06, + "loss": 1.0014, + "step": 3045 + }, + { + "epoch": 0.22891928453329324, + "grad_norm": 1.6904830703812823, + "learning_rate": 3.5992903898475757e-06, + "loss": 0.947, + "step": 3046 + }, + { + "epoch": 0.2289944385991282, + "grad_norm": 1.1723246411140593, + "learning_rate": 3.5989980069839304e-06, + "loss": 0.897, + "step": 3047 + }, + { + "epoch": 0.22906959266496318, + "grad_norm": 1.6163376496306474, + "learning_rate": 3.5987055293735305e-06, + "loss": 1.0282, + "step": 3048 + }, + { + "epoch": 0.22914474673079813, + "grad_norm": 1.6495337434377468, + "learning_rate": 3.5984129570337056e-06, + "loss": 1.0058, + "step": 3049 + }, + { + "epoch": 0.2292199007966331, + "grad_norm": 1.8249483467003633, + "learning_rate": 3.598120289981793e-06, + "loss": 0.9909, + "step": 3050 + }, + { + "epoch": 0.22929505486246807, + "grad_norm": 1.827406195804513, + "learning_rate": 3.597827528235133e-06, + "loss": 1.0137, + "step": 3051 + }, + { + "epoch": 0.229370208928303, + "grad_norm": 2.191507972899479, + "learning_rate": 3.597534671811074e-06, + "loss": 0.9274, + "step": 3052 + }, + { + "epoch": 0.22944536299413798, + "grad_norm": 2.109788190145018, + "learning_rate": 3.5972417207269675e-06, + "loss": 1.0286, + "step": 3053 + }, + { + "epoch": 0.22952051705997295, + "grad_norm": 2.183382791827721, + "learning_rate": 3.596948675000173e-06, + "loss": 1.0218, + "step": 3054 + }, + { + "epoch": 0.2295956711258079, + "grad_norm": 2.079434090070837, + "learning_rate": 3.596655534648055e-06, + "loss": 0.932, + "step": 3055 + }, + { + "epoch": 0.22967082519164286, + "grad_norm": 1.990538269740625, + "learning_rate": 3.596362299687982e-06, + "loss": 0.9995, + "step": 3056 + }, + { + "epoch": 0.22974597925747783, + "grad_norm": 1.5592010768000024, + "learning_rate": 3.59606897013733e-06, + "loss": 0.9199, + "step": 3057 + }, + { + "epoch": 0.2298211333233128, + "grad_norm": 1.400546651791144, + "learning_rate": 3.59577554601348e-06, + "loss": 0.8459, + "step": 3058 + }, + { + "epoch": 0.22989628738914775, + "grad_norm": 1.7812443114072747, + "learning_rate": 3.595482027333818e-06, + "loss": 0.8893, + "step": 3059 + }, + { + "epoch": 0.22997144145498272, + "grad_norm": 1.331267241245431, + "learning_rate": 3.5951884141157365e-06, + "loss": 0.8755, + "step": 3060 + }, + { + "epoch": 0.2300465955208177, + "grad_norm": 1.5715864337548244, + "learning_rate": 3.5948947063766334e-06, + "loss": 1.0121, + "step": 3061 + }, + { + "epoch": 0.23012174958665263, + "grad_norm": 1.6240144822268876, + "learning_rate": 3.5946009041339114e-06, + "loss": 1.0272, + "step": 3062 + }, + { + "epoch": 0.2301969036524876, + "grad_norm": 1.6805848319636076, + "learning_rate": 3.5943070074049797e-06, + "loss": 0.9129, + "step": 3063 + }, + { + "epoch": 0.23027205771832257, + "grad_norm": 2.1606943519150184, + "learning_rate": 3.5940130162072525e-06, + "loss": 1.0079, + "step": 3064 + }, + { + "epoch": 0.2303472117841575, + "grad_norm": 1.4475841094568656, + "learning_rate": 3.593718930558151e-06, + "loss": 0.987, + "step": 3065 + }, + { + "epoch": 0.23042236584999248, + "grad_norm": 1.4047086109446283, + "learning_rate": 3.5934247504750995e-06, + "loss": 0.9091, + "step": 3066 + }, + { + "epoch": 0.23049751991582745, + "grad_norm": 1.7549060444257303, + "learning_rate": 3.59313047597553e-06, + "loss": 0.8937, + "step": 3067 + }, + { + "epoch": 0.2305726739816624, + "grad_norm": 1.882258416793271, + "learning_rate": 3.5928361070768788e-06, + "loss": 0.893, + "step": 3068 + }, + { + "epoch": 0.23064782804749737, + "grad_norm": 2.201055650595928, + "learning_rate": 3.59254164379659e-06, + "loss": 1.027, + "step": 3069 + }, + { + "epoch": 0.23072298211333234, + "grad_norm": 0.8739725961229885, + "learning_rate": 3.5922470861521098e-06, + "loss": 0.8054, + "step": 3070 + }, + { + "epoch": 0.23079813617916728, + "grad_norm": 1.6133345255702845, + "learning_rate": 3.5919524341608923e-06, + "loss": 1.0065, + "step": 3071 + }, + { + "epoch": 0.23087329024500225, + "grad_norm": 0.8378624544543266, + "learning_rate": 3.5916576878403975e-06, + "loss": 0.819, + "step": 3072 + }, + { + "epoch": 0.23094844431083722, + "grad_norm": 1.5131635046010374, + "learning_rate": 3.59136284720809e-06, + "loss": 0.9533, + "step": 3073 + }, + { + "epoch": 0.2310235983766722, + "grad_norm": 1.626287830332067, + "learning_rate": 3.59106791228144e-06, + "loss": 1.0387, + "step": 3074 + }, + { + "epoch": 0.23109875244250713, + "grad_norm": 0.813596462467215, + "learning_rate": 3.5907728830779236e-06, + "loss": 0.8495, + "step": 3075 + }, + { + "epoch": 0.2311739065083421, + "grad_norm": 1.43015035839153, + "learning_rate": 3.5904777596150222e-06, + "loss": 0.9832, + "step": 3076 + }, + { + "epoch": 0.23124906057417707, + "grad_norm": 1.4740744321500712, + "learning_rate": 3.5901825419102238e-06, + "loss": 0.9391, + "step": 3077 + }, + { + "epoch": 0.23132421464001202, + "grad_norm": 1.7112802658581177, + "learning_rate": 3.58988722998102e-06, + "loss": 0.9307, + "step": 3078 + }, + { + "epoch": 0.231399368705847, + "grad_norm": 1.9926600475479992, + "learning_rate": 3.58959182384491e-06, + "loss": 0.9477, + "step": 3079 + }, + { + "epoch": 0.23147452277168196, + "grad_norm": 1.5984562221113057, + "learning_rate": 3.5892963235193968e-06, + "loss": 0.9792, + "step": 3080 + }, + { + "epoch": 0.2315496768375169, + "grad_norm": 1.9219014235750222, + "learning_rate": 3.589000729021991e-06, + "loss": 0.844, + "step": 3081 + }, + { + "epoch": 0.23162483090335187, + "grad_norm": 1.5464867769373476, + "learning_rate": 3.5887050403702073e-06, + "loss": 0.9526, + "step": 3082 + }, + { + "epoch": 0.23169998496918684, + "grad_norm": 2.8759344489700993, + "learning_rate": 3.588409257581567e-06, + "loss": 0.911, + "step": 3083 + }, + { + "epoch": 0.23177513903502178, + "grad_norm": 1.8912186754535816, + "learning_rate": 3.5881133806735956e-06, + "loss": 1.0423, + "step": 3084 + }, + { + "epoch": 0.23185029310085675, + "grad_norm": 1.9092800884725616, + "learning_rate": 3.587817409663824e-06, + "loss": 0.9977, + "step": 3085 + }, + { + "epoch": 0.23192544716669172, + "grad_norm": 0.6767142585952377, + "learning_rate": 3.5875213445697917e-06, + "loss": 0.8052, + "step": 3086 + }, + { + "epoch": 0.2320006012325267, + "grad_norm": 2.00393765965881, + "learning_rate": 3.587225185409041e-06, + "loss": 0.965, + "step": 3087 + }, + { + "epoch": 0.23207575529836164, + "grad_norm": 1.362815668128201, + "learning_rate": 3.5869289321991195e-06, + "loss": 0.9425, + "step": 3088 + }, + { + "epoch": 0.2321509093641966, + "grad_norm": 1.5895403448302334, + "learning_rate": 3.586632584957582e-06, + "loss": 0.9231, + "step": 3089 + }, + { + "epoch": 0.23222606343003158, + "grad_norm": 1.4805030516825217, + "learning_rate": 3.5863361437019885e-06, + "loss": 1.0188, + "step": 3090 + }, + { + "epoch": 0.23230121749586652, + "grad_norm": 1.2460807393031579, + "learning_rate": 3.5860396084499043e-06, + "loss": 1.024, + "step": 3091 + }, + { + "epoch": 0.2323763715617015, + "grad_norm": 1.470557703374423, + "learning_rate": 3.5857429792188996e-06, + "loss": 1.094, + "step": 3092 + }, + { + "epoch": 0.23245152562753646, + "grad_norm": 0.9042663569069147, + "learning_rate": 3.585446256026551e-06, + "loss": 0.793, + "step": 3093 + }, + { + "epoch": 0.2325266796933714, + "grad_norm": 0.8500850003881842, + "learning_rate": 3.5851494388904406e-06, + "loss": 0.8508, + "step": 3094 + }, + { + "epoch": 0.23260183375920637, + "grad_norm": 1.490584066216821, + "learning_rate": 3.5848525278281564e-06, + "loss": 0.9306, + "step": 3095 + }, + { + "epoch": 0.23267698782504134, + "grad_norm": 2.0122506929070085, + "learning_rate": 3.5845555228572907e-06, + "loss": 0.8664, + "step": 3096 + }, + { + "epoch": 0.23275214189087629, + "grad_norm": 1.7057912412418788, + "learning_rate": 3.5842584239954426e-06, + "loss": 0.9964, + "step": 3097 + }, + { + "epoch": 0.23282729595671126, + "grad_norm": 1.7328087547819013, + "learning_rate": 3.5839612312602166e-06, + "loss": 0.94, + "step": 3098 + }, + { + "epoch": 0.23290245002254623, + "grad_norm": 1.4833746500726064, + "learning_rate": 3.5836639446692223e-06, + "loss": 1.0028, + "step": 3099 + }, + { + "epoch": 0.23297760408838117, + "grad_norm": 0.8653402844917859, + "learning_rate": 3.5833665642400747e-06, + "loss": 0.813, + "step": 3100 + }, + { + "epoch": 0.23305275815421614, + "grad_norm": 1.6424579372274724, + "learning_rate": 3.5830690899903954e-06, + "loss": 0.9028, + "step": 3101 + }, + { + "epoch": 0.2331279122200511, + "grad_norm": 1.5998331549198912, + "learning_rate": 3.582771521937811e-06, + "loss": 0.9688, + "step": 3102 + }, + { + "epoch": 0.23320306628588608, + "grad_norm": 1.511212641251875, + "learning_rate": 3.582473860099952e-06, + "loss": 0.9604, + "step": 3103 + }, + { + "epoch": 0.23327822035172102, + "grad_norm": 1.3279982113625075, + "learning_rate": 3.582176104494458e-06, + "loss": 0.9592, + "step": 3104 + }, + { + "epoch": 0.233353374417556, + "grad_norm": 2.00392171693897, + "learning_rate": 3.581878255138971e-06, + "loss": 0.9667, + "step": 3105 + }, + { + "epoch": 0.23342852848339096, + "grad_norm": 1.4982797772373142, + "learning_rate": 3.5815803120511395e-06, + "loss": 0.9376, + "step": 3106 + }, + { + "epoch": 0.2335036825492259, + "grad_norm": 1.759766776606345, + "learning_rate": 3.5812822752486187e-06, + "loss": 0.9977, + "step": 3107 + }, + { + "epoch": 0.23357883661506088, + "grad_norm": 1.761154795037026, + "learning_rate": 3.5809841447490674e-06, + "loss": 1.0152, + "step": 3108 + }, + { + "epoch": 0.23365399068089585, + "grad_norm": 1.5331121527051492, + "learning_rate": 3.5806859205701523e-06, + "loss": 1.0419, + "step": 3109 + }, + { + "epoch": 0.2337291447467308, + "grad_norm": 1.8681800789752199, + "learning_rate": 3.5803876027295433e-06, + "loss": 0.993, + "step": 3110 + }, + { + "epoch": 0.23380429881256576, + "grad_norm": 0.8988978326152407, + "learning_rate": 3.580089191244917e-06, + "loss": 0.83, + "step": 3111 + }, + { + "epoch": 0.23387945287840073, + "grad_norm": 1.7569329668088134, + "learning_rate": 3.5797906861339556e-06, + "loss": 1.0626, + "step": 3112 + }, + { + "epoch": 0.23395460694423567, + "grad_norm": 1.6420311850677458, + "learning_rate": 3.579492087414347e-06, + "loss": 1.0443, + "step": 3113 + }, + { + "epoch": 0.23402976101007064, + "grad_norm": 1.5986911248050755, + "learning_rate": 3.5791933951037834e-06, + "loss": 0.9525, + "step": 3114 + }, + { + "epoch": 0.2341049150759056, + "grad_norm": 1.5895817421254819, + "learning_rate": 3.5788946092199643e-06, + "loss": 0.9726, + "step": 3115 + }, + { + "epoch": 0.23418006914174055, + "grad_norm": 1.74717443421387, + "learning_rate": 3.578595729780593e-06, + "loss": 0.9108, + "step": 3116 + }, + { + "epoch": 0.23425522320757552, + "grad_norm": 1.5593931208534269, + "learning_rate": 3.5782967568033805e-06, + "loss": 0.9652, + "step": 3117 + }, + { + "epoch": 0.2343303772734105, + "grad_norm": 1.3318033778580718, + "learning_rate": 3.5779976903060412e-06, + "loss": 0.9413, + "step": 3118 + }, + { + "epoch": 0.23440553133924547, + "grad_norm": 1.4543877314267282, + "learning_rate": 3.5776985303062965e-06, + "loss": 0.9963, + "step": 3119 + }, + { + "epoch": 0.2344806854050804, + "grad_norm": 1.6513274834911318, + "learning_rate": 3.5773992768218724e-06, + "loss": 1.0432, + "step": 3120 + }, + { + "epoch": 0.23455583947091538, + "grad_norm": 1.734300079102166, + "learning_rate": 3.577099929870501e-06, + "loss": 0.9926, + "step": 3121 + }, + { + "epoch": 0.23463099353675035, + "grad_norm": 1.3814192396368066, + "learning_rate": 3.5768004894699192e-06, + "loss": 0.8372, + "step": 3122 + }, + { + "epoch": 0.2347061476025853, + "grad_norm": 1.7081987591593248, + "learning_rate": 3.57650095563787e-06, + "loss": 0.9473, + "step": 3123 + }, + { + "epoch": 0.23478130166842026, + "grad_norm": 0.8424032554206727, + "learning_rate": 3.5762013283921033e-06, + "loss": 0.7858, + "step": 3124 + }, + { + "epoch": 0.23485645573425523, + "grad_norm": 1.4889813876401032, + "learning_rate": 3.5759016077503716e-06, + "loss": 1.0269, + "step": 3125 + }, + { + "epoch": 0.23493160980009017, + "grad_norm": 1.5423969996112512, + "learning_rate": 3.5756017937304356e-06, + "loss": 0.9403, + "step": 3126 + }, + { + "epoch": 0.23500676386592514, + "grad_norm": 1.7829554909174747, + "learning_rate": 3.57530188635006e-06, + "loss": 1.0231, + "step": 3127 + }, + { + "epoch": 0.23508191793176011, + "grad_norm": 1.3483008128670115, + "learning_rate": 3.5750018856270153e-06, + "loss": 1.0069, + "step": 3128 + }, + { + "epoch": 0.23515707199759506, + "grad_norm": 2.6774018886054276, + "learning_rate": 3.5747017915790776e-06, + "loss": 1.0515, + "step": 3129 + }, + { + "epoch": 0.23523222606343003, + "grad_norm": 1.7189472432099144, + "learning_rate": 3.5744016042240287e-06, + "loss": 0.971, + "step": 3130 + }, + { + "epoch": 0.235307380129265, + "grad_norm": 1.5751732897416253, + "learning_rate": 3.574101323579656e-06, + "loss": 1.0101, + "step": 3131 + }, + { + "epoch": 0.23538253419509997, + "grad_norm": 1.530358580537589, + "learning_rate": 3.5738009496637523e-06, + "loss": 0.9959, + "step": 3132 + }, + { + "epoch": 0.2354576882609349, + "grad_norm": 1.6489163105618976, + "learning_rate": 3.573500482494116e-06, + "loss": 1.002, + "step": 3133 + }, + { + "epoch": 0.23553284232676988, + "grad_norm": 1.9817221021905844, + "learning_rate": 3.573199922088551e-06, + "loss": 0.9571, + "step": 3134 + }, + { + "epoch": 0.23560799639260485, + "grad_norm": 1.3474522532990378, + "learning_rate": 3.5728992684648657e-06, + "loss": 0.9032, + "step": 3135 + }, + { + "epoch": 0.2356831504584398, + "grad_norm": 1.521563507098548, + "learning_rate": 3.572598521640876e-06, + "loss": 0.9672, + "step": 3136 + }, + { + "epoch": 0.23575830452427476, + "grad_norm": 1.722479986398417, + "learning_rate": 3.572297681634402e-06, + "loss": 0.9943, + "step": 3137 + }, + { + "epoch": 0.23583345859010973, + "grad_norm": 1.261840342045247, + "learning_rate": 3.57199674846327e-06, + "loss": 0.9971, + "step": 3138 + }, + { + "epoch": 0.23590861265594468, + "grad_norm": 1.6049624199196217, + "learning_rate": 3.5716957221453106e-06, + "loss": 0.9394, + "step": 3139 + }, + { + "epoch": 0.23598376672177965, + "grad_norm": 1.68528637690911, + "learning_rate": 3.571394602698362e-06, + "loss": 0.9836, + "step": 3140 + }, + { + "epoch": 0.23605892078761462, + "grad_norm": 1.4884588981594384, + "learning_rate": 3.5710933901402652e-06, + "loss": 1.0053, + "step": 3141 + }, + { + "epoch": 0.23613407485344956, + "grad_norm": 1.4271061842079722, + "learning_rate": 3.570792084488869e-06, + "loss": 0.974, + "step": 3142 + }, + { + "epoch": 0.23620922891928453, + "grad_norm": 1.4983234572942594, + "learning_rate": 3.5704906857620268e-06, + "loss": 0.9901, + "step": 3143 + }, + { + "epoch": 0.2362843829851195, + "grad_norm": 1.434433693609539, + "learning_rate": 3.5701891939775974e-06, + "loss": 0.9899, + "step": 3144 + }, + { + "epoch": 0.23635953705095444, + "grad_norm": 1.4840323604849568, + "learning_rate": 3.5698876091534465e-06, + "loss": 0.9451, + "step": 3145 + }, + { + "epoch": 0.2364346911167894, + "grad_norm": 1.3443815388469498, + "learning_rate": 3.5695859313074425e-06, + "loss": 0.995, + "step": 3146 + }, + { + "epoch": 0.23650984518262438, + "grad_norm": 1.5699885399515576, + "learning_rate": 3.5692841604574617e-06, + "loss": 0.943, + "step": 3147 + }, + { + "epoch": 0.23658499924845935, + "grad_norm": 2.21891031894938, + "learning_rate": 3.568982296621386e-06, + "loss": 0.9696, + "step": 3148 + }, + { + "epoch": 0.2366601533142943, + "grad_norm": 1.5190744375306409, + "learning_rate": 3.5686803398171007e-06, + "loss": 0.982, + "step": 3149 + }, + { + "epoch": 0.23673530738012927, + "grad_norm": 1.7976662262097218, + "learning_rate": 3.5683782900624986e-06, + "loss": 1.0222, + "step": 3150 + }, + { + "epoch": 0.23681046144596424, + "grad_norm": 1.9992658936762122, + "learning_rate": 3.5680761473754767e-06, + "loss": 0.939, + "step": 3151 + }, + { + "epoch": 0.23688561551179918, + "grad_norm": 1.489359467585896, + "learning_rate": 3.5677739117739385e-06, + "loss": 0.9818, + "step": 3152 + }, + { + "epoch": 0.23696076957763415, + "grad_norm": 1.9455574831228961, + "learning_rate": 3.5674715832757927e-06, + "loss": 0.994, + "step": 3153 + }, + { + "epoch": 0.23703592364346912, + "grad_norm": 1.6863757026257506, + "learning_rate": 3.5671691618989533e-06, + "loss": 1.0108, + "step": 3154 + }, + { + "epoch": 0.23711107770930406, + "grad_norm": 1.6574286819235517, + "learning_rate": 3.56686664766134e-06, + "loss": 1.0188, + "step": 3155 + }, + { + "epoch": 0.23718623177513903, + "grad_norm": 1.3625785962454973, + "learning_rate": 3.5665640405808785e-06, + "loss": 0.9503, + "step": 3156 + }, + { + "epoch": 0.237261385840974, + "grad_norm": 1.437654486937966, + "learning_rate": 3.566261340675498e-06, + "loss": 1.0086, + "step": 3157 + }, + { + "epoch": 0.23733653990680895, + "grad_norm": 1.6265184937019974, + "learning_rate": 3.5659585479631357e-06, + "loss": 0.9638, + "step": 3158 + }, + { + "epoch": 0.23741169397264392, + "grad_norm": 1.4823526358737684, + "learning_rate": 3.565655662461733e-06, + "loss": 1.0793, + "step": 3159 + }, + { + "epoch": 0.2374868480384789, + "grad_norm": 1.5344553504928875, + "learning_rate": 3.565352684189237e-06, + "loss": 0.9456, + "step": 3160 + }, + { + "epoch": 0.23756200210431383, + "grad_norm": 1.9942177394262723, + "learning_rate": 3.5650496131636006e-06, + "loss": 0.9992, + "step": 3161 + }, + { + "epoch": 0.2376371561701488, + "grad_norm": 1.3528618272515263, + "learning_rate": 3.564746449402781e-06, + "loss": 0.9183, + "step": 3162 + }, + { + "epoch": 0.23771231023598377, + "grad_norm": 2.028758002246016, + "learning_rate": 3.5644431929247432e-06, + "loss": 1.0107, + "step": 3163 + }, + { + "epoch": 0.23778746430181874, + "grad_norm": 1.4044639265208763, + "learning_rate": 3.5641398437474546e-06, + "loss": 1.0702, + "step": 3164 + }, + { + "epoch": 0.23786261836765368, + "grad_norm": 1.7357708923981316, + "learning_rate": 3.563836401888892e-06, + "loss": 0.8732, + "step": 3165 + }, + { + "epoch": 0.23793777243348865, + "grad_norm": 2.0883634369129838, + "learning_rate": 3.5635328673670335e-06, + "loss": 0.9747, + "step": 3166 + }, + { + "epoch": 0.23801292649932362, + "grad_norm": 1.7322063462499984, + "learning_rate": 3.5632292401998657e-06, + "loss": 0.9862, + "step": 3167 + }, + { + "epoch": 0.23808808056515857, + "grad_norm": 0.7657708398920485, + "learning_rate": 3.562925520405379e-06, + "loss": 0.7954, + "step": 3168 + }, + { + "epoch": 0.23816323463099354, + "grad_norm": 1.4518744717554715, + "learning_rate": 3.562621708001571e-06, + "loss": 1.0689, + "step": 3169 + }, + { + "epoch": 0.2382383886968285, + "grad_norm": 1.4796333472263057, + "learning_rate": 3.5623178030064426e-06, + "loss": 0.951, + "step": 3170 + }, + { + "epoch": 0.23831354276266345, + "grad_norm": 1.259405134801285, + "learning_rate": 3.562013805438002e-06, + "loss": 1.0429, + "step": 3171 + }, + { + "epoch": 0.23838869682849842, + "grad_norm": 1.5410574749000878, + "learning_rate": 3.5617097153142623e-06, + "loss": 1.0315, + "step": 3172 + }, + { + "epoch": 0.2384638508943334, + "grad_norm": 1.6166061609619191, + "learning_rate": 3.5614055326532416e-06, + "loss": 0.9104, + "step": 3173 + }, + { + "epoch": 0.23853900496016833, + "grad_norm": 2.066275637560482, + "learning_rate": 3.561101257472964e-06, + "loss": 0.9674, + "step": 3174 + }, + { + "epoch": 0.2386141590260033, + "grad_norm": 1.459001369964605, + "learning_rate": 3.560796889791459e-06, + "loss": 1.0679, + "step": 3175 + }, + { + "epoch": 0.23868931309183827, + "grad_norm": 4.614795665577641, + "learning_rate": 3.5604924296267616e-06, + "loss": 1.013, + "step": 3176 + }, + { + "epoch": 0.23876446715767324, + "grad_norm": 1.7762932995344742, + "learning_rate": 3.5601878769969123e-06, + "loss": 1.0658, + "step": 3177 + }, + { + "epoch": 0.2388396212235082, + "grad_norm": 1.9031705528282183, + "learning_rate": 3.559883231919957e-06, + "loss": 0.927, + "step": 3178 + }, + { + "epoch": 0.23891477528934316, + "grad_norm": 1.770656337494386, + "learning_rate": 3.559578494413947e-06, + "loss": 1.0198, + "step": 3179 + }, + { + "epoch": 0.23898992935517813, + "grad_norm": 0.8984246957944593, + "learning_rate": 3.559273664496939e-06, + "loss": 0.8141, + "step": 3180 + }, + { + "epoch": 0.23906508342101307, + "grad_norm": 1.9004934347792517, + "learning_rate": 3.5589687421869957e-06, + "loss": 0.8165, + "step": 3181 + }, + { + "epoch": 0.23914023748684804, + "grad_norm": 1.5674991742729676, + "learning_rate": 3.558663727502185e-06, + "loss": 0.9411, + "step": 3182 + }, + { + "epoch": 0.239215391552683, + "grad_norm": 3.000376518781529, + "learning_rate": 3.5583586204605796e-06, + "loss": 0.9859, + "step": 3183 + }, + { + "epoch": 0.23929054561851795, + "grad_norm": 1.8874023336443189, + "learning_rate": 3.5580534210802587e-06, + "loss": 0.9789, + "step": 3184 + }, + { + "epoch": 0.23936569968435292, + "grad_norm": 1.5827754024816432, + "learning_rate": 3.5577481293793063e-06, + "loss": 0.9959, + "step": 3185 + }, + { + "epoch": 0.2394408537501879, + "grad_norm": 1.703544145155986, + "learning_rate": 3.5574427453758124e-06, + "loss": 0.9096, + "step": 3186 + }, + { + "epoch": 0.23951600781602284, + "grad_norm": 3.025467853779466, + "learning_rate": 3.557137269087872e-06, + "loss": 1.0018, + "step": 3187 + }, + { + "epoch": 0.2395911618818578, + "grad_norm": 1.7863812073128669, + "learning_rate": 3.5568317005335852e-06, + "loss": 1.0393, + "step": 3188 + }, + { + "epoch": 0.23966631594769278, + "grad_norm": 1.638879774179634, + "learning_rate": 3.556526039731059e-06, + "loss": 0.8927, + "step": 3189 + }, + { + "epoch": 0.23974147001352772, + "grad_norm": 1.4196397587690106, + "learning_rate": 3.5562202866984045e-06, + "loss": 0.9396, + "step": 3190 + }, + { + "epoch": 0.2398166240793627, + "grad_norm": 1.6831556843348567, + "learning_rate": 3.555914441453739e-06, + "loss": 0.9899, + "step": 3191 + }, + { + "epoch": 0.23989177814519766, + "grad_norm": 1.4006009549107306, + "learning_rate": 3.555608504015185e-06, + "loss": 0.9598, + "step": 3192 + }, + { + "epoch": 0.23996693221103263, + "grad_norm": 1.6985667413777272, + "learning_rate": 3.5553024744008697e-06, + "loss": 0.9666, + "step": 3193 + }, + { + "epoch": 0.24004208627686757, + "grad_norm": 1.6210990446073175, + "learning_rate": 3.5549963526289276e-06, + "loss": 0.9027, + "step": 3194 + }, + { + "epoch": 0.24011724034270254, + "grad_norm": 1.7040538529764822, + "learning_rate": 3.5546901387174975e-06, + "loss": 1.0769, + "step": 3195 + }, + { + "epoch": 0.2401923944085375, + "grad_norm": 1.8848125550747306, + "learning_rate": 3.554383832684723e-06, + "loss": 1.0362, + "step": 3196 + }, + { + "epoch": 0.24026754847437246, + "grad_norm": 1.5117729236447535, + "learning_rate": 3.554077434548754e-06, + "loss": 0.9583, + "step": 3197 + }, + { + "epoch": 0.24034270254020743, + "grad_norm": 1.521384082862671, + "learning_rate": 3.5537709443277465e-06, + "loss": 0.8647, + "step": 3198 + }, + { + "epoch": 0.2404178566060424, + "grad_norm": 2.2092403732354247, + "learning_rate": 3.55346436203986e-06, + "loss": 1.0437, + "step": 3199 + }, + { + "epoch": 0.24049301067187734, + "grad_norm": 1.6511532076955193, + "learning_rate": 3.5531576877032627e-06, + "loss": 1.024, + "step": 3200 + }, + { + "epoch": 0.2405681647377123, + "grad_norm": 2.195541641386029, + "learning_rate": 3.552850921336124e-06, + "loss": 0.8846, + "step": 3201 + }, + { + "epoch": 0.24064331880354728, + "grad_norm": 1.9696572650451132, + "learning_rate": 3.5525440629566223e-06, + "loss": 0.9783, + "step": 3202 + }, + { + "epoch": 0.24071847286938222, + "grad_norm": 1.379957624973467, + "learning_rate": 3.5522371125829395e-06, + "loss": 1.0657, + "step": 3203 + }, + { + "epoch": 0.2407936269352172, + "grad_norm": 0.8977841862941851, + "learning_rate": 3.551930070233264e-06, + "loss": 0.8636, + "step": 3204 + }, + { + "epoch": 0.24086878100105216, + "grad_norm": 1.5310979495169519, + "learning_rate": 3.551622935925789e-06, + "loss": 0.9292, + "step": 3205 + }, + { + "epoch": 0.2409439350668871, + "grad_norm": 1.461321204076559, + "learning_rate": 3.5513157096787143e-06, + "loss": 0.904, + "step": 3206 + }, + { + "epoch": 0.24101908913272208, + "grad_norm": 1.4732032791981593, + "learning_rate": 3.551008391510242e-06, + "loss": 0.9302, + "step": 3207 + }, + { + "epoch": 0.24109424319855705, + "grad_norm": 1.9651860873606684, + "learning_rate": 3.5507009814385846e-06, + "loss": 1.0402, + "step": 3208 + }, + { + "epoch": 0.24116939726439202, + "grad_norm": 1.5513929414634553, + "learning_rate": 3.550393479481955e-06, + "loss": 1.0114, + "step": 3209 + }, + { + "epoch": 0.24124455133022696, + "grad_norm": 1.7981379838725977, + "learning_rate": 3.550085885658576e-06, + "loss": 0.9543, + "step": 3210 + }, + { + "epoch": 0.24131970539606193, + "grad_norm": 1.69853677327907, + "learning_rate": 3.5497781999866715e-06, + "loss": 0.9643, + "step": 3211 + }, + { + "epoch": 0.2413948594618969, + "grad_norm": 1.474496920510301, + "learning_rate": 3.5494704224844746e-06, + "loss": 0.8587, + "step": 3212 + }, + { + "epoch": 0.24147001352773184, + "grad_norm": 1.7214214717693712, + "learning_rate": 3.549162553170222e-06, + "loss": 0.8973, + "step": 3213 + }, + { + "epoch": 0.2415451675935668, + "grad_norm": 1.6034326566728219, + "learning_rate": 3.548854592062156e-06, + "loss": 0.9252, + "step": 3214 + }, + { + "epoch": 0.24162032165940178, + "grad_norm": 1.3072987539800909, + "learning_rate": 3.548546539178524e-06, + "loss": 1.0037, + "step": 3215 + }, + { + "epoch": 0.24169547572523672, + "grad_norm": 1.4038377204858774, + "learning_rate": 3.548238394537581e-06, + "loss": 1.0077, + "step": 3216 + }, + { + "epoch": 0.2417706297910717, + "grad_norm": 1.838781548287117, + "learning_rate": 3.5479301581575827e-06, + "loss": 0.9968, + "step": 3217 + }, + { + "epoch": 0.24184578385690667, + "grad_norm": 1.3958662465708291, + "learning_rate": 3.547621830056796e-06, + "loss": 1.0432, + "step": 3218 + }, + { + "epoch": 0.2419209379227416, + "grad_norm": 1.3437696943392348, + "learning_rate": 3.5473134102534895e-06, + "loss": 0.9606, + "step": 3219 + }, + { + "epoch": 0.24199609198857658, + "grad_norm": 1.706521942489388, + "learning_rate": 3.5470048987659387e-06, + "loss": 0.9267, + "step": 3220 + }, + { + "epoch": 0.24207124605441155, + "grad_norm": 1.4939454593226984, + "learning_rate": 3.5466962956124235e-06, + "loss": 0.9465, + "step": 3221 + }, + { + "epoch": 0.24214640012024652, + "grad_norm": 1.7636223973022438, + "learning_rate": 3.54638760081123e-06, + "loss": 0.9763, + "step": 3222 + }, + { + "epoch": 0.24222155418608146, + "grad_norm": 1.6221137457262658, + "learning_rate": 3.5460788143806505e-06, + "loss": 1.0129, + "step": 3223 + }, + { + "epoch": 0.24229670825191643, + "grad_norm": 1.3858462745550093, + "learning_rate": 3.54576993633898e-06, + "loss": 0.9894, + "step": 3224 + }, + { + "epoch": 0.2423718623177514, + "grad_norm": 1.731443453560233, + "learning_rate": 3.545460966704522e-06, + "loss": 0.9661, + "step": 3225 + }, + { + "epoch": 0.24244701638358634, + "grad_norm": 1.374723580058851, + "learning_rate": 3.5451519054955836e-06, + "loss": 0.9744, + "step": 3226 + }, + { + "epoch": 0.24252217044942131, + "grad_norm": 1.7858600856659939, + "learning_rate": 3.544842752730478e-06, + "loss": 1.0302, + "step": 3227 + }, + { + "epoch": 0.24259732451525629, + "grad_norm": 1.5956098139361852, + "learning_rate": 3.5445335084275235e-06, + "loss": 0.8689, + "step": 3228 + }, + { + "epoch": 0.24267247858109123, + "grad_norm": 1.5656342923008533, + "learning_rate": 3.5442241726050444e-06, + "loss": 0.9612, + "step": 3229 + }, + { + "epoch": 0.2427476326469262, + "grad_norm": 1.6445745974375456, + "learning_rate": 3.5439147452813696e-06, + "loss": 0.9808, + "step": 3230 + }, + { + "epoch": 0.24282278671276117, + "grad_norm": 1.4955343689739626, + "learning_rate": 3.5436052264748348e-06, + "loss": 0.8815, + "step": 3231 + }, + { + "epoch": 0.2428979407785961, + "grad_norm": 2.237181175149459, + "learning_rate": 3.543295616203779e-06, + "loss": 1.0117, + "step": 3232 + }, + { + "epoch": 0.24297309484443108, + "grad_norm": 1.3896334681505544, + "learning_rate": 3.5429859144865486e-06, + "loss": 0.9633, + "step": 3233 + }, + { + "epoch": 0.24304824891026605, + "grad_norm": 2.0207135928620468, + "learning_rate": 3.542676121341494e-06, + "loss": 0.8151, + "step": 3234 + }, + { + "epoch": 0.243123402976101, + "grad_norm": 1.4069746057947594, + "learning_rate": 3.5423662367869716e-06, + "loss": 0.9204, + "step": 3235 + }, + { + "epoch": 0.24319855704193596, + "grad_norm": 1.7233979235313102, + "learning_rate": 3.542056260841344e-06, + "loss": 0.9038, + "step": 3236 + }, + { + "epoch": 0.24327371110777093, + "grad_norm": 1.505376716180102, + "learning_rate": 3.5417461935229777e-06, + "loss": 0.913, + "step": 3237 + }, + { + "epoch": 0.2433488651736059, + "grad_norm": 1.3582720612044317, + "learning_rate": 3.5414360348502463e-06, + "loss": 0.824, + "step": 3238 + }, + { + "epoch": 0.24342401923944085, + "grad_norm": 1.4453962971907106, + "learning_rate": 3.5411257848415266e-06, + "loss": 0.9211, + "step": 3239 + }, + { + "epoch": 0.24349917330527582, + "grad_norm": 1.5607915311729263, + "learning_rate": 3.5408154435152034e-06, + "loss": 0.9794, + "step": 3240 + }, + { + "epoch": 0.2435743273711108, + "grad_norm": 1.644039705804046, + "learning_rate": 3.5405050108896645e-06, + "loss": 0.8296, + "step": 3241 + }, + { + "epoch": 0.24364948143694573, + "grad_norm": 1.6346832947411662, + "learning_rate": 3.5401944869833046e-06, + "loss": 0.9178, + "step": 3242 + }, + { + "epoch": 0.2437246355027807, + "grad_norm": 1.7500354218304077, + "learning_rate": 3.539883871814524e-06, + "loss": 1.0187, + "step": 3243 + }, + { + "epoch": 0.24379978956861567, + "grad_norm": 1.8611051700111316, + "learning_rate": 3.5395731654017277e-06, + "loss": 0.9745, + "step": 3244 + }, + { + "epoch": 0.2438749436344506, + "grad_norm": 1.8417012097785714, + "learning_rate": 3.539262367763325e-06, + "loss": 0.9276, + "step": 3245 + }, + { + "epoch": 0.24395009770028558, + "grad_norm": 1.4162150485918557, + "learning_rate": 3.5389514789177334e-06, + "loss": 0.9389, + "step": 3246 + }, + { + "epoch": 0.24402525176612055, + "grad_norm": 1.3948318627595453, + "learning_rate": 3.5386404988833732e-06, + "loss": 0.9991, + "step": 3247 + }, + { + "epoch": 0.2441004058319555, + "grad_norm": 1.3201227898273544, + "learning_rate": 3.538329427678672e-06, + "loss": 0.9129, + "step": 3248 + }, + { + "epoch": 0.24417555989779047, + "grad_norm": 0.977579335117681, + "learning_rate": 3.5380182653220613e-06, + "loss": 0.8224, + "step": 3249 + }, + { + "epoch": 0.24425071396362544, + "grad_norm": 1.4946242802525602, + "learning_rate": 3.5377070118319788e-06, + "loss": 0.9549, + "step": 3250 + }, + { + "epoch": 0.24432586802946038, + "grad_norm": 1.6991079009281846, + "learning_rate": 3.5373956672268683e-06, + "loss": 0.9848, + "step": 3251 + }, + { + "epoch": 0.24440102209529535, + "grad_norm": 1.6062836817909654, + "learning_rate": 3.5370842315251766e-06, + "loss": 1.0225, + "step": 3252 + }, + { + "epoch": 0.24447617616113032, + "grad_norm": 1.71905660495465, + "learning_rate": 3.5367727047453583e-06, + "loss": 1.0384, + "step": 3253 + }, + { + "epoch": 0.2445513302269653, + "grad_norm": 1.5091335063951472, + "learning_rate": 3.536461086905873e-06, + "loss": 1.0784, + "step": 3254 + }, + { + "epoch": 0.24462648429280023, + "grad_norm": 1.5825572705263622, + "learning_rate": 3.536149378025185e-06, + "loss": 0.9503, + "step": 3255 + }, + { + "epoch": 0.2447016383586352, + "grad_norm": 0.8576428035349872, + "learning_rate": 3.5358375781217634e-06, + "loss": 0.8382, + "step": 3256 + }, + { + "epoch": 0.24477679242447017, + "grad_norm": 3.484448624590012, + "learning_rate": 3.5355256872140846e-06, + "loss": 0.9531, + "step": 3257 + }, + { + "epoch": 0.24485194649030512, + "grad_norm": 1.435266168877711, + "learning_rate": 3.535213705320629e-06, + "loss": 0.8696, + "step": 3258 + }, + { + "epoch": 0.2449271005561401, + "grad_norm": 1.8690069423850715, + "learning_rate": 3.534901632459882e-06, + "loss": 0.9188, + "step": 3259 + }, + { + "epoch": 0.24500225462197506, + "grad_norm": 1.6086816868360774, + "learning_rate": 3.5345894686503366e-06, + "loss": 0.8914, + "step": 3260 + }, + { + "epoch": 0.24507740868781, + "grad_norm": 1.8451952684650665, + "learning_rate": 3.5342772139104884e-06, + "loss": 1.0216, + "step": 3261 + }, + { + "epoch": 0.24515256275364497, + "grad_norm": 2.269160911822671, + "learning_rate": 3.5339648682588397e-06, + "loss": 0.8908, + "step": 3262 + }, + { + "epoch": 0.24522771681947994, + "grad_norm": 2.103935405247301, + "learning_rate": 3.5336524317138993e-06, + "loss": 0.9332, + "step": 3263 + }, + { + "epoch": 0.24530287088531488, + "grad_norm": 1.4963738799221216, + "learning_rate": 3.5333399042941797e-06, + "loss": 0.968, + "step": 3264 + }, + { + "epoch": 0.24537802495114985, + "grad_norm": 1.333232841082286, + "learning_rate": 3.5330272860181985e-06, + "loss": 0.936, + "step": 3265 + }, + { + "epoch": 0.24545317901698482, + "grad_norm": 1.368229757579167, + "learning_rate": 3.532714576904481e-06, + "loss": 0.9563, + "step": 3266 + }, + { + "epoch": 0.2455283330828198, + "grad_norm": 0.7772278435715659, + "learning_rate": 3.5324017769715548e-06, + "loss": 0.8577, + "step": 3267 + }, + { + "epoch": 0.24560348714865474, + "grad_norm": 3.40641728261646, + "learning_rate": 3.532088886237956e-06, + "loss": 0.845, + "step": 3268 + }, + { + "epoch": 0.2456786412144897, + "grad_norm": 1.425888768152792, + "learning_rate": 3.5317759047222235e-06, + "loss": 0.9745, + "step": 3269 + }, + { + "epoch": 0.24575379528032468, + "grad_norm": 1.4692111001712638, + "learning_rate": 3.531462832442903e-06, + "loss": 0.9986, + "step": 3270 + }, + { + "epoch": 0.24582894934615962, + "grad_norm": 1.4867977247646145, + "learning_rate": 3.531149669418546e-06, + "loss": 0.935, + "step": 3271 + }, + { + "epoch": 0.2459041034119946, + "grad_norm": 2.2109725464108925, + "learning_rate": 3.530836415667708e-06, + "loss": 0.8444, + "step": 3272 + }, + { + "epoch": 0.24597925747782956, + "grad_norm": 1.830457715185609, + "learning_rate": 3.53052307120895e-06, + "loss": 1.0685, + "step": 3273 + }, + { + "epoch": 0.2460544115436645, + "grad_norm": 1.3874193168069662, + "learning_rate": 3.5302096360608385e-06, + "loss": 0.9781, + "step": 3274 + }, + { + "epoch": 0.24612956560949947, + "grad_norm": 1.7167490322143084, + "learning_rate": 3.5298961102419477e-06, + "loss": 0.905, + "step": 3275 + }, + { + "epoch": 0.24620471967533444, + "grad_norm": 1.2964944396011922, + "learning_rate": 3.5295824937708537e-06, + "loss": 1.0303, + "step": 3276 + }, + { + "epoch": 0.2462798737411694, + "grad_norm": 1.8612024639097753, + "learning_rate": 3.5292687866661396e-06, + "loss": 1.0834, + "step": 3277 + }, + { + "epoch": 0.24635502780700436, + "grad_norm": 1.8458446307155911, + "learning_rate": 3.528954988946394e-06, + "loss": 0.8783, + "step": 3278 + }, + { + "epoch": 0.24643018187283933, + "grad_norm": 1.5571880072176216, + "learning_rate": 3.5286411006302107e-06, + "loss": 0.9687, + "step": 3279 + }, + { + "epoch": 0.24650533593867427, + "grad_norm": 1.4486728678619005, + "learning_rate": 3.528327121736188e-06, + "loss": 0.9396, + "step": 3280 + }, + { + "epoch": 0.24658049000450924, + "grad_norm": 1.6023515943821578, + "learning_rate": 3.5280130522829317e-06, + "loss": 0.882, + "step": 3281 + }, + { + "epoch": 0.2466556440703442, + "grad_norm": 1.6009346824679282, + "learning_rate": 3.5276988922890503e-06, + "loss": 0.9268, + "step": 3282 + }, + { + "epoch": 0.24673079813617918, + "grad_norm": 1.7610358629620317, + "learning_rate": 3.52738464177316e-06, + "loss": 1.0079, + "step": 3283 + }, + { + "epoch": 0.24680595220201412, + "grad_norm": 1.4091884430357549, + "learning_rate": 3.527070300753881e-06, + "loss": 1.0043, + "step": 3284 + }, + { + "epoch": 0.2468811062678491, + "grad_norm": 1.6713526970545924, + "learning_rate": 3.526755869249839e-06, + "loss": 0.9371, + "step": 3285 + }, + { + "epoch": 0.24695626033368406, + "grad_norm": 1.5245288558836383, + "learning_rate": 3.5264413472796653e-06, + "loss": 0.8741, + "step": 3286 + }, + { + "epoch": 0.247031414399519, + "grad_norm": 1.5697224585505718, + "learning_rate": 3.5261267348619964e-06, + "loss": 0.9974, + "step": 3287 + }, + { + "epoch": 0.24710656846535398, + "grad_norm": 1.6565160717563046, + "learning_rate": 3.5258120320154755e-06, + "loss": 0.9369, + "step": 3288 + }, + { + "epoch": 0.24718172253118895, + "grad_norm": 1.4404022938337973, + "learning_rate": 3.5254972387587483e-06, + "loss": 0.8826, + "step": 3289 + }, + { + "epoch": 0.2472568765970239, + "grad_norm": 1.5178648139054534, + "learning_rate": 3.525182355110468e-06, + "loss": 0.8935, + "step": 3290 + }, + { + "epoch": 0.24733203066285886, + "grad_norm": 1.501843670451983, + "learning_rate": 3.524867381089293e-06, + "loss": 0.935, + "step": 3291 + }, + { + "epoch": 0.24740718472869383, + "grad_norm": 1.5074366758341062, + "learning_rate": 3.524552316713887e-06, + "loss": 0.9688, + "step": 3292 + }, + { + "epoch": 0.24748233879452877, + "grad_norm": 1.3722944952112937, + "learning_rate": 3.5242371620029176e-06, + "loss": 0.973, + "step": 3293 + }, + { + "epoch": 0.24755749286036374, + "grad_norm": 1.689490310171518, + "learning_rate": 3.5239219169750604e-06, + "loss": 0.9655, + "step": 3294 + }, + { + "epoch": 0.2476326469261987, + "grad_norm": 0.8104779317558941, + "learning_rate": 3.5236065816489938e-06, + "loss": 0.7327, + "step": 3295 + }, + { + "epoch": 0.24770780099203366, + "grad_norm": 0.9175506613670328, + "learning_rate": 3.5232911560434023e-06, + "loss": 0.7956, + "step": 3296 + }, + { + "epoch": 0.24778295505786863, + "grad_norm": 2.190593276290724, + "learning_rate": 3.5229756401769775e-06, + "loss": 1.0369, + "step": 3297 + }, + { + "epoch": 0.2478581091237036, + "grad_norm": 2.1576782140581443, + "learning_rate": 3.522660034068414e-06, + "loss": 0.8958, + "step": 3298 + }, + { + "epoch": 0.24793326318953857, + "grad_norm": 2.1987018960328406, + "learning_rate": 3.5223443377364133e-06, + "loss": 0.9575, + "step": 3299 + }, + { + "epoch": 0.2480084172553735, + "grad_norm": 2.0560604483479423, + "learning_rate": 3.5220285511996802e-06, + "loss": 0.9295, + "step": 3300 + }, + { + "epoch": 0.24808357132120848, + "grad_norm": 1.5878681026423953, + "learning_rate": 3.521712674476928e-06, + "loss": 0.9944, + "step": 3301 + }, + { + "epoch": 0.24815872538704345, + "grad_norm": 0.8220549469842062, + "learning_rate": 3.521396707586872e-06, + "loss": 0.7649, + "step": 3302 + }, + { + "epoch": 0.2482338794528784, + "grad_norm": 0.8690464954624016, + "learning_rate": 3.521080650548236e-06, + "loss": 0.7913, + "step": 3303 + }, + { + "epoch": 0.24830903351871336, + "grad_norm": 2.2337633909775088, + "learning_rate": 3.5207645033797464e-06, + "loss": 0.9823, + "step": 3304 + }, + { + "epoch": 0.24838418758454833, + "grad_norm": 1.7057655233531954, + "learning_rate": 3.5204482661001373e-06, + "loss": 0.9205, + "step": 3305 + }, + { + "epoch": 0.24845934165038328, + "grad_norm": 1.527017231647266, + "learning_rate": 3.5201319387281455e-06, + "loss": 0.9303, + "step": 3306 + }, + { + "epoch": 0.24853449571621825, + "grad_norm": 1.4331710982363788, + "learning_rate": 3.519815521282515e-06, + "loss": 1.0164, + "step": 3307 + }, + { + "epoch": 0.24860964978205322, + "grad_norm": 1.4079932110606512, + "learning_rate": 3.519499013781996e-06, + "loss": 1.0152, + "step": 3308 + }, + { + "epoch": 0.24868480384788816, + "grad_norm": 1.534804908208925, + "learning_rate": 3.5191824162453417e-06, + "loss": 1.0266, + "step": 3309 + }, + { + "epoch": 0.24875995791372313, + "grad_norm": 1.8338719212644854, + "learning_rate": 3.5188657286913115e-06, + "loss": 0.9819, + "step": 3310 + }, + { + "epoch": 0.2488351119795581, + "grad_norm": 2.381062850205945, + "learning_rate": 3.5185489511386712e-06, + "loss": 0.9804, + "step": 3311 + }, + { + "epoch": 0.24891026604539307, + "grad_norm": 2.190689377648237, + "learning_rate": 3.5182320836061906e-06, + "loss": 0.971, + "step": 3312 + }, + { + "epoch": 0.248985420111228, + "grad_norm": 1.65212015573548, + "learning_rate": 3.517915126112645e-06, + "loss": 0.915, + "step": 3313 + }, + { + "epoch": 0.24906057417706298, + "grad_norm": 1.9425065973039726, + "learning_rate": 3.517598078676816e-06, + "loss": 0.9323, + "step": 3314 + }, + { + "epoch": 0.24913572824289795, + "grad_norm": 1.6984136669237946, + "learning_rate": 3.517280941317489e-06, + "loss": 0.9899, + "step": 3315 + }, + { + "epoch": 0.2492108823087329, + "grad_norm": 1.6527117236033813, + "learning_rate": 3.5169637140534573e-06, + "loss": 1.039, + "step": 3316 + }, + { + "epoch": 0.24928603637456787, + "grad_norm": 1.5971576686843052, + "learning_rate": 3.5166463969035157e-06, + "loss": 0.8356, + "step": 3317 + }, + { + "epoch": 0.24936119044040284, + "grad_norm": 1.6381104590837787, + "learning_rate": 3.5163289898864675e-06, + "loss": 0.8759, + "step": 3318 + }, + { + "epoch": 0.24943634450623778, + "grad_norm": 0.9178968036134237, + "learning_rate": 3.5160114930211203e-06, + "loss": 0.7751, + "step": 3319 + }, + { + "epoch": 0.24951149857207275, + "grad_norm": 1.4683472101347481, + "learning_rate": 3.5156939063262875e-06, + "loss": 0.9254, + "step": 3320 + }, + { + "epoch": 0.24958665263790772, + "grad_norm": 1.5703939540101477, + "learning_rate": 3.515376229820787e-06, + "loss": 0.9517, + "step": 3321 + }, + { + "epoch": 0.24966180670374266, + "grad_norm": 1.604968213393467, + "learning_rate": 3.5150584635234416e-06, + "loss": 0.9098, + "step": 3322 + }, + { + "epoch": 0.24973696076957763, + "grad_norm": 1.5523702710386273, + "learning_rate": 3.5147406074530805e-06, + "loss": 0.9332, + "step": 3323 + }, + { + "epoch": 0.2498121148354126, + "grad_norm": 0.8794654254200545, + "learning_rate": 3.5144226616285384e-06, + "loss": 0.8405, + "step": 3324 + }, + { + "epoch": 0.24988726890124754, + "grad_norm": 1.77970973562763, + "learning_rate": 3.5141046260686537e-06, + "loss": 0.951, + "step": 3325 + }, + { + "epoch": 0.24996242296708251, + "grad_norm": 1.4699042634215598, + "learning_rate": 3.5137865007922726e-06, + "loss": 0.984, + "step": 3326 + }, + { + "epoch": 0.25003757703291746, + "grad_norm": 1.4583843585579135, + "learning_rate": 3.5134682858182448e-06, + "loss": 0.9573, + "step": 3327 + }, + { + "epoch": 0.2501127310987524, + "flos": 64135202090880.0, + "grad_norm": 1.3792087990278767, + "learning_rate": 3.5131499811654253e-06, + "loss": 0.8154, + "num_input_tokens_seen": 76350, + "step": 3328 + }, + { + "epoch": 0.2501878851645874, + "flos": 25923786963840.0, + "grad_norm": 1.432700645548556, + "learning_rate": 3.5128315868526755e-06, + "loss": 0.9078, + "num_input_tokens_seen": 101755, + "step": 3329 + }, + { + "epoch": 0.25026303923042237, + "flos": 21510657181440.0, + "grad_norm": 1.528032776715712, + "learning_rate": 3.512513102898861e-06, + "loss": 0.9169, + "num_input_tokens_seen": 126055, + "step": 3330 + }, + { + "epoch": 0.25033819329625734, + "flos": 25886696014080.0, + "grad_norm": 1.7474630904768018, + "learning_rate": 3.512194529322853e-06, + "loss": 0.878, + "num_input_tokens_seen": 152060, + "step": 3331 + }, + { + "epoch": 0.2504133473620923, + "flos": 20805470058240.0, + "grad_norm": 1.788022334469229, + "learning_rate": 3.511875866143529e-06, + "loss": 0.981, + "num_input_tokens_seen": 175375, + "step": 3332 + }, + { + "epoch": 0.2504885014279272, + "flos": 22040984067840.0, + "grad_norm": 1.2937424756259766, + "learning_rate": 3.511557113379771e-06, + "loss": 0.9839, + "num_input_tokens_seen": 199885, + "step": 3333 + }, + { + "epoch": 0.2505636554937622, + "flos": 24724071014400.0, + "grad_norm": 1.8064788251027006, + "learning_rate": 3.511238271050465e-06, + "loss": 0.9607, + "num_input_tokens_seen": 223240, + "step": 3334 + }, + { + "epoch": 0.25063880955959716, + "flos": 22641219135360.0, + "grad_norm": 1.8061820443117615, + "learning_rate": 3.510919339174505e-06, + "loss": 1.0413, + "num_input_tokens_seen": 245950, + "step": 3335 + }, + { + "epoch": 0.25071396362543213, + "flos": 17128369382400.0, + "grad_norm": 1.68517574322019, + "learning_rate": 3.5106003177707882e-06, + "loss": 0.9274, + "num_input_tokens_seen": 266965, + "step": 3336 + }, + { + "epoch": 0.2507891176912671, + "flos": 25143627841920.0, + "grad_norm": 1.6468622717465025, + "learning_rate": 3.5102812068582183e-06, + "loss": 0.9627, + "num_input_tokens_seen": 290355, + "step": 3337 + }, + { + "epoch": 0.2508642717571021, + "flos": 20728989688320.0, + "grad_norm": 1.8074432734535044, + "learning_rate": 3.509962006455704e-06, + "loss": 1.0328, + "num_input_tokens_seen": 309705, + "step": 3338 + }, + { + "epoch": 0.25093942582293705, + "flos": 23057651479680.0, + "grad_norm": 1.781917848369563, + "learning_rate": 3.5096427165821583e-06, + "loss": 1.0565, + "num_input_tokens_seen": 330365, + "step": 3339 + }, + { + "epoch": 0.25101457988877196, + "flos": 24583069503360.0, + "grad_norm": 1.3550612307012495, + "learning_rate": 3.509323337256501e-06, + "loss": 0.9181, + "num_input_tokens_seen": 356995, + "step": 3340 + }, + { + "epoch": 0.25108973395460693, + "flos": 59551459839360.0, + "grad_norm": 0.9220501038369597, + "learning_rate": 3.5090038684976563e-06, + "loss": 0.8474, + "num_input_tokens_seen": 427310, + "step": 3341 + }, + { + "epoch": 0.2511648880204419, + "flos": 24470791073280.0, + "grad_norm": 1.457502927957149, + "learning_rate": 3.5086843103245542e-06, + "loss": 0.9407, + "num_input_tokens_seen": 453320, + "step": 3342 + }, + { + "epoch": 0.25124004208627687, + "flos": 17132966323200.0, + "grad_norm": 1.5389321363977106, + "learning_rate": 3.508364662756129e-06, + "loss": 1.0898, + "num_input_tokens_seen": 474225, + "step": 3343 + }, + { + "epoch": 0.25131519615211184, + "flos": 18473827438080.0, + "grad_norm": 1.6910914597279458, + "learning_rate": 3.5080449258113224e-06, + "loss": 0.954, + "num_input_tokens_seen": 495230, + "step": 3344 + }, + { + "epoch": 0.2513903502179468, + "flos": 12723571555200.0, + "grad_norm": 1.6118784571562867, + "learning_rate": 3.5077250995090786e-06, + "loss": 0.9899, + "num_input_tokens_seen": 516385, + "step": 3345 + }, + { + "epoch": 0.2514655042837817, + "flos": 24546984134400.0, + "grad_norm": 1.4127739362172338, + "learning_rate": 3.5074051838683497e-06, + "loss": 1.0262, + "num_input_tokens_seen": 541265, + "step": 3346 + }, + { + "epoch": 0.2515406583496167, + "flos": 22359287940480.0, + "grad_norm": 1.664352678241661, + "learning_rate": 3.507085178908091e-06, + "loss": 1.0617, + "num_input_tokens_seen": 564900, + "step": 3347 + }, + { + "epoch": 0.25161581241545167, + "flos": 26837333913600.0, + "grad_norm": 2.0404574128249395, + "learning_rate": 3.506765084647265e-06, + "loss": 1.0078, + "num_input_tokens_seen": 589595, + "step": 3348 + }, + { + "epoch": 0.25169096648128664, + "flos": 15651929710080.0, + "grad_norm": 1.4651304244224257, + "learning_rate": 3.506444901104837e-06, + "loss": 0.9896, + "num_input_tokens_seen": 608840, + "step": 3349 + }, + { + "epoch": 0.2517661205471216, + "flos": 27756806607360.0, + "grad_norm": 1.7778285444812691, + "learning_rate": 3.506124628299781e-06, + "loss": 0.8008, + "num_input_tokens_seen": 635355, + "step": 3350 + }, + { + "epoch": 0.2518412746129566, + "flos": 21864866855040.0, + "grad_norm": 1.7363443963883614, + "learning_rate": 3.505804266251073e-06, + "loss": 1.0005, + "num_input_tokens_seen": 658650, + "step": 3351 + }, + { + "epoch": 0.25191642867879155, + "flos": 62653457168640.0, + "grad_norm": 0.815845241992241, + "learning_rate": 3.505483814977696e-06, + "loss": 0.7703, + "num_input_tokens_seen": 730830, + "step": 3352 + }, + { + "epoch": 0.25199158274462646, + "flos": 16428784780800.0, + "grad_norm": 1.5687181720316525, + "learning_rate": 3.5051632744986384e-06, + "loss": 0.998, + "num_input_tokens_seen": 752275, + "step": 3353 + }, + { + "epoch": 0.25206673681046143, + "flos": 21366387532800.0, + "grad_norm": 2.273373422670061, + "learning_rate": 3.5048426448328926e-06, + "loss": 1.0049, + "num_input_tokens_seen": 774980, + "step": 3354 + }, + { + "epoch": 0.2521418908762964, + "flos": 21229659740160.0, + "grad_norm": 1.4811846529553676, + "learning_rate": 3.504521925999458e-06, + "loss": 0.979, + "num_input_tokens_seen": 798585, + "step": 3355 + }, + { + "epoch": 0.2522170449421314, + "flos": 64890814072320.0, + "grad_norm": 0.7899396572124328, + "learning_rate": 3.5042011180173386e-06, + "loss": 0.7308, + "num_input_tokens_seen": 875145, + "step": 3356 + }, + { + "epoch": 0.25229219900796634, + "flos": 19959101856000.0, + "grad_norm": 2.0473586814095417, + "learning_rate": 3.5038802209055424e-06, + "loss": 0.9645, + "num_input_tokens_seen": 898500, + "step": 3357 + }, + { + "epoch": 0.2523673530738013, + "flos": 15440660881920.0, + "grad_norm": 1.8065531941470976, + "learning_rate": 3.5035592346830846e-06, + "loss": 1.0143, + "num_input_tokens_seen": 920170, + "step": 3358 + }, + { + "epoch": 0.25244250713963623, + "flos": 68098517642880.0, + "grad_norm": 1.0579761237711598, + "learning_rate": 3.5032381593689843e-06, + "loss": 0.8457, + "num_input_tokens_seen": 998925, + "step": 3359 + }, + { + "epoch": 0.2525176612054712, + "flos": 24830172305280.0, + "grad_norm": 1.5158356982200984, + "learning_rate": 3.502916994982267e-06, + "loss": 0.8436, + "num_input_tokens_seen": 1024630, + "step": 3360 + }, + { + "epoch": 0.25259281527130617, + "flos": 14133873974400.0, + "grad_norm": 1.6848625482477901, + "learning_rate": 3.502595741541963e-06, + "loss": 0.969, + "num_input_tokens_seen": 1046720, + "step": 3361 + }, + { + "epoch": 0.25266796933714114, + "flos": 36021609993600.0, + "grad_norm": 1.5972155123952085, + "learning_rate": 3.502274399067107e-06, + "loss": 0.9559, + "num_input_tokens_seen": 1076775, + "step": 3362 + }, + { + "epoch": 0.2527431234029761, + "flos": 16670105493120.0, + "grad_norm": 1.9911860922348021, + "learning_rate": 3.5019529675767403e-06, + "loss": 0.8877, + "num_input_tokens_seen": 1094810, + "step": 3363 + }, + { + "epoch": 0.2528182774688111, + "flos": 25887522026880.0, + "grad_norm": 1.7821971316063108, + "learning_rate": 3.501631447089909e-06, + "loss": 0.9927, + "num_input_tokens_seen": 1118390, + "step": 3364 + }, + { + "epoch": 0.25289343153464605, + "flos": 15405652920960.0, + "grad_norm": 1.9134443880985477, + "learning_rate": 3.5013098376256645e-06, + "loss": 1.0535, + "num_input_tokens_seen": 1139605, + "step": 3365 + }, + { + "epoch": 0.25296858560048097, + "flos": 36335424666240.0, + "grad_norm": 1.328199721085689, + "learning_rate": 3.5009881392030633e-06, + "loss": 0.9872, + "num_input_tokens_seen": 1165810, + "step": 3366 + }, + { + "epoch": 0.25304373966631594, + "flos": 25288400280960.0, + "grad_norm": 1.5522168334622248, + "learning_rate": 3.5006663518411666e-06, + "loss": 1.0042, + "num_input_tokens_seen": 1189390, + "step": 3367 + }, + { + "epoch": 0.2531188937321509, + "flos": 14841359568000.0, + "grad_norm": 1.6631231387822152, + "learning_rate": 3.500344475559043e-06, + "loss": 1.0129, + "num_input_tokens_seen": 1212300, + "step": 3368 + }, + { + "epoch": 0.2531940477979859, + "flos": 24512227568640.0, + "grad_norm": 1.5131950956524787, + "learning_rate": 3.5000225103757634e-06, + "loss": 1.0225, + "num_input_tokens_seen": 1236780, + "step": 3369 + }, + { + "epoch": 0.25326920186382085, + "flos": 26206615998720.0, + "grad_norm": 1.7093098880843134, + "learning_rate": 3.499700456310406e-06, + "loss": 0.989, + "num_input_tokens_seen": 1260635, + "step": 3370 + }, + { + "epoch": 0.2533443559296558, + "flos": 23731996619520.0, + "grad_norm": 1.5141907030436146, + "learning_rate": 3.499378313382054e-06, + "loss": 0.9703, + "num_input_tokens_seen": 1284260, + "step": 3371 + }, + { + "epoch": 0.25341950999549073, + "flos": 58801288573440.0, + "grad_norm": 0.9342334104192512, + "learning_rate": 3.4990560816097954e-06, + "loss": 0.88, + "num_input_tokens_seen": 1354160, + "step": 3372 + }, + { + "epoch": 0.2534946640613257, + "flos": 21512093725440.0, + "grad_norm": 1.3610995085647746, + "learning_rate": 3.4987337610127237e-06, + "loss": 1.0033, + "num_input_tokens_seen": 1378435, + "step": 3373 + }, + { + "epoch": 0.2535698181271607, + "flos": 17240109108480.0, + "grad_norm": 1.7057017860447528, + "learning_rate": 3.498411351609938e-06, + "loss": 1.0307, + "num_input_tokens_seen": 1400515, + "step": 3374 + }, + { + "epoch": 0.25364497219299564, + "flos": 31920140067840.0, + "grad_norm": 1.5307456762516625, + "learning_rate": 3.4980888534205414e-06, + "loss": 1.0046, + "num_input_tokens_seen": 1424970, + "step": 3375 + }, + { + "epoch": 0.2537201262588306, + "flos": 19851635848320.0, + "grad_norm": 1.6864974964755466, + "learning_rate": 3.4977662664636443e-06, + "loss": 0.9199, + "num_input_tokens_seen": 1448295, + "step": 3376 + }, + { + "epoch": 0.2537952803246656, + "flos": 33368646758400.0, + "grad_norm": 1.5620249979898186, + "learning_rate": 3.4974435907583597e-06, + "loss": 0.8739, + "num_input_tokens_seen": 1471655, + "step": 3377 + }, + { + "epoch": 0.2538704343905005, + "flos": 16034862464640.0, + "grad_norm": 1.527134717603273, + "learning_rate": 3.497120826323809e-06, + "loss": 0.9902, + "num_input_tokens_seen": 1492820, + "step": 3378 + }, + { + "epoch": 0.25394558845633547, + "flos": 25496867848320.0, + "grad_norm": 1.6414234216764136, + "learning_rate": 3.496797973179116e-06, + "loss": 0.9501, + "num_input_tokens_seen": 1515990, + "step": 3379 + }, + { + "epoch": 0.25402074252217044, + "flos": 27230394303360.0, + "grad_norm": 1.4998245931427858, + "learning_rate": 3.4964750313434114e-06, + "loss": 1.0299, + "num_input_tokens_seen": 1538800, + "step": 3380 + }, + { + "epoch": 0.2540958965880054, + "flos": 51266193500160.0, + "grad_norm": 0.7974326482571865, + "learning_rate": 3.496152000835831e-06, + "loss": 0.714, + "num_input_tokens_seen": 1615125, + "step": 3381 + }, + { + "epoch": 0.2541710506538404, + "flos": 22429555257600.0, + "grad_norm": 1.441936695116675, + "learning_rate": 3.495828881675516e-06, + "loss": 0.9481, + "num_input_tokens_seen": 1639420, + "step": 3382 + }, + { + "epoch": 0.25424620471967535, + "flos": 29241937370880.0, + "grad_norm": 1.3149960715811762, + "learning_rate": 3.4955056738816113e-06, + "loss": 0.9203, + "num_input_tokens_seen": 1665930, + "step": 3383 + }, + { + "epoch": 0.2543213587855103, + "flos": 58363774945920.0, + "grad_norm": 0.8691882859781467, + "learning_rate": 3.4951823774732686e-06, + "loss": 0.8019, + "num_input_tokens_seen": 1739565, + "step": 3384 + }, + { + "epoch": 0.25439651285134524, + "flos": 18158540307840.0, + "grad_norm": 1.8988952084956927, + "learning_rate": 3.4948589924696447e-06, + "loss": 0.953, + "num_input_tokens_seen": 1761365, + "step": 3385 + }, + { + "epoch": 0.2544716669171802, + "flos": 31005874846080.0, + "grad_norm": 1.4635107052968326, + "learning_rate": 3.4945355188899013e-06, + "loss": 1.0647, + "num_input_tokens_seen": 1787655, + "step": 3386 + }, + { + "epoch": 0.2545468209830152, + "flos": 22747428167040.0, + "grad_norm": 1.7070987085766232, + "learning_rate": 3.494211956753206e-06, + "loss": 0.8817, + "num_input_tokens_seen": 1811220, + "step": 3387 + }, + { + "epoch": 0.25462197504885015, + "flos": 16006354865280.0, + "grad_norm": 1.6135748718404461, + "learning_rate": 3.49388830607873e-06, + "loss": 1.006, + "num_input_tokens_seen": 1834165, + "step": 3388 + }, + { + "epoch": 0.2546971291146851, + "flos": 36476713486080.0, + "grad_norm": 1.6229173078360248, + "learning_rate": 3.493564566885651e-06, + "loss": 0.9826, + "num_input_tokens_seen": 1861045, + "step": 3389 + }, + { + "epoch": 0.2547722831805201, + "flos": 21365310124800.0, + "grad_norm": 1.5020439209795957, + "learning_rate": 3.4932407391931527e-06, + "loss": 0.9556, + "num_input_tokens_seen": 1882695, + "step": 3390 + }, + { + "epoch": 0.254847437246355, + "flos": 22712384292480.0, + "grad_norm": 1.818361603994673, + "learning_rate": 3.4929168230204226e-06, + "loss": 0.997, + "num_input_tokens_seen": 1904555, + "step": 3391 + }, + { + "epoch": 0.25492259131219, + "flos": 20027681233920.0, + "grad_norm": 1.4773984288708326, + "learning_rate": 3.4925928183866534e-06, + "loss": 1.0075, + "num_input_tokens_seen": 1930080, + "step": 3392 + }, + { + "epoch": 0.25499774537802494, + "flos": 19499832385920.0, + "grad_norm": 1.5242504591701043, + "learning_rate": 3.492268725311045e-06, + "loss": 0.9935, + "num_input_tokens_seen": 1952910, + "step": 3393 + }, + { + "epoch": 0.2550728994438599, + "flos": 23452794858240.0, + "grad_norm": 2.163249830663232, + "learning_rate": 3.4919445438128e-06, + "loss": 1.0451, + "num_input_tokens_seen": 1976470, + "step": 3394 + }, + { + "epoch": 0.2551480535096949, + "flos": 20629819722240.0, + "grad_norm": 1.5967967512514325, + "learning_rate": 3.491620273911128e-06, + "loss": 0.8381, + "num_input_tokens_seen": 1999195, + "step": 3395 + }, + { + "epoch": 0.25522320757552985, + "flos": 24123907774080.0, + "grad_norm": 1.4004829833833097, + "learning_rate": 3.491295915625243e-06, + "loss": 1.045, + "num_input_tokens_seen": 2022145, + "step": 3396 + }, + { + "epoch": 0.2552983616413648, + "flos": 28040533482240.0, + "grad_norm": 1.660590045576871, + "learning_rate": 3.490971468974364e-06, + "loss": 1.071, + "num_input_tokens_seen": 2046140, + "step": 3397 + }, + { + "epoch": 0.25537351570719974, + "flos": 20099959712640.0, + "grad_norm": 1.6700305925639467, + "learning_rate": 3.490646933977716e-06, + "loss": 1.0287, + "num_input_tokens_seen": 2069500, + "step": 3398 + }, + { + "epoch": 0.2554486697730347, + "flos": 29595321031680.0, + "grad_norm": 1.4327927516295977, + "learning_rate": 3.49032231065453e-06, + "loss": 0.8873, + "num_input_tokens_seen": 2095315, + "step": 3399 + }, + { + "epoch": 0.2555238238388697, + "flos": 18722618179200.0, + "grad_norm": 1.5284460436500888, + "learning_rate": 3.48999759902404e-06, + "loss": 0.945, + "num_input_tokens_seen": 2117085, + "step": 3400 + }, + { + "epoch": 0.25559897790470465, + "flos": 21794276315520.0, + "grad_norm": 1.4747455054093563, + "learning_rate": 3.4896727991054856e-06, + "loss": 0.9838, + "num_input_tokens_seen": 2140450, + "step": 3401 + }, + { + "epoch": 0.2556741319705396, + "flos": 33617509326720.0, + "grad_norm": 1.5900882467336561, + "learning_rate": 3.4893479109181144e-06, + "loss": 0.8203, + "num_input_tokens_seen": 2166260, + "step": 3402 + }, + { + "epoch": 0.2557492860363746, + "flos": 26170782024960.0, + "grad_norm": 1.7724013473540057, + "learning_rate": 3.489022934481176e-06, + "loss": 0.9496, + "num_input_tokens_seen": 2190730, + "step": 3403 + }, + { + "epoch": 0.2558244401022095, + "flos": 23202100696320.0, + "grad_norm": 1.6413993824411603, + "learning_rate": 3.4886978698139275e-06, + "loss": 1.0028, + "num_input_tokens_seen": 2212675, + "step": 3404 + }, + { + "epoch": 0.2558995941680445, + "flos": 25499633195520.0, + "grad_norm": 1.4650514175713867, + "learning_rate": 3.4883727169356293e-06, + "loss": 0.9828, + "num_input_tokens_seen": 2238495, + "step": 3405 + }, + { + "epoch": 0.25597474823387945, + "flos": 25677043297920.0, + "grad_norm": 1.284200991609764, + "learning_rate": 3.4880474758655485e-06, + "loss": 1.0244, + "num_input_tokens_seen": 2263020, + "step": 3406 + }, + { + "epoch": 0.2560499022997144, + "flos": 64584828564480.0, + "grad_norm": 0.8710951527126634, + "learning_rate": 3.487722146622956e-06, + "loss": 0.8947, + "num_input_tokens_seen": 2342200, + "step": 3407 + }, + { + "epoch": 0.2561250563655494, + "flos": 25677043297920.0, + "grad_norm": 1.5185957433909436, + "learning_rate": 3.48739672922713e-06, + "loss": 1.0368, + "num_input_tokens_seen": 2366820, + "step": 3408 + }, + { + "epoch": 0.25620021043138436, + "flos": 24970671025920.0, + "grad_norm": 1.656064652922347, + "learning_rate": 3.4870712236973524e-06, + "loss": 1.0462, + "num_input_tokens_seen": 2392135, + "step": 3409 + }, + { + "epoch": 0.2562753644972193, + "flos": 20874121263360.0, + "grad_norm": 1.4169842794080671, + "learning_rate": 3.4867456300529096e-06, + "loss": 1.0035, + "num_input_tokens_seen": 2415015, + "step": 3410 + }, + { + "epoch": 0.25635051856305424, + "flos": 36512511546240.0, + "grad_norm": 1.9299973459176112, + "learning_rate": 3.4864199483130957e-06, + "loss": 0.8422, + "num_input_tokens_seen": 2442870, + "step": 3411 + }, + { + "epoch": 0.2564256726288892, + "flos": 21229623826560.0, + "grad_norm": 1.5163639272987264, + "learning_rate": 3.4860941784972077e-06, + "loss": 0.9923, + "num_input_tokens_seen": 2466580, + "step": 3412 + }, + { + "epoch": 0.2565008266947242, + "flos": 24470791073280.0, + "grad_norm": 1.4557781814030242, + "learning_rate": 3.485768320624549e-06, + "loss": 1.0528, + "num_input_tokens_seen": 2488370, + "step": 3413 + }, + { + "epoch": 0.25657598076055915, + "flos": 20347134341760.0, + "grad_norm": 1.5017237295866581, + "learning_rate": 3.485442374714428e-06, + "loss": 0.9626, + "num_input_tokens_seen": 2511975, + "step": 3414 + }, + { + "epoch": 0.2566511348263941, + "flos": 14276599338240.0, + "grad_norm": 1.7399404772081415, + "learning_rate": 3.485116340786158e-06, + "loss": 1.0421, + "num_input_tokens_seen": 2533925, + "step": 3415 + }, + { + "epoch": 0.2567262888922291, + "flos": 21052249637760.0, + "grad_norm": 2.3128841184225952, + "learning_rate": 3.4847902188590582e-06, + "loss": 0.9427, + "num_input_tokens_seen": 2556685, + "step": 3416 + }, + { + "epoch": 0.256801442958064, + "flos": 45755059800960.0, + "grad_norm": 1.5354764570433106, + "learning_rate": 3.484464008952452e-06, + "loss": 0.946, + "num_input_tokens_seen": 2585415, + "step": 3417 + }, + { + "epoch": 0.256876597023899, + "flos": 19210036112640.0, + "grad_norm": 3.2502621765026, + "learning_rate": 3.484137711085669e-06, + "loss": 1.0027, + "num_input_tokens_seen": 2607150, + "step": 3418 + }, + { + "epoch": 0.25695175108973395, + "flos": 22111466866560.0, + "grad_norm": 1.3803600827652673, + "learning_rate": 3.4838113252780435e-06, + "loss": 1.0865, + "num_input_tokens_seen": 2632140, + "step": 3419 + }, + { + "epoch": 0.2570269051555689, + "flos": 22568904743040.0, + "grad_norm": 1.3386250728443656, + "learning_rate": 3.4834848515489154e-06, + "loss": 0.9223, + "num_input_tokens_seen": 2657255, + "step": 3420 + }, + { + "epoch": 0.2571020592214039, + "flos": 22359934385280.0, + "grad_norm": 1.445355388932903, + "learning_rate": 3.4831582899176286e-06, + "loss": 0.9899, + "num_input_tokens_seen": 2681905, + "step": 3421 + }, + { + "epoch": 0.25717721328723886, + "flos": 21017816294400.0, + "grad_norm": 1.4584167274972202, + "learning_rate": 3.4828316404035345e-06, + "loss": 1.032, + "num_input_tokens_seen": 2705265, + "step": 3422 + }, + { + "epoch": 0.2572523673530738, + "flos": 12539086473600.0, + "grad_norm": 1.7625703297103616, + "learning_rate": 3.4825049030259868e-06, + "loss": 0.9752, + "num_input_tokens_seen": 2725045, + "step": 3423 + }, + { + "epoch": 0.25732752141890874, + "flos": 25534712983680.0, + "grad_norm": 4.048936709002839, + "learning_rate": 3.482178077804347e-06, + "loss": 0.9022, + "num_input_tokens_seen": 2749860, + "step": 3424 + }, + { + "epoch": 0.2574026754847437, + "flos": 16919901815040.0, + "grad_norm": 6.388414816843871, + "learning_rate": 3.48185116475798e-06, + "loss": 0.9895, + "num_input_tokens_seen": 2769240, + "step": 3425 + }, + { + "epoch": 0.2574778295505787, + "flos": 21401323666560.0, + "grad_norm": 1.6004695978062804, + "learning_rate": 3.481524163906258e-06, + "loss": 0.9386, + "num_input_tokens_seen": 2790370, + "step": 3426 + }, + { + "epoch": 0.25755298361641366, + "flos": 31038368855040.0, + "grad_norm": 1.5199985128947562, + "learning_rate": 3.4811970752685555e-06, + "loss": 0.9189, + "num_input_tokens_seen": 2817145, + "step": 3427 + }, + { + "epoch": 0.2576281376822486, + "flos": 64581955476480.0, + "grad_norm": 0.8128100317213925, + "learning_rate": 3.4808698988642547e-06, + "loss": 0.7986, + "num_input_tokens_seen": 2895310, + "step": 3428 + }, + { + "epoch": 0.2577032917480836, + "flos": 24508564381440.0, + "grad_norm": 1.5158887810585928, + "learning_rate": 3.4805426347127416e-06, + "loss": 1.0208, + "num_input_tokens_seen": 2922405, + "step": 3429 + }, + { + "epoch": 0.2577784458139185, + "flos": 20559121441920.0, + "grad_norm": 3.8550009709417603, + "learning_rate": 3.4802152828334083e-06, + "loss": 0.9698, + "num_input_tokens_seen": 2946495, + "step": 3430 + }, + { + "epoch": 0.2578535998797535, + "flos": 22889471172480.0, + "grad_norm": 1.8277997762804932, + "learning_rate": 3.479887843245651e-06, + "loss": 1.0113, + "num_input_tokens_seen": 2970455, + "step": 3431 + }, + { + "epoch": 0.25792875394558845, + "flos": 26523124191360.0, + "grad_norm": 6.530724609269567, + "learning_rate": 3.4795603159688725e-06, + "loss": 0.984, + "num_input_tokens_seen": 2993685, + "step": 3432 + }, + { + "epoch": 0.2580039080114234, + "flos": 30435871230720.0, + "grad_norm": 1.569430582816126, + "learning_rate": 3.4792327010224794e-06, + "loss": 0.7712, + "num_input_tokens_seen": 3019010, + "step": 3433 + }, + { + "epoch": 0.2580790620772584, + "flos": 18511241610240.0, + "grad_norm": 1.6424806545517614, + "learning_rate": 3.478904998425884e-06, + "loss": 0.8471, + "num_input_tokens_seen": 3043170, + "step": 3434 + }, + { + "epoch": 0.25815421614309336, + "flos": 36971170485120.0, + "grad_norm": 1.7390959812721742, + "learning_rate": 3.478577208198505e-06, + "loss": 0.9601, + "num_input_tokens_seen": 3068785, + "step": 3435 + }, + { + "epoch": 0.2582293702089283, + "flos": 19353946625280.0, + "grad_norm": 1.9856468390157431, + "learning_rate": 3.478249330359764e-06, + "loss": 1.0038, + "num_input_tokens_seen": 3091535, + "step": 3436 + }, + { + "epoch": 0.25830452427476325, + "flos": 25676181371520.0, + "grad_norm": 1.4993527923504044, + "learning_rate": 3.4779213649290907e-06, + "loss": 0.921, + "num_input_tokens_seen": 3118465, + "step": 3437 + }, + { + "epoch": 0.2583796783405982, + "flos": 27017940326400.0, + "grad_norm": 1.5921763710778565, + "learning_rate": 3.4775933119259162e-06, + "loss": 0.953, + "num_input_tokens_seen": 3142050, + "step": 3438 + }, + { + "epoch": 0.2584548324064332, + "flos": 27864488096640.0, + "grad_norm": 3.9716082515628144, + "learning_rate": 3.47726517136968e-06, + "loss": 0.9362, + "num_input_tokens_seen": 3168070, + "step": 3439 + }, + { + "epoch": 0.25852998647226816, + "flos": 60265338053760.0, + "grad_norm": 0.848063568836511, + "learning_rate": 3.4769369432798258e-06, + "loss": 0.8202, + "num_input_tokens_seen": 3238645, + "step": 3440 + }, + { + "epoch": 0.25860514053810313, + "flos": 20064341220480.0, + "grad_norm": 1.7641994156258194, + "learning_rate": 3.4766086276758014e-06, + "loss": 0.9772, + "num_input_tokens_seen": 3261510, + "step": 3441 + }, + { + "epoch": 0.2586802946039381, + "flos": 23805172938240.0, + "grad_norm": 1.2941154627220568, + "learning_rate": 3.4762802245770627e-06, + "loss": 0.84, + "num_input_tokens_seen": 3287865, + "step": 3442 + }, + { + "epoch": 0.258755448669773, + "flos": 67348769541120.0, + "grad_norm": 0.9180387105052017, + "learning_rate": 3.4759517340030674e-06, + "loss": 0.7976, + "num_input_tokens_seen": 3361560, + "step": 3443 + }, + { + "epoch": 0.258830602735608, + "flos": 21049520204160.0, + "grad_norm": 1.4302106223350854, + "learning_rate": 3.475623155973279e-06, + "loss": 0.8437, + "num_input_tokens_seen": 3385480, + "step": 3444 + }, + { + "epoch": 0.25890575680144295, + "flos": 15935081967360.0, + "grad_norm": 1.67986683442636, + "learning_rate": 3.4752944905071687e-06, + "loss": 0.8923, + "num_input_tokens_seen": 3408370, + "step": 3445 + }, + { + "epoch": 0.2589809108672779, + "flos": 36441813265920.0, + "grad_norm": 1.5667021892795032, + "learning_rate": 3.474965737624211e-06, + "loss": 0.8334, + "num_input_tokens_seen": 3434550, + "step": 3446 + }, + { + "epoch": 0.2590560649331129, + "flos": 22888609246080.0, + "grad_norm": 1.8491534564437602, + "learning_rate": 3.474636897343885e-06, + "loss": 0.8567, + "num_input_tokens_seen": 3457875, + "step": 3447 + }, + { + "epoch": 0.25913121899894787, + "flos": 59277493664640.0, + "grad_norm": 1.2260057193971183, + "learning_rate": 3.474307969685676e-06, + "loss": 0.917, + "num_input_tokens_seen": 3488505, + "step": 3448 + }, + { + "epoch": 0.2592063730647828, + "flos": 24754051071360.0, + "grad_norm": 1.3255620124381091, + "learning_rate": 3.473978954669074e-06, + "loss": 0.9801, + "num_input_tokens_seen": 3512960, + "step": 3449 + }, + { + "epoch": 0.25928152713061775, + "flos": 63613576258560.0, + "grad_norm": 0.7926272704641336, + "learning_rate": 3.473649852313575e-06, + "loss": 0.8552, + "num_input_tokens_seen": 3586655, + "step": 3450 + }, + { + "epoch": 0.2593566811964527, + "flos": 22993273992960.0, + "grad_norm": 1.9406999189624385, + "learning_rate": 3.4733206626386794e-06, + "loss": 1.0664, + "num_input_tokens_seen": 3609515, + "step": 3451 + }, + { + "epoch": 0.2594318352622877, + "flos": 25251309331200.0, + "grad_norm": 1.742433675984213, + "learning_rate": 3.472991385663893e-06, + "loss": 0.9603, + "num_input_tokens_seen": 3635145, + "step": 3452 + }, + { + "epoch": 0.25950698932812266, + "flos": 31104470194560.0, + "grad_norm": 1.9806528099950922, + "learning_rate": 3.4726620214087264e-06, + "loss": 1.0362, + "num_input_tokens_seen": 3656740, + "step": 3453 + }, + { + "epoch": 0.25958214339395763, + "flos": 25782246748800.0, + "grad_norm": 2.0005858278595716, + "learning_rate": 3.4723325698926953e-06, + "loss": 0.8007, + "num_input_tokens_seen": 3680960, + "step": 3454 + }, + { + "epoch": 0.2596572974597926, + "flos": 27155386391040.0, + "grad_norm": 1.1779913424253081, + "learning_rate": 3.4720030311353216e-06, + "loss": 0.9725, + "num_input_tokens_seen": 3707315, + "step": 3455 + }, + { + "epoch": 0.2597324515256275, + "flos": 17523979637760.0, + "grad_norm": 1.903511581870642, + "learning_rate": 3.4716734051561324e-06, + "loss": 0.9995, + "num_input_tokens_seen": 3727550, + "step": 3456 + }, + { + "epoch": 0.2598076055914625, + "flos": 25992258600960.0, + "grad_norm": 1.4358506899755274, + "learning_rate": 3.471343691974658e-06, + "loss": 0.9644, + "num_input_tokens_seen": 3752905, + "step": 3457 + }, + { + "epoch": 0.25988275965729746, + "flos": 19640762069760.0, + "grad_norm": 1.610986069872992, + "learning_rate": 3.471013891610436e-06, + "loss": 1.083, + "num_input_tokens_seen": 3776955, + "step": 3458 + }, + { + "epoch": 0.2599579137231324, + "flos": 19500119694720.0, + "grad_norm": 1.866542561129353, + "learning_rate": 3.4706840040830076e-06, + "loss": 1.0151, + "num_input_tokens_seen": 3799110, + "step": 3459 + }, + { + "epoch": 0.2600330677889674, + "flos": 16146997240320.0, + "grad_norm": 1.6751561277133473, + "learning_rate": 3.4703540294119204e-06, + "loss": 1.0348, + "num_input_tokens_seen": 3819965, + "step": 3460 + }, + { + "epoch": 0.26010822185480237, + "flos": 28147388958720.0, + "grad_norm": 1.866158301520921, + "learning_rate": 3.4700239676167264e-06, + "loss": 0.9016, + "num_input_tokens_seen": 3845735, + "step": 3461 + }, + { + "epoch": 0.2601833759206373, + "flos": 18399825106560.0, + "grad_norm": 1.4745819695559739, + "learning_rate": 3.4696938187169836e-06, + "loss": 1.0201, + "num_input_tokens_seen": 3868035, + "step": 3462 + }, + { + "epoch": 0.26025852998647225, + "flos": 14585278366080.0, + "grad_norm": 1.9754589151229507, + "learning_rate": 3.469363582732254e-06, + "loss": 0.9537, + "num_input_tokens_seen": 3884295, + "step": 3463 + }, + { + "epoch": 0.2603336840523072, + "flos": 25040543293440.0, + "grad_norm": 1.8637301945801321, + "learning_rate": 3.4690332596821065e-06, + "loss": 0.9027, + "num_input_tokens_seen": 3906880, + "step": 3464 + }, + { + "epoch": 0.2604088381181422, + "flos": 12892254652800.0, + "grad_norm": 1.5783148925144377, + "learning_rate": 3.468702849586112e-06, + "loss": 0.9404, + "num_input_tokens_seen": 3928165, + "step": 3465 + }, + { + "epoch": 0.26048399218397716, + "flos": 67122424800000.0, + "grad_norm": 0.8415677141494791, + "learning_rate": 3.4683723524638494e-06, + "loss": 0.7464, + "num_input_tokens_seen": 4007330, + "step": 3466 + }, + { + "epoch": 0.26055914624981213, + "flos": 22779850348800.0, + "grad_norm": 1.4437366039100508, + "learning_rate": 3.4680417683349024e-06, + "loss": 1.0545, + "num_input_tokens_seen": 4031290, + "step": 3467 + }, + { + "epoch": 0.26063430031564705, + "flos": 59890306291200.0, + "grad_norm": 0.8971794608515664, + "learning_rate": 3.46771109721886e-06, + "loss": 0.9014, + "num_input_tokens_seen": 4112705, + "step": 3468 + }, + { + "epoch": 0.260709454381482, + "flos": 22958804736000.0, + "grad_norm": 1.4997599727593456, + "learning_rate": 3.467380339135314e-06, + "loss": 0.9639, + "num_input_tokens_seen": 4137100, + "step": 3469 + }, + { + "epoch": 0.260784608447317, + "flos": 22888034628480.0, + "grad_norm": 1.6073535705367263, + "learning_rate": 3.4670494941038642e-06, + "loss": 0.9527, + "num_input_tokens_seen": 4159440, + "step": 3470 + }, + { + "epoch": 0.26085976251315196, + "flos": 63334913201280.0, + "grad_norm": 0.8753961960403768, + "learning_rate": 3.466718562144114e-06, + "loss": 0.8068, + "num_input_tokens_seen": 4237890, + "step": 3471 + }, + { + "epoch": 0.26093491657898693, + "flos": 66989539762560.0, + "grad_norm": 0.7077843885755125, + "learning_rate": 3.4663875432756726e-06, + "loss": 0.8089, + "num_input_tokens_seen": 4319240, + "step": 3472 + }, + { + "epoch": 0.2610100706448219, + "flos": 18194051059200.0, + "grad_norm": 1.499589704984226, + "learning_rate": 3.466056437518154e-06, + "loss": 0.9886, + "num_input_tokens_seen": 4341565, + "step": 3473 + }, + { + "epoch": 0.26108522471065687, + "flos": 23695085237760.0, + "grad_norm": 1.666685302948073, + "learning_rate": 3.465725244891178e-06, + "loss": 1.0063, + "num_input_tokens_seen": 4366860, + "step": 3474 + }, + { + "epoch": 0.2611603787764918, + "flos": 24049330824960.0, + "grad_norm": 1.3583853616091994, + "learning_rate": 3.465393965414368e-06, + "loss": 0.975, + "num_input_tokens_seen": 4390935, + "step": 3475 + }, + { + "epoch": 0.26123553284232676, + "flos": 28358298650880.0, + "grad_norm": 1.5878657753138985, + "learning_rate": 3.4650625991073543e-06, + "loss": 0.9527, + "num_input_tokens_seen": 4416835, + "step": 3476 + }, + { + "epoch": 0.2613106869081617, + "flos": 23938417111680.0, + "grad_norm": 1.522797748534149, + "learning_rate": 3.464731145989772e-06, + "loss": 0.9381, + "num_input_tokens_seen": 4438145, + "step": 3477 + }, + { + "epoch": 0.2613858409739967, + "flos": 21159069200640.0, + "grad_norm": 1.7093407831480971, + "learning_rate": 3.46439960608126e-06, + "loss": 1.0605, + "num_input_tokens_seen": 4461010, + "step": 3478 + }, + { + "epoch": 0.26146099503983167, + "flos": 19535558618880.0, + "grad_norm": 1.5915346643458104, + "learning_rate": 3.464067979401464e-06, + "loss": 0.9881, + "num_input_tokens_seen": 4482345, + "step": 3479 + }, + { + "epoch": 0.26153614910566664, + "flos": 21512057811840.0, + "grad_norm": 1.6804956665437185, + "learning_rate": 3.4637362659700337e-06, + "loss": 0.9514, + "num_input_tokens_seen": 4506085, + "step": 3480 + }, + { + "epoch": 0.26161130317150155, + "flos": 17733309131520.0, + "grad_norm": 2.0952917088011165, + "learning_rate": 3.463404465806625e-06, + "loss": 0.9462, + "num_input_tokens_seen": 4526455, + "step": 3481 + }, + { + "epoch": 0.2616864572373365, + "flos": 18406038159360.0, + "grad_norm": 1.4204234164901843, + "learning_rate": 3.4630725789308974e-06, + "loss": 1.0057, + "num_input_tokens_seen": 4546915, + "step": 3482 + }, + { + "epoch": 0.2617616113031715, + "flos": 18546393225600.0, + "grad_norm": 2.0858614904149424, + "learning_rate": 3.4627406053625175e-06, + "loss": 0.9872, + "num_input_tokens_seen": 4569745, + "step": 3483 + }, + { + "epoch": 0.26183676536900646, + "flos": 20946687050880.0, + "grad_norm": 1.678284169752044, + "learning_rate": 3.462408545121155e-06, + "loss": 0.9233, + "num_input_tokens_seen": 4592820, + "step": 3484 + }, + { + "epoch": 0.26191191943484143, + "flos": 32700299189760.0, + "grad_norm": 1.5970728772135683, + "learning_rate": 3.462076398226487e-06, + "loss": 0.9368, + "num_input_tokens_seen": 4616230, + "step": 3485 + }, + { + "epoch": 0.2619870735006764, + "flos": 16393920474240.0, + "grad_norm": 1.7448030689683007, + "learning_rate": 3.4617441646981935e-06, + "loss": 1.0183, + "num_input_tokens_seen": 4636425, + "step": 3486 + }, + { + "epoch": 0.2620622275665114, + "flos": 16640017695360.0, + "grad_norm": 1.3111793140454235, + "learning_rate": 3.461411844555961e-06, + "loss": 1.0255, + "num_input_tokens_seen": 4659345, + "step": 3487 + }, + { + "epoch": 0.2621373816323463, + "flos": 20660123001600.0, + "grad_norm": 1.3669835401714723, + "learning_rate": 3.46107943781948e-06, + "loss": 0.9913, + "num_input_tokens_seen": 4682700, + "step": 3488 + }, + { + "epoch": 0.26221253569818126, + "flos": 14770194410880.0, + "grad_norm": 3.311392401012584, + "learning_rate": 3.460746944508448e-06, + "loss": 0.9251, + "num_input_tokens_seen": 4702455, + "step": 3489 + }, + { + "epoch": 0.26228768976401623, + "flos": 24935806719360.0, + "grad_norm": 1.4070358200041388, + "learning_rate": 3.4604143646425655e-06, + "loss": 0.9194, + "num_input_tokens_seen": 4727995, + "step": 3490 + }, + { + "epoch": 0.2623628438298512, + "flos": 29171095436160.0, + "grad_norm": 1.6179978556388914, + "learning_rate": 3.46008169824154e-06, + "loss": 0.9721, + "num_input_tokens_seen": 4754090, + "step": 3491 + }, + { + "epoch": 0.26243799789568617, + "flos": 19465183560960.0, + "grad_norm": 1.6239566387814466, + "learning_rate": 3.4597489453250824e-06, + "loss": 0.9944, + "num_input_tokens_seen": 4774735, + "step": 3492 + }, + { + "epoch": 0.26251315196152114, + "flos": 12546017798400.0, + "grad_norm": 2.9746367865919354, + "learning_rate": 3.4594161059129102e-06, + "loss": 0.8353, + "num_input_tokens_seen": 4793060, + "step": 3493 + }, + { + "epoch": 0.26258830602735606, + "flos": 20205881435520.0, + "grad_norm": 1.6233935852279335, + "learning_rate": 3.4590831800247457e-06, + "loss": 0.9098, + "num_input_tokens_seen": 4816380, + "step": 3494 + }, + { + "epoch": 0.262663460093191, + "flos": 29242332420480.0, + "grad_norm": 1.4276496246274337, + "learning_rate": 3.458750167680315e-06, + "loss": 0.8268, + "num_input_tokens_seen": 4844285, + "step": 3495 + }, + { + "epoch": 0.262738614159026, + "flos": 41171810376960.0, + "grad_norm": 1.2077172561341334, + "learning_rate": 3.458417068899351e-06, + "loss": 0.9475, + "num_input_tokens_seen": 4873620, + "step": 3496 + }, + { + "epoch": 0.26281376822486097, + "flos": 29275832010240.0, + "grad_norm": 1.3820505575380766, + "learning_rate": 3.4580838837015915e-06, + "loss": 0.903, + "num_input_tokens_seen": 4899845, + "step": 3497 + }, + { + "epoch": 0.26288892229069594, + "flos": 34183346964480.0, + "grad_norm": 1.5800524734282153, + "learning_rate": 3.4577506121067784e-06, + "loss": 0.908, + "num_input_tokens_seen": 4924365, + "step": 3498 + }, + { + "epoch": 0.2629640763565309, + "flos": 10393473219840.0, + "grad_norm": 1.8361787028847742, + "learning_rate": 3.457417254134659e-06, + "loss": 0.9077, + "num_input_tokens_seen": 4945640, + "step": 3499 + }, + { + "epoch": 0.2630392304223659, + "flos": 22883365860480.0, + "grad_norm": 1.9055964490854216, + "learning_rate": 3.457083809804986e-06, + "loss": 0.9139, + "num_input_tokens_seen": 4968695, + "step": 3500 + }, + { + "epoch": 0.2631143844882008, + "flos": 30154837875840.0, + "grad_norm": 1.652108610840937, + "learning_rate": 3.456750279137519e-06, + "loss": 0.9692, + "num_input_tokens_seen": 4992990, + "step": 3501 + }, + { + "epoch": 0.26318953855403576, + "flos": 17947235566080.0, + "grad_norm": 1.367094985011584, + "learning_rate": 3.4564166621520193e-06, + "loss": 0.8643, + "num_input_tokens_seen": 5015910, + "step": 3502 + }, + { + "epoch": 0.26326469261987073, + "flos": 25461141615360.0, + "grad_norm": 1.5228223292211445, + "learning_rate": 3.456082958868255e-06, + "loss": 1.0058, + "num_input_tokens_seen": 5040255, + "step": 3503 + }, + { + "epoch": 0.2633398466857057, + "flos": 25040974256640.0, + "grad_norm": 2.0173339696018773, + "learning_rate": 3.455749169306e-06, + "loss": 0.8765, + "num_input_tokens_seen": 5064190, + "step": 3504 + }, + { + "epoch": 0.2634150007515407, + "flos": 22494866497920.0, + "grad_norm": 1.9837285588337046, + "learning_rate": 3.455415293485032e-06, + "loss": 1.0279, + "num_input_tokens_seen": 5086165, + "step": 3505 + }, + { + "epoch": 0.26349015481737564, + "flos": 28074823171200.0, + "grad_norm": 1.3463870498901411, + "learning_rate": 3.455081331425135e-06, + "loss": 0.8893, + "num_input_tokens_seen": 5110790, + "step": 3506 + }, + { + "epoch": 0.26356530888321056, + "flos": 19740075690240.0, + "grad_norm": 1.6575125434596487, + "learning_rate": 3.4547472831460973e-06, + "loss": 0.8916, + "num_input_tokens_seen": 5132785, + "step": 3507 + }, + { + "epoch": 0.26364046294904553, + "flos": 22492639854720.0, + "grad_norm": 1.4920555975171144, + "learning_rate": 3.4544131486677124e-06, + "loss": 1.058, + "num_input_tokens_seen": 5157170, + "step": 3508 + }, + { + "epoch": 0.2637156170148805, + "flos": 24510072752640.0, + "grad_norm": 1.6010827066906756, + "learning_rate": 3.454078928009779e-06, + "loss": 0.9999, + "num_input_tokens_seen": 5183715, + "step": 3509 + }, + { + "epoch": 0.26379077108071547, + "flos": 25250160096000.0, + "grad_norm": 3.2138796504430185, + "learning_rate": 3.4537446211921008e-06, + "loss": 0.9734, + "num_input_tokens_seen": 5207145, + "step": 3510 + }, + { + "epoch": 0.26386592514655044, + "flos": 19747186583040.0, + "grad_norm": 1.5543159539297353, + "learning_rate": 3.4534102282344876e-06, + "loss": 1.0342, + "num_input_tokens_seen": 5230055, + "step": 3511 + }, + { + "epoch": 0.2639410792123854, + "flos": 21896534851200.0, + "grad_norm": 1.5871394070772606, + "learning_rate": 3.453075749156753e-06, + "loss": 1.0204, + "num_input_tokens_seen": 5253055, + "step": 3512 + }, + { + "epoch": 0.2640162332782203, + "flos": 24539729587200.0, + "grad_norm": 1.809102919814459, + "learning_rate": 3.4527411839787152e-06, + "loss": 0.967, + "num_input_tokens_seen": 5276020, + "step": 3513 + }, + { + "epoch": 0.2640913873440553, + "flos": 45266313064320.0, + "grad_norm": 1.3709788478079374, + "learning_rate": 3.4524065327202e-06, + "loss": 0.8844, + "num_input_tokens_seen": 5304590, + "step": 3514 + }, + { + "epoch": 0.26416654140989027, + "flos": 23236534039680.0, + "grad_norm": 1.4343625535611957, + "learning_rate": 3.4520717954010356e-06, + "loss": 0.9288, + "num_input_tokens_seen": 5329195, + "step": 3515 + }, + { + "epoch": 0.26424169547572524, + "flos": 23168529279360.0, + "grad_norm": 1.7990603882056786, + "learning_rate": 3.4517369720410576e-06, + "loss": 0.9937, + "num_input_tokens_seen": 5354370, + "step": 3516 + }, + { + "epoch": 0.2643168495415602, + "flos": 18299434078080.0, + "grad_norm": 1.6317587194430412, + "learning_rate": 3.4514020626601044e-06, + "loss": 1.0467, + "num_input_tokens_seen": 5377075, + "step": 3517 + }, + { + "epoch": 0.2643920036073952, + "flos": 21405705125760.0, + "grad_norm": 1.4822657004052615, + "learning_rate": 3.451067067278021e-06, + "loss": 0.956, + "num_input_tokens_seen": 5401020, + "step": 3518 + }, + { + "epoch": 0.26446715767323015, + "flos": 20170334770560.0, + "grad_norm": 1.3782206011286315, + "learning_rate": 3.4507319859146585e-06, + "loss": 1.0422, + "num_input_tokens_seen": 5424550, + "step": 3519 + }, + { + "epoch": 0.26454231173906506, + "flos": 18794106558720.0, + "grad_norm": 1.4321844257319243, + "learning_rate": 3.4503968185898696e-06, + "loss": 0.9379, + "num_input_tokens_seen": 5446875, + "step": 3520 + }, + { + "epoch": 0.26461746580490003, + "flos": 21684009047040.0, + "grad_norm": 2.1186092993906063, + "learning_rate": 3.450061565323516e-06, + "loss": 0.9415, + "num_input_tokens_seen": 5470060, + "step": 3521 + }, + { + "epoch": 0.264692619870735, + "flos": 20517074415360.0, + "grad_norm": 1.4184492153668058, + "learning_rate": 3.449726226135461e-06, + "loss": 1.0371, + "num_input_tokens_seen": 5491110, + "step": 3522 + }, + { + "epoch": 0.26476777393657, + "flos": 22635796181760.0, + "grad_norm": 1.7010246414606471, + "learning_rate": 3.4493908010455762e-06, + "loss": 0.9592, + "num_input_tokens_seen": 5513220, + "step": 3523 + }, + { + "epoch": 0.26484292800240494, + "flos": 15017189472000.0, + "grad_norm": 1.6640637894746324, + "learning_rate": 3.4490552900737363e-06, + "loss": 0.9906, + "num_input_tokens_seen": 5535815, + "step": 3524 + }, + { + "epoch": 0.2649180820682399, + "flos": 20874264917760.0, + "grad_norm": 1.498763687542159, + "learning_rate": 3.448719693239822e-06, + "loss": 1.0041, + "num_input_tokens_seen": 5560950, + "step": 3525 + }, + { + "epoch": 0.2649932361340748, + "flos": 23417607329280.0, + "grad_norm": 1.491412535744595, + "learning_rate": 3.448384010563718e-06, + "loss": 0.8703, + "num_input_tokens_seen": 5583470, + "step": 3526 + }, + { + "epoch": 0.2650683901999098, + "flos": 21159105114240.0, + "grad_norm": 1.3122572447303733, + "learning_rate": 3.4480482420653153e-06, + "loss": 0.9348, + "num_input_tokens_seen": 5606115, + "step": 3527 + }, + { + "epoch": 0.26514354426574477, + "flos": 15405940229760.0, + "grad_norm": 1.356663158437206, + "learning_rate": 3.4477123877645093e-06, + "loss": 0.9754, + "num_input_tokens_seen": 5628015, + "step": 3528 + }, + { + "epoch": 0.26521869833157974, + "flos": 27511391744640.0, + "grad_norm": 1.9254944996008594, + "learning_rate": 3.4473764476812004e-06, + "loss": 1.0552, + "num_input_tokens_seen": 5649450, + "step": 3529 + }, + { + "epoch": 0.2652938523974147, + "flos": 16463864568960.0, + "grad_norm": 1.5660187582059342, + "learning_rate": 3.447040421835295e-06, + "loss": 0.9173, + "num_input_tokens_seen": 5672480, + "step": 3530 + }, + { + "epoch": 0.2653690064632497, + "flos": 16711039198080.0, + "grad_norm": 1.5954615808187078, + "learning_rate": 3.446704310246703e-06, + "loss": 0.8526, + "num_input_tokens_seen": 5694470, + "step": 3531 + }, + { + "epoch": 0.26544416052908465, + "flos": 18089134917120.0, + "grad_norm": 1.506125262148734, + "learning_rate": 3.4463681129353413e-06, + "loss": 0.9983, + "num_input_tokens_seen": 5718575, + "step": 3532 + }, + { + "epoch": 0.26551931459491956, + "flos": 21017708553600.0, + "grad_norm": 1.2302379573363613, + "learning_rate": 3.4460318299211304e-06, + "loss": 0.9707, + "num_input_tokens_seen": 5742315, + "step": 3533 + }, + { + "epoch": 0.26559446866075453, + "flos": 21187792281600.0, + "grad_norm": 1.7360264244617487, + "learning_rate": 3.4456954612239964e-06, + "loss": 0.98, + "num_input_tokens_seen": 5763950, + "step": 3534 + }, + { + "epoch": 0.2656696227265895, + "flos": 22323777189120.0, + "grad_norm": 1.3421605046391873, + "learning_rate": 3.44535900686387e-06, + "loss": 0.9042, + "num_input_tokens_seen": 5787425, + "step": 3535 + }, + { + "epoch": 0.2657447767924245, + "flos": 18124250618880.0, + "grad_norm": 1.7530049682433626, + "learning_rate": 3.4450224668606884e-06, + "loss": 1.0223, + "num_input_tokens_seen": 5810360, + "step": 3536 + }, + { + "epoch": 0.26581993085825945, + "flos": 26276560093440.0, + "grad_norm": 1.5985481738485017, + "learning_rate": 3.444685841234392e-06, + "loss": 1.0209, + "num_input_tokens_seen": 5835795, + "step": 3537 + }, + { + "epoch": 0.2658950849240944, + "flos": 25351484878080.0, + "grad_norm": 1.5104113699283024, + "learning_rate": 3.444349130004927e-06, + "loss": 1.0168, + "num_input_tokens_seen": 5858365, + "step": 3538 + }, + { + "epoch": 0.26597023898992933, + "flos": 15334667331840.0, + "grad_norm": 1.499755601046138, + "learning_rate": 3.4440123331922457e-06, + "loss": 1.0168, + "num_input_tokens_seen": 5879315, + "step": 3539 + }, + { + "epoch": 0.2660453930557643, + "flos": 26524417080960.0, + "grad_norm": 1.539671801484531, + "learning_rate": 3.443675450816304e-06, + "loss": 0.9396, + "num_input_tokens_seen": 5903825, + "step": 3540 + }, + { + "epoch": 0.26612054712159927, + "flos": 25853160510720.0, + "grad_norm": 1.7052612905448719, + "learning_rate": 3.4433384828970636e-06, + "loss": 0.9453, + "num_input_tokens_seen": 5926995, + "step": 3541 + }, + { + "epoch": 0.26619570118743424, + "flos": 28782236937600.0, + "grad_norm": 1.3082827070285725, + "learning_rate": 3.443001429454491e-06, + "loss": 0.9931, + "num_input_tokens_seen": 5954555, + "step": 3542 + }, + { + "epoch": 0.2662708552532692, + "flos": 12723356073600.0, + "grad_norm": 1.7775051840104321, + "learning_rate": 3.4426642905085585e-06, + "loss": 0.9307, + "num_input_tokens_seen": 5977615, + "step": 3543 + }, + { + "epoch": 0.2663460093191042, + "flos": 22288517832960.0, + "grad_norm": 1.7782114447043795, + "learning_rate": 3.4423270660792422e-06, + "loss": 0.9474, + "num_input_tokens_seen": 6001795, + "step": 3544 + }, + { + "epoch": 0.26642116338493915, + "flos": 24717857961600.0, + "grad_norm": 5.4143642324943615, + "learning_rate": 3.4419897561865242e-06, + "loss": 1.0024, + "num_input_tokens_seen": 6023195, + "step": 3545 + }, + { + "epoch": 0.26649631745077407, + "flos": 16852435758720.0, + "grad_norm": 1.8982038314873915, + "learning_rate": 3.4416523608503914e-06, + "loss": 0.9406, + "num_input_tokens_seen": 6043260, + "step": 3546 + }, + { + "epoch": 0.26657147151660904, + "flos": 20833869916800.0, + "grad_norm": 1.6223072336006121, + "learning_rate": 3.4413148800908364e-06, + "loss": 1.0603, + "num_input_tokens_seen": 6062890, + "step": 3547 + }, + { + "epoch": 0.266646625582444, + "flos": 23200269102720.0, + "grad_norm": 1.5643292972169336, + "learning_rate": 3.4409773139278546e-06, + "loss": 1.0269, + "num_input_tokens_seen": 6083190, + "step": 3548 + }, + { + "epoch": 0.266721779648279, + "flos": 20235969233280.0, + "grad_norm": 1.513995051185653, + "learning_rate": 3.44063966238145e-06, + "loss": 0.9974, + "num_input_tokens_seen": 6105190, + "step": 3549 + }, + { + "epoch": 0.26679693371411395, + "flos": 21617297176320.0, + "grad_norm": 1.5051385759955225, + "learning_rate": 3.440301925471628e-06, + "loss": 0.9845, + "num_input_tokens_seen": 6130425, + "step": 3550 + }, + { + "epoch": 0.2668720877799489, + "flos": 28037444912640.0, + "grad_norm": 1.4594843364004433, + "learning_rate": 3.439964103218402e-06, + "loss": 0.8901, + "num_input_tokens_seen": 6154545, + "step": 3551 + }, + { + "epoch": 0.26694724184578383, + "flos": 21900126211200.0, + "grad_norm": 1.8183895317551675, + "learning_rate": 3.439626195641789e-06, + "loss": 1.0215, + "num_input_tokens_seen": 6177995, + "step": 3552 + }, + { + "epoch": 0.2670223959116188, + "flos": 22888752900480.0, + "grad_norm": 1.4423905813412141, + "learning_rate": 3.4392882027618113e-06, + "loss": 0.9461, + "num_input_tokens_seen": 6200790, + "step": 3553 + }, + { + "epoch": 0.2670975499774538, + "flos": 20732616961920.0, + "grad_norm": 2.0283183142367984, + "learning_rate": 3.438950124598496e-06, + "loss": 0.7702, + "num_input_tokens_seen": 6220135, + "step": 3554 + }, + { + "epoch": 0.26717270404328874, + "flos": 18547147411200.0, + "grad_norm": 1.3866874154002697, + "learning_rate": 3.438611961171875e-06, + "loss": 0.8899, + "num_input_tokens_seen": 6243430, + "step": 3555 + }, + { + "epoch": 0.2672478581091237, + "flos": 17770543735680.0, + "grad_norm": 1.5749868029086935, + "learning_rate": 3.4382737125019874e-06, + "loss": 1.0379, + "num_input_tokens_seen": 6266640, + "step": 3556 + }, + { + "epoch": 0.2673230121749587, + "flos": 17098030189440.0, + "grad_norm": 1.7376473974923925, + "learning_rate": 3.4379353786088748e-06, + "loss": 1.0939, + "num_input_tokens_seen": 6287195, + "step": 3557 + }, + { + "epoch": 0.2673981662407936, + "flos": 29378449681920.0, + "grad_norm": 1.723900793062257, + "learning_rate": 3.437596959512585e-06, + "loss": 1.0069, + "num_input_tokens_seen": 6310170, + "step": 3558 + }, + { + "epoch": 0.26747332030662857, + "flos": 20418155844480.0, + "grad_norm": 1.5548580977036943, + "learning_rate": 3.4372584552331694e-06, + "loss": 0.9541, + "num_input_tokens_seen": 6332275, + "step": 3559 + }, + { + "epoch": 0.26754847437246354, + "flos": 23627475527040.0, + "grad_norm": 1.428875885604839, + "learning_rate": 3.4369198657906875e-06, + "loss": 0.9419, + "num_input_tokens_seen": 6355355, + "step": 3560 + }, + { + "epoch": 0.2676236284382985, + "flos": 46954021564800.0, + "grad_norm": 1.491160733540135, + "learning_rate": 3.4365811912052013e-06, + "loss": 0.9909, + "num_input_tokens_seen": 6380500, + "step": 3561 + }, + { + "epoch": 0.2676987825041335, + "flos": 19994540780160.0, + "grad_norm": 1.705987257517444, + "learning_rate": 3.4362424314967777e-06, + "loss": 0.9785, + "num_input_tokens_seen": 6402500, + "step": 3562 + }, + { + "epoch": 0.26777393656996845, + "flos": 65146500224640.0, + "grad_norm": 0.856242557131308, + "learning_rate": 3.4359035866854907e-06, + "loss": 0.8499, + "num_input_tokens_seen": 6481055, + "step": 3563 + }, + { + "epoch": 0.2678490906358034, + "flos": 29343010757760.0, + "grad_norm": 1.561186734483732, + "learning_rate": 3.435564656791418e-06, + "loss": 0.9381, + "num_input_tokens_seen": 6507350, + "step": 3564 + }, + { + "epoch": 0.26792424470163834, + "flos": 22881354698880.0, + "grad_norm": 1.592645673930972, + "learning_rate": 3.435225641834642e-06, + "loss": 1.0665, + "num_input_tokens_seen": 6528110, + "step": 3565 + }, + { + "epoch": 0.2679993987674733, + "flos": 24087678750720.0, + "grad_norm": 1.1843264486783973, + "learning_rate": 3.434886541835251e-06, + "loss": 0.9734, + "num_input_tokens_seen": 6555205, + "step": 3566 + }, + { + "epoch": 0.2680745528333083, + "flos": 20170550252160.0, + "grad_norm": 1.3492755394201237, + "learning_rate": 3.434547356813338e-06, + "loss": 0.9937, + "num_input_tokens_seen": 6580430, + "step": 3567 + }, + { + "epoch": 0.26814970689914325, + "flos": 21475649220480.0, + "grad_norm": 1.5544636627544088, + "learning_rate": 3.4342080867890006e-06, + "loss": 0.9627, + "num_input_tokens_seen": 6603795, + "step": 3568 + }, + { + "epoch": 0.2682248609649782, + "flos": 21053686181760.0, + "grad_norm": 1.9934251119967568, + "learning_rate": 3.4338687317823425e-06, + "loss": 1.018, + "num_input_tokens_seen": 6627755, + "step": 3569 + }, + { + "epoch": 0.2683000150308132, + "flos": 29346422549760.0, + "grad_norm": 1.3135059225778754, + "learning_rate": 3.4335292918134713e-06, + "loss": 0.9822, + "num_input_tokens_seen": 6653450, + "step": 3570 + }, + { + "epoch": 0.2683751690966481, + "flos": 20516320229760.0, + "grad_norm": 1.517419990931576, + "learning_rate": 3.4331897669024996e-06, + "loss": 1.0398, + "num_input_tokens_seen": 6675625, + "step": 3571 + }, + { + "epoch": 0.2684503231624831, + "flos": 21546814377600.0, + "grad_norm": 1.9617707811122411, + "learning_rate": 3.432850157069546e-06, + "loss": 1.0482, + "num_input_tokens_seen": 6698540, + "step": 3572 + }, + { + "epoch": 0.26852547722831804, + "flos": 22536159338880.0, + "grad_norm": 1.5429441039615885, + "learning_rate": 3.4325104623347345e-06, + "loss": 1.0158, + "num_input_tokens_seen": 6721115, + "step": 3573 + }, + { + "epoch": 0.268600631294153, + "flos": 23627726922240.0, + "grad_norm": 1.321609988149788, + "learning_rate": 3.432170682718193e-06, + "loss": 0.949, + "num_input_tokens_seen": 6744360, + "step": 3574 + }, + { + "epoch": 0.268675785359988, + "flos": 19430247427200.0, + "grad_norm": 1.4132587403655226, + "learning_rate": 3.431830818240054e-06, + "loss": 0.8746, + "num_input_tokens_seen": 6766690, + "step": 3575 + }, + { + "epoch": 0.26875093942582295, + "flos": 20098451341440.0, + "grad_norm": 1.3612594696640314, + "learning_rate": 3.431490868920456e-06, + "loss": 0.9569, + "num_input_tokens_seen": 6791120, + "step": 3576 + }, + { + "epoch": 0.2688260934916579, + "flos": 23029718497920.0, + "grad_norm": 1.51733836436541, + "learning_rate": 3.4311508347795427e-06, + "loss": 0.99, + "num_input_tokens_seen": 6815625, + "step": 3577 + }, + { + "epoch": 0.26890124755749284, + "flos": 29628425571840.0, + "grad_norm": 1.5987810498429464, + "learning_rate": 3.430810715837462e-06, + "loss": 0.9158, + "num_input_tokens_seen": 6841770, + "step": 3578 + }, + { + "epoch": 0.2689764016233278, + "flos": 25465020284160.0, + "grad_norm": 1.611253031512502, + "learning_rate": 3.4304705121143674e-06, + "loss": 0.9753, + "num_input_tokens_seen": 6864670, + "step": 3579 + }, + { + "epoch": 0.2690515556891628, + "flos": 14806100211840.0, + "grad_norm": 1.5262081676002162, + "learning_rate": 3.4301302236304174e-06, + "loss": 0.9724, + "num_input_tokens_seen": 6884670, + "step": 3580 + }, + { + "epoch": 0.26912670975499775, + "flos": 20594991329280.0, + "grad_norm": 1.706733101443333, + "learning_rate": 3.4297898504057754e-06, + "loss": 0.9558, + "num_input_tokens_seen": 6909450, + "step": 3581 + }, + { + "epoch": 0.2692018638208327, + "flos": 24049977269760.0, + "grad_norm": 1.333532119873341, + "learning_rate": 3.4294493924606095e-06, + "loss": 1.0304, + "num_input_tokens_seen": 6935710, + "step": 3582 + }, + { + "epoch": 0.2692770178866677, + "flos": 18688328490240.0, + "grad_norm": 1.465900172447727, + "learning_rate": 3.429108849815094e-06, + "loss": 1.0695, + "num_input_tokens_seen": 6957955, + "step": 3583 + }, + { + "epoch": 0.2693521719525026, + "flos": 28712292842880.0, + "grad_norm": 1.4688128194158112, + "learning_rate": 3.428768222489406e-06, + "loss": 0.895, + "num_input_tokens_seen": 6983360, + "step": 3584 + }, + { + "epoch": 0.2694273260183376, + "flos": 21365920656000.0, + "grad_norm": 1.8137286395675818, + "learning_rate": 3.4284275105037298e-06, + "loss": 0.9853, + "num_input_tokens_seen": 7007775, + "step": 3585 + }, + { + "epoch": 0.26950248008417255, + "flos": 26806491930240.0, + "grad_norm": 2.150278973777491, + "learning_rate": 3.4280867138782544e-06, + "loss": 0.9627, + "num_input_tokens_seen": 7032285, + "step": 3586 + }, + { + "epoch": 0.2695776341500075, + "flos": 23347555493760.0, + "grad_norm": 1.5892657604771225, + "learning_rate": 3.427745832633172e-06, + "loss": 0.9917, + "num_input_tokens_seen": 7056530, + "step": 3587 + }, + { + "epoch": 0.2696527882158425, + "flos": 22006550724480.0, + "grad_norm": 1.4050131126552141, + "learning_rate": 3.4274048667886826e-06, + "loss": 0.9606, + "num_input_tokens_seen": 7081065, + "step": 3588 + }, + { + "epoch": 0.26972794228167746, + "flos": 23876769058560.0, + "grad_norm": 1.4325096750515198, + "learning_rate": 3.4270638163649884e-06, + "loss": 1.0287, + "num_input_tokens_seen": 7105475, + "step": 3589 + }, + { + "epoch": 0.2698030963475124, + "flos": 19465865919360.0, + "grad_norm": 1.646263895171663, + "learning_rate": 3.4267226813822983e-06, + "loss": 0.9416, + "num_input_tokens_seen": 7126575, + "step": 3590 + }, + { + "epoch": 0.26987825041334734, + "flos": 25851185262720.0, + "grad_norm": 1.3021158697513682, + "learning_rate": 3.426381461860826e-06, + "loss": 1.0008, + "num_input_tokens_seen": 7151865, + "step": 3591 + }, + { + "epoch": 0.2699534044791823, + "flos": 26417884826880.0, + "grad_norm": 1.5061072159200877, + "learning_rate": 3.4260401578207904e-06, + "loss": 0.9716, + "num_input_tokens_seen": 7177680, + "step": 3592 + }, + { + "epoch": 0.2700285585450173, + "flos": 18264605685120.0, + "grad_norm": 1.3201196744161412, + "learning_rate": 3.425698769282415e-06, + "loss": 0.9887, + "num_input_tokens_seen": 7200625, + "step": 3593 + }, + { + "epoch": 0.27010371261085225, + "flos": 19639756488960.0, + "grad_norm": 1.4974654560212495, + "learning_rate": 3.4253572962659276e-06, + "loss": 0.986, + "num_input_tokens_seen": 7225365, + "step": 3594 + }, + { + "epoch": 0.2701788666766872, + "flos": 15617280885120.0, + "grad_norm": 1.6645136834931729, + "learning_rate": 3.425015738791563e-06, + "loss": 0.906, + "num_input_tokens_seen": 7247235, + "step": 3595 + }, + { + "epoch": 0.2702540207425222, + "flos": 20906579358720.0, + "grad_norm": 1.8454228587020494, + "learning_rate": 3.424674096879559e-06, + "loss": 0.9247, + "num_input_tokens_seen": 7268035, + "step": 3596 + }, + { + "epoch": 0.2703291748083571, + "flos": 17094618397440.0, + "grad_norm": 1.4912343601068565, + "learning_rate": 3.424332370550159e-06, + "loss": 0.9812, + "num_input_tokens_seen": 7289940, + "step": 3597 + }, + { + "epoch": 0.2704043288741921, + "flos": 18649082724480.0, + "grad_norm": 1.5980406207864661, + "learning_rate": 3.4239905598236115e-06, + "loss": 0.9863, + "num_input_tokens_seen": 7312415, + "step": 3598 + }, + { + "epoch": 0.27047948294002705, + "flos": 38236484983680.0, + "grad_norm": 1.3425675445841188, + "learning_rate": 3.423648664720171e-06, + "loss": 1.0063, + "num_input_tokens_seen": 7337735, + "step": 3599 + }, + { + "epoch": 0.270554637005862, + "flos": 19782230457600.0, + "grad_norm": 1.5136073732476767, + "learning_rate": 3.4233066852600958e-06, + "loss": 1.042, + "num_input_tokens_seen": 7359790, + "step": 3600 + }, + { + "epoch": 0.270629791071697, + "flos": 21829032881280.0, + "grad_norm": 1.4310781250746372, + "learning_rate": 3.422964621463649e-06, + "loss": 0.9698, + "num_input_tokens_seen": 7385035, + "step": 3601 + }, + { + "epoch": 0.27070494513753196, + "flos": 19358866788480.0, + "grad_norm": 2.318623142936597, + "learning_rate": 3.4226224733511e-06, + "loss": 1.038, + "num_input_tokens_seen": 7406425, + "step": 3602 + }, + { + "epoch": 0.2707800992033669, + "flos": 19923555191040.0, + "grad_norm": 1.811820659263388, + "learning_rate": 3.4222802409427216e-06, + "loss": 0.918, + "num_input_tokens_seen": 7427600, + "step": 3603 + }, + { + "epoch": 0.27085525326920185, + "flos": 21759304268160.0, + "grad_norm": 1.4559741237772348, + "learning_rate": 3.421937924258792e-06, + "loss": 1.0167, + "num_input_tokens_seen": 7452980, + "step": 3604 + }, + { + "epoch": 0.2709304073350368, + "flos": 25677079211520.0, + "grad_norm": 1.3690523344949899, + "learning_rate": 3.421595523319596e-06, + "loss": 1.0577, + "num_input_tokens_seen": 7477160, + "step": 3605 + }, + { + "epoch": 0.2710055614008718, + "flos": 20905573777920.0, + "grad_norm": 1.5362748979455125, + "learning_rate": 3.421253038145421e-06, + "loss": 1.0081, + "num_input_tokens_seen": 7498775, + "step": 3606 + }, + { + "epoch": 0.27108071546670676, + "flos": 20135219068800.0, + "grad_norm": 1.3290582855825115, + "learning_rate": 3.420910468756562e-06, + "loss": 0.8845, + "num_input_tokens_seen": 7523515, + "step": 3607 + }, + { + "epoch": 0.2711558695325417, + "flos": 22147803630720.0, + "grad_norm": 1.5323828574334748, + "learning_rate": 3.4205678151733162e-06, + "loss": 0.9393, + "num_input_tokens_seen": 7547895, + "step": 3608 + }, + { + "epoch": 0.2712310235983767, + "flos": 14346794828160.0, + "grad_norm": 1.549550526575891, + "learning_rate": 3.420225077415988e-06, + "loss": 0.975, + "num_input_tokens_seen": 7569675, + "step": 3609 + }, + { + "epoch": 0.2713061776642116, + "flos": 21787093595520.0, + "grad_norm": 1.4297883045116797, + "learning_rate": 3.4198822555048856e-06, + "loss": 1.0774, + "num_input_tokens_seen": 7591210, + "step": 3610 + }, + { + "epoch": 0.2713813317300466, + "flos": 25921524407040.0, + "grad_norm": 1.4224493522544146, + "learning_rate": 3.419539349460322e-06, + "loss": 1.0629, + "num_input_tokens_seen": 7613570, + "step": 3611 + }, + { + "epoch": 0.27145648579588155, + "flos": 17699486319360.0, + "grad_norm": 1.8897084938776922, + "learning_rate": 3.4191963593026163e-06, + "loss": 1.0486, + "num_input_tokens_seen": 7633520, + "step": 3612 + }, + { + "epoch": 0.2715316398617165, + "flos": 13994021698560.0, + "grad_norm": 1.9465679728733742, + "learning_rate": 3.4188532850520924e-06, + "loss": 0.9953, + "num_input_tokens_seen": 7653725, + "step": 3613 + }, + { + "epoch": 0.2716067939275515, + "flos": 15934830572160.0, + "grad_norm": 2.5760503571072313, + "learning_rate": 3.4185101267290773e-06, + "loss": 0.975, + "num_input_tokens_seen": 7673475, + "step": 3614 + }, + { + "epoch": 0.27168194799338646, + "flos": 26556587867520.0, + "grad_norm": 1.3532303238330978, + "learning_rate": 3.418166884353906e-06, + "loss": 0.9796, + "num_input_tokens_seen": 7697695, + "step": 3615 + }, + { + "epoch": 0.2717571020592214, + "flos": 23273337680640.0, + "grad_norm": 1.522717036545405, + "learning_rate": 3.4178235579469154e-06, + "loss": 1.0015, + "num_input_tokens_seen": 7721370, + "step": 3616 + }, + { + "epoch": 0.27183225612505635, + "flos": 23735013361920.0, + "grad_norm": 1.3642211559616542, + "learning_rate": 3.417480147528451e-06, + "loss": 0.9815, + "num_input_tokens_seen": 7745615, + "step": 3617 + }, + { + "epoch": 0.2719074101908913, + "flos": 17276948663040.0, + "grad_norm": 1.6524381137172814, + "learning_rate": 3.4171366531188596e-06, + "loss": 0.9911, + "num_input_tokens_seen": 7767955, + "step": 3618 + }, + { + "epoch": 0.2719825642567263, + "flos": 19605646368000.0, + "grad_norm": 1.4424727301441345, + "learning_rate": 3.4167930747384947e-06, + "loss": 0.9606, + "num_input_tokens_seen": 7790800, + "step": 3619 + }, + { + "epoch": 0.27205771832256126, + "flos": 25217881568640.0, + "grad_norm": 1.5821136829893419, + "learning_rate": 3.416449412407715e-06, + "loss": 0.973, + "num_input_tokens_seen": 7813830, + "step": 3620 + }, + { + "epoch": 0.27213287238839623, + "flos": 14629659776640.0, + "grad_norm": 1.3315353589177945, + "learning_rate": 3.4161056661468834e-06, + "loss": 1.1114, + "num_input_tokens_seen": 7835740, + "step": 3621 + }, + { + "epoch": 0.2722080264542312, + "flos": 30226003032960.0, + "grad_norm": 1.4734981164366843, + "learning_rate": 3.4157618359763687e-06, + "loss": 0.9243, + "num_input_tokens_seen": 7859995, + "step": 3622 + }, + { + "epoch": 0.2722831805200661, + "flos": 24617682414720.0, + "grad_norm": 1.4039207665781377, + "learning_rate": 3.4154179219165435e-06, + "loss": 1.0005, + "num_input_tokens_seen": 7883105, + "step": 3623 + }, + { + "epoch": 0.2723583345859011, + "flos": 18364745318400.0, + "grad_norm": 1.7890738940813138, + "learning_rate": 3.415073923987787e-06, + "loss": 0.9531, + "num_input_tokens_seen": 7904205, + "step": 3624 + }, + { + "epoch": 0.27243348865173606, + "flos": 21794240401920.0, + "grad_norm": 1.7591124750365172, + "learning_rate": 3.4147298422104815e-06, + "loss": 0.9583, + "num_input_tokens_seen": 7926490, + "step": 3625 + }, + { + "epoch": 0.272508642717571, + "flos": 41841486748800.0, + "grad_norm": 1.3881606651234737, + "learning_rate": 3.4143856766050157e-06, + "loss": 0.9002, + "num_input_tokens_seen": 7953830, + "step": 3626 + }, + { + "epoch": 0.272583796783406, + "flos": 19000527050880.0, + "grad_norm": 1.5355172157577086, + "learning_rate": 3.4140414271917825e-06, + "loss": 1.0358, + "num_input_tokens_seen": 7974800, + "step": 3627 + }, + { + "epoch": 0.27265895084924097, + "flos": 25500602862720.0, + "grad_norm": 1.8655779297480153, + "learning_rate": 3.4136970939911793e-06, + "loss": 1.0015, + "num_input_tokens_seen": 8000230, + "step": 3628 + }, + { + "epoch": 0.2727341049150759, + "flos": 68977308026880.0, + "grad_norm": 0.86091977774489, + "learning_rate": 3.413352677023611e-06, + "loss": 0.803, + "num_input_tokens_seen": 8069530, + "step": 3629 + }, + { + "epoch": 0.27280925898091085, + "flos": 19282027282560.0, + "grad_norm": 2.341379925513355, + "learning_rate": 3.4130081763094836e-06, + "loss": 0.9524, + "num_input_tokens_seen": 8091705, + "step": 3630 + }, + { + "epoch": 0.2728844130467458, + "flos": 34889180532480.0, + "grad_norm": 1.476039894908774, + "learning_rate": 3.4126635918692114e-06, + "loss": 0.9759, + "num_input_tokens_seen": 8117490, + "step": 3631 + }, + { + "epoch": 0.2729595671125808, + "flos": 22464886440960.0, + "grad_norm": 1.4252556655147557, + "learning_rate": 3.412318923723212e-06, + "loss": 0.9688, + "num_input_tokens_seen": 8141645, + "step": 3632 + }, + { + "epoch": 0.27303472117841576, + "flos": 16182005201280.0, + "grad_norm": 2.0623652096254763, + "learning_rate": 3.411974171891908e-06, + "loss": 1.015, + "num_input_tokens_seen": 8161800, + "step": 3633 + }, + { + "epoch": 0.27310987524425073, + "flos": 16992036639360.0, + "grad_norm": 1.4380218553752462, + "learning_rate": 3.4116293363957276e-06, + "loss": 0.9754, + "num_input_tokens_seen": 8183950, + "step": 3634 + }, + { + "epoch": 0.2731850293100857, + "flos": 26064968042880.0, + "grad_norm": 1.8301783058975905, + "learning_rate": 3.4112844172551034e-06, + "loss": 0.8783, + "num_input_tokens_seen": 8209405, + "step": 3635 + }, + { + "epoch": 0.2732601833759206, + "flos": 25533527834880.0, + "grad_norm": 1.2806630534694152, + "learning_rate": 3.410939414490474e-06, + "loss": 0.9489, + "num_input_tokens_seen": 8234140, + "step": 3636 + }, + { + "epoch": 0.2733353374417556, + "flos": 16111666056960.0, + "grad_norm": 1.7585707999807463, + "learning_rate": 3.4105943281222804e-06, + "loss": 1.0374, + "num_input_tokens_seen": 8256975, + "step": 3637 + }, + { + "epoch": 0.27341049150759056, + "flos": 33618191685120.0, + "grad_norm": 1.6643884903650774, + "learning_rate": 3.4102491581709717e-06, + "loss": 0.9921, + "num_input_tokens_seen": 8282780, + "step": 3638 + }, + { + "epoch": 0.27348564557342553, + "flos": 20840765328000.0, + "grad_norm": 2.1709442339762925, + "learning_rate": 3.4099039046570006e-06, + "loss": 1.1161, + "num_input_tokens_seen": 8304390, + "step": 3639 + }, + { + "epoch": 0.2735607996392605, + "flos": 21512057811840.0, + "grad_norm": 1.7004839516846766, + "learning_rate": 3.4095585676008234e-06, + "loss": 0.8898, + "num_input_tokens_seen": 8327625, + "step": 3640 + }, + { + "epoch": 0.27363595370509547, + "flos": 64675127871360.0, + "grad_norm": 0.8120456305399971, + "learning_rate": 3.4092131470229045e-06, + "loss": 0.8339, + "num_input_tokens_seen": 8406865, + "step": 3641 + }, + { + "epoch": 0.2737111077709304, + "flos": 17135623929600.0, + "grad_norm": 1.4694940833056667, + "learning_rate": 3.40886764294371e-06, + "loss": 1.0281, + "num_input_tokens_seen": 8428850, + "step": 3642 + }, + { + "epoch": 0.27378626183676535, + "flos": 61501139372160.0, + "grad_norm": 0.7430251363863822, + "learning_rate": 3.4085220553837133e-06, + "loss": 0.8138, + "num_input_tokens_seen": 8497955, + "step": 3643 + }, + { + "epoch": 0.2738614159026003, + "flos": 21649072913280.0, + "grad_norm": 1.6182933472559378, + "learning_rate": 3.40817638436339e-06, + "loss": 0.9298, + "num_input_tokens_seen": 8519650, + "step": 3644 + }, + { + "epoch": 0.2739365699684353, + "flos": 28959539299200.0, + "grad_norm": 1.7276006831020057, + "learning_rate": 3.407830629903224e-06, + "loss": 1.0019, + "num_input_tokens_seen": 8543505, + "step": 3645 + }, + { + "epoch": 0.27401172403427027, + "flos": 23873572748160.0, + "grad_norm": 1.5024821883683466, + "learning_rate": 3.4074847920237032e-06, + "loss": 0.9707, + "num_input_tokens_seen": 8568230, + "step": 3646 + }, + { + "epoch": 0.27408687810010524, + "flos": 23837379638400.0, + "grad_norm": 1.7659736221607276, + "learning_rate": 3.407138870745318e-06, + "loss": 0.9842, + "num_input_tokens_seen": 8590195, + "step": 3647 + }, + { + "epoch": 0.27416203216594015, + "flos": 24581489304960.0, + "grad_norm": 1.8514151856613, + "learning_rate": 3.4067928660885665e-06, + "loss": 1.0063, + "num_input_tokens_seen": 8615425, + "step": 3648 + }, + { + "epoch": 0.2742371862317751, + "flos": 21547496736000.0, + "grad_norm": 1.3988656182525459, + "learning_rate": 3.406446778073951e-06, + "loss": 0.9658, + "num_input_tokens_seen": 8639855, + "step": 3649 + }, + { + "epoch": 0.2743123402976101, + "flos": 20841663168000.0, + "grad_norm": 1.3731548760644838, + "learning_rate": 3.4061006067219776e-06, + "loss": 0.9418, + "num_input_tokens_seen": 8663975, + "step": 3650 + }, + { + "epoch": 0.27438749436344506, + "flos": 11483316950400.0, + "grad_norm": 2.094768105830765, + "learning_rate": 3.40575435205316e-06, + "loss": 0.9711, + "num_input_tokens_seen": 8683840, + "step": 3651 + }, + { + "epoch": 0.27446264842928003, + "flos": 18053372770560.0, + "grad_norm": 1.3336847954696762, + "learning_rate": 3.405408014088013e-06, + "loss": 1.0216, + "num_input_tokens_seen": 8705595, + "step": 3652 + }, + { + "epoch": 0.274537802495115, + "flos": 22323023003520.0, + "grad_norm": 1.6494023541158596, + "learning_rate": 3.40506159284706e-06, + "loss": 1.0029, + "num_input_tokens_seen": 8730540, + "step": 3653 + }, + { + "epoch": 0.27461295656095, + "flos": 15723023040000.0, + "grad_norm": 1.5193781052325837, + "learning_rate": 3.4047150883508274e-06, + "loss": 0.9894, + "num_input_tokens_seen": 8751230, + "step": 3654 + }, + { + "epoch": 0.2746881106267849, + "flos": 36511793274240.0, + "grad_norm": 1.717490757968696, + "learning_rate": 3.4043685006198465e-06, + "loss": 0.9039, + "num_input_tokens_seen": 8774825, + "step": 3655 + }, + { + "epoch": 0.27476326469261986, + "flos": 30544845609600.0, + "grad_norm": 1.498670704249603, + "learning_rate": 3.4040218296746544e-06, + "loss": 0.9866, + "num_input_tokens_seen": 8798440, + "step": 3656 + }, + { + "epoch": 0.2748384187584548, + "flos": 17770795130880.0, + "grad_norm": 1.9368033233309963, + "learning_rate": 3.403675075535793e-06, + "loss": 0.9827, + "num_input_tokens_seen": 8819595, + "step": 3657 + }, + { + "epoch": 0.2749135728242898, + "flos": 13641104914560.0, + "grad_norm": 1.649705588922111, + "learning_rate": 3.403328238223808e-06, + "loss": 0.8788, + "num_input_tokens_seen": 8840000, + "step": 3658 + }, + { + "epoch": 0.27498872689012477, + "flos": 24758863493760.0, + "grad_norm": 1.5302739049038054, + "learning_rate": 3.4029813177592504e-06, + "loss": 0.9363, + "num_input_tokens_seen": 8862925, + "step": 3659 + }, + { + "epoch": 0.27506388095595974, + "flos": 14876511183360.0, + "grad_norm": 1.3241057221103987, + "learning_rate": 3.402634314162678e-06, + "loss": 0.938, + "num_input_tokens_seen": 8885545, + "step": 3660 + }, + { + "epoch": 0.27513903502179465, + "flos": 25110343733760.0, + "grad_norm": 1.5628760838898812, + "learning_rate": 3.4022872274546517e-06, + "loss": 0.9556, + "num_input_tokens_seen": 8908795, + "step": 3661 + }, + { + "epoch": 0.2752141890876296, + "flos": 27764815340160.0, + "grad_norm": 1.258426208521219, + "learning_rate": 3.4019400576557377e-06, + "loss": 1.0616, + "num_input_tokens_seen": 8936920, + "step": 3662 + }, + { + "epoch": 0.2752893431534646, + "flos": 20660984928000.0, + "grad_norm": 1.5894579970050056, + "learning_rate": 3.4015928047865056e-06, + "loss": 1.0352, + "num_input_tokens_seen": 8959675, + "step": 3663 + }, + { + "epoch": 0.27536449721929956, + "flos": 16499914024320.0, + "grad_norm": 1.3989993623618908, + "learning_rate": 3.401245468867534e-06, + "loss": 0.9683, + "num_input_tokens_seen": 8981825, + "step": 3664 + }, + { + "epoch": 0.27543965128513453, + "flos": 24548133369600.0, + "grad_norm": 1.3319057282403182, + "learning_rate": 3.4008980499194025e-06, + "loss": 1.0079, + "num_input_tokens_seen": 9007735, + "step": 3665 + }, + { + "epoch": 0.2755148053509695, + "flos": 25853196424320.0, + "grad_norm": 1.484810935806624, + "learning_rate": 3.4005505479626965e-06, + "loss": 0.9211, + "num_input_tokens_seen": 9031130, + "step": 3666 + }, + { + "epoch": 0.2755899594168045, + "flos": 24017662828800.0, + "grad_norm": 1.550739816261073, + "learning_rate": 3.4002029630180074e-06, + "loss": 0.9477, + "num_input_tokens_seen": 9055110, + "step": 3667 + }, + { + "epoch": 0.2756651134826394, + "flos": 29978002391040.0, + "grad_norm": 2.812767948526341, + "learning_rate": 3.399855295105932e-06, + "loss": 1.0243, + "num_input_tokens_seen": 9076160, + "step": 3668 + }, + { + "epoch": 0.27574026754847436, + "flos": 19747042928640.0, + "grad_norm": 1.4874949543330425, + "learning_rate": 3.3995075442470694e-06, + "loss": 0.9537, + "num_input_tokens_seen": 9099970, + "step": 3669 + }, + { + "epoch": 0.27581542161430933, + "flos": 21900808569600.0, + "grad_norm": 1.369550483076765, + "learning_rate": 3.3991597104620253e-06, + "loss": 0.9619, + "num_input_tokens_seen": 9125340, + "step": 3670 + }, + { + "epoch": 0.2758905756801443, + "flos": 23417463674880.0, + "grad_norm": 1.203515819139722, + "learning_rate": 3.3988117937714114e-06, + "loss": 1.0425, + "num_input_tokens_seen": 9149845, + "step": 3671 + }, + { + "epoch": 0.27596572974597927, + "flos": 26311352572800.0, + "grad_norm": 2.7874765027343096, + "learning_rate": 3.398463794195842e-06, + "loss": 0.9473, + "num_input_tokens_seen": 9174470, + "step": 3672 + }, + { + "epoch": 0.27604088381181424, + "flos": 55554087841920.0, + "grad_norm": 0.738875982124813, + "learning_rate": 3.3981157117559376e-06, + "loss": 0.8139, + "num_input_tokens_seen": 9254685, + "step": 3673 + }, + { + "epoch": 0.27611603787764916, + "flos": 20946866618880.0, + "grad_norm": 2.105487660622114, + "learning_rate": 3.397767546472323e-06, + "loss": 0.9426, + "num_input_tokens_seen": 9278005, + "step": 3674 + }, + { + "epoch": 0.2761911919434841, + "flos": 64888443774720.0, + "grad_norm": 0.8128514996727091, + "learning_rate": 3.39741929836563e-06, + "loss": 0.7591, + "num_input_tokens_seen": 9353825, + "step": 3675 + }, + { + "epoch": 0.2762663460093191, + "flos": 17346785016960.0, + "grad_norm": 1.7159449489183811, + "learning_rate": 3.3970709674564918e-06, + "loss": 1.0028, + "num_input_tokens_seen": 9375165, + "step": 3676 + }, + { + "epoch": 0.27634150007515407, + "flos": 15928222469760.0, + "grad_norm": 1.6863886387800617, + "learning_rate": 3.3967225537655492e-06, + "loss": 1.0085, + "num_input_tokens_seen": 9396050, + "step": 3677 + }, + { + "epoch": 0.27641665414098904, + "flos": 16215684359040.0, + "grad_norm": 1.9586951212446695, + "learning_rate": 3.396374057313447e-06, + "loss": 0.9531, + "num_input_tokens_seen": 9416405, + "step": 3678 + }, + { + "epoch": 0.276491808206824, + "flos": 21865082336640.0, + "grad_norm": 1.5213731130090398, + "learning_rate": 3.396025478120835e-06, + "loss": 0.9238, + "num_input_tokens_seen": 9438085, + "step": 3679 + }, + { + "epoch": 0.276566962272659, + "flos": 20205378645120.0, + "grad_norm": 1.4619106376776017, + "learning_rate": 3.395676816208367e-06, + "loss": 0.9572, + "num_input_tokens_seen": 9461800, + "step": 3680 + }, + { + "epoch": 0.2766421163384939, + "flos": 24759438111360.0, + "grad_norm": 1.4500058239786981, + "learning_rate": 3.3953280715967036e-06, + "loss": 0.9984, + "num_input_tokens_seen": 9486150, + "step": 3681 + }, + { + "epoch": 0.27671727040432886, + "flos": 19036253283840.0, + "grad_norm": 1.6607680685589923, + "learning_rate": 3.394979244306509e-06, + "loss": 0.8725, + "num_input_tokens_seen": 9507340, + "step": 3682 + }, + { + "epoch": 0.27679242447016383, + "flos": 20345482316160.0, + "grad_norm": 1.3599758793290684, + "learning_rate": 3.3946303343584523e-06, + "loss": 0.8872, + "num_input_tokens_seen": 9532395, + "step": 3683 + }, + { + "epoch": 0.2768675785359988, + "flos": 21476223838080.0, + "grad_norm": 1.5424560467657493, + "learning_rate": 3.3942813417732083e-06, + "loss": 0.9925, + "num_input_tokens_seen": 9556475, + "step": 3684 + }, + { + "epoch": 0.2769427326018338, + "flos": 13429189641600.0, + "grad_norm": 1.5284919812981357, + "learning_rate": 3.3939322665714548e-06, + "loss": 0.998, + "num_input_tokens_seen": 9578915, + "step": 3685 + }, + { + "epoch": 0.27701788666766874, + "flos": 20170873474560.0, + "grad_norm": 1.6553558239084305, + "learning_rate": 3.3935831087738774e-06, + "loss": 0.9567, + "num_input_tokens_seen": 9602395, + "step": 3686 + }, + { + "epoch": 0.27709304073350366, + "flos": 18825415418880.0, + "grad_norm": 1.3456192765599213, + "learning_rate": 3.3932338684011646e-06, + "loss": 0.9602, + "num_input_tokens_seen": 9625475, + "step": 3687 + }, + { + "epoch": 0.27716819479933863, + "flos": 17276194477440.0, + "grad_norm": 1.4687792186671476, + "learning_rate": 3.3928845454740097e-06, + "loss": 1.0459, + "num_input_tokens_seen": 9648385, + "step": 3688 + }, + { + "epoch": 0.2772433488651736, + "flos": 26097246570240.0, + "grad_norm": 1.5893640942786298, + "learning_rate": 3.3925351400131118e-06, + "loss": 1.0426, + "num_input_tokens_seen": 9672565, + "step": 3689 + }, + { + "epoch": 0.27731850293100857, + "flos": 24682419037440.0, + "grad_norm": 1.7527309634483876, + "learning_rate": 3.392185652039175e-06, + "loss": 0.9391, + "num_input_tokens_seen": 9695515, + "step": 3690 + }, + { + "epoch": 0.27739365699684354, + "flos": 29100289415040.0, + "grad_norm": 1.5501040269952324, + "learning_rate": 3.3918360815729066e-06, + "loss": 1.0278, + "num_input_tokens_seen": 9723375, + "step": 3691 + }, + { + "epoch": 0.2774688110626785, + "flos": 23483744582400.0, + "grad_norm": 1.5087426829693977, + "learning_rate": 3.391486428635021e-06, + "loss": 1.0029, + "num_input_tokens_seen": 9747715, + "step": 3692 + }, + { + "epoch": 0.2775439651285134, + "flos": 21012608822400.0, + "grad_norm": 1.6777948406367973, + "learning_rate": 3.391136693246236e-06, + "loss": 0.8745, + "num_input_tokens_seen": 9771425, + "step": 3693 + }, + { + "epoch": 0.2776191191943484, + "flos": 18864804839040.0, + "grad_norm": 2.3835241036881816, + "learning_rate": 3.390786875427275e-06, + "loss": 1.0394, + "num_input_tokens_seen": 9793630, + "step": 3694 + }, + { + "epoch": 0.27769427326018337, + "flos": 12582390476160.0, + "grad_norm": 1.564884578726405, + "learning_rate": 3.3904369751988657e-06, + "loss": 1.0044, + "num_input_tokens_seen": 9814540, + "step": 3695 + }, + { + "epoch": 0.27776942732601834, + "flos": 20945250506880.0, + "grad_norm": 1.4396715761881025, + "learning_rate": 3.3900869925817416e-06, + "loss": 0.9868, + "num_input_tokens_seen": 9837435, + "step": 3696 + }, + { + "epoch": 0.2778445813918533, + "flos": 34641467199360.0, + "grad_norm": 1.6505400640784373, + "learning_rate": 3.3897369275966404e-06, + "loss": 0.9096, + "num_input_tokens_seen": 9862100, + "step": 3697 + }, + { + "epoch": 0.2779197354576883, + "flos": 21158997373440.0, + "grad_norm": 1.4536815110897587, + "learning_rate": 3.389386780264304e-06, + "loss": 0.9564, + "num_input_tokens_seen": 9886135, + "step": 3698 + }, + { + "epoch": 0.27799488952352325, + "flos": 20801196339840.0, + "grad_norm": 1.6399092793377346, + "learning_rate": 3.389036550605481e-06, + "loss": 0.9962, + "num_input_tokens_seen": 9910655, + "step": 3699 + }, + { + "epoch": 0.27807004358935816, + "flos": 20912145966720.0, + "grad_norm": 1.5439338080443852, + "learning_rate": 3.3886862386409233e-06, + "loss": 0.9716, + "num_input_tokens_seen": 9932235, + "step": 3700 + }, + { + "epoch": 0.27814519765519313, + "flos": 18299829127680.0, + "grad_norm": 1.5452448414492879, + "learning_rate": 3.3883358443913883e-06, + "loss": 1.0278, + "num_input_tokens_seen": 9954925, + "step": 3701 + }, + { + "epoch": 0.2782203517210281, + "flos": 20767086218880.0, + "grad_norm": 1.3211255250760645, + "learning_rate": 3.387985367877638e-06, + "loss": 0.976, + "num_input_tokens_seen": 9977830, + "step": 3702 + }, + { + "epoch": 0.2782955057868631, + "flos": 18053588252160.0, + "grad_norm": 1.6016936457833086, + "learning_rate": 3.38763480912044e-06, + "loss": 0.9199, + "num_input_tokens_seen": 10000130, + "step": 3703 + }, + { + "epoch": 0.27837065985269804, + "flos": 22111466866560.0, + "grad_norm": 2.5247813328397384, + "learning_rate": 3.3872841681405654e-06, + "loss": 1.0161, + "num_input_tokens_seen": 10023560, + "step": 3704 + }, + { + "epoch": 0.278445813918533, + "flos": 17343265484160.0, + "grad_norm": 1.5826323697622307, + "learning_rate": 3.3869334449587925e-06, + "loss": 0.9809, + "num_input_tokens_seen": 10044075, + "step": 3705 + }, + { + "epoch": 0.27852096798436793, + "flos": 33688387175040.0, + "grad_norm": 1.690615920855105, + "learning_rate": 3.3865826395959018e-06, + "loss": 0.9162, + "num_input_tokens_seen": 10067185, + "step": 3706 + }, + { + "epoch": 0.2785961220502029, + "flos": 16709387172480.0, + "grad_norm": 1.554842303826165, + "learning_rate": 3.38623175207268e-06, + "loss": 1.0084, + "num_input_tokens_seen": 10089380, + "step": 3707 + }, + { + "epoch": 0.27867127611603787, + "flos": 16994011887360.0, + "grad_norm": 1.7405828090738833, + "learning_rate": 3.3858807824099182e-06, + "loss": 1.0297, + "num_input_tokens_seen": 10112270, + "step": 3708 + }, + { + "epoch": 0.27874643018187284, + "flos": 20418084017280.0, + "grad_norm": 1.5789184841642776, + "learning_rate": 3.385529730628414e-06, + "loss": 0.9351, + "num_input_tokens_seen": 10135725, + "step": 3709 + }, + { + "epoch": 0.2788215842477078, + "flos": 20417904449280.0, + "grad_norm": 1.8797543964864745, + "learning_rate": 3.385178596748967e-06, + "loss": 1.0811, + "num_input_tokens_seen": 10159000, + "step": 3710 + }, + { + "epoch": 0.2788967383135428, + "flos": 19957162521600.0, + "grad_norm": 2.396259424771125, + "learning_rate": 3.3848273807923836e-06, + "loss": 0.9867, + "num_input_tokens_seen": 10181865, + "step": 3711 + }, + { + "epoch": 0.27897189237937775, + "flos": 25253033184000.0, + "grad_norm": 1.5476486891398635, + "learning_rate": 3.384476082779476e-06, + "loss": 1.0019, + "num_input_tokens_seen": 10206645, + "step": 3712 + }, + { + "epoch": 0.27904704644521267, + "flos": 20417832622080.0, + "grad_norm": 1.6513981559642408, + "learning_rate": 3.3841247027310584e-06, + "loss": 1.0274, + "num_input_tokens_seen": 10229330, + "step": 3713 + }, + { + "epoch": 0.27912220051104764, + "flos": 19923483363840.0, + "grad_norm": 1.7184137188781157, + "learning_rate": 3.3837732406679524e-06, + "loss": 0.9771, + "num_input_tokens_seen": 10251315, + "step": 3714 + }, + { + "epoch": 0.2791973545768826, + "flos": 22112005570560.0, + "grad_norm": 1.4852459410384173, + "learning_rate": 3.3834216966109827e-06, + "loss": 0.9548, + "num_input_tokens_seen": 10274630, + "step": 3715 + }, + { + "epoch": 0.2792725086427176, + "flos": 63080555852160.0, + "grad_norm": 0.8057892673190482, + "learning_rate": 3.3830700705809802e-06, + "loss": 0.8101, + "num_input_tokens_seen": 10351255, + "step": 3716 + }, + { + "epoch": 0.27934766270855255, + "flos": 19888044439680.0, + "grad_norm": 1.7723106131372928, + "learning_rate": 3.38271836259878e-06, + "loss": 1.0389, + "num_input_tokens_seen": 10371990, + "step": 3717 + }, + { + "epoch": 0.2794228167743875, + "flos": 20628706400640.0, + "grad_norm": 1.381479385799284, + "learning_rate": 3.382366572685222e-06, + "loss": 0.9211, + "num_input_tokens_seen": 10396765, + "step": 3718 + }, + { + "epoch": 0.27949797084022243, + "flos": 22604020444800.0, + "grad_norm": 1.5813267775882693, + "learning_rate": 3.3820147008611512e-06, + "loss": 1.0123, + "num_input_tokens_seen": 10421320, + "step": 3719 + }, + { + "epoch": 0.2795731249060574, + "flos": 20770174788480.0, + "grad_norm": 1.4169854572117906, + "learning_rate": 3.3816627471474166e-06, + "loss": 0.9672, + "num_input_tokens_seen": 10445520, + "step": 3720 + }, + { + "epoch": 0.2796482789718924, + "flos": 24194354659200.0, + "grad_norm": 1.986586052306568, + "learning_rate": 3.381310711564874e-06, + "loss": 0.8889, + "num_input_tokens_seen": 10466005, + "step": 3721 + }, + { + "epoch": 0.27972343303772734, + "flos": 22102703948160.0, + "grad_norm": 1.5851147401355972, + "learning_rate": 3.380958594134382e-06, + "loss": 1.0561, + "num_input_tokens_seen": 10485270, + "step": 3722 + }, + { + "epoch": 0.2797985871035623, + "flos": 36583461221760.0, + "grad_norm": 1.5411023404437856, + "learning_rate": 3.380606394876806e-06, + "loss": 0.9312, + "num_input_tokens_seen": 10511760, + "step": 3723 + }, + { + "epoch": 0.2798737411693973, + "flos": 19817848949760.0, + "grad_norm": 1.357719509509376, + "learning_rate": 3.380254113813014e-06, + "loss": 1.012, + "num_input_tokens_seen": 10535010, + "step": 3724 + }, + { + "epoch": 0.27994889523523225, + "flos": 22041630512640.0, + "grad_norm": 1.556174329273819, + "learning_rate": 3.3799017509638805e-06, + "loss": 1.043, + "num_input_tokens_seen": 10558330, + "step": 3725 + }, + { + "epoch": 0.28002404930106717, + "flos": 35131650480000.0, + "grad_norm": 1.7474321190088682, + "learning_rate": 3.3795493063502836e-06, + "loss": 0.8352, + "num_input_tokens_seen": 10584470, + "step": 3726 + }, + { + "epoch": 0.28009920336690214, + "flos": 17311274265600.0, + "grad_norm": 1.7382016967399971, + "learning_rate": 3.3791967799931085e-06, + "loss": 0.9701, + "num_input_tokens_seen": 10605010, + "step": 3727 + }, + { + "epoch": 0.2801743574327371, + "flos": 40254061536000.0, + "grad_norm": 1.404479331948514, + "learning_rate": 3.3788441719132425e-06, + "loss": 0.9248, + "num_input_tokens_seen": 10633440, + "step": 3728 + }, + { + "epoch": 0.2802495114985721, + "flos": 19781763580800.0, + "grad_norm": 1.6323447401632185, + "learning_rate": 3.37849148213158e-06, + "loss": 0.9779, + "num_input_tokens_seen": 10654990, + "step": 3729 + }, + { + "epoch": 0.28032466556440705, + "flos": 21864651373440.0, + "grad_norm": 1.4820729926971792, + "learning_rate": 3.3781387106690175e-06, + "loss": 0.9804, + "num_input_tokens_seen": 10678395, + "step": 3730 + }, + { + "epoch": 0.280399819630242, + "flos": 20982305543040.0, + "grad_norm": 2.0598936084939456, + "learning_rate": 3.37778585754646e-06, + "loss": 0.9874, + "num_input_tokens_seen": 10701515, + "step": 3731 + }, + { + "epoch": 0.28047497369607693, + "flos": 18582263112960.0, + "grad_norm": 1.8368932955032913, + "learning_rate": 3.3774329227848144e-06, + "loss": 0.8517, + "num_input_tokens_seen": 10723050, + "step": 3732 + }, + { + "epoch": 0.2805501277619119, + "flos": 24406557240960.0, + "grad_norm": 1.4110474874182186, + "learning_rate": 3.3770799064049927e-06, + "loss": 1.0598, + "num_input_tokens_seen": 10746845, + "step": 3733 + }, + { + "epoch": 0.2806252818277469, + "flos": 26309305497600.0, + "grad_norm": 1.3622624216393722, + "learning_rate": 3.3767268084279143e-06, + "loss": 0.9702, + "num_input_tokens_seen": 10772030, + "step": 3734 + }, + { + "epoch": 0.28070043589358185, + "flos": 28706187530880.0, + "grad_norm": 1.5850680368544183, + "learning_rate": 3.376373628874501e-06, + "loss": 0.8873, + "num_input_tokens_seen": 10797960, + "step": 3735 + }, + { + "epoch": 0.2807755899594168, + "flos": 19111728072960.0, + "grad_norm": 1.6256231433662744, + "learning_rate": 3.3760203677656786e-06, + "loss": 0.9485, + "num_input_tokens_seen": 10821165, + "step": 3736 + }, + { + "epoch": 0.2808507440252518, + "flos": 16358553377280.0, + "grad_norm": 1.7347623375369943, + "learning_rate": 3.3756670251223813e-06, + "loss": 0.9436, + "num_input_tokens_seen": 10842950, + "step": 3737 + }, + { + "epoch": 0.2809258980910867, + "flos": 19355095860480.0, + "grad_norm": 1.625071890781303, + "learning_rate": 3.375313600965544e-06, + "loss": 0.9776, + "num_input_tokens_seen": 10864700, + "step": 3738 + }, + { + "epoch": 0.28100105215692167, + "flos": 24547558752000.0, + "grad_norm": 1.6684609609305858, + "learning_rate": 3.3749600953161102e-06, + "loss": 1.0117, + "num_input_tokens_seen": 10888445, + "step": 3739 + }, + { + "epoch": 0.28107620622275664, + "flos": 21124204894080.0, + "grad_norm": 1.4554894192467305, + "learning_rate": 3.3746065081950253e-06, + "loss": 1.0365, + "num_input_tokens_seen": 10911265, + "step": 3740 + }, + { + "epoch": 0.2811513602885916, + "flos": 20312126380800.0, + "grad_norm": 2.4310033246499745, + "learning_rate": 3.374252839623241e-06, + "loss": 0.9915, + "num_input_tokens_seen": 10933230, + "step": 3741 + }, + { + "epoch": 0.2812265143544266, + "flos": 23595843444480.0, + "grad_norm": 1.3215685856774422, + "learning_rate": 3.373899089621714e-06, + "loss": 1.0327, + "num_input_tokens_seen": 10959585, + "step": 3742 + }, + { + "epoch": 0.28130166842026155, + "flos": 24018093792000.0, + "grad_norm": 1.4909583816553187, + "learning_rate": 3.3735452582114046e-06, + "loss": 0.8981, + "num_input_tokens_seen": 10985870, + "step": 3743 + }, + { + "epoch": 0.2813768224860965, + "flos": 64060176026880.0, + "grad_norm": 0.7982541286171779, + "learning_rate": 3.373191345413279e-06, + "loss": 0.7961, + "num_input_tokens_seen": 11063810, + "step": 3744 + }, + { + "epoch": 0.28145197655193144, + "flos": 24438332977920.0, + "grad_norm": 1.4554439622423747, + "learning_rate": 3.3728373512483083e-06, + "loss": 1.0242, + "num_input_tokens_seen": 11088535, + "step": 3745 + }, + { + "epoch": 0.2815271306177664, + "flos": 28927260771840.0, + "grad_norm": 2.7511875016195066, + "learning_rate": 3.3724832757374674e-06, + "loss": 0.9781, + "num_input_tokens_seen": 11117020, + "step": 3746 + }, + { + "epoch": 0.2816022846836014, + "flos": 29486238912000.0, + "grad_norm": 1.7580399429678941, + "learning_rate": 3.3721291189017363e-06, + "loss": 1.0469, + "num_input_tokens_seen": 11141705, + "step": 3747 + }, + { + "epoch": 0.28167743874943635, + "flos": 20837999980800.0, + "grad_norm": 1.3925963583845336, + "learning_rate": 3.371774880762101e-06, + "loss": 1.0313, + "num_input_tokens_seen": 11164965, + "step": 3748 + }, + { + "epoch": 0.2817525928152713, + "flos": 21828996967680.0, + "grad_norm": 1.4397174483824102, + "learning_rate": 3.3714205613395513e-06, + "loss": 1.0209, + "num_input_tokens_seen": 11190080, + "step": 3749 + }, + { + "epoch": 0.2818277468811063, + "flos": 14417169886080.0, + "grad_norm": 1.2581031889903325, + "learning_rate": 3.371066160655082e-06, + "loss": 0.9488, + "num_input_tokens_seen": 11211900, + "step": 3750 + }, + { + "epoch": 0.2819029009469412, + "flos": 21652520618880.0, + "grad_norm": 1.3818331337620628, + "learning_rate": 3.3707116787296918e-06, + "loss": 0.9596, + "num_input_tokens_seen": 11237215, + "step": 3751 + }, + { + "epoch": 0.2819780550127762, + "flos": 16005349284480.0, + "grad_norm": 1.8270125305422573, + "learning_rate": 3.3703571155843866e-06, + "loss": 1.0691, + "num_input_tokens_seen": 11257930, + "step": 3752 + }, + { + "epoch": 0.28205320907861114, + "flos": 36088645086720.0, + "grad_norm": 1.4536510868813595, + "learning_rate": 3.370002471240174e-06, + "loss": 0.904, + "num_input_tokens_seen": 11284255, + "step": 3753 + }, + { + "epoch": 0.2821283631444461, + "flos": 25393855127040.0, + "grad_norm": 1.4047409862141542, + "learning_rate": 3.36964774571807e-06, + "loss": 0.97, + "num_input_tokens_seen": 11308515, + "step": 3754 + }, + { + "epoch": 0.2822035172102811, + "flos": 34777368979200.0, + "grad_norm": 1.4118831975731836, + "learning_rate": 3.3692929390390914e-06, + "loss": 0.946, + "num_input_tokens_seen": 11333105, + "step": 3755 + }, + { + "epoch": 0.28227867127611606, + "flos": 19464537116160.0, + "grad_norm": 1.766831382534242, + "learning_rate": 3.3689380512242627e-06, + "loss": 0.9604, + "num_input_tokens_seen": 11354650, + "step": 3756 + }, + { + "epoch": 0.282353825341951, + "flos": 22675436997120.0, + "grad_norm": 1.5560470080985194, + "learning_rate": 3.3685830822946134e-06, + "loss": 0.8252, + "num_input_tokens_seen": 11379575, + "step": 3757 + }, + { + "epoch": 0.28242897940778594, + "flos": 20802776538240.0, + "grad_norm": 1.3662207100704806, + "learning_rate": 3.3682280322711753e-06, + "loss": 1.0447, + "num_input_tokens_seen": 11402850, + "step": 3758 + }, + { + "epoch": 0.2825041334736209, + "flos": 23171617848960.0, + "grad_norm": 1.6075142779391032, + "learning_rate": 3.367872901174987e-06, + "loss": 0.9832, + "num_input_tokens_seen": 11426885, + "step": 3759 + }, + { + "epoch": 0.2825792875394559, + "flos": 19535486791680.0, + "grad_norm": 1.5275254397607299, + "learning_rate": 3.367517689027091e-06, + "loss": 0.8913, + "num_input_tokens_seen": 11450025, + "step": 3760 + }, + { + "epoch": 0.28265444160529085, + "flos": 21618230929920.0, + "grad_norm": 1.4373993216378567, + "learning_rate": 3.3671623958485354e-06, + "loss": 0.9653, + "num_input_tokens_seen": 11472380, + "step": 3761 + }, + { + "epoch": 0.2827295956711258, + "flos": 15858673424640.0, + "grad_norm": 2.9705401996023575, + "learning_rate": 3.3668070216603736e-06, + "loss": 1.0584, + "num_input_tokens_seen": 11492510, + "step": 3762 + }, + { + "epoch": 0.2828047497369608, + "flos": 15087815925120.0, + "grad_norm": 1.6084105184873776, + "learning_rate": 3.366451566483661e-06, + "loss": 0.9015, + "num_input_tokens_seen": 11515615, + "step": 3763 + }, + { + "epoch": 0.2828799038027957, + "flos": 20700338434560.0, + "grad_norm": 1.3207173827066563, + "learning_rate": 3.366096030339461e-06, + "loss": 1.0075, + "num_input_tokens_seen": 11538565, + "step": 3764 + }, + { + "epoch": 0.2829550578686307, + "flos": 16641238757760.0, + "grad_norm": 2.4266333319746396, + "learning_rate": 3.3657404132488403e-06, + "loss": 0.9784, + "num_input_tokens_seen": 11559830, + "step": 3765 + }, + { + "epoch": 0.28303021193446565, + "flos": 14593682148480.0, + "grad_norm": 1.5731798250801414, + "learning_rate": 3.3653847152328694e-06, + "loss": 1.0383, + "num_input_tokens_seen": 11581045, + "step": 3766 + }, + { + "epoch": 0.2831053660003006, + "flos": 23025121557120.0, + "grad_norm": 1.3881584752910159, + "learning_rate": 3.3650289363126266e-06, + "loss": 1.012, + "num_input_tokens_seen": 11604385, + "step": 3767 + }, + { + "epoch": 0.2831805200661356, + "flos": 23382132491520.0, + "grad_norm": 1.5257826532141605, + "learning_rate": 3.3646730765091916e-06, + "loss": 0.9558, + "num_input_tokens_seen": 11627335, + "step": 3768 + }, + { + "epoch": 0.28325567413197056, + "flos": 66477592840320.0, + "grad_norm": 0.7620763868331992, + "learning_rate": 3.3643171358436513e-06, + "loss": 0.8207, + "num_input_tokens_seen": 11708380, + "step": 3769 + }, + { + "epoch": 0.28333082819780553, + "flos": 34075234512000.0, + "grad_norm": 1.300386752772847, + "learning_rate": 3.3639611143370967e-06, + "loss": 0.8721, + "num_input_tokens_seen": 11736875, + "step": 3770 + }, + { + "epoch": 0.28340598226364044, + "flos": 17416549543680.0, + "grad_norm": 1.3194669465048001, + "learning_rate": 3.3636050120106233e-06, + "loss": 1.0347, + "num_input_tokens_seen": 11760380, + "step": 3771 + }, + { + "epoch": 0.2834811363294754, + "flos": 28323901221120.0, + "grad_norm": 1.3885926815581286, + "learning_rate": 3.363248828885331e-06, + "loss": 0.9461, + "num_input_tokens_seen": 11789040, + "step": 3772 + }, + { + "epoch": 0.2835562903953104, + "flos": 13816862991360.0, + "grad_norm": 1.9976843302710878, + "learning_rate": 3.362892564982325e-06, + "loss": 0.9051, + "num_input_tokens_seen": 11809300, + "step": 3773 + }, + { + "epoch": 0.28363144446114535, + "flos": 20629676067840.0, + "grad_norm": 1.681576823974807, + "learning_rate": 3.3625362203227167e-06, + "loss": 0.9758, + "num_input_tokens_seen": 11832580, + "step": 3774 + }, + { + "epoch": 0.2837065985269803, + "flos": 21753809487360.0, + "grad_norm": 1.4209153846175782, + "learning_rate": 3.3621797949276188e-06, + "loss": 0.9729, + "num_input_tokens_seen": 11856505, + "step": 3775 + }, + { + "epoch": 0.2837817525928153, + "flos": 26276667834240.0, + "grad_norm": 1.4584499130881445, + "learning_rate": 3.3618232888181524e-06, + "loss": 0.9688, + "num_input_tokens_seen": 11880925, + "step": 3776 + }, + { + "epoch": 0.2838569066586502, + "flos": 21294216794880.0, + "grad_norm": 1.7573176386763791, + "learning_rate": 3.3614667020154415e-06, + "loss": 1.0766, + "num_input_tokens_seen": 11902745, + "step": 3777 + }, + { + "epoch": 0.2839320607244852, + "flos": 23524175496960.0, + "grad_norm": 1.6202525702551616, + "learning_rate": 3.3611100345406146e-06, + "loss": 0.9079, + "num_input_tokens_seen": 11923805, + "step": 3778 + }, + { + "epoch": 0.28400721479032015, + "flos": 25815638597760.0, + "grad_norm": 1.7157429833483386, + "learning_rate": 3.3607532864148063e-06, + "loss": 0.9766, + "num_input_tokens_seen": 11945530, + "step": 3779 + }, + { + "epoch": 0.2840823688561551, + "flos": 22818018706560.0, + "grad_norm": 1.6301656926665078, + "learning_rate": 3.3603964576591553e-06, + "loss": 0.8698, + "num_input_tokens_seen": 11971440, + "step": 3780 + }, + { + "epoch": 0.2841575229219901, + "flos": 19676596043520.0, + "grad_norm": 1.620432302815271, + "learning_rate": 3.360039548294805e-06, + "loss": 1.0501, + "num_input_tokens_seen": 11994825, + "step": 3781 + }, + { + "epoch": 0.28423267698782506, + "flos": 19673615214720.0, + "grad_norm": 1.5297756493081547, + "learning_rate": 3.3596825583429033e-06, + "loss": 0.918, + "num_input_tokens_seen": 12020670, + "step": 3782 + }, + { + "epoch": 0.28430783105366, + "flos": 12334605315840.0, + "grad_norm": 1.7225087074535843, + "learning_rate": 3.3593254878246035e-06, + "loss": 0.9617, + "num_input_tokens_seen": 12041630, + "step": 3783 + }, + { + "epoch": 0.28438298511949495, + "flos": 13282406040960.0, + "grad_norm": 1.7536691620100857, + "learning_rate": 3.358968336761063e-06, + "loss": 0.9642, + "num_input_tokens_seen": 12063655, + "step": 3784 + }, + { + "epoch": 0.2844581391853299, + "flos": 23453010339840.0, + "grad_norm": 1.5245029733890951, + "learning_rate": 3.3586111051734455e-06, + "loss": 0.9687, + "num_input_tokens_seen": 12087720, + "step": 3785 + }, + { + "epoch": 0.2845332932511649, + "flos": 27157541207040.0, + "grad_norm": 1.5357746305681073, + "learning_rate": 3.358253793082917e-06, + "loss": 0.9619, + "num_input_tokens_seen": 12114320, + "step": 3786 + }, + { + "epoch": 0.28460844731699986, + "flos": 21229156949760.0, + "grad_norm": 2.2648763537815264, + "learning_rate": 3.3578964005106496e-06, + "loss": 0.9157, + "num_input_tokens_seen": 12138490, + "step": 3787 + }, + { + "epoch": 0.2846836013828348, + "flos": 16217767347840.0, + "grad_norm": 1.4994590101602094, + "learning_rate": 3.3575389274778214e-06, + "loss": 1.0342, + "num_input_tokens_seen": 12160510, + "step": 3788 + }, + { + "epoch": 0.2847587554486698, + "flos": 21153897642240.0, + "grad_norm": 1.6172816299833999, + "learning_rate": 3.3571813740056135e-06, + "loss": 1.0447, + "num_input_tokens_seen": 12182130, + "step": 3789 + }, + { + "epoch": 0.2848339095145047, + "flos": 27267844389120.0, + "grad_norm": 1.387760082990974, + "learning_rate": 3.356823740115212e-06, + "loss": 0.9508, + "num_input_tokens_seen": 12210865, + "step": 3790 + }, + { + "epoch": 0.2849090635803397, + "flos": 16567272339840.0, + "grad_norm": 1.6069069613553837, + "learning_rate": 3.3564660258278085e-06, + "loss": 0.8978, + "num_input_tokens_seen": 12231670, + "step": 3791 + }, + { + "epoch": 0.28498421764617465, + "flos": 19534624865280.0, + "grad_norm": 1.443330963226792, + "learning_rate": 3.3561082311645982e-06, + "loss": 1.0334, + "num_input_tokens_seen": 12254445, + "step": 3792 + }, + { + "epoch": 0.2850593717120096, + "flos": 19005770436480.0, + "grad_norm": 1.6566663974375413, + "learning_rate": 3.3557503561467832e-06, + "loss": 0.9873, + "num_input_tokens_seen": 12276050, + "step": 3793 + }, + { + "epoch": 0.2851345257778446, + "flos": 43924230887040.0, + "grad_norm": 1.6642742469995966, + "learning_rate": 3.3553924007955673e-06, + "loss": 0.8617, + "num_input_tokens_seen": 12303905, + "step": 3794 + }, + { + "epoch": 0.28520967984367956, + "flos": 23559039803520.0, + "grad_norm": 1.6329461001992704, + "learning_rate": 3.355034365132162e-06, + "loss": 0.8576, + "num_input_tokens_seen": 12328365, + "step": 3795 + }, + { + "epoch": 0.2852848339095145, + "flos": 24264837457920.0, + "grad_norm": 1.4680151216939057, + "learning_rate": 3.354676249177781e-06, + "loss": 1.0369, + "num_input_tokens_seen": 12351900, + "step": 3796 + }, + { + "epoch": 0.28535998797534945, + "flos": 18652781825280.0, + "grad_norm": 1.7745062342770963, + "learning_rate": 3.354318052953646e-06, + "loss": 0.9638, + "num_input_tokens_seen": 12373810, + "step": 3797 + }, + { + "epoch": 0.2854351420411844, + "flos": 22532855287680.0, + "grad_norm": 1.5628786772576624, + "learning_rate": 3.3539597764809794e-06, + "loss": 0.8793, + "num_input_tokens_seen": 12396000, + "step": 3798 + }, + { + "epoch": 0.2855102961070194, + "flos": 13817904485760.0, + "grad_norm": 1.8221154841056597, + "learning_rate": 3.3536014197810115e-06, + "loss": 0.9641, + "num_input_tokens_seen": 12417275, + "step": 3799 + }, + { + "epoch": 0.28558545017285436, + "flos": 19960861622400.0, + "grad_norm": 1.524611895845261, + "learning_rate": 3.3532429828749768e-06, + "loss": 0.9007, + "num_input_tokens_seen": 12441130, + "step": 3800 + }, + { + "epoch": 0.28566060423868933, + "flos": 27476096474880.0, + "grad_norm": 1.3011123190093377, + "learning_rate": 3.3528844657841128e-06, + "loss": 1.0083, + "num_input_tokens_seen": 12469600, + "step": 3801 + }, + { + "epoch": 0.2857357583045243, + "flos": 22177460465280.0, + "grad_norm": 1.9092815245178256, + "learning_rate": 3.352525868529664e-06, + "loss": 0.9738, + "num_input_tokens_seen": 12492330, + "step": 3802 + }, + { + "epoch": 0.2858109123703592, + "flos": 24334135107840.0, + "grad_norm": 1.3924938888004785, + "learning_rate": 3.352167191132878e-06, + "loss": 0.999, + "num_input_tokens_seen": 12516420, + "step": 3803 + }, + { + "epoch": 0.2858860664361942, + "flos": 19322960987520.0, + "grad_norm": 1.7441205987923294, + "learning_rate": 3.3518084336150084e-06, + "loss": 1.0339, + "num_input_tokens_seen": 12536810, + "step": 3804 + }, + { + "epoch": 0.28596122050202916, + "flos": 21864507719040.0, + "grad_norm": 1.3343348119611422, + "learning_rate": 3.3514495959973125e-06, + "loss": 0.9356, + "num_input_tokens_seen": 12562155, + "step": 3805 + }, + { + "epoch": 0.2860363745678641, + "flos": 18018005673600.0, + "grad_norm": 1.659833522170094, + "learning_rate": 3.3510906783010536e-06, + "loss": 0.9645, + "num_input_tokens_seen": 12585195, + "step": 3806 + }, + { + "epoch": 0.2861115286336991, + "flos": 21794599537920.0, + "grad_norm": 1.6914972364519791, + "learning_rate": 3.3507316805474976e-06, + "loss": 0.8609, + "num_input_tokens_seen": 12608870, + "step": 3807 + }, + { + "epoch": 0.28618668269953407, + "flos": 18088237077120.0, + "grad_norm": 1.2692414407792771, + "learning_rate": 3.3503726027579175e-06, + "loss": 0.9656, + "num_input_tokens_seen": 12632800, + "step": 3808 + }, + { + "epoch": 0.286261836765369, + "flos": 26766276497280.0, + "grad_norm": 1.2827084312089903, + "learning_rate": 3.3500134449535894e-06, + "loss": 0.9536, + "num_input_tokens_seen": 12657370, + "step": 3809 + }, + { + "epoch": 0.28633699083120395, + "flos": 20312126380800.0, + "grad_norm": 1.6809249223308398, + "learning_rate": 3.3496542071557955e-06, + "loss": 0.8563, + "num_input_tokens_seen": 12681095, + "step": 3810 + }, + { + "epoch": 0.2864121448970389, + "flos": 24053640456960.0, + "grad_norm": 1.3175569932398297, + "learning_rate": 3.3492948893858217e-06, + "loss": 0.9313, + "num_input_tokens_seen": 12706160, + "step": 3811 + }, + { + "epoch": 0.2864872989628739, + "flos": 13888064062080.0, + "grad_norm": 1.877601408372212, + "learning_rate": 3.3489354916649593e-06, + "loss": 0.8951, + "num_input_tokens_seen": 12725245, + "step": 3812 + }, + { + "epoch": 0.28656245302870886, + "flos": 18759170424960.0, + "grad_norm": 1.7713543592896133, + "learning_rate": 3.348576014014503e-06, + "loss": 0.9931, + "num_input_tokens_seen": 12749435, + "step": 3813 + }, + { + "epoch": 0.28663760709454383, + "flos": 21368829657600.0, + "grad_norm": 1.9073621288585023, + "learning_rate": 3.3482164564557537e-06, + "loss": 0.9579, + "num_input_tokens_seen": 12773040, + "step": 3814 + }, + { + "epoch": 0.2867127611603788, + "flos": 16252667568000.0, + "grad_norm": 1.7397555496339996, + "learning_rate": 3.3478568190100173e-06, + "loss": 0.9843, + "num_input_tokens_seen": 12795100, + "step": 3815 + }, + { + "epoch": 0.2867879152262137, + "flos": 23099842160640.0, + "grad_norm": 1.6493934643612815, + "learning_rate": 3.3474971016986024e-06, + "loss": 0.8791, + "num_input_tokens_seen": 12818585, + "step": 3816 + }, + { + "epoch": 0.2868630692920487, + "flos": 16887587374080.0, + "grad_norm": 1.4212952993617023, + "learning_rate": 3.3471373045428248e-06, + "loss": 1.0446, + "num_input_tokens_seen": 12841700, + "step": 3817 + }, + { + "epoch": 0.28693822335788366, + "flos": 49107751292160.0, + "grad_norm": 1.5011228491356134, + "learning_rate": 3.346777427564003e-06, + "loss": 0.8849, + "num_input_tokens_seen": 12870170, + "step": 3818 + }, + { + "epoch": 0.28701337742371863, + "flos": 23947395511680.0, + "grad_norm": 1.4435427991800505, + "learning_rate": 3.3464174707834618e-06, + "loss": 1.0329, + "num_input_tokens_seen": 12894280, + "step": 3819 + }, + { + "epoch": 0.2870885314895536, + "flos": 25288400280960.0, + "grad_norm": 1.5746900737943546, + "learning_rate": 3.34605743422253e-06, + "loss": 0.9549, + "num_input_tokens_seen": 12918200, + "step": 3820 + }, + { + "epoch": 0.28716368555538857, + "flos": 14593933543680.0, + "grad_norm": 1.4985470887843588, + "learning_rate": 3.34569731790254e-06, + "loss": 0.8841, + "num_input_tokens_seen": 12940260, + "step": 3821 + }, + { + "epoch": 0.2872388396212235, + "flos": 25994736639360.0, + "grad_norm": 1.5678285909598355, + "learning_rate": 3.3453371218448318e-06, + "loss": 1.0236, + "num_input_tokens_seen": 12964370, + "step": 3822 + }, + { + "epoch": 0.28731399368705846, + "flos": 15120345847680.0, + "grad_norm": 1.8518690668294915, + "learning_rate": 3.3449768460707465e-06, + "loss": 0.9563, + "num_input_tokens_seen": 12987010, + "step": 3823 + }, + { + "epoch": 0.2873891477528934, + "flos": 57403584028800.0, + "grad_norm": 0.8764789888308439, + "learning_rate": 3.344616490601633e-06, + "loss": 0.789, + "num_input_tokens_seen": 13057370, + "step": 3824 + }, + { + "epoch": 0.2874643018187284, + "flos": 22218573738240.0, + "grad_norm": 1.5569010556191152, + "learning_rate": 3.3442560554588444e-06, + "loss": 0.9789, + "num_input_tokens_seen": 13081230, + "step": 3825 + }, + { + "epoch": 0.28753945588456337, + "flos": 24900367795200.0, + "grad_norm": 1.8077718296603151, + "learning_rate": 3.3438955406637365e-06, + "loss": 0.9724, + "num_input_tokens_seen": 13105945, + "step": 3826 + }, + { + "epoch": 0.28761460995039834, + "flos": 19958706806400.0, + "grad_norm": 1.5022154183409238, + "learning_rate": 3.3435349462376713e-06, + "loss": 0.9284, + "num_input_tokens_seen": 13130120, + "step": 3827 + }, + { + "epoch": 0.28768976401623325, + "flos": 21474859121280.0, + "grad_norm": 1.2654729681183752, + "learning_rate": 3.343174272202017e-06, + "loss": 0.941, + "num_input_tokens_seen": 13154885, + "step": 3828 + }, + { + "epoch": 0.2877649180820682, + "flos": 14170067084160.0, + "grad_norm": 1.4019426287504997, + "learning_rate": 3.3428135185781425e-06, + "loss": 0.9264, + "num_input_tokens_seen": 13176275, + "step": 3829 + }, + { + "epoch": 0.2878400721479032, + "flos": 21158925546240.0, + "grad_norm": 1.905205049367681, + "learning_rate": 3.3424526853874252e-06, + "loss": 1.0053, + "num_input_tokens_seen": 13198740, + "step": 3830 + }, + { + "epoch": 0.28791522621373816, + "flos": 30086833115520.0, + "grad_norm": 1.5339889942057872, + "learning_rate": 3.342091772651246e-06, + "loss": 0.8356, + "num_input_tokens_seen": 13225335, + "step": 3831 + }, + { + "epoch": 0.28799038027957313, + "flos": 20064951751680.0, + "grad_norm": 1.564093959312854, + "learning_rate": 3.34173078039099e-06, + "loss": 0.9852, + "num_input_tokens_seen": 13249775, + "step": 3832 + }, + { + "epoch": 0.2880655343454081, + "flos": 30935643442560.0, + "grad_norm": 1.6167210445238456, + "learning_rate": 3.341369708628047e-06, + "loss": 1.1029, + "num_input_tokens_seen": 13275365, + "step": 3833 + }, + { + "epoch": 0.2881406884112431, + "flos": 64539010609920.0, + "grad_norm": 0.8409170699854026, + "learning_rate": 3.341008557383813e-06, + "loss": 0.7733, + "num_input_tokens_seen": 13353290, + "step": 3834 + }, + { + "epoch": 0.288215842477078, + "flos": 60820868488320.0, + "grad_norm": 0.8210345042710899, + "learning_rate": 3.3406473266796865e-06, + "loss": 0.7843, + "num_input_tokens_seen": 13429520, + "step": 3835 + }, + { + "epoch": 0.28829099654291296, + "flos": 19711855399680.0, + "grad_norm": 1.6744420673854408, + "learning_rate": 3.3402860165370724e-06, + "loss": 0.9777, + "num_input_tokens_seen": 13450870, + "step": 3836 + }, + { + "epoch": 0.28836615060874793, + "flos": 19005447214080.0, + "grad_norm": 1.537597113542894, + "learning_rate": 3.3399246269773796e-06, + "loss": 1.0083, + "num_input_tokens_seen": 13473065, + "step": 3837 + }, + { + "epoch": 0.2884413046745829, + "flos": 25218132963840.0, + "grad_norm": 1.5770754203782644, + "learning_rate": 3.3395631580220213e-06, + "loss": 0.9283, + "num_input_tokens_seen": 13497960, + "step": 3838 + }, + { + "epoch": 0.28851645874041787, + "flos": 26626280567040.0, + "grad_norm": 1.7287893304881294, + "learning_rate": 3.3392016096924168e-06, + "loss": 0.9701, + "num_input_tokens_seen": 13521825, + "step": 3839 + }, + { + "epoch": 0.28859161280625284, + "flos": 21440641259520.0, + "grad_norm": 1.750487804545786, + "learning_rate": 3.3388399820099887e-06, + "loss": 0.9743, + "num_input_tokens_seen": 13544685, + "step": 3840 + }, + { + "epoch": 0.28866676687208775, + "flos": 24615886734720.0, + "grad_norm": 1.3285929584736347, + "learning_rate": 3.3384782749961646e-06, + "loss": 1.0016, + "num_input_tokens_seen": 13569620, + "step": 3841 + }, + { + "epoch": 0.2887419209379227, + "flos": 28994583173760.0, + "grad_norm": 1.5306242617959314, + "learning_rate": 3.3381164886723777e-06, + "loss": 0.9462, + "num_input_tokens_seen": 13594015, + "step": 3842 + }, + { + "epoch": 0.2888170750037577, + "flos": 23206589896320.0, + "grad_norm": 1.8767889389647117, + "learning_rate": 3.337754623060065e-06, + "loss": 0.9842, + "num_input_tokens_seen": 13615420, + "step": 3843 + }, + { + "epoch": 0.28889222906959267, + "flos": 12969704689920.0, + "grad_norm": 1.870203176568289, + "learning_rate": 3.337392678180668e-06, + "loss": 1.0277, + "num_input_tokens_seen": 13635350, + "step": 3844 + }, + { + "epoch": 0.28896738313542764, + "flos": 24512766272640.0, + "grad_norm": 1.4858293335511863, + "learning_rate": 3.3370306540556336e-06, + "loss": 1.0366, + "num_input_tokens_seen": 13661550, + "step": 3845 + }, + { + "epoch": 0.2890425372012626, + "flos": 22110209890560.0, + "grad_norm": 1.347900456955543, + "learning_rate": 3.336668550706413e-06, + "loss": 1.0029, + "num_input_tokens_seen": 13685825, + "step": 3846 + }, + { + "epoch": 0.2891176912670976, + "flos": 19776879331200.0, + "grad_norm": 2.894447397518136, + "learning_rate": 3.3363063681544628e-06, + "loss": 0.9533, + "num_input_tokens_seen": 13708970, + "step": 3847 + }, + { + "epoch": 0.2891928453329325, + "flos": 27617133899520.0, + "grad_norm": 1.5496159169948718, + "learning_rate": 3.335944106421243e-06, + "loss": 0.9697, + "num_input_tokens_seen": 13733260, + "step": 3848 + }, + { + "epoch": 0.28926799939876746, + "flos": 29099355661440.0, + "grad_norm": 1.6499094244632022, + "learning_rate": 3.3355817655282188e-06, + "loss": 0.9679, + "num_input_tokens_seen": 13755870, + "step": 3849 + }, + { + "epoch": 0.28934315346460243, + "flos": 9643401895680.0, + "grad_norm": 2.0889602076866125, + "learning_rate": 3.3352193454968607e-06, + "loss": 0.9759, + "num_input_tokens_seen": 13770065, + "step": 3850 + }, + { + "epoch": 0.2894183075304374, + "flos": 32240814238080.0, + "grad_norm": 1.361874529561438, + "learning_rate": 3.334856846348644e-06, + "loss": 0.9529, + "num_input_tokens_seen": 13798895, + "step": 3851 + }, + { + "epoch": 0.2894934615962724, + "flos": 24401708904960.0, + "grad_norm": 2.0908261280789624, + "learning_rate": 3.3344942681050477e-06, + "loss": 0.7747, + "num_input_tokens_seen": 13822585, + "step": 3852 + }, + { + "epoch": 0.28956861566210734, + "flos": 17734745675520.0, + "grad_norm": 1.7137113060120113, + "learning_rate": 3.3341316107875552e-06, + "loss": 0.9861, + "num_input_tokens_seen": 13845605, + "step": 3853 + }, + { + "epoch": 0.28964376972794226, + "flos": 15899571216000.0, + "grad_norm": 1.5137550380380396, + "learning_rate": 3.3337688744176564e-06, + "loss": 1.0117, + "num_input_tokens_seen": 13868795, + "step": 3854 + }, + { + "epoch": 0.2897189237937772, + "flos": 17806054487040.0, + "grad_norm": 2.1297817284945078, + "learning_rate": 3.3334060590168447e-06, + "loss": 0.9512, + "num_input_tokens_seen": 13890320, + "step": 3855 + }, + { + "epoch": 0.2897940778596122, + "flos": 24472730407680.0, + "grad_norm": 1.554286349122635, + "learning_rate": 3.333043164606618e-06, + "loss": 1.0384, + "num_input_tokens_seen": 13913350, + "step": 3856 + }, + { + "epoch": 0.28986923192544717, + "flos": 15681406976640.0, + "grad_norm": 1.517040103309345, + "learning_rate": 3.332680191208479e-06, + "loss": 0.8926, + "num_input_tokens_seen": 13935615, + "step": 3857 + }, + { + "epoch": 0.28994438599128214, + "flos": 25111062005760.0, + "grad_norm": 1.64352581975725, + "learning_rate": 3.3323171388439353e-06, + "loss": 1.0316, + "num_input_tokens_seen": 13957525, + "step": 3858 + }, + { + "epoch": 0.2900195400571171, + "flos": 23803987789440.0, + "grad_norm": 1.3953850533993712, + "learning_rate": 3.3319540075344996e-06, + "loss": 0.8786, + "num_input_tokens_seen": 13983815, + "step": 3859 + }, + { + "epoch": 0.2900946941229521, + "flos": 21617656312320.0, + "grad_norm": 2.5893029386201176, + "learning_rate": 3.331590797301689e-06, + "loss": 0.8972, + "num_input_tokens_seen": 14007995, + "step": 3860 + }, + { + "epoch": 0.290169848188787, + "flos": 16641023276160.0, + "grad_norm": 1.3970406658067347, + "learning_rate": 3.331227508167024e-06, + "loss": 0.9663, + "num_input_tokens_seen": 14031585, + "step": 3861 + }, + { + "epoch": 0.29024500225462196, + "flos": 19535055828480.0, + "grad_norm": 1.5318177396857522, + "learning_rate": 3.330864140152032e-06, + "loss": 1.0836, + "num_input_tokens_seen": 14054260, + "step": 3862 + }, + { + "epoch": 0.29032015632045693, + "flos": 24540699254400.0, + "grad_norm": 1.3295561316534956, + "learning_rate": 3.3305006932782435e-06, + "loss": 1.0159, + "num_input_tokens_seen": 14078630, + "step": 3863 + }, + { + "epoch": 0.2903953103862919, + "flos": 31780754668800.0, + "grad_norm": 1.8864271993049686, + "learning_rate": 3.3301371675671935e-06, + "loss": 0.901, + "num_input_tokens_seen": 14104360, + "step": 3864 + }, + { + "epoch": 0.2904704644521269, + "flos": 53356551321600.0, + "grad_norm": 0.9296178390947288, + "learning_rate": 3.329773563040423e-06, + "loss": 0.8072, + "num_input_tokens_seen": 14181040, + "step": 3865 + }, + { + "epoch": 0.29054561851796185, + "flos": 20770174788480.0, + "grad_norm": 1.6667953044203625, + "learning_rate": 3.3294098797194776e-06, + "loss": 0.9352, + "num_input_tokens_seen": 14205370, + "step": 3866 + }, + { + "epoch": 0.29062077258379676, + "flos": 19005483127680.0, + "grad_norm": 1.6609140618752851, + "learning_rate": 3.3290461176259054e-06, + "loss": 0.9491, + "num_input_tokens_seen": 14226810, + "step": 3867 + }, + { + "epoch": 0.29069592664963173, + "flos": 66062345644800.0, + "grad_norm": 0.9235891472603845, + "learning_rate": 3.3286822767812618e-06, + "loss": 0.8162, + "num_input_tokens_seen": 14306350, + "step": 3868 + }, + { + "epoch": 0.2907710807154667, + "flos": 22499499352320.0, + "grad_norm": 1.5981190950704545, + "learning_rate": 3.3283183572071054e-06, + "loss": 0.9628, + "num_input_tokens_seen": 14328925, + "step": 3869 + }, + { + "epoch": 0.29084623478130167, + "flos": 14911447317120.0, + "grad_norm": 2.0641847145463026, + "learning_rate": 3.3279543589249998e-06, + "loss": 1.031, + "num_input_tokens_seen": 14350785, + "step": 3870 + }, + { + "epoch": 0.29092138884713664, + "flos": 24827155562880.0, + "grad_norm": 1.6869776588457543, + "learning_rate": 3.3275902819565127e-06, + "loss": 0.9234, + "num_input_tokens_seen": 14374410, + "step": 3871 + }, + { + "epoch": 0.2909965429129716, + "flos": 16987403784960.0, + "grad_norm": 2.00003945788561, + "learning_rate": 3.3272261263232195e-06, + "loss": 0.8793, + "num_input_tokens_seen": 14396635, + "step": 3872 + }, + { + "epoch": 0.2910716969788065, + "flos": 25816105474560.0, + "grad_norm": 2.1523012132820334, + "learning_rate": 3.326861892046694e-06, + "loss": 0.9381, + "num_input_tokens_seen": 14419840, + "step": 3873 + }, + { + "epoch": 0.2911468510446415, + "flos": 23312080656000.0, + "grad_norm": 1.574397604652459, + "learning_rate": 3.3264975791485218e-06, + "loss": 0.997, + "num_input_tokens_seen": 14444465, + "step": 3874 + }, + { + "epoch": 0.29122200511047647, + "flos": 21336694784640.0, + "grad_norm": 1.6774823284781772, + "learning_rate": 3.3261331876502884e-06, + "loss": 0.9149, + "num_input_tokens_seen": 14470095, + "step": 3875 + }, + { + "epoch": 0.29129715917631144, + "flos": 15370321737600.0, + "grad_norm": 1.6928520102535203, + "learning_rate": 3.325768717573585e-06, + "loss": 0.9956, + "num_input_tokens_seen": 14489930, + "step": 3876 + }, + { + "epoch": 0.2913723132421464, + "flos": 21647528628480.0, + "grad_norm": 1.5417076655254527, + "learning_rate": 3.325404168940009e-06, + "loss": 0.9923, + "num_input_tokens_seen": 14512605, + "step": 3877 + }, + { + "epoch": 0.2914474673079814, + "flos": 21682464762240.0, + "grad_norm": 1.3846605957833626, + "learning_rate": 3.3250395417711605e-06, + "loss": 1.0258, + "num_input_tokens_seen": 14537300, + "step": 3878 + }, + { + "epoch": 0.29152262137381635, + "flos": 19462238645760.0, + "grad_norm": 1.441815325986611, + "learning_rate": 3.3246748360886453e-06, + "loss": 1.0391, + "num_input_tokens_seen": 14560305, + "step": 3879 + }, + { + "epoch": 0.29159777543965126, + "flos": 33300354689280.0, + "grad_norm": 1.3156189870810207, + "learning_rate": 3.324310051914073e-06, + "loss": 0.966, + "num_input_tokens_seen": 14588135, + "step": 3880 + }, + { + "epoch": 0.29167292950548623, + "flos": 26877262037760.0, + "grad_norm": 1.372839704487574, + "learning_rate": 3.323945189269059e-06, + "loss": 0.9675, + "num_input_tokens_seen": 14613665, + "step": 3881 + }, + { + "epoch": 0.2917480835713212, + "flos": 17558951685120.0, + "grad_norm": 1.5051417440536932, + "learning_rate": 3.323580248175223e-06, + "loss": 0.9112, + "num_input_tokens_seen": 14635725, + "step": 3882 + }, + { + "epoch": 0.2918232376371562, + "flos": 18364960800000.0, + "grad_norm": 1.7075297241081246, + "learning_rate": 3.3232152286541898e-06, + "loss": 0.9525, + "num_input_tokens_seen": 14658245, + "step": 3883 + }, + { + "epoch": 0.29189839170299114, + "flos": 21370733078400.0, + "grad_norm": 1.4007241079458905, + "learning_rate": 3.3228501307275866e-06, + "loss": 0.9169, + "num_input_tokens_seen": 14681640, + "step": 3884 + }, + { + "epoch": 0.2919735457688261, + "flos": 25747669751040.0, + "grad_norm": 1.4330543106020126, + "learning_rate": 3.3224849544170475e-06, + "loss": 0.8964, + "num_input_tokens_seen": 14705035, + "step": 3885 + }, + { + "epoch": 0.29204869983466103, + "flos": 17306102707200.0, + "grad_norm": 1.748281657046459, + "learning_rate": 3.3221196997442107e-06, + "loss": 0.892, + "num_input_tokens_seen": 14726375, + "step": 3886 + }, + { + "epoch": 0.292123853900496, + "flos": 20982664679040.0, + "grad_norm": 1.694819717159464, + "learning_rate": 3.3217543667307196e-06, + "loss": 0.9421, + "num_input_tokens_seen": 14749325, + "step": 3887 + }, + { + "epoch": 0.29219900796633097, + "flos": 16919327197440.0, + "grad_norm": 1.3946535713980217, + "learning_rate": 3.3213889553982206e-06, + "loss": 1.0414, + "num_input_tokens_seen": 14771890, + "step": 3888 + }, + { + "epoch": 0.29227416203216594, + "flos": 60102041996160.0, + "grad_norm": 0.8614265921917238, + "learning_rate": 3.321023465768366e-06, + "loss": 0.8457, + "num_input_tokens_seen": 14853395, + "step": 3889 + }, + { + "epoch": 0.2923493160980009, + "flos": 21083450757120.0, + "grad_norm": 1.4696059472154235, + "learning_rate": 3.320657897862812e-06, + "loss": 0.9887, + "num_input_tokens_seen": 14875300, + "step": 3890 + }, + { + "epoch": 0.2924244701638359, + "flos": 39865023469440.0, + "grad_norm": 1.5382990613751877, + "learning_rate": 3.320292251703221e-06, + "loss": 0.8405, + "num_input_tokens_seen": 14903595, + "step": 3891 + }, + { + "epoch": 0.29249962422967085, + "flos": 16778864390400.0, + "grad_norm": 1.458049265124687, + "learning_rate": 3.3199265273112583e-06, + "loss": 0.8665, + "num_input_tokens_seen": 14926895, + "step": 3892 + }, + { + "epoch": 0.29257477829550577, + "flos": 23664602390400.0, + "grad_norm": 1.821292662425229, + "learning_rate": 3.3195607247085945e-06, + "loss": 0.841, + "num_input_tokens_seen": 14950660, + "step": 3893 + }, + { + "epoch": 0.29264993236134074, + "flos": 24828771674880.0, + "grad_norm": 1.731823874978256, + "learning_rate": 3.319194843916905e-06, + "loss": 0.8413, + "num_input_tokens_seen": 14976425, + "step": 3894 + }, + { + "epoch": 0.2927250864271757, + "flos": 24017411433600.0, + "grad_norm": 1.3802644346339512, + "learning_rate": 3.3188288849578694e-06, + "loss": 0.9539, + "num_input_tokens_seen": 15001340, + "step": 3895 + }, + { + "epoch": 0.2928002404930107, + "flos": 18473432388480.0, + "grad_norm": 1.719087810264154, + "learning_rate": 3.318462847853172e-06, + "loss": 1.0149, + "num_input_tokens_seen": 15022500, + "step": 3896 + }, + { + "epoch": 0.29287539455884565, + "flos": 65056229032320.0, + "grad_norm": 0.8060052322733748, + "learning_rate": 3.3180967326245018e-06, + "loss": 0.7822, + "num_input_tokens_seen": 15094730, + "step": 3897 + }, + { + "epoch": 0.2929505486246806, + "flos": 23170791836160.0, + "grad_norm": 1.559099005544929, + "learning_rate": 3.3177305392935536e-06, + "loss": 1.0007, + "num_input_tokens_seen": 15118620, + "step": 3898 + }, + { + "epoch": 0.29302570269051553, + "flos": 24300420036480.0, + "grad_norm": 1.4462396895322183, + "learning_rate": 3.317364267882025e-06, + "loss": 0.9863, + "num_input_tokens_seen": 15142335, + "step": 3899 + }, + { + "epoch": 0.2931008567563505, + "flos": 18187622524800.0, + "grad_norm": 1.9338398478175063, + "learning_rate": 3.3169979184116182e-06, + "loss": 1.0734, + "num_input_tokens_seen": 15165840, + "step": 3900 + }, + { + "epoch": 0.2931760108221855, + "flos": 21193789852800.0, + "grad_norm": 1.3619702437955228, + "learning_rate": 3.3166314909040427e-06, + "loss": 0.9589, + "num_input_tokens_seen": 15188930, + "step": 3901 + }, + { + "epoch": 0.29325116488802044, + "flos": 24969306309120.0, + "grad_norm": 1.4161440034746715, + "learning_rate": 3.316264985381009e-06, + "loss": 0.9477, + "num_input_tokens_seen": 15214450, + "step": 3902 + }, + { + "epoch": 0.2933263189538554, + "flos": 18437706155520.0, + "grad_norm": 1.4741870805970465, + "learning_rate": 3.315898401864235e-06, + "loss": 0.9569, + "num_input_tokens_seen": 15236590, + "step": 3903 + }, + { + "epoch": 0.2934014730196904, + "flos": 23621729351040.0, + "grad_norm": 2.9618356595603283, + "learning_rate": 3.315531740375441e-06, + "loss": 0.9084, + "num_input_tokens_seen": 15258565, + "step": 3904 + }, + { + "epoch": 0.29347662708552535, + "flos": 12612083224320.0, + "grad_norm": 1.630733352634642, + "learning_rate": 3.3151650009363544e-06, + "loss": 0.9565, + "num_input_tokens_seen": 15275985, + "step": 3905 + }, + { + "epoch": 0.29355178115136027, + "flos": 18405750850560.0, + "grad_norm": 1.296352832875754, + "learning_rate": 3.3147981835687054e-06, + "loss": 0.942, + "num_input_tokens_seen": 15298515, + "step": 3906 + }, + { + "epoch": 0.29362693521719524, + "flos": 20839652006400.0, + "grad_norm": 1.212926547736322, + "learning_rate": 3.314431288294229e-06, + "loss": 1.0107, + "num_input_tokens_seen": 15323140, + "step": 3907 + }, + { + "epoch": 0.2937020892830302, + "flos": 22006155674880.0, + "grad_norm": 2.0203066136155665, + "learning_rate": 3.314064315134666e-06, + "loss": 0.945, + "num_input_tokens_seen": 15345940, + "step": 3908 + }, + { + "epoch": 0.2937772433488652, + "flos": 22605852038400.0, + "grad_norm": 1.4682832341446135, + "learning_rate": 3.31369726411176e-06, + "loss": 0.8986, + "num_input_tokens_seen": 15371185, + "step": 3909 + }, + { + "epoch": 0.29385239741470015, + "flos": 20063587034880.0, + "grad_norm": 1.600628911035564, + "learning_rate": 3.313330135247261e-06, + "loss": 1.0127, + "num_input_tokens_seen": 15394325, + "step": 3910 + }, + { + "epoch": 0.2939275514805351, + "flos": 25147039633920.0, + "grad_norm": 1.355601103639308, + "learning_rate": 3.312962928562922e-06, + "loss": 0.8827, + "num_input_tokens_seen": 15421260, + "step": 3911 + }, + { + "epoch": 0.29400270554637004, + "flos": 27477640759680.0, + "grad_norm": 1.3332225137592217, + "learning_rate": 3.312595644080502e-06, + "loss": 0.8326, + "num_input_tokens_seen": 15449170, + "step": 3912 + }, + { + "epoch": 0.294077859612205, + "flos": 18334980743040.0, + "grad_norm": 1.462772053903238, + "learning_rate": 3.312228281821764e-06, + "loss": 0.9281, + "num_input_tokens_seen": 15471430, + "step": 3913 + }, + { + "epoch": 0.29415301367804, + "flos": 21123055658880.0, + "grad_norm": 1.6394412175024, + "learning_rate": 3.311860841808475e-06, + "loss": 0.979, + "num_input_tokens_seen": 15492775, + "step": 3914 + }, + { + "epoch": 0.29422816774387495, + "flos": 21935529221760.0, + "grad_norm": 1.553163481145439, + "learning_rate": 3.311493324062408e-06, + "loss": 0.8978, + "num_input_tokens_seen": 15515305, + "step": 3915 + }, + { + "epoch": 0.2943033218097099, + "flos": 16780157280000.0, + "grad_norm": 1.4660652459211532, + "learning_rate": 3.3111257286053394e-06, + "loss": 0.9176, + "num_input_tokens_seen": 15537135, + "step": 3916 + }, + { + "epoch": 0.2943784758755449, + "flos": 17479741881600.0, + "grad_norm": 2.732029895602145, + "learning_rate": 3.310758055459051e-06, + "loss": 0.9179, + "num_input_tokens_seen": 15557795, + "step": 3917 + }, + { + "epoch": 0.2944536299413798, + "flos": 23099913987840.0, + "grad_norm": 1.4442019758330626, + "learning_rate": 3.3103903046453282e-06, + "loss": 1.007, + "num_input_tokens_seen": 15582290, + "step": 3918 + }, + { + "epoch": 0.2945287840072148, + "flos": 21579559781760.0, + "grad_norm": 1.5233612823983038, + "learning_rate": 3.3100224761859626e-06, + "loss": 0.9877, + "num_input_tokens_seen": 15605485, + "step": 3919 + }, + { + "epoch": 0.29460393807304974, + "flos": 20450793507840.0, + "grad_norm": 1.6076431585611761, + "learning_rate": 3.309654570102748e-06, + "loss": 0.9941, + "num_input_tokens_seen": 15629270, + "step": 3920 + }, + { + "epoch": 0.2946790921388847, + "flos": 24897889756800.0, + "grad_norm": 1.4585431719992286, + "learning_rate": 3.309286586417486e-06, + "loss": 0.9095, + "num_input_tokens_seen": 15652620, + "step": 3921 + }, + { + "epoch": 0.2947542462047197, + "flos": 27789408357120.0, + "grad_norm": 1.644565899046728, + "learning_rate": 3.3089185251519797e-06, + "loss": 0.894, + "num_input_tokens_seen": 15678420, + "step": 3922 + }, + { + "epoch": 0.29482940027055465, + "flos": 16570181341440.0, + "grad_norm": 1.4949417659172837, + "learning_rate": 3.3085503863280387e-06, + "loss": 0.9513, + "num_input_tokens_seen": 15699900, + "step": 3923 + }, + { + "epoch": 0.2949045543363896, + "flos": 30087623214720.0, + "grad_norm": 3.418443047448991, + "learning_rate": 3.3081821699674763e-06, + "loss": 0.9521, + "num_input_tokens_seen": 15723305, + "step": 3924 + }, + { + "epoch": 0.29497970840222454, + "flos": 19817956690560.0, + "grad_norm": 2.1123797455819058, + "learning_rate": 3.307813876092111e-06, + "loss": 0.9553, + "num_input_tokens_seen": 15746455, + "step": 3925 + }, + { + "epoch": 0.2950548624680595, + "flos": 19034170295040.0, + "grad_norm": 1.956247991493827, + "learning_rate": 3.307445504723766e-06, + "loss": 1.0295, + "num_input_tokens_seen": 15765835, + "step": 3926 + }, + { + "epoch": 0.2951300165338945, + "flos": 58822218403200.0, + "grad_norm": 0.8135285835988785, + "learning_rate": 3.307077055884268e-06, + "loss": 0.8444, + "num_input_tokens_seen": 15843315, + "step": 3927 + }, + { + "epoch": 0.29520517059972945, + "flos": 25007043703680.0, + "grad_norm": 1.598401555676128, + "learning_rate": 3.3067085295954497e-06, + "loss": 1.0242, + "num_input_tokens_seen": 15868060, + "step": 3928 + }, + { + "epoch": 0.2952803246655644, + "flos": 16393597251840.0, + "grad_norm": 2.276338384798744, + "learning_rate": 3.306339925879147e-06, + "loss": 0.9128, + "num_input_tokens_seen": 15889470, + "step": 3929 + }, + { + "epoch": 0.2953554787313994, + "flos": 22324387720320.0, + "grad_norm": 1.4129119744874932, + "learning_rate": 3.305971244757201e-06, + "loss": 0.9776, + "num_input_tokens_seen": 15914920, + "step": 3930 + }, + { + "epoch": 0.2954306327972343, + "flos": 21931937861760.0, + "grad_norm": 1.5649023280271035, + "learning_rate": 3.305602486251458e-06, + "loss": 0.9854, + "num_input_tokens_seen": 15935470, + "step": 3931 + }, + { + "epoch": 0.2955057868630693, + "flos": 31924270131840.0, + "grad_norm": 1.667903997160477, + "learning_rate": 3.3052336503837686e-06, + "loss": 0.8921, + "num_input_tokens_seen": 15960605, + "step": 3932 + }, + { + "epoch": 0.29558094092890425, + "flos": 15546977654400.0, + "grad_norm": 1.4969268949109993, + "learning_rate": 3.304864737175987e-06, + "loss": 0.9701, + "num_input_tokens_seen": 15982240, + "step": 3933 + }, + { + "epoch": 0.2956560949947392, + "flos": 20417724881280.0, + "grad_norm": 1.234252633898457, + "learning_rate": 3.3044957466499736e-06, + "loss": 0.872, + "num_input_tokens_seen": 16005705, + "step": 3934 + }, + { + "epoch": 0.2957312490605742, + "flos": 21473243009280.0, + "grad_norm": 1.5330983120025723, + "learning_rate": 3.3041266788275913e-06, + "loss": 0.9804, + "num_input_tokens_seen": 16029520, + "step": 3935 + }, + { + "epoch": 0.29580640312640916, + "flos": 21299747489280.0, + "grad_norm": 1.5904258018726327, + "learning_rate": 3.303757533730709e-06, + "loss": 1.0161, + "num_input_tokens_seen": 16049985, + "step": 3936 + }, + { + "epoch": 0.2958815571922441, + "flos": 21793450302720.0, + "grad_norm": 1.515306419632557, + "learning_rate": 3.3033883113812017e-06, + "loss": 0.9808, + "num_input_tokens_seen": 16074095, + "step": 3937 + }, + { + "epoch": 0.29595671125807904, + "flos": 16536250788480.0, + "grad_norm": 1.7000037473749483, + "learning_rate": 3.303019011800946e-06, + "loss": 0.9772, + "num_input_tokens_seen": 16096350, + "step": 3938 + }, + { + "epoch": 0.296031865323914, + "flos": 20346559724160.0, + "grad_norm": 1.8295413668267515, + "learning_rate": 3.302649635011823e-06, + "loss": 1.0181, + "num_input_tokens_seen": 16115815, + "step": 3939 + }, + { + "epoch": 0.296107019389749, + "flos": 21861095927040.0, + "grad_norm": 1.5778144304214972, + "learning_rate": 3.302280181035722e-06, + "loss": 1.0088, + "num_input_tokens_seen": 16140165, + "step": 3940 + }, + { + "epoch": 0.29618217345558395, + "flos": 18193332787200.0, + "grad_norm": 2.0036710188369984, + "learning_rate": 3.301910649894533e-06, + "loss": 0.9542, + "num_input_tokens_seen": 16159055, + "step": 3941 + }, + { + "epoch": 0.2962573275214189, + "flos": 18046764668160.0, + "grad_norm": 1.7320700786059866, + "learning_rate": 3.3015410416101527e-06, + "loss": 0.9483, + "num_input_tokens_seen": 16181670, + "step": 3942 + }, + { + "epoch": 0.2963324815872539, + "flos": 19531859518080.0, + "grad_norm": 1.551088163431697, + "learning_rate": 3.301171356204482e-06, + "loss": 1.0443, + "num_input_tokens_seen": 16202875, + "step": 3943 + }, + { + "epoch": 0.2964076356530888, + "flos": 36369247478400.0, + "grad_norm": 1.5059911766705656, + "learning_rate": 3.300801593699425e-06, + "loss": 0.8896, + "num_input_tokens_seen": 16229960, + "step": 3944 + }, + { + "epoch": 0.2964827897189238, + "flos": 66210853098240.0, + "grad_norm": 0.7318451912638236, + "learning_rate": 3.300431754116894e-06, + "loss": 0.7667, + "num_input_tokens_seen": 16315800, + "step": 3945 + }, + { + "epoch": 0.29655794378475875, + "flos": 21475182343680.0, + "grad_norm": 1.831108007036179, + "learning_rate": 3.3000618374788e-06, + "loss": 0.8149, + "num_input_tokens_seen": 16337440, + "step": 3946 + }, + { + "epoch": 0.2966330978505937, + "flos": 21938222741760.0, + "grad_norm": 1.437952550887802, + "learning_rate": 3.299691843807065e-06, + "loss": 1.0161, + "num_input_tokens_seen": 16365070, + "step": 3947 + }, + { + "epoch": 0.2967082519164287, + "flos": 21088299093120.0, + "grad_norm": 2.027261544914834, + "learning_rate": 3.2993217731236118e-06, + "loss": 0.9898, + "num_input_tokens_seen": 16385775, + "step": 3948 + }, + { + "epoch": 0.29678340598226366, + "flos": 19888978193280.0, + "grad_norm": 1.4713979348843678, + "learning_rate": 3.2989516254503677e-06, + "loss": 0.792, + "num_input_tokens_seen": 16409770, + "step": 3949 + }, + { + "epoch": 0.29685856004809863, + "flos": 35205509157120.0, + "grad_norm": 1.5723841273034322, + "learning_rate": 3.298581400809266e-06, + "loss": 0.877, + "num_input_tokens_seen": 16436925, + "step": 3950 + }, + { + "epoch": 0.29693371411393354, + "flos": 20131735449600.0, + "grad_norm": 1.6885333958995283, + "learning_rate": 3.298211099222243e-06, + "loss": 0.9292, + "num_input_tokens_seen": 16459825, + "step": 3951 + }, + { + "epoch": 0.2970088681797685, + "flos": 25076341353600.0, + "grad_norm": 1.5799376398165936, + "learning_rate": 3.2978407207112416e-06, + "loss": 0.9554, + "num_input_tokens_seen": 16486280, + "step": 3952 + }, + { + "epoch": 0.2970840222456035, + "flos": 46819017624960.0, + "grad_norm": 1.896072798755829, + "learning_rate": 3.297470265298208e-06, + "loss": 0.8618, + "num_input_tokens_seen": 16514155, + "step": 3953 + }, + { + "epoch": 0.29715917631143846, + "flos": 21653238890880.0, + "grad_norm": 1.8703999357890826, + "learning_rate": 3.2970997330050923e-06, + "loss": 0.9458, + "num_input_tokens_seen": 16537925, + "step": 3954 + }, + { + "epoch": 0.2972343303772734, + "flos": 23835799440000.0, + "grad_norm": 1.4923441540039846, + "learning_rate": 3.2967291238538507e-06, + "loss": 0.9053, + "num_input_tokens_seen": 16561070, + "step": 3955 + }, + { + "epoch": 0.2973094844431084, + "flos": 20452589187840.0, + "grad_norm": 2.473725821293525, + "learning_rate": 3.296358437866443e-06, + "loss": 0.9454, + "num_input_tokens_seen": 16581985, + "step": 3956 + }, + { + "epoch": 0.2973846385089433, + "flos": 24334853379840.0, + "grad_norm": 1.6762804830511737, + "learning_rate": 3.2959876750648338e-06, + "loss": 0.9765, + "num_input_tokens_seen": 16605255, + "step": 3957 + }, + { + "epoch": 0.2974597925747783, + "flos": 20766403860480.0, + "grad_norm": 1.712525071009312, + "learning_rate": 3.2956168354709923e-06, + "loss": 0.9645, + "num_input_tokens_seen": 16629050, + "step": 3958 + }, + { + "epoch": 0.29753494664061325, + "flos": 17064027809280.0, + "grad_norm": 1.7948034535923363, + "learning_rate": 3.295245919106892e-06, + "loss": 0.8991, + "num_input_tokens_seen": 16651660, + "step": 3959 + }, + { + "epoch": 0.2976101007064482, + "flos": 20735956926720.0, + "grad_norm": 1.6032606100983153, + "learning_rate": 3.294874925994511e-06, + "loss": 1.0135, + "num_input_tokens_seen": 16675065, + "step": 3960 + }, + { + "epoch": 0.2976852547722832, + "flos": 33617509326720.0, + "grad_norm": 1.7226968704243921, + "learning_rate": 3.2945038561558324e-06, + "loss": 0.8737, + "num_input_tokens_seen": 16701935, + "step": 3961 + }, + { + "epoch": 0.29776040883811816, + "flos": 22641003653760.0, + "grad_norm": 1.4998749045025368, + "learning_rate": 3.2941327096128435e-06, + "loss": 0.9901, + "num_input_tokens_seen": 16726135, + "step": 3962 + }, + { + "epoch": 0.2978355629039531, + "flos": 66019221210240.0, + "grad_norm": 0.861285634625489, + "learning_rate": 3.2937614863875353e-06, + "loss": 0.8867, + "num_input_tokens_seen": 16805225, + "step": 3963 + }, + { + "epoch": 0.29791071696978805, + "flos": 24653372734080.0, + "grad_norm": 1.4149342584334004, + "learning_rate": 3.293390186501906e-06, + "loss": 0.8978, + "num_input_tokens_seen": 16830270, + "step": 3964 + }, + { + "epoch": 0.297985871035623, + "flos": 21405812866560.0, + "grad_norm": 1.77924937280084, + "learning_rate": 3.2930188099779546e-06, + "loss": 0.882, + "num_input_tokens_seen": 16853165, + "step": 3965 + }, + { + "epoch": 0.298061025101458, + "flos": 22288625573760.0, + "grad_norm": 1.496277003606283, + "learning_rate": 3.292647356837688e-06, + "loss": 1.0125, + "num_input_tokens_seen": 16877695, + "step": 3966 + }, + { + "epoch": 0.29813617916729296, + "flos": 23205656142720.0, + "grad_norm": 1.7625257584874228, + "learning_rate": 3.2922758271031147e-06, + "loss": 1.0133, + "num_input_tokens_seen": 16897215, + "step": 3967 + }, + { + "epoch": 0.29821133323312793, + "flos": 28112309170560.0, + "grad_norm": 1.762910494786441, + "learning_rate": 3.2919042207962506e-06, + "loss": 0.8801, + "num_input_tokens_seen": 16923190, + "step": 3968 + }, + { + "epoch": 0.2982864872989629, + "flos": 19888331748480.0, + "grad_norm": 1.6346603232027803, + "learning_rate": 3.2915325379391147e-06, + "loss": 0.9462, + "num_input_tokens_seen": 16946775, + "step": 3969 + }, + { + "epoch": 0.2983616413647978, + "flos": 19959245510400.0, + "grad_norm": 1.4154888567605473, + "learning_rate": 3.2911607785537297e-06, + "loss": 0.9935, + "num_input_tokens_seen": 16971095, + "step": 3970 + }, + { + "epoch": 0.2984367954306328, + "flos": 24582171663360.0, + "grad_norm": 1.4005198824723113, + "learning_rate": 3.290788942662125e-06, + "loss": 1.0337, + "num_input_tokens_seen": 16994300, + "step": 3971 + }, + { + "epoch": 0.29851194949646775, + "flos": 21511662762240.0, + "grad_norm": 1.657482192690375, + "learning_rate": 3.290417030286333e-06, + "loss": 0.9937, + "num_input_tokens_seen": 17018090, + "step": 3972 + }, + { + "epoch": 0.2985871035623027, + "flos": 14629121072640.0, + "grad_norm": 1.54101121560978, + "learning_rate": 3.2900450414483897e-06, + "loss": 0.9187, + "num_input_tokens_seen": 17039605, + "step": 3973 + }, + { + "epoch": 0.2986622576281377, + "flos": 24435028926720.0, + "grad_norm": 1.5686385308895834, + "learning_rate": 3.2896729761703386e-06, + "loss": 0.9302, + "num_input_tokens_seen": 17063155, + "step": 3974 + }, + { + "epoch": 0.29873741169397267, + "flos": 19923698845440.0, + "grad_norm": 1.4833507818988851, + "learning_rate": 3.2893008344742244e-06, + "loss": 0.8793, + "num_input_tokens_seen": 17085925, + "step": 3975 + }, + { + "epoch": 0.2988125657598076, + "flos": 21156411594240.0, + "grad_norm": 1.4995605301808086, + "learning_rate": 3.288928616382099e-06, + "loss": 0.9786, + "num_input_tokens_seen": 17111705, + "step": 3976 + }, + { + "epoch": 0.29888771982564255, + "flos": 21928633810560.0, + "grad_norm": 1.514739259496985, + "learning_rate": 3.288556321916018e-06, + "loss": 0.9437, + "num_input_tokens_seen": 17134035, + "step": 3977 + }, + { + "epoch": 0.2989628738914775, + "flos": 21717867772800.0, + "grad_norm": 1.790044656548133, + "learning_rate": 3.2881839510980403e-06, + "loss": 1.0504, + "num_input_tokens_seen": 17156010, + "step": 3978 + }, + { + "epoch": 0.2990380279573125, + "flos": 21969926651520.0, + "grad_norm": 1.4432909875755806, + "learning_rate": 3.2878115039502304e-06, + "loss": 0.9075, + "num_input_tokens_seen": 17182945, + "step": 3979 + }, + { + "epoch": 0.29911318202314746, + "flos": 19180702500480.0, + "grad_norm": 1.8164227884831758, + "learning_rate": 3.2874389804946575e-06, + "loss": 0.9671, + "num_input_tokens_seen": 17205115, + "step": 3980 + }, + { + "epoch": 0.29918833608898243, + "flos": 16853477253120.0, + "grad_norm": 1.4272654711843151, + "learning_rate": 3.287066380753395e-06, + "loss": 1.0274, + "num_input_tokens_seen": 17226120, + "step": 3981 + }, + { + "epoch": 0.2992634901548174, + "flos": 16354890190080.0, + "grad_norm": 1.7094865333671605, + "learning_rate": 3.2866937047485216e-06, + "loss": 1.0321, + "num_input_tokens_seen": 17245410, + "step": 3982 + }, + { + "epoch": 0.2993386442206523, + "flos": 21971111800320.0, + "grad_norm": 1.5586016279513857, + "learning_rate": 3.2863209525021186e-06, + "loss": 0.9411, + "num_input_tokens_seen": 17269940, + "step": 3983 + }, + { + "epoch": 0.2994137982864873, + "flos": 25076449094400.0, + "grad_norm": 1.3546405647709285, + "learning_rate": 3.285948124036274e-06, + "loss": 0.9717, + "num_input_tokens_seen": 17295485, + "step": 3984 + }, + { + "epoch": 0.29948895235232226, + "flos": 32452765424640.0, + "grad_norm": 1.544104358605231, + "learning_rate": 3.2855752193730786e-06, + "loss": 1.0003, + "num_input_tokens_seen": 17324390, + "step": 3985 + }, + { + "epoch": 0.2995641064181572, + "flos": 26062705486080.0, + "grad_norm": 1.5410618841579475, + "learning_rate": 3.2852022385346283e-06, + "loss": 0.9916, + "num_input_tokens_seen": 17348375, + "step": 3986 + }, + { + "epoch": 0.2996392604839922, + "flos": 44136182073600.0, + "grad_norm": 1.307510109492181, + "learning_rate": 3.2848291815430245e-06, + "loss": 0.9654, + "num_input_tokens_seen": 17375010, + "step": 3987 + }, + { + "epoch": 0.29971441454982717, + "flos": 20341855042560.0, + "grad_norm": 1.405942798544317, + "learning_rate": 3.2844560484203717e-06, + "loss": 1.0161, + "num_input_tokens_seen": 17399295, + "step": 3988 + }, + { + "epoch": 0.2997895686156621, + "flos": 20804284909440.0, + "grad_norm": 1.3681763914588816, + "learning_rate": 3.2840828391887792e-06, + "loss": 0.86, + "num_input_tokens_seen": 17423650, + "step": 3989 + }, + { + "epoch": 0.29986472268149705, + "flos": 29346709858560.0, + "grad_norm": 1.6150351681733266, + "learning_rate": 3.2837095538703613e-06, + "loss": 0.8823, + "num_input_tokens_seen": 17449690, + "step": 3990 + }, + { + "epoch": 0.299939876747332, + "flos": 24716816467200.0, + "grad_norm": 1.441280866095552, + "learning_rate": 3.283336192487237e-06, + "loss": 0.8614, + "num_input_tokens_seen": 17471485, + "step": 3991 + }, + { + "epoch": 0.300015030813167, + "flos": 21370840819200.0, + "grad_norm": 1.5766706658299854, + "learning_rate": 3.282962755061529e-06, + "loss": 0.9538, + "num_input_tokens_seen": 17495825, + "step": 3992 + }, + { + "epoch": 0.30009018487900196, + "flos": 20029441000320.0, + "grad_norm": 1.6153011654164458, + "learning_rate": 3.2825892416153656e-06, + "loss": 1.0508, + "num_input_tokens_seen": 17517800, + "step": 3993 + }, + { + "epoch": 0.30016533894483693, + "flos": 24442822177920.0, + "grad_norm": 1.2877547132509122, + "learning_rate": 3.282215652170877e-06, + "loss": 1.0304, + "num_input_tokens_seen": 17545070, + "step": 3994 + }, + { + "epoch": 0.3002404930106719, + "flos": 20947261668480.0, + "grad_norm": 1.6318075199266713, + "learning_rate": 3.2818419867502024e-06, + "loss": 0.9886, + "num_input_tokens_seen": 17569160, + "step": 3995 + }, + { + "epoch": 0.3003156470765068, + "flos": 14452572896640.0, + "grad_norm": 1.8495668368095497, + "learning_rate": 3.2814682453754805e-06, + "loss": 0.9782, + "num_input_tokens_seen": 17591290, + "step": 3996 + }, + { + "epoch": 0.3003908011423418, + "flos": 24900188227200.0, + "grad_norm": 1.7269631175623805, + "learning_rate": 3.281094428068858e-06, + "loss": 0.9743, + "num_input_tokens_seen": 17616410, + "step": 3997 + }, + { + "epoch": 0.30046595520817676, + "flos": 21015589651200.0, + "grad_norm": 1.5343069585875462, + "learning_rate": 3.280720534852486e-06, + "loss": 1.0099, + "num_input_tokens_seen": 17640150, + "step": 3998 + }, + { + "epoch": 0.30054110927401173, + "flos": 65372737224960.0, + "grad_norm": 0.8074027610550172, + "learning_rate": 3.2803465657485175e-06, + "loss": 0.8214, + "num_input_tokens_seen": 17710290, + "step": 3999 + }, + { + "epoch": 0.3006162633398467, + "flos": 20805721453440.0, + "grad_norm": 1.3390618536824077, + "learning_rate": 3.279972520779112e-06, + "loss": 0.9605, + "num_input_tokens_seen": 17735540, + "step": 4000 + }, + { + "epoch": 0.30069141740568167, + "flos": 23630097219840.0, + "grad_norm": 1.6670957330905067, + "learning_rate": 3.279598399966433e-06, + "loss": 0.906, + "num_input_tokens_seen": 17759430, + "step": 4001 + }, + { + "epoch": 0.3007665714715166, + "flos": 20417688967680.0, + "grad_norm": 1.7879203915592397, + "learning_rate": 3.27922420333265e-06, + "loss": 0.9837, + "num_input_tokens_seen": 17782105, + "step": 4002 + }, + { + "epoch": 0.30084172553735156, + "flos": 25640957928960.0, + "grad_norm": 1.529626647549311, + "learning_rate": 3.278849930899934e-06, + "loss": 0.903, + "num_input_tokens_seen": 17804935, + "step": 4003 + }, + { + "epoch": 0.3009168796031865, + "flos": 22500145797120.0, + "grad_norm": 1.5082816545404303, + "learning_rate": 3.278475582690462e-06, + "loss": 0.9989, + "num_input_tokens_seen": 17829190, + "step": 4004 + }, + { + "epoch": 0.3009920336690215, + "flos": 19181995390080.0, + "grad_norm": 1.317923466707806, + "learning_rate": 3.2781011587264173e-06, + "loss": 0.9456, + "num_input_tokens_seen": 17853700, + "step": 4005 + }, + { + "epoch": 0.30106718773485647, + "flos": 62889326812800.0, + "grad_norm": 0.8408041851699032, + "learning_rate": 3.2777266590299835e-06, + "loss": 0.842, + "num_input_tokens_seen": 17926150, + "step": 4006 + }, + { + "epoch": 0.30114234180069144, + "flos": 33935669544960.0, + "grad_norm": 1.4559755156680112, + "learning_rate": 3.277352083623353e-06, + "loss": 0.9157, + "num_input_tokens_seen": 17952695, + "step": 4007 + }, + { + "epoch": 0.30121749586652635, + "flos": 19606220985600.0, + "grad_norm": 1.7835830668105646, + "learning_rate": 3.2769774325287197e-06, + "loss": 0.9554, + "num_input_tokens_seen": 17974980, + "step": 4008 + }, + { + "epoch": 0.3012926499323613, + "flos": 22708864759680.0, + "grad_norm": 1.6312664688857508, + "learning_rate": 3.2766027057682844e-06, + "loss": 1.0319, + "num_input_tokens_seen": 17998550, + "step": 4009 + }, + { + "epoch": 0.3013678039981963, + "flos": 17453281357440.0, + "grad_norm": 1.342991015752877, + "learning_rate": 3.27622790336425e-06, + "loss": 0.9376, + "num_input_tokens_seen": 18022120, + "step": 4010 + }, + { + "epoch": 0.30144295806403126, + "flos": 26132577753600.0, + "grad_norm": 1.7238968514565105, + "learning_rate": 3.2758530253388255e-06, + "loss": 0.9269, + "num_input_tokens_seen": 18046155, + "step": 4011 + }, + { + "epoch": 0.30151811212986623, + "flos": 66861603002880.0, + "grad_norm": 0.782967396409344, + "learning_rate": 3.2754780717142233e-06, + "loss": 0.7907, + "num_input_tokens_seen": 18121425, + "step": 4012 + }, + { + "epoch": 0.3015932661957012, + "flos": 18335268051840.0, + "grad_norm": 1.6836224971715534, + "learning_rate": 3.2751030425126616e-06, + "loss": 1.0335, + "num_input_tokens_seen": 18144605, + "step": 4013 + }, + { + "epoch": 0.3016684202615362, + "flos": 20237405777280.0, + "grad_norm": 1.7014059788198095, + "learning_rate": 3.2747279377563616e-06, + "loss": 0.9884, + "num_input_tokens_seen": 18167115, + "step": 4014 + }, + { + "epoch": 0.3017435743273711, + "flos": 20172130450560.0, + "grad_norm": 1.6103573365856914, + "learning_rate": 3.2743527574675507e-06, + "loss": 1.0142, + "num_input_tokens_seen": 18192535, + "step": 4015 + }, + { + "epoch": 0.30181872839320606, + "flos": 23099734419840.0, + "grad_norm": 1.3128168995072294, + "learning_rate": 3.2739775016684584e-06, + "loss": 1.0001, + "num_input_tokens_seen": 18219075, + "step": 4016 + }, + { + "epoch": 0.30189388245904103, + "flos": 20805541885440.0, + "grad_norm": 1.9306707309729298, + "learning_rate": 3.2736021703813214e-06, + "loss": 0.9911, + "num_input_tokens_seen": 18243150, + "step": 4017 + }, + { + "epoch": 0.301969036524876, + "flos": 24083440945920.0, + "grad_norm": 1.5562084943146595, + "learning_rate": 3.2732267636283782e-06, + "loss": 0.9787, + "num_input_tokens_seen": 18264860, + "step": 4018 + }, + { + "epoch": 0.30204419059071097, + "flos": 68667731159040.0, + "grad_norm": 0.9084893872717672, + "learning_rate": 3.2728512814318744e-06, + "loss": 0.8585, + "num_input_tokens_seen": 18346615, + "step": 4019 + }, + { + "epoch": 0.30211934465654594, + "flos": 29135656512000.0, + "grad_norm": 1.4137093540355752, + "learning_rate": 3.2724757238140572e-06, + "loss": 0.94, + "num_input_tokens_seen": 18371615, + "step": 4020 + }, + { + "epoch": 0.30219449872238086, + "flos": 18688328490240.0, + "grad_norm": 1.6422153575795613, + "learning_rate": 3.2721000907971813e-06, + "loss": 1.0242, + "num_input_tokens_seen": 18393070, + "step": 4021 + }, + { + "epoch": 0.3022696527882158, + "flos": 22536015684480.0, + "grad_norm": 1.5249864577645784, + "learning_rate": 3.2717243824035037e-06, + "loss": 0.8649, + "num_input_tokens_seen": 18417330, + "step": 4022 + }, + { + "epoch": 0.3023448068540508, + "flos": 27582664642560.0, + "grad_norm": 1.7611457248031208, + "learning_rate": 3.2713485986552865e-06, + "loss": 0.931, + "num_input_tokens_seen": 18439005, + "step": 4023 + }, + { + "epoch": 0.30241996091988577, + "flos": 28575241827840.0, + "grad_norm": 1.7904276074963033, + "learning_rate": 3.2709727395747974e-06, + "loss": 0.9779, + "num_input_tokens_seen": 18466695, + "step": 4024 + }, + { + "epoch": 0.30249511498572074, + "flos": 65956495749120.0, + "grad_norm": 0.8044335788716895, + "learning_rate": 3.2705968051843053e-06, + "loss": 0.7355, + "num_input_tokens_seen": 18537045, + "step": 4025 + }, + { + "epoch": 0.3025702690515557, + "flos": 19112518172160.0, + "grad_norm": 1.3283017209230932, + "learning_rate": 3.270220795506088e-06, + "loss": 0.9704, + "num_input_tokens_seen": 18559570, + "step": 4026 + }, + { + "epoch": 0.3026454231173907, + "flos": 21123594362880.0, + "grad_norm": 1.5570862633556182, + "learning_rate": 3.269844710562424e-06, + "loss": 1.0164, + "num_input_tokens_seen": 18582895, + "step": 4027 + }, + { + "epoch": 0.3027205771832256, + "flos": 16359056167680.0, + "grad_norm": 2.064324005910747, + "learning_rate": 3.269468550375599e-06, + "loss": 0.9133, + "num_input_tokens_seen": 18605075, + "step": 4028 + }, + { + "epoch": 0.30279573124906056, + "flos": 28563713562240.0, + "grad_norm": 1.5859534352070428, + "learning_rate": 3.2690923149679008e-06, + "loss": 0.9301, + "num_input_tokens_seen": 18628940, + "step": 4029 + }, + { + "epoch": 0.30287088531489553, + "flos": 21159069200640.0, + "grad_norm": 2.3619196224312815, + "learning_rate": 3.268716004361623e-06, + "loss": 0.9353, + "num_input_tokens_seen": 18652540, + "step": 4030 + }, + { + "epoch": 0.3029460393807305, + "flos": 20276400147840.0, + "grad_norm": 1.6020621660055323, + "learning_rate": 3.2683396185790644e-06, + "loss": 1.0021, + "num_input_tokens_seen": 18675490, + "step": 4031 + }, + { + "epoch": 0.3030211934465655, + "flos": 21085569659520.0, + "grad_norm": 1.2810166309238975, + "learning_rate": 3.2679631576425265e-06, + "loss": 0.9701, + "num_input_tokens_seen": 18699705, + "step": 4032 + }, + { + "epoch": 0.30309634751240044, + "flos": 21012285600000.0, + "grad_norm": 1.8717453365601562, + "learning_rate": 3.267586621574315e-06, + "loss": 0.9496, + "num_input_tokens_seen": 18721335, + "step": 4033 + }, + { + "epoch": 0.30317150157823536, + "flos": 19283391999360.0, + "grad_norm": 1.3656971683244634, + "learning_rate": 3.2672100103967434e-06, + "loss": 0.9027, + "num_input_tokens_seen": 18744595, + "step": 4034 + }, + { + "epoch": 0.30324665564407033, + "flos": 23310177235200.0, + "grad_norm": 1.6504323739634446, + "learning_rate": 3.266833324132126e-06, + "loss": 1.014, + "num_input_tokens_seen": 18768115, + "step": 4035 + }, + { + "epoch": 0.3033218097099053, + "flos": 15334739159040.0, + "grad_norm": 1.7264243368388514, + "learning_rate": 3.2664565628027833e-06, + "loss": 1.0571, + "num_input_tokens_seen": 18788605, + "step": 4036 + }, + { + "epoch": 0.30339696377574027, + "flos": 22640895912960.0, + "grad_norm": 1.6028284047617463, + "learning_rate": 3.2660797264310393e-06, + "loss": 0.923, + "num_input_tokens_seen": 18811590, + "step": 4037 + }, + { + "epoch": 0.30347211784157524, + "flos": 18864948493440.0, + "grad_norm": 1.451118232074445, + "learning_rate": 3.2657028150392236e-06, + "loss": 0.9901, + "num_input_tokens_seen": 18834455, + "step": 4038 + }, + { + "epoch": 0.3035472719074102, + "flos": 24965391726720.0, + "grad_norm": 1.4103900578222528, + "learning_rate": 3.2653258286496696e-06, + "loss": 0.9959, + "num_input_tokens_seen": 18857750, + "step": 4039 + }, + { + "epoch": 0.3036224259732452, + "flos": 13923000195840.0, + "grad_norm": 1.5972214086042165, + "learning_rate": 3.2649487672847143e-06, + "loss": 0.9638, + "num_input_tokens_seen": 18880050, + "step": 4040 + }, + { + "epoch": 0.3036975800390801, + "flos": 13994057612160.0, + "grad_norm": 1.4162696768327243, + "learning_rate": 3.264571630966701e-06, + "loss": 1.0098, + "num_input_tokens_seen": 18900410, + "step": 4041 + }, + { + "epoch": 0.30377273410491507, + "flos": 24759043061760.0, + "grad_norm": 1.4286980044965305, + "learning_rate": 3.2641944197179767e-06, + "loss": 0.9562, + "num_input_tokens_seen": 18925025, + "step": 4042 + }, + { + "epoch": 0.30384788817075004, + "flos": 29272025168640.0, + "grad_norm": 1.7869374677180356, + "learning_rate": 3.2638171335608914e-06, + "loss": 0.9014, + "num_input_tokens_seen": 18949180, + "step": 4043 + }, + { + "epoch": 0.303923042236585, + "flos": 15264328187520.0, + "grad_norm": 1.518018503988518, + "learning_rate": 3.263439772517802e-06, + "loss": 0.9967, + "num_input_tokens_seen": 18970495, + "step": 4044 + }, + { + "epoch": 0.30399819630242, + "flos": 23452938512640.0, + "grad_norm": 2.269156814126059, + "learning_rate": 3.2630623366110673e-06, + "loss": 0.8977, + "num_input_tokens_seen": 18992435, + "step": 4045 + }, + { + "epoch": 0.30407335036825495, + "flos": 25564693040640.0, + "grad_norm": 1.4942228805738602, + "learning_rate": 3.2626848258630533e-06, + "loss": 0.839, + "num_input_tokens_seen": 19014465, + "step": 4046 + }, + { + "epoch": 0.30414850443408986, + "flos": 33653522868480.0, + "grad_norm": 1.347907709075235, + "learning_rate": 3.2623072402961283e-06, + "loss": 0.8488, + "num_input_tokens_seen": 19040470, + "step": 4047 + }, + { + "epoch": 0.30422365849992483, + "flos": 25426169568000.0, + "grad_norm": 1.8470225628609103, + "learning_rate": 3.2619295799326657e-06, + "loss": 0.9108, + "num_input_tokens_seen": 19062710, + "step": 4048 + }, + { + "epoch": 0.3042988125657598, + "flos": 16323796811520.0, + "grad_norm": 1.6714618207957026, + "learning_rate": 3.2615518447950425e-06, + "loss": 1.0074, + "num_input_tokens_seen": 19083505, + "step": 4049 + }, + { + "epoch": 0.3043739666315948, + "flos": 20061504046080.0, + "grad_norm": 1.6863684215772208, + "learning_rate": 3.2611740349056424e-06, + "loss": 0.9493, + "num_input_tokens_seen": 19106585, + "step": 4050 + }, + { + "epoch": 0.30444912069742974, + "flos": 23944809732480.0, + "grad_norm": 1.2629094130407275, + "learning_rate": 3.2607961502868507e-06, + "loss": 1.011, + "num_input_tokens_seen": 19130060, + "step": 4051 + }, + { + "epoch": 0.3045242747632647, + "flos": 22177675946880.0, + "grad_norm": 1.339573068218541, + "learning_rate": 3.2604181909610595e-06, + "loss": 0.9677, + "num_input_tokens_seen": 19154965, + "step": 4052 + }, + { + "epoch": 0.3045994288290996, + "flos": 26665310851200.0, + "grad_norm": 1.4306139003044758, + "learning_rate": 3.2600401569506646e-06, + "loss": 0.9245, + "num_input_tokens_seen": 19180495, + "step": 4053 + }, + { + "epoch": 0.3046745828949346, + "flos": 24794051022720.0, + "grad_norm": 1.835144456067925, + "learning_rate": 3.2596620482780647e-06, + "loss": 0.8979, + "num_input_tokens_seen": 19204805, + "step": 4054 + }, + { + "epoch": 0.30474973696076957, + "flos": 26098288064640.0, + "grad_norm": 1.761573396425381, + "learning_rate": 3.2592838649656648e-06, + "loss": 0.9828, + "num_input_tokens_seen": 19229205, + "step": 4055 + }, + { + "epoch": 0.30482489102660454, + "flos": 21900090297600.0, + "grad_norm": 1.6285045686540007, + "learning_rate": 3.2589056070358743e-06, + "loss": 0.9877, + "num_input_tokens_seen": 19252300, + "step": 4056 + }, + { + "epoch": 0.3049000450924395, + "flos": 23700113141760.0, + "grad_norm": 1.5320721968528863, + "learning_rate": 3.258527274511105e-06, + "loss": 0.9609, + "num_input_tokens_seen": 19277215, + "step": 4057 + }, + { + "epoch": 0.3049751991582745, + "flos": 20593734353280.0, + "grad_norm": 1.3556504801091895, + "learning_rate": 3.2581488674137766e-06, + "loss": 1.04, + "num_input_tokens_seen": 19301720, + "step": 4058 + }, + { + "epoch": 0.30505035322410945, + "flos": 17135336620800.0, + "grad_norm": 1.4893538647303424, + "learning_rate": 3.2577703857663094e-06, + "loss": 0.899, + "num_input_tokens_seen": 19324920, + "step": 4059 + }, + { + "epoch": 0.30512550728994436, + "flos": 22500361278720.0, + "grad_norm": 1.7710612505676728, + "learning_rate": 3.2573918295911306e-06, + "loss": 0.965, + "num_input_tokens_seen": 19347375, + "step": 4060 + }, + { + "epoch": 0.30520066135577933, + "flos": 28253777558400.0, + "grad_norm": 1.4161184128911055, + "learning_rate": 3.257013198910671e-06, + "loss": 0.9202, + "num_input_tokens_seen": 19373170, + "step": 4061 + }, + { + "epoch": 0.3052758154216143, + "flos": 20841663168000.0, + "grad_norm": 2.184065711482595, + "learning_rate": 3.256634493747366e-06, + "loss": 1.0063, + "num_input_tokens_seen": 19397325, + "step": 4062 + }, + { + "epoch": 0.3053509694874493, + "flos": 23876481749760.0, + "grad_norm": 1.8519864140294426, + "learning_rate": 3.256255714123655e-06, + "loss": 0.9376, + "num_input_tokens_seen": 19420040, + "step": 4063 + }, + { + "epoch": 0.30542612355328425, + "flos": 21472452910080.0, + "grad_norm": 1.357628193276832, + "learning_rate": 3.255876860061983e-06, + "loss": 0.9483, + "num_input_tokens_seen": 19442625, + "step": 4064 + }, + { + "epoch": 0.3055012776191192, + "flos": 40606044566400.0, + "grad_norm": 1.4919400634129953, + "learning_rate": 3.2554979315847977e-06, + "loss": 0.9693, + "num_input_tokens_seen": 19469355, + "step": 4065 + }, + { + "epoch": 0.30557643168495413, + "flos": 22817659570560.0, + "grad_norm": 1.9948620962221237, + "learning_rate": 3.255118928714552e-06, + "loss": 0.8577, + "num_input_tokens_seen": 19491215, + "step": 4066 + }, + { + "epoch": 0.3056515857507891, + "flos": 53128934006400.0, + "grad_norm": 1.6067984984091024, + "learning_rate": 3.2547398514737038e-06, + "loss": 0.8875, + "num_input_tokens_seen": 19519900, + "step": 4067 + }, + { + "epoch": 0.30572673981662407, + "flos": 25429653187200.0, + "grad_norm": 2.007923998872418, + "learning_rate": 3.2543606998847145e-06, + "loss": 0.9241, + "num_input_tokens_seen": 19545340, + "step": 4068 + }, + { + "epoch": 0.30580189388245904, + "flos": 17664191049600.0, + "grad_norm": 1.9597800436966468, + "learning_rate": 3.253981473970051e-06, + "loss": 0.9406, + "num_input_tokens_seen": 19567335, + "step": 4069 + }, + { + "epoch": 0.305877047948294, + "flos": 38731552513920.0, + "grad_norm": 1.4510716523054648, + "learning_rate": 3.253602173752183e-06, + "loss": 0.8979, + "num_input_tokens_seen": 19594060, + "step": 4070 + }, + { + "epoch": 0.305952202014129, + "flos": 21439204715520.0, + "grad_norm": 1.4279228117407723, + "learning_rate": 3.253222799253586e-06, + "loss": 0.9218, + "num_input_tokens_seen": 19617470, + "step": 4071 + }, + { + "epoch": 0.30602735607996395, + "flos": 30548221488000.0, + "grad_norm": 2.1311254291541823, + "learning_rate": 3.2528433504967394e-06, + "loss": 0.8705, + "num_input_tokens_seen": 19641920, + "step": 4072 + }, + { + "epoch": 0.30610251014579887, + "flos": 21229049208960.0, + "grad_norm": 2.3879023986755796, + "learning_rate": 3.252463827504126e-06, + "loss": 1.0163, + "num_input_tokens_seen": 19663500, + "step": 4073 + }, + { + "epoch": 0.30617766421163384, + "flos": 27650597575680.0, + "grad_norm": 1.5730856326803577, + "learning_rate": 3.2520842302982356e-06, + "loss": 0.9415, + "num_input_tokens_seen": 19687270, + "step": 4074 + }, + { + "epoch": 0.3062528182774688, + "flos": 26769508721280.0, + "grad_norm": 1.6990535962753963, + "learning_rate": 3.2517045589015602e-06, + "loss": 0.9171, + "num_input_tokens_seen": 19711885, + "step": 4075 + }, + { + "epoch": 0.3063279723433038, + "flos": 30860815098240.0, + "grad_norm": 1.431726055016304, + "learning_rate": 3.251324813336596e-06, + "loss": 0.9043, + "num_input_tokens_seen": 19736065, + "step": 4076 + }, + { + "epoch": 0.30640312640913875, + "flos": 23986856759040.0, + "grad_norm": 1.2874689135455863, + "learning_rate": 3.2509449936258452e-06, + "loss": 0.9345, + "num_input_tokens_seen": 19764015, + "step": 4077 + }, + { + "epoch": 0.3064782804749737, + "flos": 19887793044480.0, + "grad_norm": 1.581986942377228, + "learning_rate": 3.2505650997918127e-06, + "loss": 1.0463, + "num_input_tokens_seen": 19785000, + "step": 4078 + }, + { + "epoch": 0.30655343454080863, + "flos": 24192594892800.0, + "grad_norm": 1.3849767784898694, + "learning_rate": 3.2501851318570103e-06, + "loss": 0.8844, + "num_input_tokens_seen": 19809390, + "step": 4079 + }, + { + "epoch": 0.3066285886066436, + "flos": 18189382291200.0, + "grad_norm": 2.9272397000686015, + "learning_rate": 3.249805089843951e-06, + "loss": 0.9989, + "num_input_tokens_seen": 19831470, + "step": 4080 + }, + { + "epoch": 0.3067037426724786, + "flos": 15193989043200.0, + "grad_norm": 1.5502984282630248, + "learning_rate": 3.249424973775155e-06, + "loss": 0.9237, + "num_input_tokens_seen": 19853495, + "step": 4081 + }, + { + "epoch": 0.30677889673831354, + "flos": 33050127404160.0, + "grad_norm": 1.4108649082034144, + "learning_rate": 3.249044783673144e-06, + "loss": 0.89, + "num_input_tokens_seen": 19878530, + "step": 4082 + }, + { + "epoch": 0.3068540508041485, + "flos": 32347382405760.0, + "grad_norm": 1.3963114598065758, + "learning_rate": 3.2486645195604466e-06, + "loss": 0.9872, + "num_input_tokens_seen": 19905300, + "step": 4083 + }, + { + "epoch": 0.3069292048699835, + "flos": 23205799797120.0, + "grad_norm": 1.8721388285420932, + "learning_rate": 3.2482841814595954e-06, + "loss": 0.9539, + "num_input_tokens_seen": 19928165, + "step": 4084 + }, + { + "epoch": 0.30700435893581846, + "flos": 20378622769920.0, + "grad_norm": 1.5007470178150324, + "learning_rate": 3.247903769393127e-06, + "loss": 0.9624, + "num_input_tokens_seen": 19950085, + "step": 4085 + }, + { + "epoch": 0.30707951300165337, + "flos": 24583464552960.0, + "grad_norm": 1.454151816233068, + "learning_rate": 3.247523283383581e-06, + "loss": 1.0056, + "num_input_tokens_seen": 19976435, + "step": 4086 + }, + { + "epoch": 0.30715466706748834, + "flos": 21088406833920.0, + "grad_norm": 1.369513750590121, + "learning_rate": 3.2471427234535034e-06, + "loss": 1.0187, + "num_input_tokens_seen": 20000955, + "step": 4087 + }, + { + "epoch": 0.3072298211333233, + "flos": 16529786340480.0, + "grad_norm": 2.020766332505359, + "learning_rate": 3.246762089625444e-06, + "loss": 0.8721, + "num_input_tokens_seen": 20021655, + "step": 4088 + }, + { + "epoch": 0.3073049751991583, + "flos": 19075965926400.0, + "grad_norm": 1.3143875083019794, + "learning_rate": 3.2463813819219565e-06, + "loss": 1.0563, + "num_input_tokens_seen": 20045760, + "step": 4089 + }, + { + "epoch": 0.30738012926499325, + "flos": 25111852104960.0, + "grad_norm": 1.6510179760634183, + "learning_rate": 3.2460006003655993e-06, + "loss": 1.0888, + "num_input_tokens_seen": 20068450, + "step": 4090 + }, + { + "epoch": 0.3074552833308282, + "flos": 30298425166080.0, + "grad_norm": 1.7117426123233324, + "learning_rate": 3.245619744978936e-06, + "loss": 0.8816, + "num_input_tokens_seen": 20095725, + "step": 4091 + }, + { + "epoch": 0.30753043739666314, + "flos": 55677627544320.0, + "grad_norm": 1.616671272481014, + "learning_rate": 3.2452388157845322e-06, + "loss": 0.8555, + "num_input_tokens_seen": 20123890, + "step": 4092 + }, + { + "epoch": 0.3076055914624981, + "flos": 19740255258240.0, + "grad_norm": 1.6505947370546266, + "learning_rate": 3.2448578128049607e-06, + "loss": 1.1055, + "num_input_tokens_seen": 20145140, + "step": 4093 + }, + { + "epoch": 0.3076807455283331, + "flos": 64605291517440.0, + "grad_norm": 0.959188839622711, + "learning_rate": 3.2444767360627964e-06, + "loss": 0.883, + "num_input_tokens_seen": 20222685, + "step": 4094 + }, + { + "epoch": 0.30775589959416805, + "flos": 20771395850880.0, + "grad_norm": 1.5606495390197563, + "learning_rate": 3.2440955855806203e-06, + "loss": 0.9222, + "num_input_tokens_seen": 20244635, + "step": 4095 + }, + { + "epoch": 0.307831053660003, + "flos": 22535081930880.0, + "grad_norm": 1.5982447801681052, + "learning_rate": 3.2437143613810173e-06, + "loss": 0.9496, + "num_input_tokens_seen": 20267830, + "step": 4096 + }, + { + "epoch": 0.307906207725838, + "flos": 20064341220480.0, + "grad_norm": 2.0833019508541284, + "learning_rate": 3.2433330634865757e-06, + "loss": 1.0377, + "num_input_tokens_seen": 20291275, + "step": 4097 + }, + { + "epoch": 0.3079813617916729, + "flos": 44665503379200.0, + "grad_norm": 1.5001633873010742, + "learning_rate": 3.242951691919888e-06, + "loss": 0.8362, + "num_input_tokens_seen": 20319300, + "step": 4098 + }, + { + "epoch": 0.3080565158575079, + "flos": 22958840649600.0, + "grad_norm": 1.6118221011287988, + "learning_rate": 3.242570246703554e-06, + "loss": 0.8528, + "num_input_tokens_seen": 20342970, + "step": 4099 + }, + { + "epoch": 0.30813166992334284, + "flos": 20629281018240.0, + "grad_norm": 1.5461161900492226, + "learning_rate": 3.242188727860174e-06, + "loss": 0.9313, + "num_input_tokens_seen": 20365365, + "step": 4100 + }, + { + "epoch": 0.3082068239891778, + "flos": 26027517957120.0, + "grad_norm": 1.9144928117618625, + "learning_rate": 3.2418071354123566e-06, + "loss": 0.9038, + "num_input_tokens_seen": 20390320, + "step": 4101 + }, + { + "epoch": 0.3082819780550128, + "flos": 23417894638080.0, + "grad_norm": 1.6088781978861995, + "learning_rate": 3.2414254693827098e-06, + "loss": 0.9692, + "num_input_tokens_seen": 20413405, + "step": 4102 + }, + { + "epoch": 0.30835713212084775, + "flos": 21582540610560.0, + "grad_norm": 1.3638507779207811, + "learning_rate": 3.2410437297938512e-06, + "loss": 0.9235, + "num_input_tokens_seen": 20436480, + "step": 4103 + }, + { + "epoch": 0.3084322861866827, + "flos": 37288827912960.0, + "grad_norm": 1.613921179405531, + "learning_rate": 3.240661916668399e-06, + "loss": 0.9194, + "num_input_tokens_seen": 20462215, + "step": 4104 + }, + { + "epoch": 0.30850744025251764, + "flos": 53505517852800.0, + "grad_norm": 0.9983900640976746, + "learning_rate": 3.2402800300289773e-06, + "loss": 0.915, + "num_input_tokens_seen": 20522315, + "step": 4105 + }, + { + "epoch": 0.3085825943183526, + "flos": 16711721556480.0, + "grad_norm": 1.7584583367604114, + "learning_rate": 3.2398980698982143e-06, + "loss": 0.8207, + "num_input_tokens_seen": 20545240, + "step": 4106 + }, + { + "epoch": 0.3086577483841876, + "flos": 27829695617280.0, + "grad_norm": 1.8241477805470308, + "learning_rate": 3.2395160362987432e-06, + "loss": 0.9898, + "num_input_tokens_seen": 20569550, + "step": 4107 + }, + { + "epoch": 0.30873290245002255, + "flos": 21370230288000.0, + "grad_norm": 1.6378472928138437, + "learning_rate": 3.2391339292532004e-06, + "loss": 0.9945, + "num_input_tokens_seen": 20593540, + "step": 4108 + }, + { + "epoch": 0.3088080565158575, + "flos": 17488361145600.0, + "grad_norm": 1.532494875851103, + "learning_rate": 3.2387517487842273e-06, + "loss": 1.0605, + "num_input_tokens_seen": 20615060, + "step": 4109 + }, + { + "epoch": 0.3088832105816925, + "flos": 21264883182720.0, + "grad_norm": 2.147073985821644, + "learning_rate": 3.2383694949144693e-06, + "loss": 0.9745, + "num_input_tokens_seen": 20640115, + "step": 4110 + }, + { + "epoch": 0.3089583646475274, + "flos": 17656721020800.0, + "grad_norm": 1.5419186652486925, + "learning_rate": 3.2379871676665767e-06, + "loss": 1.0434, + "num_input_tokens_seen": 20662505, + "step": 4111 + }, + { + "epoch": 0.3090335187133624, + "flos": 25252674048000.0, + "grad_norm": 1.5373930420927644, + "learning_rate": 3.237604767063204e-06, + "loss": 0.9912, + "num_input_tokens_seen": 20689055, + "step": 4112 + }, + { + "epoch": 0.30910867277919735, + "flos": 22598346096000.0, + "grad_norm": 1.6397544367631762, + "learning_rate": 3.23722229312701e-06, + "loss": 0.9639, + "num_input_tokens_seen": 20711900, + "step": 4113 + }, + { + "epoch": 0.3091838268450323, + "flos": 20027322097920.0, + "grad_norm": 1.3427847123106191, + "learning_rate": 3.2368397458806573e-06, + "loss": 0.9314, + "num_input_tokens_seen": 20737495, + "step": 4114 + }, + { + "epoch": 0.3092589809108673, + "flos": 33124201562880.0, + "grad_norm": 5.4807412480143185, + "learning_rate": 3.236457125346814e-06, + "loss": 0.9107, + "num_input_tokens_seen": 20765880, + "step": 4115 + }, + { + "epoch": 0.30933413497670226, + "flos": 27401447698560.0, + "grad_norm": 1.775907099334568, + "learning_rate": 3.23607443154815e-06, + "loss": 0.9432, + "num_input_tokens_seen": 20785040, + "step": 4116 + }, + { + "epoch": 0.3094092890425372, + "flos": 20029225518720.0, + "grad_norm": 1.8072772579736405, + "learning_rate": 3.2356916645073444e-06, + "loss": 0.9915, + "num_input_tokens_seen": 20807755, + "step": 4117 + }, + { + "epoch": 0.30948444310837214, + "flos": 26342445951360.0, + "grad_norm": 1.5167423337708394, + "learning_rate": 3.2353088242470744e-06, + "loss": 0.8138, + "num_input_tokens_seen": 20832215, + "step": 4118 + }, + { + "epoch": 0.3095595971742071, + "flos": 18154230675840.0, + "grad_norm": 1.4260695910546874, + "learning_rate": 3.2349259107900267e-06, + "loss": 0.9433, + "num_input_tokens_seen": 20855460, + "step": 4119 + }, + { + "epoch": 0.3096347512400421, + "flos": 20276256493440.0, + "grad_norm": 1.796115482525721, + "learning_rate": 3.2345429241588902e-06, + "loss": 0.9338, + "num_input_tokens_seen": 20876355, + "step": 4120 + }, + { + "epoch": 0.30970990530587705, + "flos": 23239622609280.0, + "grad_norm": 1.4223592583297897, + "learning_rate": 3.234159864376358e-06, + "loss": 0.953, + "num_input_tokens_seen": 20900235, + "step": 4121 + }, + { + "epoch": 0.309785059371712, + "flos": 14805561507840.0, + "grad_norm": 1.9608160946411533, + "learning_rate": 3.233776731465128e-06, + "loss": 0.8745, + "num_input_tokens_seen": 20920480, + "step": 4122 + }, + { + "epoch": 0.309860213437547, + "flos": 21508681933440.0, + "grad_norm": 1.6882438433102895, + "learning_rate": 3.233393525447902e-06, + "loss": 0.893, + "num_input_tokens_seen": 20944165, + "step": 4123 + }, + { + "epoch": 0.3099353675033819, + "flos": 29168976533760.0, + "grad_norm": 1.7574684993259615, + "learning_rate": 3.2330102463473867e-06, + "loss": 0.9765, + "num_input_tokens_seen": 20964510, + "step": 4124 + }, + { + "epoch": 0.3100105215692169, + "flos": 21824220458880.0, + "grad_norm": 1.6707571695651406, + "learning_rate": 3.2326268941862927e-06, + "loss": 0.9886, + "num_input_tokens_seen": 20988250, + "step": 4125 + }, + { + "epoch": 0.31008567563505185, + "flos": 29135441030400.0, + "grad_norm": 1.407953544435531, + "learning_rate": 3.2322434689873353e-06, + "loss": 1.0353, + "num_input_tokens_seen": 21011645, + "step": 4126 + }, + { + "epoch": 0.3101608297008868, + "flos": 22353398110080.0, + "grad_norm": 1.3949105737933938, + "learning_rate": 3.231859970773234e-06, + "loss": 0.9936, + "num_input_tokens_seen": 21037140, + "step": 4127 + }, + { + "epoch": 0.3102359837667218, + "flos": 22252073328000.0, + "grad_norm": 1.7685546875, + "learning_rate": 3.231476399566712e-06, + "loss": 0.9125, + "num_input_tokens_seen": 21060290, + "step": 4128 + }, + { + "epoch": 0.31031113783255676, + "flos": 25041441133440.0, + "grad_norm": 1.3908444777892277, + "learning_rate": 3.231092755390498e-06, + "loss": 1.0621, + "num_input_tokens_seen": 21086065, + "step": 4129 + }, + { + "epoch": 0.31038629189839173, + "flos": 58054162164480.0, + "grad_norm": 0.8634333756990679, + "learning_rate": 3.230709038267324e-06, + "loss": 0.7521, + "num_input_tokens_seen": 21161750, + "step": 4130 + }, + { + "epoch": 0.31046144596422665, + "flos": 27370390233600.0, + "grad_norm": 1.805775830289691, + "learning_rate": 3.2303252482199265e-06, + "loss": 0.9356, + "num_input_tokens_seen": 21184190, + "step": 4131 + }, + { + "epoch": 0.3105366000300616, + "flos": 30864478285440.0, + "grad_norm": 1.6296627931559868, + "learning_rate": 3.2299413852710466e-06, + "loss": 0.8783, + "num_input_tokens_seen": 21206610, + "step": 4132 + }, + { + "epoch": 0.3106117540958966, + "flos": 19635554597760.0, + "grad_norm": 2.1825464019561864, + "learning_rate": 3.2295574494434307e-06, + "loss": 0.9679, + "num_input_tokens_seen": 21227870, + "step": 4133 + }, + { + "epoch": 0.31068690816173156, + "flos": 22849866270720.0, + "grad_norm": 1.71813551582124, + "learning_rate": 3.229173440759827e-06, + "loss": 1.0174, + "num_input_tokens_seen": 21249700, + "step": 4134 + }, + { + "epoch": 0.3107620622275665, + "flos": 23843089900800.0, + "grad_norm": 1.519914201272472, + "learning_rate": 3.22878935924299e-06, + "loss": 0.9989, + "num_input_tokens_seen": 21274860, + "step": 4135 + }, + { + "epoch": 0.3108372162934015, + "flos": 28424040854400.0, + "grad_norm": 2.740997143036593, + "learning_rate": 3.2284052049156783e-06, + "loss": 0.9208, + "num_input_tokens_seen": 21296745, + "step": 4136 + }, + { + "epoch": 0.3109123703592364, + "flos": 20064592615680.0, + "grad_norm": 1.7432627417578646, + "learning_rate": 3.2280209778006545e-06, + "loss": 0.989, + "num_input_tokens_seen": 21319150, + "step": 4137 + }, + { + "epoch": 0.3109875244250714, + "flos": 25780810204800.0, + "grad_norm": 1.6292340065518085, + "learning_rate": 3.227636677920685e-06, + "loss": 1.0139, + "num_input_tokens_seen": 21345290, + "step": 4138 + }, + { + "epoch": 0.31106267849090635, + "flos": 67535445352320.0, + "grad_norm": 1.1186122745579121, + "learning_rate": 3.227252305298542e-06, + "loss": 0.9376, + "num_input_tokens_seen": 21427300, + "step": 4139 + }, + { + "epoch": 0.3111378325567413, + "flos": 21088658229120.0, + "grad_norm": 2.114105571525157, + "learning_rate": 3.2268678599570002e-06, + "loss": 1.0253, + "num_input_tokens_seen": 21452665, + "step": 4140 + }, + { + "epoch": 0.3112129866225763, + "flos": 18687825699840.0, + "grad_norm": 2.1424720758779445, + "learning_rate": 3.2264833419188397e-06, + "loss": 0.9775, + "num_input_tokens_seen": 21471810, + "step": 4141 + }, + { + "epoch": 0.31128814068841126, + "flos": 21194795433600.0, + "grad_norm": 1.8387313038947162, + "learning_rate": 3.2260987512068443e-06, + "loss": 0.8137, + "num_input_tokens_seen": 21498360, + "step": 4142 + }, + { + "epoch": 0.3113632947542462, + "flos": 17064889735680.0, + "grad_norm": 1.418957074215706, + "learning_rate": 3.225714087843803e-06, + "loss": 1.0094, + "num_input_tokens_seen": 21521145, + "step": 4143 + }, + { + "epoch": 0.31143844882008115, + "flos": 20206707448320.0, + "grad_norm": 1.7110373411267512, + "learning_rate": 3.225329351852509e-06, + "loss": 0.99, + "num_input_tokens_seen": 21544790, + "step": 4144 + }, + { + "epoch": 0.3115136028859161, + "flos": 24759294456960.0, + "grad_norm": 1.8461882422984328, + "learning_rate": 3.2249445432557584e-06, + "loss": 0.9649, + "num_input_tokens_seen": 21570090, + "step": 4145 + }, + { + "epoch": 0.3115887569517511, + "flos": 18898304428800.0, + "grad_norm": 1.5684559033475725, + "learning_rate": 3.224559662076353e-06, + "loss": 0.9336, + "num_input_tokens_seen": 21590870, + "step": 4146 + }, + { + "epoch": 0.31166391101758606, + "flos": 27653578404480.0, + "grad_norm": 1.515399483488996, + "learning_rate": 3.224174708337098e-06, + "loss": 0.9555, + "num_input_tokens_seen": 21619045, + "step": 4147 + }, + { + "epoch": 0.31173906508342103, + "flos": 24228931656960.0, + "grad_norm": 1.6034210586076645, + "learning_rate": 3.2237896820608047e-06, + "loss": 0.9837, + "num_input_tokens_seen": 21642830, + "step": 4148 + }, + { + "epoch": 0.311814219149256, + "flos": 22352320702080.0, + "grad_norm": 1.393944967001805, + "learning_rate": 3.223404583270286e-06, + "loss": 1.0702, + "num_input_tokens_seen": 21664735, + "step": 4149 + }, + { + "epoch": 0.3118893732150909, + "flos": 18295375841280.0, + "grad_norm": 1.5346281196355591, + "learning_rate": 3.223019411988361e-06, + "loss": 1.0385, + "num_input_tokens_seen": 21685935, + "step": 4150 + }, + { + "epoch": 0.3119645272809259, + "flos": 28746618445440.0, + "grad_norm": 2.0073126382199096, + "learning_rate": 3.2226341682378525e-06, + "loss": 0.9807, + "num_input_tokens_seen": 21706995, + "step": 4151 + }, + { + "epoch": 0.31203968134676086, + "flos": 21547676304000.0, + "grad_norm": 1.3548284331933997, + "learning_rate": 3.2222488520415876e-06, + "loss": 0.9894, + "num_input_tokens_seen": 21732440, + "step": 4152 + }, + { + "epoch": 0.3121148354125958, + "flos": 16958465222400.0, + "grad_norm": 1.912956447610055, + "learning_rate": 3.221863463422399e-06, + "loss": 1.0406, + "num_input_tokens_seen": 21752965, + "step": 4153 + }, + { + "epoch": 0.3121899894784308, + "flos": 16250476838400.0, + "grad_norm": 1.5119985868507926, + "learning_rate": 3.2214780024031204e-06, + "loss": 1.0033, + "num_input_tokens_seen": 21774235, + "step": 4154 + }, + { + "epoch": 0.31226514354426577, + "flos": 24331405674240.0, + "grad_norm": 1.6635647597966337, + "learning_rate": 3.221092469006593e-06, + "loss": 0.898, + "num_input_tokens_seen": 21799400, + "step": 4155 + }, + { + "epoch": 0.3123402976101007, + "flos": 19182462266880.0, + "grad_norm": 1.5847246013820009, + "learning_rate": 3.220706863255661e-06, + "loss": 0.9976, + "num_input_tokens_seen": 21822095, + "step": 4156 + }, + { + "epoch": 0.31241545167593565, + "flos": 21229049208960.0, + "grad_norm": 1.4431825356560182, + "learning_rate": 3.220321185173173e-06, + "loss": 0.993, + "num_input_tokens_seen": 21843650, + "step": 4157 + }, + { + "epoch": 0.3124906057417706, + "flos": 21370445769600.0, + "grad_norm": 1.4971630766932176, + "learning_rate": 3.219935434781982e-06, + "loss": 0.9302, + "num_input_tokens_seen": 21867035, + "step": 4158 + }, + { + "epoch": 0.3125657598076056, + "flos": 21752049720960.0, + "grad_norm": 1.5785935716610993, + "learning_rate": 3.2195496121049447e-06, + "loss": 1.0588, + "num_input_tokens_seen": 21887855, + "step": 4159 + }, + { + "epoch": 0.31264091387344056, + "flos": 22887603665280.0, + "grad_norm": 1.6163963557254422, + "learning_rate": 3.219163717164923e-06, + "loss": 0.9513, + "num_input_tokens_seen": 21913205, + "step": 4160 + }, + { + "epoch": 0.31271606793927553, + "flos": 17558736203520.0, + "grad_norm": 1.9082734671896044, + "learning_rate": 3.218777749984782e-06, + "loss": 1.027, + "num_input_tokens_seen": 21933975, + "step": 4161 + }, + { + "epoch": 0.3127912220051105, + "flos": 27968219089920.0, + "grad_norm": 1.6876895939319851, + "learning_rate": 3.2183917105873934e-06, + "loss": 0.822, + "num_input_tokens_seen": 21959515, + "step": 4162 + }, + { + "epoch": 0.3128663760709454, + "flos": 22252576118400.0, + "grad_norm": 1.6375572223711503, + "learning_rate": 3.21800559899563e-06, + "loss": 0.912, + "num_input_tokens_seen": 21982160, + "step": 4163 + }, + { + "epoch": 0.3129415301367804, + "flos": 35730556744320.0, + "grad_norm": 1.933968062727864, + "learning_rate": 3.217619415232371e-06, + "loss": 1.0438, + "num_input_tokens_seen": 22005110, + "step": 4164 + }, + { + "epoch": 0.31301668420261536, + "flos": 19535199482880.0, + "grad_norm": 2.054566470450699, + "learning_rate": 3.217233159320498e-06, + "loss": 1.0179, + "num_input_tokens_seen": 22024930, + "step": 4165 + }, + { + "epoch": 0.31309183826845033, + "flos": 23701190549760.0, + "grad_norm": 1.7760491990929972, + "learning_rate": 3.2168468312829005e-06, + "loss": 0.9003, + "num_input_tokens_seen": 22047680, + "step": 4166 + }, + { + "epoch": 0.3131669923342853, + "flos": 16880763790080.0, + "grad_norm": 1.53425668854644, + "learning_rate": 3.2164604311424677e-06, + "loss": 0.9523, + "num_input_tokens_seen": 22067460, + "step": 4167 + }, + { + "epoch": 0.31324214640012027, + "flos": 68694478992000.0, + "grad_norm": 0.8580214157432873, + "learning_rate": 3.2160739589220968e-06, + "loss": 0.8255, + "num_input_tokens_seen": 22151450, + "step": 4168 + }, + { + "epoch": 0.3133173004659552, + "flos": 25742103143040.0, + "grad_norm": 1.7349414072394362, + "learning_rate": 3.2156874146446872e-06, + "loss": 0.9527, + "num_input_tokens_seen": 22176215, + "step": 4169 + }, + { + "epoch": 0.31339245453179015, + "flos": 22041379117440.0, + "grad_norm": 2.2950973378734334, + "learning_rate": 3.2153007983331437e-06, + "loss": 1.0527, + "num_input_tokens_seen": 22201705, + "step": 4170 + }, + { + "epoch": 0.3134676085976251, + "flos": 13852553310720.0, + "grad_norm": 2.2263993052858124, + "learning_rate": 3.214914110010373e-06, + "loss": 0.9697, + "num_input_tokens_seen": 22221990, + "step": 4171 + }, + { + "epoch": 0.3135427626634601, + "flos": 22535584721280.0, + "grad_norm": 1.6974454900784648, + "learning_rate": 3.214527349699289e-06, + "loss": 1.0094, + "num_input_tokens_seen": 22245330, + "step": 4172 + }, + { + "epoch": 0.31361791672929507, + "flos": 17197020587520.0, + "grad_norm": 1.9039266842703104, + "learning_rate": 3.21414051742281e-06, + "loss": 0.9722, + "num_input_tokens_seen": 22264575, + "step": 4173 + }, + { + "epoch": 0.31369307079513004, + "flos": 24157551018240.0, + "grad_norm": 2.457306329463233, + "learning_rate": 3.2137536132038552e-06, + "loss": 1.0009, + "num_input_tokens_seen": 22284475, + "step": 4174 + }, + { + "epoch": 0.31376822486096495, + "flos": 16146243054720.0, + "grad_norm": 1.7288871064227302, + "learning_rate": 3.213366637065351e-06, + "loss": 0.9975, + "num_input_tokens_seen": 22307455, + "step": 4175 + }, + { + "epoch": 0.3138433789267999, + "flos": 21403981272960.0, + "grad_norm": 1.532695788374101, + "learning_rate": 3.212979589030228e-06, + "loss": 1.0368, + "num_input_tokens_seen": 22330345, + "step": 4176 + }, + { + "epoch": 0.3139185329926349, + "flos": 18931229400960.0, + "grad_norm": 1.7300896842951077, + "learning_rate": 3.212592469121419e-06, + "loss": 1.0596, + "num_input_tokens_seen": 22351410, + "step": 4177 + }, + { + "epoch": 0.31399368705846986, + "flos": 18264857080320.0, + "grad_norm": 1.5909467240626074, + "learning_rate": 3.2122052773618625e-06, + "loss": 1.0756, + "num_input_tokens_seen": 22375555, + "step": 4178 + }, + { + "epoch": 0.31406884112430483, + "flos": 25747095133440.0, + "grad_norm": 1.7777089844082608, + "learning_rate": 3.211818013774502e-06, + "loss": 1.0217, + "num_input_tokens_seen": 22399990, + "step": 4179 + }, + { + "epoch": 0.3141439951901398, + "flos": 21265062750720.0, + "grad_norm": 1.4187939977335542, + "learning_rate": 3.211430678382284e-06, + "loss": 1.0135, + "num_input_tokens_seen": 22425195, + "step": 4180 + }, + { + "epoch": 0.3142191492559748, + "flos": 45758938469760.0, + "grad_norm": 1.449755874998158, + "learning_rate": 3.211043271208159e-06, + "loss": 0.8976, + "num_input_tokens_seen": 22455150, + "step": 4181 + }, + { + "epoch": 0.3142943033218097, + "flos": 50804158682880.0, + "grad_norm": 0.7907416351893827, + "learning_rate": 3.2106557922750826e-06, + "loss": 0.7987, + "num_input_tokens_seen": 22522540, + "step": 4182 + }, + { + "epoch": 0.31436945738764466, + "flos": 28993613506560.0, + "grad_norm": 1.7931055752276144, + "learning_rate": 3.210268241606015e-06, + "loss": 0.9047, + "num_input_tokens_seen": 22546535, + "step": 4183 + }, + { + "epoch": 0.3144446114534796, + "flos": 55743413460480.0, + "grad_norm": 0.8316844880240232, + "learning_rate": 3.2098806192239196e-06, + "loss": 0.8564, + "num_input_tokens_seen": 22622750, + "step": 4184 + }, + { + "epoch": 0.3145197655193146, + "flos": 25429725014400.0, + "grad_norm": 2.0104727728991625, + "learning_rate": 3.209492925151765e-06, + "loss": 0.8361, + "num_input_tokens_seen": 22647205, + "step": 4185 + }, + { + "epoch": 0.31459491958514957, + "flos": 12934768556160.0, + "grad_norm": 1.7562553459982093, + "learning_rate": 3.209105159412522e-06, + "loss": 0.924, + "num_input_tokens_seen": 22668930, + "step": 4186 + }, + { + "epoch": 0.31467007365098454, + "flos": 22462659797760.0, + "grad_norm": 1.3578764449361822, + "learning_rate": 3.2087173220291695e-06, + "loss": 1.0435, + "num_input_tokens_seen": 22692385, + "step": 4187 + }, + { + "epoch": 0.31474522771681945, + "flos": 18970762475520.0, + "grad_norm": 1.6548963089006183, + "learning_rate": 3.2083294130246865e-06, + "loss": 0.9353, + "num_input_tokens_seen": 22715300, + "step": 4188 + }, + { + "epoch": 0.3148203817826544, + "flos": 22459319832960.0, + "grad_norm": 1.3654447510785581, + "learning_rate": 3.207941432422059e-06, + "loss": 0.9905, + "num_input_tokens_seen": 22740450, + "step": 4189 + }, + { + "epoch": 0.3148955358484894, + "flos": 23945240695680.0, + "grad_norm": 1.6773802768832646, + "learning_rate": 3.2075533802442757e-06, + "loss": 0.9095, + "num_input_tokens_seen": 22767665, + "step": 4190 + }, + { + "epoch": 0.31497068991432436, + "flos": 57079318498560.0, + "grad_norm": 0.8865433237921764, + "learning_rate": 3.2071652565143313e-06, + "loss": 0.879, + "num_input_tokens_seen": 22841765, + "step": 4191 + }, + { + "epoch": 0.31504584398015933, + "flos": 13496476129920.0, + "grad_norm": 1.6099508792183983, + "learning_rate": 3.206777061255223e-06, + "loss": 0.9789, + "num_input_tokens_seen": 22863840, + "step": 4192 + }, + { + "epoch": 0.3151209980459943, + "flos": 20488279507200.0, + "grad_norm": 1.9654973120524237, + "learning_rate": 3.206388794489952e-06, + "loss": 0.9691, + "num_input_tokens_seen": 22887720, + "step": 4193 + }, + { + "epoch": 0.3151961521118293, + "flos": 24123117674880.0, + "grad_norm": 1.7209957363290853, + "learning_rate": 3.2060004562415258e-06, + "loss": 0.9995, + "num_input_tokens_seen": 22908930, + "step": 4194 + }, + { + "epoch": 0.3152713061776642, + "flos": 27546794755200.0, + "grad_norm": 1.4617579803014087, + "learning_rate": 3.205612046532954e-06, + "loss": 0.8641, + "num_input_tokens_seen": 22935015, + "step": 4195 + }, + { + "epoch": 0.31534646024349916, + "flos": 21512524688640.0, + "grad_norm": 1.718232302293309, + "learning_rate": 3.2052235653872525e-06, + "loss": 1.0739, + "num_input_tokens_seen": 22958170, + "step": 4196 + }, + { + "epoch": 0.31542161430933413, + "flos": 20205665953920.0, + "grad_norm": 2.0174902040983524, + "learning_rate": 3.2048350128274395e-06, + "loss": 0.9853, + "num_input_tokens_seen": 22981375, + "step": 4197 + }, + { + "epoch": 0.3154967683751691, + "flos": 22288481919360.0, + "grad_norm": 1.465673349066961, + "learning_rate": 3.2044463888765384e-06, + "loss": 1.0182, + "num_input_tokens_seen": 23003840, + "step": 4198 + }, + { + "epoch": 0.31557192244100407, + "flos": 19888224007680.0, + "grad_norm": 1.4215612379711267, + "learning_rate": 3.204057693557576e-06, + "loss": 1.0303, + "num_input_tokens_seen": 23026650, + "step": 4199 + }, + { + "epoch": 0.31564707650683904, + "flos": 22812990802560.0, + "grad_norm": 1.3622352062814505, + "learning_rate": 3.203668926893585e-06, + "loss": 0.9846, + "num_input_tokens_seen": 23051620, + "step": 4200 + }, + { + "epoch": 0.31572223057267396, + "flos": 55721513963520.0, + "grad_norm": 0.8157862449073691, + "learning_rate": 3.2032800889076014e-06, + "loss": 0.8306, + "num_input_tokens_seen": 23130900, + "step": 4201 + }, + { + "epoch": 0.3157973846385089, + "flos": 20521491788160.0, + "grad_norm": 1.4861238316658016, + "learning_rate": 3.2028911796226642e-06, + "loss": 0.9071, + "num_input_tokens_seen": 23154305, + "step": 4202 + }, + { + "epoch": 0.3158725387043439, + "flos": 67783697389440.0, + "grad_norm": 0.9130531019538841, + "learning_rate": 3.2025021990618193e-06, + "loss": 0.8682, + "num_input_tokens_seen": 23228890, + "step": 4203 + }, + { + "epoch": 0.31594769277017887, + "flos": 21938366396160.0, + "grad_norm": 1.3027369371929451, + "learning_rate": 3.2021131472481135e-06, + "loss": 0.9271, + "num_input_tokens_seen": 23255950, + "step": 4204 + }, + { + "epoch": 0.31602284683601384, + "flos": 20168754572160.0, + "grad_norm": 1.3336746951065517, + "learning_rate": 3.2017240242046005e-06, + "loss": 0.9469, + "num_input_tokens_seen": 23279670, + "step": 4205 + }, + { + "epoch": 0.3160980009018488, + "flos": 22247404560000.0, + "grad_norm": 1.9172601886577532, + "learning_rate": 3.2013348299543382e-06, + "loss": 1.0245, + "num_input_tokens_seen": 23301490, + "step": 4206 + }, + { + "epoch": 0.3161731549676838, + "flos": 38940163735680.0, + "grad_norm": 1.5908819084817765, + "learning_rate": 3.2009455645203864e-06, + "loss": 0.9125, + "num_input_tokens_seen": 23325125, + "step": 4207 + }, + { + "epoch": 0.3162483090335187, + "flos": 20206276485120.0, + "grad_norm": 1.8127006551234097, + "learning_rate": 3.2005562279258113e-06, + "loss": 0.9947, + "num_input_tokens_seen": 23348120, + "step": 4208 + }, + { + "epoch": 0.31632346309935366, + "flos": 15794116369920.0, + "grad_norm": 2.4901856421395805, + "learning_rate": 3.200166820193682e-06, + "loss": 0.8811, + "num_input_tokens_seen": 23371045, + "step": 4209 + }, + { + "epoch": 0.31639861716518863, + "flos": 21052608773760.0, + "grad_norm": 1.8089091806551933, + "learning_rate": 3.1997773413470736e-06, + "loss": 1.0015, + "num_input_tokens_seen": 23393255, + "step": 4210 + }, + { + "epoch": 0.3164737712310236, + "flos": 15476638510080.0, + "grad_norm": 1.552947177339883, + "learning_rate": 3.1993877914090632e-06, + "loss": 0.9779, + "num_input_tokens_seen": 23415865, + "step": 4211 + }, + { + "epoch": 0.3165489252968586, + "flos": 27011009001600.0, + "grad_norm": 2.2755565926595813, + "learning_rate": 3.198998170402733e-06, + "loss": 1.0108, + "num_input_tokens_seen": 23439640, + "step": 4212 + }, + { + "epoch": 0.31662407936269354, + "flos": 14805561507840.0, + "grad_norm": 1.941701947111352, + "learning_rate": 3.1986084783511708e-06, + "loss": 0.9262, + "num_input_tokens_seen": 23461810, + "step": 4213 + }, + { + "epoch": 0.31669923342852846, + "flos": 24866006279040.0, + "grad_norm": 1.5480426273292485, + "learning_rate": 3.198218715277466e-06, + "loss": 0.9915, + "num_input_tokens_seen": 23485415, + "step": 4214 + }, + { + "epoch": 0.31677438749436343, + "flos": 28393450266240.0, + "grad_norm": 1.5779195972889186, + "learning_rate": 3.1978288812047136e-06, + "loss": 0.9351, + "num_input_tokens_seen": 23509580, + "step": 4215 + }, + { + "epoch": 0.3168495415601984, + "flos": 27510529818240.0, + "grad_norm": 1.1831012872172546, + "learning_rate": 3.1974389761560137e-06, + "loss": 0.9916, + "num_input_tokens_seen": 23533525, + "step": 4216 + }, + { + "epoch": 0.31692469562603337, + "flos": 28253669817600.0, + "grad_norm": 1.443558819816124, + "learning_rate": 3.197049000154469e-06, + "loss": 0.9535, + "num_input_tokens_seen": 23559205, + "step": 4217 + }, + { + "epoch": 0.31699984969186834, + "flos": 40710745226880.0, + "grad_norm": 1.4543356825461085, + "learning_rate": 3.1966589532231876e-06, + "loss": 0.8447, + "num_input_tokens_seen": 23587865, + "step": 4218 + }, + { + "epoch": 0.3170750037577033, + "flos": 18582981384960.0, + "grad_norm": 1.4809569507552318, + "learning_rate": 3.1962688353852805e-06, + "loss": 1.0881, + "num_input_tokens_seen": 23609250, + "step": 4219 + }, + { + "epoch": 0.3171501578235382, + "flos": 21264416305920.0, + "grad_norm": 1.5556295789786874, + "learning_rate": 3.195878646663864e-06, + "loss": 0.9253, + "num_input_tokens_seen": 23633225, + "step": 4220 + }, + { + "epoch": 0.3172253118893732, + "flos": 21229587912960.0, + "grad_norm": 1.622105955757734, + "learning_rate": 3.195488387082059e-06, + "loss": 0.9571, + "num_input_tokens_seen": 23653255, + "step": 4221 + }, + { + "epoch": 0.31730046595520817, + "flos": 23806286259840.0, + "grad_norm": 1.4673732229706704, + "learning_rate": 3.1950980566629886e-06, + "loss": 0.8541, + "num_input_tokens_seen": 23677145, + "step": 4222 + }, + { + "epoch": 0.31737562002104314, + "flos": 55562591566080.0, + "grad_norm": 0.9050154333748163, + "learning_rate": 3.194707655429782e-06, + "loss": 0.8567, + "num_input_tokens_seen": 23758205, + "step": 4223 + }, + { + "epoch": 0.3174507740868781, + "flos": 21511375453440.0, + "grad_norm": 2.1773202818887984, + "learning_rate": 3.1943171834055723e-06, + "loss": 1.0503, + "num_input_tokens_seen": 23781500, + "step": 4224 + }, + { + "epoch": 0.3175259281527131, + "flos": 23488377436800.0, + "grad_norm": 1.4998683871703569, + "learning_rate": 3.1939266406134963e-06, + "loss": 0.9804, + "num_input_tokens_seen": 23806310, + "step": 4225 + }, + { + "epoch": 0.31760108221854805, + "flos": 23731637483520.0, + "grad_norm": 1.8024299009027944, + "learning_rate": 3.1935360270766945e-06, + "loss": 0.9704, + "num_input_tokens_seen": 23829095, + "step": 4226 + }, + { + "epoch": 0.31767623628438296, + "flos": 19140846203520.0, + "grad_norm": 1.6653816116890086, + "learning_rate": 3.1931453428183125e-06, + "loss": 1.0283, + "num_input_tokens_seen": 23851640, + "step": 4227 + }, + { + "epoch": 0.31775139035021793, + "flos": 23523564965760.0, + "grad_norm": 1.528862707641778, + "learning_rate": 3.1927545878615005e-06, + "loss": 1.0245, + "num_input_tokens_seen": 23875400, + "step": 4228 + }, + { + "epoch": 0.3178265444160529, + "flos": 21865082336640.0, + "grad_norm": 1.5652181252763473, + "learning_rate": 3.1923637622294123e-06, + "loss": 1.0184, + "num_input_tokens_seen": 23900030, + "step": 4229 + }, + { + "epoch": 0.3179016984818879, + "flos": 24047355576960.0, + "grad_norm": 1.5909506953360417, + "learning_rate": 3.191972865945205e-06, + "loss": 0.9254, + "num_input_tokens_seen": 23923315, + "step": 4230 + }, + { + "epoch": 0.31797685254772284, + "flos": 24759509938560.0, + "grad_norm": 1.3704532529254616, + "learning_rate": 3.191581899032041e-06, + "loss": 0.9711, + "num_input_tokens_seen": 23947205, + "step": 4231 + }, + { + "epoch": 0.3180520066135578, + "flos": 21088586401920.0, + "grad_norm": 5.058242703190977, + "learning_rate": 3.1911908615130862e-06, + "loss": 0.9253, + "num_input_tokens_seen": 23972250, + "step": 4232 + }, + { + "epoch": 0.31812716067939273, + "flos": 26338962332160.0, + "grad_norm": 3.357851472524603, + "learning_rate": 3.1907997534115118e-06, + "loss": 0.9842, + "num_input_tokens_seen": 23993075, + "step": 4233 + }, + { + "epoch": 0.3182023147452277, + "flos": 19001999508480.0, + "grad_norm": 1.6116816462333814, + "learning_rate": 3.190408574750492e-06, + "loss": 1.0152, + "num_input_tokens_seen": 24018035, + "step": 4234 + }, + { + "epoch": 0.31827746881106267, + "flos": 19535163569280.0, + "grad_norm": 1.5358786399949063, + "learning_rate": 3.1900173255532057e-06, + "loss": 1.022, + "num_input_tokens_seen": 24039575, + "step": 4235 + }, + { + "epoch": 0.31835262287689764, + "flos": 16920476432640.0, + "grad_norm": 1.4848977222606785, + "learning_rate": 3.1896260058428364e-06, + "loss": 0.9224, + "num_input_tokens_seen": 24063105, + "step": 4236 + }, + { + "epoch": 0.3184277769427326, + "flos": 23767112321280.0, + "grad_norm": 1.4323045302304693, + "learning_rate": 3.1892346156425704e-06, + "loss": 1.0022, + "num_input_tokens_seen": 24087400, + "step": 4237 + }, + { + "epoch": 0.3185029310085676, + "flos": 23346190776960.0, + "grad_norm": 2.7915915768521558, + "learning_rate": 3.1888431549755998e-06, + "loss": 1.0178, + "num_input_tokens_seen": 24106130, + "step": 4238 + }, + { + "epoch": 0.31857808507440255, + "flos": 35170142060160.0, + "grad_norm": 1.5277098390134913, + "learning_rate": 3.18845162386512e-06, + "loss": 0.9415, + "num_input_tokens_seen": 24132280, + "step": 4239 + }, + { + "epoch": 0.31865323914023747, + "flos": 27406439688960.0, + "grad_norm": 1.3287943780042901, + "learning_rate": 3.1880600223343303e-06, + "loss": 0.9817, + "num_input_tokens_seen": 24157500, + "step": 4240 + }, + { + "epoch": 0.31872839320607244, + "flos": 12864501239040.0, + "grad_norm": 2.3748220075867805, + "learning_rate": 3.187668350406435e-06, + "loss": 1.0239, + "num_input_tokens_seen": 24179430, + "step": 4241 + }, + { + "epoch": 0.3188035472719074, + "flos": 25076700489600.0, + "grad_norm": 1.653560542112282, + "learning_rate": 3.187276608104642e-06, + "loss": 1.0251, + "num_input_tokens_seen": 24202340, + "step": 4242 + }, + { + "epoch": 0.3188787013377424, + "flos": 20770533924480.0, + "grad_norm": 1.6760861135041933, + "learning_rate": 3.1868847954521635e-06, + "loss": 0.9782, + "num_input_tokens_seen": 24227855, + "step": 4243 + }, + { + "epoch": 0.31895385540357735, + "flos": 13217669418240.0, + "grad_norm": 2.057157243288539, + "learning_rate": 3.1864929124722162e-06, + "loss": 0.9611, + "num_input_tokens_seen": 24250785, + "step": 4244 + }, + { + "epoch": 0.3190290094694123, + "flos": 17558879857920.0, + "grad_norm": 1.5675243468272249, + "learning_rate": 3.1861009591880206e-06, + "loss": 1.007, + "num_input_tokens_seen": 24272095, + "step": 4245 + }, + { + "epoch": 0.31910416353524723, + "flos": 18053049548160.0, + "grad_norm": 1.621409043133857, + "learning_rate": 3.1857089356228015e-06, + "loss": 0.9243, + "num_input_tokens_seen": 24294100, + "step": 4246 + }, + { + "epoch": 0.3191793176010822, + "flos": 27227772610560.0, + "grad_norm": 1.4387091651245412, + "learning_rate": 3.185316841799787e-06, + "loss": 0.9656, + "num_input_tokens_seen": 24319250, + "step": 4247 + }, + { + "epoch": 0.3192544716669172, + "flos": 17271202487040.0, + "grad_norm": 1.3803520079873068, + "learning_rate": 3.1849246777422108e-06, + "loss": 0.9892, + "num_input_tokens_seen": 24342940, + "step": 4248 + }, + { + "epoch": 0.31932962573275214, + "flos": 25641424805760.0, + "grad_norm": 1.5744699888206892, + "learning_rate": 3.1845324434733104e-06, + "loss": 0.9655, + "num_input_tokens_seen": 24367600, + "step": 4249 + }, + { + "epoch": 0.3194047797985871, + "flos": 30118321543680.0, + "grad_norm": 1.4298949377772672, + "learning_rate": 3.1841401390163263e-06, + "loss": 0.9776, + "num_input_tokens_seen": 24394815, + "step": 4250 + }, + { + "epoch": 0.3194799338644221, + "flos": 21300573502080.0, + "grad_norm": 1.6216412090931374, + "learning_rate": 3.183747764394505e-06, + "loss": 0.9777, + "num_input_tokens_seen": 24417115, + "step": 4251 + }, + { + "epoch": 0.31955508793025705, + "flos": 17522471266560.0, + "grad_norm": 1.625462686251817, + "learning_rate": 3.1833553196310956e-06, + "loss": 0.9887, + "num_input_tokens_seen": 24439015, + "step": 4252 + }, + { + "epoch": 0.31963024199609197, + "flos": 20629352845440.0, + "grad_norm": 1.540552879641765, + "learning_rate": 3.1829628047493523e-06, + "loss": 0.8898, + "num_input_tokens_seen": 24461330, + "step": 4253 + }, + { + "epoch": 0.31970539606192694, + "flos": 12934624901760.0, + "grad_norm": 1.5693477874669284, + "learning_rate": 3.1825702197725325e-06, + "loss": 0.9328, + "num_input_tokens_seen": 24482600, + "step": 4254 + }, + { + "epoch": 0.3197805501277619, + "flos": 15264148619520.0, + "grad_norm": 1.459336244775114, + "learning_rate": 3.1821775647239e-06, + "loss": 0.9141, + "num_input_tokens_seen": 24504480, + "step": 4255 + }, + { + "epoch": 0.3198557041935969, + "flos": 33829927390080.0, + "grad_norm": 1.7485519276283077, + "learning_rate": 3.1817848396267188e-06, + "loss": 0.9468, + "num_input_tokens_seen": 24529630, + "step": 4256 + }, + { + "epoch": 0.31993085825943185, + "flos": 20235322788480.0, + "grad_norm": 1.4731638713979471, + "learning_rate": 3.181392044504261e-06, + "loss": 1.0613, + "num_input_tokens_seen": 24551080, + "step": 4257 + }, + { + "epoch": 0.3200060123252668, + "flos": 19140307499520.0, + "grad_norm": 1.6472837197909045, + "learning_rate": 3.1809991793798e-06, + "loss": 0.8836, + "num_input_tokens_seen": 24573300, + "step": 4258 + }, + { + "epoch": 0.32008116639110173, + "flos": 22954207795200.0, + "grad_norm": 1.5190592917899113, + "learning_rate": 3.180606244276616e-06, + "loss": 0.9486, + "num_input_tokens_seen": 24596415, + "step": 4259 + }, + { + "epoch": 0.3201563204569367, + "flos": 18546788275200.0, + "grad_norm": 1.454503625707215, + "learning_rate": 3.180213239217991e-06, + "loss": 0.9095, + "num_input_tokens_seen": 24619265, + "step": 4260 + }, + { + "epoch": 0.3202314745227717, + "flos": 17239642231680.0, + "grad_norm": 1.953604067223179, + "learning_rate": 3.1798201642272123e-06, + "loss": 0.9184, + "num_input_tokens_seen": 24639620, + "step": 4261 + }, + { + "epoch": 0.32030662858860665, + "flos": 60371331603840.0, + "grad_norm": 0.8304774860773088, + "learning_rate": 3.179427019327571e-06, + "loss": 0.795, + "num_input_tokens_seen": 24714825, + "step": 4262 + }, + { + "epoch": 0.3203817826544416, + "flos": 16640987362560.0, + "grad_norm": 1.500406845867147, + "learning_rate": 3.179033804542363e-06, + "loss": 0.9928, + "num_input_tokens_seen": 24738015, + "step": 4263 + }, + { + "epoch": 0.3204569367202766, + "flos": 19641480341760.0, + "grad_norm": 2.026764480066668, + "learning_rate": 3.178640519894886e-06, + "loss": 0.9951, + "num_input_tokens_seen": 24761120, + "step": 4264 + }, + { + "epoch": 0.3205320907861115, + "flos": 22428980640000.0, + "grad_norm": 1.8861149360731564, + "learning_rate": 3.178247165408446e-06, + "loss": 1.0363, + "num_input_tokens_seen": 24784545, + "step": 4265 + }, + { + "epoch": 0.32060724485194647, + "flos": 21652197396480.0, + "grad_norm": 1.3268165875234943, + "learning_rate": 3.1778537411063487e-06, + "loss": 0.9639, + "num_input_tokens_seen": 24808575, + "step": 4266 + }, + { + "epoch": 0.32068239891778144, + "flos": 14240513969280.0, + "grad_norm": 1.7471197130906964, + "learning_rate": 3.1774602470119076e-06, + "loss": 1.0517, + "num_input_tokens_seen": 24829425, + "step": 4267 + }, + { + "epoch": 0.3207575529836164, + "flos": 15577280933760.0, + "grad_norm": 1.5656317034975704, + "learning_rate": 3.177066683148438e-06, + "loss": 1.0058, + "num_input_tokens_seen": 24848805, + "step": 4268 + }, + { + "epoch": 0.3208327070494514, + "flos": 18332323136640.0, + "grad_norm": 1.5961684782260825, + "learning_rate": 3.17667304953926e-06, + "loss": 1.0228, + "num_input_tokens_seen": 24869905, + "step": 4269 + }, + { + "epoch": 0.32090786111528635, + "flos": 20837245795200.0, + "grad_norm": 1.4211705737034337, + "learning_rate": 3.176279346207698e-06, + "loss": 0.9812, + "num_input_tokens_seen": 24894610, + "step": 4270 + }, + { + "epoch": 0.3209830151811213, + "flos": 21123019745280.0, + "grad_norm": 1.6947841260383165, + "learning_rate": 3.1758855731770807e-06, + "loss": 0.9717, + "num_input_tokens_seen": 24917485, + "step": 4271 + }, + { + "epoch": 0.32105816924695624, + "flos": 67638027110400.0, + "grad_norm": 0.9001547283046516, + "learning_rate": 3.1754917304707405e-06, + "loss": 0.8462, + "num_input_tokens_seen": 24991070, + "step": 4272 + }, + { + "epoch": 0.3211333233127912, + "flos": 64304800790400.0, + "grad_norm": 0.6600434923652653, + "learning_rate": 3.1750978181120136e-06, + "loss": 0.7828, + "num_input_tokens_seen": 25075395, + "step": 4273 + }, + { + "epoch": 0.3212084773786262, + "flos": 20663965756800.0, + "grad_norm": 1.5387589983208319, + "learning_rate": 3.1747038361242417e-06, + "loss": 0.9467, + "num_input_tokens_seen": 25098450, + "step": 4274 + }, + { + "epoch": 0.32128363144446115, + "flos": 13040510711040.0, + "grad_norm": 1.8872866200471794, + "learning_rate": 3.1743097845307692e-06, + "loss": 0.9516, + "num_input_tokens_seen": 25119035, + "step": 4275 + }, + { + "epoch": 0.3213587855102961, + "flos": 31288452485760.0, + "grad_norm": 1.7454277663046056, + "learning_rate": 3.1739156633549445e-06, + "loss": 0.8615, + "num_input_tokens_seen": 25146055, + "step": 4276 + }, + { + "epoch": 0.3214339395761311, + "flos": 20912325534720.0, + "grad_norm": 1.2910705185385563, + "learning_rate": 3.1735214726201223e-06, + "loss": 0.9384, + "num_input_tokens_seen": 25170920, + "step": 4277 + }, + { + "epoch": 0.321509093641966, + "flos": 14376128440320.0, + "grad_norm": 1.6668154729534626, + "learning_rate": 3.1731272123496588e-06, + "loss": 0.7952, + "num_input_tokens_seen": 25192650, + "step": 4278 + }, + { + "epoch": 0.321584247707801, + "flos": 17276266304640.0, + "grad_norm": 1.5394299838135175, + "learning_rate": 3.1727328825669164e-06, + "loss": 0.929, + "num_input_tokens_seen": 25215545, + "step": 4279 + }, + { + "epoch": 0.32165940177363594, + "flos": 19351899550080.0, + "grad_norm": 1.6387309449089702, + "learning_rate": 3.172338483295259e-06, + "loss": 1.0083, + "num_input_tokens_seen": 25237535, + "step": 4280 + }, + { + "epoch": 0.3217345558394709, + "flos": 29661745593600.0, + "grad_norm": 1.5507707955803156, + "learning_rate": 3.1719440145580578e-06, + "loss": 0.9274, + "num_input_tokens_seen": 25263055, + "step": 4281 + }, + { + "epoch": 0.3218097099053059, + "flos": 25747418355840.0, + "grad_norm": 2.626633181463411, + "learning_rate": 3.1715494763786855e-06, + "loss": 1.0733, + "num_input_tokens_seen": 25287860, + "step": 4282 + }, + { + "epoch": 0.32188486397114086, + "flos": 13816288373760.0, + "grad_norm": 1.5263061120718224, + "learning_rate": 3.1711548687805214e-06, + "loss": 0.8534, + "num_input_tokens_seen": 25309495, + "step": 4283 + }, + { + "epoch": 0.3219600180369758, + "flos": 22817946879360.0, + "grad_norm": 1.934731073839262, + "learning_rate": 3.170760191786946e-06, + "loss": 1.0495, + "num_input_tokens_seen": 25333495, + "step": 4284 + }, + { + "epoch": 0.32203517210281074, + "flos": 22111754175360.0, + "grad_norm": 1.3310949889006112, + "learning_rate": 3.1703654454213473e-06, + "loss": 0.9933, + "num_input_tokens_seen": 25357900, + "step": 4285 + }, + { + "epoch": 0.3221103261686457, + "flos": 24301605185280.0, + "grad_norm": 2.0640983168377094, + "learning_rate": 3.169970629707113e-06, + "loss": 0.932, + "num_input_tokens_seen": 25382240, + "step": 4286 + }, + { + "epoch": 0.3221854802344807, + "flos": 23629486688640.0, + "grad_norm": 1.4849837008779487, + "learning_rate": 3.1695757446676396e-06, + "loss": 1.0834, + "num_input_tokens_seen": 25405900, + "step": 4287 + }, + { + "epoch": 0.32226063430031565, + "flos": 43747359488640.0, + "grad_norm": 1.6637797943817911, + "learning_rate": 3.169180790326324e-06, + "loss": 0.8764, + "num_input_tokens_seen": 25434605, + "step": 4288 + }, + { + "epoch": 0.3223357883661506, + "flos": 15469599444480.0, + "grad_norm": 1.4996506761058885, + "learning_rate": 3.1687857667065698e-06, + "loss": 0.974, + "num_input_tokens_seen": 25456080, + "step": 4289 + }, + { + "epoch": 0.3224109424319856, + "flos": 21617728139520.0, + "grad_norm": 2.0100934916906015, + "learning_rate": 3.1683906738317838e-06, + "loss": 0.9501, + "num_input_tokens_seen": 25480390, + "step": 4290 + }, + { + "epoch": 0.3224860964978205, + "flos": 23304107836800.0, + "grad_norm": 1.5208525808849533, + "learning_rate": 3.167995511725375e-06, + "loss": 1.0413, + "num_input_tokens_seen": 25504995, + "step": 4291 + }, + { + "epoch": 0.3225612505636555, + "flos": 21822927569280.0, + "grad_norm": 1.4992861638658805, + "learning_rate": 3.1676002804107607e-06, + "loss": 0.9293, + "num_input_tokens_seen": 25525815, + "step": 4292 + }, + { + "epoch": 0.32263640462949045, + "flos": 22953884572800.0, + "grad_norm": 1.7109312292532775, + "learning_rate": 3.167204979911359e-06, + "loss": 1.0087, + "num_input_tokens_seen": 25549505, + "step": 4293 + }, + { + "epoch": 0.3227115586953254, + "flos": 18717662102400.0, + "grad_norm": 1.8792042327419078, + "learning_rate": 3.166809610250592e-06, + "loss": 0.9228, + "num_input_tokens_seen": 25571455, + "step": 4294 + }, + { + "epoch": 0.3227867127611604, + "flos": 21688534160640.0, + "grad_norm": 1.7324917853354695, + "learning_rate": 3.1664141714518876e-06, + "loss": 0.9708, + "num_input_tokens_seen": 25593660, + "step": 4295 + }, + { + "epoch": 0.32286186682699536, + "flos": 39759353141760.0, + "grad_norm": 1.3990492521591371, + "learning_rate": 3.1660186635386773e-06, + "loss": 0.7954, + "num_input_tokens_seen": 25623235, + "step": 4296 + }, + { + "epoch": 0.32293702089283033, + "flos": 26594073866880.0, + "grad_norm": 1.7185157616264024, + "learning_rate": 3.1656230865343964e-06, + "loss": 0.825, + "num_input_tokens_seen": 25644820, + "step": 4297 + }, + { + "epoch": 0.32301217495866524, + "flos": 26735290859520.0, + "grad_norm": 1.3652494808993063, + "learning_rate": 3.165227440462484e-06, + "loss": 0.9925, + "num_input_tokens_seen": 25669795, + "step": 4298 + }, + { + "epoch": 0.3230873290245002, + "flos": 63664996734720.0, + "grad_norm": 0.9480840549165038, + "learning_rate": 3.164831725346383e-06, + "loss": 0.8976, + "num_input_tokens_seen": 25747485, + "step": 4299 + }, + { + "epoch": 0.3231624830903352, + "flos": 22040984067840.0, + "grad_norm": 1.5303017930724838, + "learning_rate": 3.1644359412095432e-06, + "loss": 1.0413, + "num_input_tokens_seen": 25771285, + "step": 4300 + }, + { + "epoch": 0.32323763715617015, + "flos": 30652886234880.0, + "grad_norm": 1.6985235085419996, + "learning_rate": 3.164040088075414e-06, + "loss": 0.9592, + "num_input_tokens_seen": 25795175, + "step": 4301 + }, + { + "epoch": 0.3233127912220051, + "flos": 23277360003840.0, + "grad_norm": 1.3605543260701618, + "learning_rate": 3.1636441659674528e-06, + "loss": 1.037, + "num_input_tokens_seen": 25821620, + "step": 4302 + }, + { + "epoch": 0.3233879452878401, + "flos": 20452876496640.0, + "grad_norm": 1.6579582473872332, + "learning_rate": 3.1632481749091185e-06, + "loss": 1.0183, + "num_input_tokens_seen": 25844350, + "step": 4303 + }, + { + "epoch": 0.323463099353675, + "flos": 30935966664960.0, + "grad_norm": 1.4978090020307433, + "learning_rate": 3.1628521149238757e-06, + "loss": 0.96, + "num_input_tokens_seen": 25871660, + "step": 4304 + }, + { + "epoch": 0.32353825341951, + "flos": 20805577799040.0, + "grad_norm": 1.5171493258324242, + "learning_rate": 3.1624559860351917e-06, + "loss": 1.0308, + "num_input_tokens_seen": 25895920, + "step": 4305 + }, + { + "epoch": 0.32361340748534495, + "flos": 22643266210560.0, + "grad_norm": 1.4695273432256348, + "learning_rate": 3.1620597882665393e-06, + "loss": 0.9021, + "num_input_tokens_seen": 25923590, + "step": 4306 + }, + { + "epoch": 0.3236885615511799, + "flos": 21787955521920.0, + "grad_norm": 1.392512465717397, + "learning_rate": 3.1616635216413952e-06, + "loss": 1.0331, + "num_input_tokens_seen": 25946180, + "step": 4307 + }, + { + "epoch": 0.3237637156170149, + "flos": 23552934491520.0, + "grad_norm": 1.3736921940062106, + "learning_rate": 3.161267186183239e-06, + "loss": 1.067, + "num_input_tokens_seen": 25970015, + "step": 4308 + }, + { + "epoch": 0.32383886968284986, + "flos": 37357335463680.0, + "grad_norm": 3.2625153990634246, + "learning_rate": 3.160870781915555e-06, + "loss": 0.9455, + "num_input_tokens_seen": 25995780, + "step": 4309 + }, + { + "epoch": 0.3239140237486848, + "flos": 20946687050880.0, + "grad_norm": 1.2645657663313994, + "learning_rate": 3.160474308861832e-06, + "loss": 1.0717, + "num_input_tokens_seen": 26021110, + "step": 4310 + }, + { + "epoch": 0.32398917781451975, + "flos": 13426388380800.0, + "grad_norm": 2.019811378397245, + "learning_rate": 3.160077767045562e-06, + "loss": 1.0666, + "num_input_tokens_seen": 26036950, + "step": 4311 + }, + { + "epoch": 0.3240643318803547, + "flos": 17700420072960.0, + "grad_norm": 1.5715760418818017, + "learning_rate": 3.1596811564902426e-06, + "loss": 1.014, + "num_input_tokens_seen": 26059300, + "step": 4312 + }, + { + "epoch": 0.3241394859461897, + "flos": 23558824321920.0, + "grad_norm": 1.591519908563587, + "learning_rate": 3.159284477219374e-06, + "loss": 0.9678, + "num_input_tokens_seen": 26082235, + "step": 4313 + }, + { + "epoch": 0.32421464001202466, + "flos": 23344431010560.0, + "grad_norm": 1.4360905247157028, + "learning_rate": 3.1588877292564606e-06, + "loss": 0.8374, + "num_input_tokens_seen": 26106700, + "step": 4314 + }, + { + "epoch": 0.3242897940778596, + "flos": 31179729502080.0, + "grad_norm": 1.3754381868762253, + "learning_rate": 3.1584909126250116e-06, + "loss": 0.9796, + "num_input_tokens_seen": 26132405, + "step": 4315 + }, + { + "epoch": 0.3243649481436946, + "flos": 18441297515520.0, + "grad_norm": 1.6869775175169182, + "learning_rate": 3.15809402734854e-06, + "loss": 0.9022, + "num_input_tokens_seen": 26155070, + "step": 4316 + }, + { + "epoch": 0.3244401022095295, + "flos": 23947575079680.0, + "grad_norm": 1.4731110293049992, + "learning_rate": 3.1576970734505624e-06, + "loss": 0.9884, + "num_input_tokens_seen": 26178240, + "step": 4317 + }, + { + "epoch": 0.3245152562753645, + "flos": 20065095406080.0, + "grad_norm": 1.4593527455236137, + "learning_rate": 3.1573000509546004e-06, + "loss": 1.0118, + "num_input_tokens_seen": 26200260, + "step": 4318 + }, + { + "epoch": 0.32459041034119945, + "flos": 32735630373120.0, + "grad_norm": 1.99033739536484, + "learning_rate": 3.1569029598841788e-06, + "loss": 0.8852, + "num_input_tokens_seen": 26226040, + "step": 4319 + }, + { + "epoch": 0.3246655644070344, + "flos": 23735085189120.0, + "grad_norm": 1.4357212505700327, + "learning_rate": 3.1565058002628268e-06, + "loss": 0.9071, + "num_input_tokens_seen": 26250550, + "step": 4320 + }, + { + "epoch": 0.3247407184728694, + "flos": 24263185432320.0, + "grad_norm": 1.5706809451442418, + "learning_rate": 3.156108572114077e-06, + "loss": 0.9746, + "num_input_tokens_seen": 26274555, + "step": 4321 + }, + { + "epoch": 0.32481587253870436, + "flos": 38524629231360.0, + "grad_norm": 1.4889267049956223, + "learning_rate": 3.155711275461468e-06, + "loss": 0.8995, + "num_input_tokens_seen": 26300865, + "step": 4322 + }, + { + "epoch": 0.3248910266045393, + "flos": 32908443534720.0, + "grad_norm": 2.4230302577945633, + "learning_rate": 3.15531391032854e-06, + "loss": 0.8623, + "num_input_tokens_seen": 26329680, + "step": 4323 + }, + { + "epoch": 0.32496618067037425, + "flos": 19711388522880.0, + "grad_norm": 1.5249469810399672, + "learning_rate": 3.1549164767388386e-06, + "loss": 1.0877, + "num_input_tokens_seen": 26353350, + "step": 4324 + }, + { + "epoch": 0.3250413347362092, + "flos": 24476573162880.0, + "grad_norm": 1.986935502985359, + "learning_rate": 3.1545189747159136e-06, + "loss": 1.0245, + "num_input_tokens_seen": 26373000, + "step": 4325 + }, + { + "epoch": 0.3251164888020442, + "flos": 10746282263040.0, + "grad_norm": 2.0074393671553756, + "learning_rate": 3.1541214042833187e-06, + "loss": 0.9268, + "num_input_tokens_seen": 26393925, + "step": 4326 + }, + { + "epoch": 0.32519164286787916, + "flos": 22640895912960.0, + "grad_norm": 1.5370660215902154, + "learning_rate": 3.153723765464611e-06, + "loss": 0.9808, + "num_input_tokens_seen": 26417720, + "step": 4327 + }, + { + "epoch": 0.32526679693371413, + "flos": 13569580621440.0, + "grad_norm": 1.7413623720428215, + "learning_rate": 3.1533260582833527e-06, + "loss": 0.9731, + "num_input_tokens_seen": 26437665, + "step": 4328 + }, + { + "epoch": 0.3253419509995491, + "flos": 20026388344320.0, + "grad_norm": 1.6505904759498233, + "learning_rate": 3.152928282763109e-06, + "loss": 0.8574, + "num_input_tokens_seen": 26462480, + "step": 4329 + }, + { + "epoch": 0.325417105065384, + "flos": 22917835117440.0, + "grad_norm": 1.8083172910245344, + "learning_rate": 3.152530438927449e-06, + "loss": 1.0414, + "num_input_tokens_seen": 26485455, + "step": 4330 + }, + { + "epoch": 0.325492259131219, + "flos": 22041486858240.0, + "grad_norm": 1.5039642719123898, + "learning_rate": 3.1521325267999476e-06, + "loss": 0.9275, + "num_input_tokens_seen": 26510695, + "step": 4331 + }, + { + "epoch": 0.32556741319705396, + "flos": 15687979165440.0, + "grad_norm": 1.665048942648472, + "learning_rate": 3.1517345464041817e-06, + "loss": 1.0571, + "num_input_tokens_seen": 26532760, + "step": 4332 + }, + { + "epoch": 0.3256425672628889, + "flos": 23205656142720.0, + "grad_norm": 1.5210196063090342, + "learning_rate": 3.1513364977637344e-06, + "loss": 1.0467, + "num_input_tokens_seen": 26558170, + "step": 4333 + }, + { + "epoch": 0.3257177213287239, + "flos": 20523502949760.0, + "grad_norm": 1.8219759302988552, + "learning_rate": 3.1509383809021905e-06, + "loss": 0.9786, + "num_input_tokens_seen": 26579970, + "step": 4334 + }, + { + "epoch": 0.32579287539455887, + "flos": 17977179709440.0, + "grad_norm": 1.506800178306251, + "learning_rate": 3.15054019584314e-06, + "loss": 1.0213, + "num_input_tokens_seen": 26601995, + "step": 4335 + }, + { + "epoch": 0.3258680294603938, + "flos": 13887992234880.0, + "grad_norm": 1.740741214947527, + "learning_rate": 3.150141942610178e-06, + "loss": 0.971, + "num_input_tokens_seen": 26621705, + "step": 4336 + }, + { + "epoch": 0.32594318352622875, + "flos": 24476178113280.0, + "grad_norm": 1.652553896695955, + "learning_rate": 3.149743621226901e-06, + "loss": 0.9968, + "num_input_tokens_seen": 26645115, + "step": 4337 + }, + { + "epoch": 0.3260183375920637, + "flos": 23241166894080.0, + "grad_norm": 1.6167915338620176, + "learning_rate": 3.149345231716912e-06, + "loss": 0.9706, + "num_input_tokens_seen": 26669670, + "step": 4338 + }, + { + "epoch": 0.3260934916578987, + "flos": 35270461261440.0, + "grad_norm": 1.6488271677556228, + "learning_rate": 3.148946774103817e-06, + "loss": 0.9998, + "num_input_tokens_seen": 26692395, + "step": 4339 + }, + { + "epoch": 0.32616864572373366, + "flos": 15794044542720.0, + "grad_norm": 1.354645756801042, + "learning_rate": 3.1485482484112257e-06, + "loss": 0.9266, + "num_input_tokens_seen": 26715095, + "step": 4340 + }, + { + "epoch": 0.32624379978956863, + "flos": 28181894129280.0, + "grad_norm": 1.80417866985507, + "learning_rate": 3.148149654662753e-06, + "loss": 0.8063, + "num_input_tokens_seen": 26737825, + "step": 4341 + }, + { + "epoch": 0.3263189538554036, + "flos": 25814453448960.0, + "grad_norm": 1.3836709068320368, + "learning_rate": 3.1477509928820165e-06, + "loss": 1.0032, + "num_input_tokens_seen": 26763625, + "step": 4342 + }, + { + "epoch": 0.3263941079212385, + "flos": 21653059322880.0, + "grad_norm": 1.8787004195502957, + "learning_rate": 3.147352263092638e-06, + "loss": 0.9461, + "num_input_tokens_seen": 26788195, + "step": 4343 + }, + { + "epoch": 0.3264692619870735, + "flos": 15935477016960.0, + "grad_norm": 1.5239188191976902, + "learning_rate": 3.1469534653182453e-06, + "loss": 1.0138, + "num_input_tokens_seen": 26810140, + "step": 4344 + }, + { + "epoch": 0.32654441605290846, + "flos": 16035616650240.0, + "grad_norm": 1.5162077894255528, + "learning_rate": 3.146554599582468e-06, + "loss": 0.9849, + "num_input_tokens_seen": 26832030, + "step": 4345 + }, + { + "epoch": 0.32661957011874343, + "flos": 20593662526080.0, + "grad_norm": 1.8326977003589213, + "learning_rate": 3.1461556659089397e-06, + "loss": 0.9723, + "num_input_tokens_seen": 26854885, + "step": 4346 + }, + { + "epoch": 0.3266947241845784, + "flos": 14231715137280.0, + "grad_norm": 1.8783186471540398, + "learning_rate": 3.145756664321299e-06, + "loss": 1.0445, + "num_input_tokens_seen": 26873055, + "step": 4347 + }, + { + "epoch": 0.32676987825041337, + "flos": 26523303759360.0, + "grad_norm": 1.4799229564460412, + "learning_rate": 3.145357594843189e-06, + "loss": 0.9945, + "num_input_tokens_seen": 26897220, + "step": 4348 + }, + { + "epoch": 0.3268450323162483, + "flos": 21648067332480.0, + "grad_norm": 1.788879518438298, + "learning_rate": 3.1449584574982556e-06, + "loss": 0.9469, + "num_input_tokens_seen": 26919560, + "step": 4349 + }, + { + "epoch": 0.32692018638208326, + "flos": 20876707042560.0, + "grad_norm": 1.5449601468868865, + "learning_rate": 3.144559252310149e-06, + "loss": 0.9322, + "num_input_tokens_seen": 26942930, + "step": 4350 + }, + { + "epoch": 0.3269953404479182, + "flos": 23664566476800.0, + "grad_norm": 1.996090047276715, + "learning_rate": 3.1441599793025243e-06, + "loss": 0.9672, + "num_input_tokens_seen": 26966835, + "step": 4351 + }, + { + "epoch": 0.3270704945137532, + "flos": 27652572823680.0, + "grad_norm": 1.5562168439527384, + "learning_rate": 3.1437606384990396e-06, + "loss": 0.901, + "num_input_tokens_seen": 26991695, + "step": 4352 + }, + { + "epoch": 0.32714564857958817, + "flos": 23347052703360.0, + "grad_norm": 1.2738930291161568, + "learning_rate": 3.1433612299233567e-06, + "loss": 0.9373, + "num_input_tokens_seen": 27015230, + "step": 4353 + }, + { + "epoch": 0.32722080264542314, + "flos": 23241597857280.0, + "grad_norm": 1.7412231924125317, + "learning_rate": 3.1429617535991427e-06, + "loss": 0.8378, + "num_input_tokens_seen": 27039845, + "step": 4354 + }, + { + "epoch": 0.32729595671125805, + "flos": 24440020917120.0, + "grad_norm": 1.6549959293326941, + "learning_rate": 3.1425622095500685e-06, + "loss": 0.8572, + "num_input_tokens_seen": 27063760, + "step": 4355 + }, + { + "epoch": 0.327371110777093, + "flos": 19495415013120.0, + "grad_norm": 1.8778076767030256, + "learning_rate": 3.1421625977998087e-06, + "loss": 1.0021, + "num_input_tokens_seen": 27083020, + "step": 4356 + }, + { + "epoch": 0.327446264842928, + "flos": 19004262065280.0, + "grad_norm": 1.4449443451350366, + "learning_rate": 3.1417629183720403e-06, + "loss": 0.9384, + "num_input_tokens_seen": 27107020, + "step": 4357 + }, + { + "epoch": 0.32752141890876296, + "flos": 21335042759040.0, + "grad_norm": 1.568614136181215, + "learning_rate": 3.1413631712904476e-06, + "loss": 0.9246, + "num_input_tokens_seen": 27129735, + "step": 4358 + }, + { + "epoch": 0.32759657297459793, + "flos": 18895287686400.0, + "grad_norm": 6.893025574789856, + "learning_rate": 3.140963356578716e-06, + "loss": 1.0822, + "num_input_tokens_seen": 27151625, + "step": 4359 + }, + { + "epoch": 0.3276717270404329, + "flos": 23028892485120.0, + "grad_norm": 1.9971344207696193, + "learning_rate": 3.1405634742605366e-06, + "loss": 1.0925, + "num_input_tokens_seen": 27174840, + "step": 4360 + }, + { + "epoch": 0.3277468811062679, + "flos": 19923339709440.0, + "grad_norm": 1.7847875301947969, + "learning_rate": 3.1401635243596043e-06, + "loss": 1.0437, + "num_input_tokens_seen": 27196590, + "step": 4361 + }, + { + "epoch": 0.3278220351721028, + "flos": 25883751098880.0, + "grad_norm": 1.69207989886863, + "learning_rate": 3.1397635068996167e-06, + "loss": 0.9255, + "num_input_tokens_seen": 27221055, + "step": 4362 + }, + { + "epoch": 0.32789718923793776, + "flos": 20629101450240.0, + "grad_norm": 1.4625035832027002, + "learning_rate": 3.139363421904277e-06, + "loss": 0.8901, + "num_input_tokens_seen": 27243565, + "step": 4363 + }, + { + "epoch": 0.32797234330377273, + "flos": 57667638049920.0, + "grad_norm": 0.7451810596932912, + "learning_rate": 3.138963269397292e-06, + "loss": 0.7693, + "num_input_tokens_seen": 27325325, + "step": 4364 + }, + { + "epoch": 0.3280474973696077, + "flos": 15615700686720.0, + "grad_norm": 1.6525701273063371, + "learning_rate": 3.1385630494023716e-06, + "loss": 1.0142, + "num_input_tokens_seen": 27345735, + "step": 4365 + }, + { + "epoch": 0.32812265143544267, + "flos": 22217783639040.0, + "grad_norm": 1.3444676811445078, + "learning_rate": 3.1381627619432307e-06, + "loss": 0.9856, + "num_input_tokens_seen": 27371275, + "step": 4366 + }, + { + "epoch": 0.32819780550127764, + "flos": 20452265965440.0, + "grad_norm": 1.6398189199311566, + "learning_rate": 3.1377624070435874e-06, + "loss": 0.9835, + "num_input_tokens_seen": 27391565, + "step": 4367 + }, + { + "epoch": 0.32827295956711255, + "flos": 18794322040320.0, + "grad_norm": 1.6252869572729134, + "learning_rate": 3.137361984727165e-06, + "loss": 0.9993, + "num_input_tokens_seen": 27413235, + "step": 4368 + }, + { + "epoch": 0.3283481136329475, + "flos": 26170530629760.0, + "grad_norm": 1.8155598298817164, + "learning_rate": 3.1369614950176903e-06, + "loss": 0.908, + "num_input_tokens_seen": 27442335, + "step": 4369 + }, + { + "epoch": 0.3284232676987825, + "flos": 22994459141760.0, + "grad_norm": 1.7477498574696775, + "learning_rate": 3.1365609379388922e-06, + "loss": 0.9921, + "num_input_tokens_seen": 27466140, + "step": 4370 + }, + { + "epoch": 0.32849842176461747, + "flos": 19497964878720.0, + "grad_norm": 1.3885555084446726, + "learning_rate": 3.1361603135145074e-06, + "loss": 0.9369, + "num_input_tokens_seen": 27490895, + "step": 4371 + }, + { + "epoch": 0.32857357583045244, + "flos": 20629855635840.0, + "grad_norm": 3.4002450630281724, + "learning_rate": 3.135759621768273e-06, + "loss": 1.0081, + "num_input_tokens_seen": 27511315, + "step": 4372 + }, + { + "epoch": 0.3286487298962874, + "flos": 18930008338560.0, + "grad_norm": 1.4836177299694857, + "learning_rate": 3.1353588627239317e-06, + "loss": 1.0266, + "num_input_tokens_seen": 27533630, + "step": 4373 + }, + { + "epoch": 0.3287238839621224, + "flos": 21689144691840.0, + "grad_norm": 1.679713191346036, + "learning_rate": 3.13495803640523e-06, + "loss": 0.9639, + "num_input_tokens_seen": 27556170, + "step": 4374 + }, + { + "epoch": 0.3287990380279573, + "flos": 34746957959040.0, + "grad_norm": 1.8815881064103956, + "learning_rate": 3.134557142835919e-06, + "loss": 0.8681, + "num_input_tokens_seen": 27585660, + "step": 4375 + }, + { + "epoch": 0.32887419209379226, + "flos": 20805901021440.0, + "grad_norm": 2.1235356614846204, + "learning_rate": 3.134156182039753e-06, + "loss": 0.9017, + "num_input_tokens_seen": 27610665, + "step": 4376 + }, + { + "epoch": 0.32894934615962723, + "flos": 24017698742400.0, + "grad_norm": 1.4114667979006548, + "learning_rate": 3.13375515404049e-06, + "loss": 0.9449, + "num_input_tokens_seen": 27633370, + "step": 4377 + }, + { + "epoch": 0.3290245002254622, + "flos": 20487740803200.0, + "grad_norm": 1.647330106504875, + "learning_rate": 3.133354058861893e-06, + "loss": 1.0183, + "num_input_tokens_seen": 27656355, + "step": 4378 + }, + { + "epoch": 0.3290996542912972, + "flos": 23099913987840.0, + "grad_norm": 1.269933548788287, + "learning_rate": 3.1329528965277275e-06, + "loss": 0.9793, + "num_input_tokens_seen": 27682940, + "step": 4379 + }, + { + "epoch": 0.32917480835713214, + "flos": 21652628359680.0, + "grad_norm": 1.7209038850054938, + "learning_rate": 3.1325516670617648e-06, + "loss": 0.9834, + "num_input_tokens_seen": 27705900, + "step": 4380 + }, + { + "epoch": 0.32924996242296706, + "flos": 17629506311040.0, + "grad_norm": 1.506802472616283, + "learning_rate": 3.132150370487779e-06, + "loss": 0.9537, + "num_input_tokens_seen": 27727900, + "step": 4381 + }, + { + "epoch": 0.329325116488802, + "flos": 12223009244160.0, + "grad_norm": 1.4723121787524822, + "learning_rate": 3.1317490068295486e-06, + "loss": 0.9479, + "num_input_tokens_seen": 27749040, + "step": 4382 + }, + { + "epoch": 0.329400270554637, + "flos": 19076468716800.0, + "grad_norm": 4.081974244518228, + "learning_rate": 3.131347576110855e-06, + "loss": 1.031, + "num_input_tokens_seen": 27771835, + "step": 4383 + }, + { + "epoch": 0.32947542462047197, + "flos": 20064520788480.0, + "grad_norm": 1.2815794637469375, + "learning_rate": 3.130946078355486e-06, + "loss": 0.9966, + "num_input_tokens_seen": 27797210, + "step": 4384 + }, + { + "epoch": 0.32955057868630694, + "flos": 20981479530240.0, + "grad_norm": 1.7931066389396977, + "learning_rate": 3.1305445135872318e-06, + "loss": 0.9979, + "num_input_tokens_seen": 27819885, + "step": 4385 + }, + { + "epoch": 0.3296257327521419, + "flos": 21085102782720.0, + "grad_norm": 1.5174808741060257, + "learning_rate": 3.1301428818298847e-06, + "loss": 0.8793, + "num_input_tokens_seen": 27843400, + "step": 4386 + }, + { + "epoch": 0.3297008868179769, + "flos": 23383173985920.0, + "grad_norm": 1.5881299428551388, + "learning_rate": 3.129741183107245e-06, + "loss": 1.0042, + "num_input_tokens_seen": 27867075, + "step": 4387 + }, + { + "epoch": 0.3297760408838118, + "flos": 22605277420800.0, + "grad_norm": 1.3410077168970147, + "learning_rate": 3.129339417443114e-06, + "loss": 0.8812, + "num_input_tokens_seen": 27892135, + "step": 4388 + }, + { + "epoch": 0.32985119494964676, + "flos": 21053901663360.0, + "grad_norm": 1.327558553736956, + "learning_rate": 3.128937584861298e-06, + "loss": 0.9746, + "num_input_tokens_seen": 27916850, + "step": 4389 + }, + { + "epoch": 0.32992634901548173, + "flos": 23841653356800.0, + "grad_norm": 1.8396957564715408, + "learning_rate": 3.128535685385607e-06, + "loss": 0.8827, + "num_input_tokens_seen": 27939960, + "step": 4390 + }, + { + "epoch": 0.3300015030813167, + "flos": 17594426522880.0, + "grad_norm": 1.486972503736258, + "learning_rate": 3.1281337190398552e-06, + "loss": 1.0064, + "num_input_tokens_seen": 27961090, + "step": 4391 + }, + { + "epoch": 0.3300766571471517, + "flos": 18829653223680.0, + "grad_norm": 1.6844325974052556, + "learning_rate": 3.1277316858478607e-06, + "loss": 0.9421, + "num_input_tokens_seen": 27984275, + "step": 4392 + }, + { + "epoch": 0.33015181121298665, + "flos": 64815518851200.0, + "grad_norm": 0.7745754848189985, + "learning_rate": 3.1273295858334454e-06, + "loss": 0.7717, + "num_input_tokens_seen": 28061445, + "step": 4393 + }, + { + "epoch": 0.33022696527882156, + "flos": 16639119855360.0, + "grad_norm": 3.7374122430711605, + "learning_rate": 3.1269274190204352e-06, + "loss": 1.0049, + "num_input_tokens_seen": 28080255, + "step": 4394 + }, + { + "epoch": 0.33030211934465653, + "flos": 17206178555520.0, + "grad_norm": 1.5632030431276607, + "learning_rate": 3.1265251854326613e-06, + "loss": 0.9835, + "num_input_tokens_seen": 28103015, + "step": 4395 + }, + { + "epoch": 0.3303772734104915, + "flos": 16358732945280.0, + "grad_norm": 1.7222902077754476, + "learning_rate": 3.126122885093955e-06, + "loss": 0.9043, + "num_input_tokens_seen": 28126075, + "step": 4396 + }, + { + "epoch": 0.33045242747632647, + "flos": 23735444325120.0, + "grad_norm": 1.3411976727093153, + "learning_rate": 3.1257205180281555e-06, + "loss": 0.984, + "num_input_tokens_seen": 28152340, + "step": 4397 + }, + { + "epoch": 0.33052758154216144, + "flos": 17664370617600.0, + "grad_norm": 2.0344451423424834, + "learning_rate": 3.125318084259105e-06, + "loss": 1.0128, + "num_input_tokens_seen": 28171735, + "step": 4398 + }, + { + "epoch": 0.3306027356079964, + "flos": 19640797983360.0, + "grad_norm": 1.8820279334748888, + "learning_rate": 3.1249155838106493e-06, + "loss": 0.9089, + "num_input_tokens_seen": 28194835, + "step": 4399 + }, + { + "epoch": 0.3306778896738313, + "flos": 27017581190400.0, + "grad_norm": 1.78184679975777, + "learning_rate": 3.1245130167066373e-06, + "loss": 0.9479, + "num_input_tokens_seen": 28217925, + "step": 4400 + }, + { + "epoch": 0.3307530437396663, + "flos": 16958501136000.0, + "grad_norm": 1.4295228957176949, + "learning_rate": 3.1241103829709234e-06, + "loss": 0.8926, + "num_input_tokens_seen": 28241310, + "step": 4401 + }, + { + "epoch": 0.33082819780550127, + "flos": 20911750917120.0, + "grad_norm": 1.4586643751522756, + "learning_rate": 3.123707682627364e-06, + "loss": 0.8813, + "num_input_tokens_seen": 28264465, + "step": 4402 + }, + { + "epoch": 0.33090335187133624, + "flos": 17590799249280.0, + "grad_norm": 1.5971993163673166, + "learning_rate": 3.1233049156998215e-06, + "loss": 0.8526, + "num_input_tokens_seen": 28286615, + "step": 4403 + }, + { + "epoch": 0.3309785059371712, + "flos": 26276236871040.0, + "grad_norm": 1.816130717719403, + "learning_rate": 3.122902082212162e-06, + "loss": 1.003, + "num_input_tokens_seen": 28309955, + "step": 4404 + }, + { + "epoch": 0.3310536600030062, + "flos": 20523610690560.0, + "grad_norm": 1.9929899506916082, + "learning_rate": 3.122499182188254e-06, + "loss": 0.9336, + "num_input_tokens_seen": 28333710, + "step": 4405 + }, + { + "epoch": 0.33112881406884115, + "flos": 21195046828800.0, + "grad_norm": 1.4173570054201583, + "learning_rate": 3.1220962156519715e-06, + "loss": 0.9593, + "num_input_tokens_seen": 28359340, + "step": 4406 + }, + { + "epoch": 0.33120396813467606, + "flos": 22853278062720.0, + "grad_norm": 1.6608649682099244, + "learning_rate": 3.121693182627191e-06, + "loss": 1.0086, + "num_input_tokens_seen": 28382260, + "step": 4407 + }, + { + "epoch": 0.33127912220051103, + "flos": 18017682451200.0, + "grad_norm": 1.9287918421520174, + "learning_rate": 3.1212900831377934e-06, + "loss": 1.0094, + "num_input_tokens_seen": 28404055, + "step": 4408 + }, + { + "epoch": 0.331354276266346, + "flos": 20841483600000.0, + "grad_norm": 1.5720361008951735, + "learning_rate": 3.1208869172076657e-06, + "loss": 0.9207, + "num_input_tokens_seen": 28428365, + "step": 4409 + }, + { + "epoch": 0.331429430332181, + "flos": 22316953605120.0, + "grad_norm": 1.9982515679568884, + "learning_rate": 3.120483684860696e-06, + "loss": 0.9234, + "num_input_tokens_seen": 28449535, + "step": 4410 + }, + { + "epoch": 0.33150458439801594, + "flos": 69856744855680.0, + "grad_norm": 0.8096518849321417, + "learning_rate": 3.1200803861207774e-06, + "loss": 0.8526, + "num_input_tokens_seen": 28524870, + "step": 4411 + }, + { + "epoch": 0.3315797384638509, + "flos": 27119336935680.0, + "grad_norm": 1.9229356893615897, + "learning_rate": 3.1196770210118063e-06, + "loss": 0.9857, + "num_input_tokens_seen": 28550690, + "step": 4412 + }, + { + "epoch": 0.33165489252968583, + "flos": 26381224840320.0, + "grad_norm": 1.5985197611026716, + "learning_rate": 3.1192735895576845e-06, + "loss": 1.009, + "num_input_tokens_seen": 28574945, + "step": 4413 + }, + { + "epoch": 0.3317300465955208, + "flos": 21547065772800.0, + "grad_norm": 1.6008529207671869, + "learning_rate": 3.118870091782316e-06, + "loss": 0.8628, + "num_input_tokens_seen": 28599190, + "step": 4414 + }, + { + "epoch": 0.33180520066135577, + "flos": 20947513063680.0, + "grad_norm": 1.7348340088650611, + "learning_rate": 3.118466527709611e-06, + "loss": 0.9246, + "num_input_tokens_seen": 28622060, + "step": 4415 + }, + { + "epoch": 0.33188035472719074, + "flos": 16499914024320.0, + "grad_norm": 1.7130514523940004, + "learning_rate": 3.1180628973634807e-06, + "loss": 0.9348, + "num_input_tokens_seen": 28643535, + "step": 4416 + }, + { + "epoch": 0.3319555087930257, + "flos": 26524057944960.0, + "grad_norm": 1.4211281292548963, + "learning_rate": 3.117659200767843e-06, + "loss": 0.9094, + "num_input_tokens_seen": 28668785, + "step": 4417 + }, + { + "epoch": 0.3320306628588607, + "flos": 27124149358080.0, + "grad_norm": 1.7024711963832957, + "learning_rate": 3.1172554379466176e-06, + "loss": 0.8835, + "num_input_tokens_seen": 28695785, + "step": 4418 + }, + { + "epoch": 0.33210581692469565, + "flos": 21331595053440.0, + "grad_norm": 1.4474230228966596, + "learning_rate": 3.1168516089237288e-06, + "loss": 0.9277, + "num_input_tokens_seen": 28719290, + "step": 4419 + }, + { + "epoch": 0.33218097099053057, + "flos": 64119022819200.0, + "grad_norm": 0.8077408720095185, + "learning_rate": 3.1164477137231054e-06, + "loss": 0.7769, + "num_input_tokens_seen": 28801195, + "step": 4420 + }, + { + "epoch": 0.33225612505636554, + "flos": 20272198256640.0, + "grad_norm": 1.5031630227718364, + "learning_rate": 3.1160437523686806e-06, + "loss": 1.0069, + "num_input_tokens_seen": 28822635, + "step": 4421 + }, + { + "epoch": 0.3323312791222005, + "flos": 21046898511360.0, + "grad_norm": 1.2998082643111475, + "learning_rate": 3.1156397248843896e-06, + "loss": 1.0129, + "num_input_tokens_seen": 28845555, + "step": 4422 + }, + { + "epoch": 0.3324064331880355, + "flos": 20695238703360.0, + "grad_norm": 1.4765099410133529, + "learning_rate": 3.1152356312941724e-06, + "loss": 0.9791, + "num_input_tokens_seen": 28870560, + "step": 4423 + }, + { + "epoch": 0.33248158725387045, + "flos": 18723516019200.0, + "grad_norm": 1.5254973741756959, + "learning_rate": 3.114831471621974e-06, + "loss": 1.0151, + "num_input_tokens_seen": 28892190, + "step": 4424 + }, + { + "epoch": 0.3325567413197054, + "flos": 25955562700800.0, + "grad_norm": 1.6145899454617212, + "learning_rate": 3.1144272458917417e-06, + "loss": 0.812, + "num_input_tokens_seen": 28914920, + "step": 4425 + }, + { + "epoch": 0.33263189538554033, + "flos": 45758040629760.0, + "grad_norm": 1.417097596468094, + "learning_rate": 3.114022954127427e-06, + "loss": 0.9097, + "num_input_tokens_seen": 28942665, + "step": 4426 + }, + { + "epoch": 0.3327070494513753, + "flos": 25493025093120.0, + "grad_norm": 1.4348172571876314, + "learning_rate": 3.1136185963529873e-06, + "loss": 0.9552, + "num_input_tokens_seen": 28966550, + "step": 4427 + }, + { + "epoch": 0.3327822035172103, + "flos": 13422832934400.0, + "grad_norm": 1.6646003391254982, + "learning_rate": 3.1132141725923812e-06, + "loss": 1.026, + "num_input_tokens_seen": 28986760, + "step": 4428 + }, + { + "epoch": 0.33285735758304524, + "flos": 22465389231360.0, + "grad_norm": 1.9760353078892345, + "learning_rate": 3.1128096828695728e-06, + "loss": 0.998, + "num_input_tokens_seen": 29008870, + "step": 4429 + }, + { + "epoch": 0.3329325116488802, + "flos": 30793780005120.0, + "grad_norm": 1.4431450340099647, + "learning_rate": 3.1124051272085286e-06, + "loss": 0.9523, + "num_input_tokens_seen": 29035880, + "step": 4430 + }, + { + "epoch": 0.3330076657147152, + "flos": 20841339945600.0, + "grad_norm": 1.4650513362027116, + "learning_rate": 3.1120005056332216e-06, + "loss": 0.984, + "num_input_tokens_seen": 29059810, + "step": 4431 + }, + { + "epoch": 0.33308281978055015, + "flos": 24402211695360.0, + "grad_norm": 1.360721359871039, + "learning_rate": 3.111595818167627e-06, + "loss": 0.958, + "num_input_tokens_seen": 29084915, + "step": 4432 + }, + { + "epoch": 0.33315797384638507, + "flos": 17347431461760.0, + "grad_norm": 1.4900854993777815, + "learning_rate": 3.111191064835723e-06, + "loss": 0.9253, + "num_input_tokens_seen": 29106585, + "step": 4433 + }, + { + "epoch": 0.33323312791222004, + "flos": 63085080965760.0, + "grad_norm": 0.9453216741447747, + "learning_rate": 3.1107862456614932e-06, + "loss": 0.7798, + "num_input_tokens_seen": 29185840, + "step": 4434 + }, + { + "epoch": 0.333308281978055, + "flos": 18543304656000.0, + "grad_norm": 1.9580430634100812, + "learning_rate": 3.1103813606689253e-06, + "loss": 0.9824, + "num_input_tokens_seen": 29207175, + "step": 4435 + }, + { + "epoch": 0.33338343604389, + "flos": 23665033353600.0, + "grad_norm": 1.6909355935949781, + "learning_rate": 3.1099764098820096e-06, + "loss": 0.9929, + "num_input_tokens_seen": 29231975, + "step": 4436 + }, + { + "epoch": 0.33345859010972495, + "grad_norm": 1.6063843132890714, + "learning_rate": 3.1095713933247416e-06, + "loss": 0.9793, + "num_input_tokens_seen": 29254490, + "step": 4437 + }, + { + "epoch": 0.3335337441755599, + "grad_norm": 2.5595919227197235, + "learning_rate": 3.1091663110211188e-06, + "loss": 0.9355, + "num_input_tokens_seen": 29278890, + "step": 4438 + }, + { + "epoch": 0.33360889824139484, + "grad_norm": 1.6852046048827405, + "learning_rate": 3.1087611629951457e-06, + "loss": 0.7964, + "num_input_tokens_seen": 29303210, + "step": 4439 + }, + { + "epoch": 0.3336840523072298, + "grad_norm": 1.5846338786669132, + "learning_rate": 3.1083559492708277e-06, + "loss": 1.0019, + "num_input_tokens_seen": 29325950, + "step": 4440 + }, + { + "epoch": 0.3337592063730648, + "grad_norm": 1.5762379761251633, + "learning_rate": 3.1079506698721752e-06, + "loss": 0.9382, + "num_input_tokens_seen": 29350775, + "step": 4441 + }, + { + "epoch": 0.33383436043889975, + "grad_norm": 0.9094387713747127, + "learning_rate": 3.107545324823203e-06, + "loss": 0.9445, + "num_input_tokens_seen": 29434125, + "step": 4442 + }, + { + "epoch": 0.3339095145047347, + "grad_norm": 1.7152435601339697, + "learning_rate": 3.1071399141479292e-06, + "loss": 0.9107, + "num_input_tokens_seen": 29460085, + "step": 4443 + }, + { + "epoch": 0.3339846685705697, + "grad_norm": 1.5927475413188361, + "learning_rate": 3.1067344378703765e-06, + "loss": 0.8955, + "num_input_tokens_seen": 29485460, + "step": 4444 + }, + { + "epoch": 0.3340598226364046, + "grad_norm": 1.2320860875322204, + "learning_rate": 3.10632889601457e-06, + "loss": 0.8985, + "num_input_tokens_seen": 29511480, + "step": 4445 + }, + { + "epoch": 0.3341349767022396, + "grad_norm": 1.7028359684103134, + "learning_rate": 3.10592328860454e-06, + "loss": 1.0154, + "num_input_tokens_seen": 29532480, + "step": 4446 + }, + { + "epoch": 0.33421013076807454, + "grad_norm": 1.6530495437102215, + "learning_rate": 3.10551761566432e-06, + "loss": 0.8769, + "num_input_tokens_seen": 29555990, + "step": 4447 + }, + { + "epoch": 0.3342852848339095, + "grad_norm": 1.6348335133519818, + "learning_rate": 3.1051118772179483e-06, + "loss": 1.0526, + "num_input_tokens_seen": 29578425, + "step": 4448 + }, + { + "epoch": 0.3343604388997445, + "grad_norm": 1.481900290503056, + "learning_rate": 3.104706073289466e-06, + "loss": 1.038, + "num_input_tokens_seen": 29600435, + "step": 4449 + }, + { + "epoch": 0.33443559296557945, + "grad_norm": 1.4498104727957368, + "learning_rate": 3.104300203902919e-06, + "loss": 1.0079, + "num_input_tokens_seen": 29625635, + "step": 4450 + }, + { + "epoch": 0.3345107470314144, + "grad_norm": 9.737394351410588, + "learning_rate": 3.1038942690823556e-06, + "loss": 0.9952, + "num_input_tokens_seen": 29648715, + "step": 4451 + }, + { + "epoch": 0.33458590109724934, + "grad_norm": 1.3512482911839299, + "learning_rate": 3.10348826885183e-06, + "loss": 0.9032, + "num_input_tokens_seen": 29673170, + "step": 4452 + }, + { + "epoch": 0.3346610551630843, + "grad_norm": 1.5224639129120219, + "learning_rate": 3.1030822032353997e-06, + "loss": 1.0943, + "num_input_tokens_seen": 29697230, + "step": 4453 + }, + { + "epoch": 0.3347362092289193, + "grad_norm": 1.545652966460344, + "learning_rate": 3.1026760722571236e-06, + "loss": 0.8984, + "num_input_tokens_seen": 29717400, + "step": 4454 + }, + { + "epoch": 0.33481136329475425, + "grad_norm": 1.3785373961668994, + "learning_rate": 3.1022698759410684e-06, + "loss": 0.9682, + "num_input_tokens_seen": 29743575, + "step": 4455 + }, + { + "epoch": 0.3348865173605892, + "grad_norm": 1.5585584134563317, + "learning_rate": 3.1018636143113022e-06, + "loss": 1.0404, + "num_input_tokens_seen": 29765610, + "step": 4456 + }, + { + "epoch": 0.3349616714264242, + "grad_norm": 2.1221318804345177, + "learning_rate": 3.1014572873918976e-06, + "loss": 0.9601, + "num_input_tokens_seen": 29790305, + "step": 4457 + }, + { + "epoch": 0.3350368254922591, + "grad_norm": 1.509905059810826, + "learning_rate": 3.101050895206931e-06, + "loss": 1.0235, + "num_input_tokens_seen": 29811315, + "step": 4458 + }, + { + "epoch": 0.3351119795580941, + "grad_norm": 1.8610360555871326, + "learning_rate": 3.100644437780482e-06, + "loss": 0.9497, + "num_input_tokens_seen": 29833220, + "step": 4459 + }, + { + "epoch": 0.33518713362392905, + "grad_norm": 1.5312071813707948, + "learning_rate": 3.100237915136636e-06, + "loss": 1.0407, + "num_input_tokens_seen": 29857985, + "step": 4460 + }, + { + "epoch": 0.335262287689764, + "grad_norm": 1.424546132672631, + "learning_rate": 3.0998313272994805e-06, + "loss": 0.838, + "num_input_tokens_seen": 29883810, + "step": 4461 + }, + { + "epoch": 0.335337441755599, + "grad_norm": 1.3145332482051955, + "learning_rate": 3.0994246742931076e-06, + "loss": 1.0078, + "num_input_tokens_seen": 29906145, + "step": 4462 + }, + { + "epoch": 0.33541259582143396, + "grad_norm": 1.6633744227596918, + "learning_rate": 3.099017956141612e-06, + "loss": 0.9805, + "num_input_tokens_seen": 29929045, + "step": 4463 + }, + { + "epoch": 0.3354877498872689, + "grad_norm": 1.9014384722167843, + "learning_rate": 3.098611172869094e-06, + "loss": 0.9657, + "num_input_tokens_seen": 29951570, + "step": 4464 + }, + { + "epoch": 0.33556290395310384, + "grad_norm": 1.357853838573359, + "learning_rate": 3.0982043244996582e-06, + "loss": 0.9075, + "num_input_tokens_seen": 29975620, + "step": 4465 + }, + { + "epoch": 0.3356380580189388, + "grad_norm": 1.8715522857300662, + "learning_rate": 3.09779741105741e-06, + "loss": 0.9334, + "num_input_tokens_seen": 29997380, + "step": 4466 + }, + { + "epoch": 0.3357132120847738, + "grad_norm": 1.7159699584803292, + "learning_rate": 3.0973904325664615e-06, + "loss": 0.9405, + "num_input_tokens_seen": 30020010, + "step": 4467 + }, + { + "epoch": 0.33578836615060875, + "grad_norm": 1.4809979220265628, + "learning_rate": 3.0969833890509282e-06, + "loss": 0.9027, + "num_input_tokens_seen": 30040525, + "step": 4468 + }, + { + "epoch": 0.3358635202164437, + "grad_norm": 1.4808451392984556, + "learning_rate": 3.096576280534928e-06, + "loss": 0.9562, + "num_input_tokens_seen": 30064020, + "step": 4469 + }, + { + "epoch": 0.3359386742822787, + "grad_norm": 1.5555935620403005, + "learning_rate": 3.096169107042584e-06, + "loss": 0.9405, + "num_input_tokens_seen": 30087020, + "step": 4470 + }, + { + "epoch": 0.3360138283481136, + "grad_norm": 1.5935724944952432, + "learning_rate": 3.0957618685980233e-06, + "loss": 0.962, + "num_input_tokens_seen": 30110270, + "step": 4471 + }, + { + "epoch": 0.3360889824139486, + "grad_norm": 0.826698730433998, + "learning_rate": 3.0953545652253763e-06, + "loss": 0.8269, + "num_input_tokens_seen": 30186425, + "step": 4472 + }, + { + "epoch": 0.33616413647978355, + "grad_norm": 1.8292617524037869, + "learning_rate": 3.094947196948776e-06, + "loss": 1.0004, + "num_input_tokens_seen": 30208160, + "step": 4473 + }, + { + "epoch": 0.3362392905456185, + "grad_norm": 1.685998106990489, + "learning_rate": 3.0945397637923617e-06, + "loss": 0.9411, + "num_input_tokens_seen": 30231050, + "step": 4474 + }, + { + "epoch": 0.3363144446114535, + "grad_norm": 1.6587547220538035, + "learning_rate": 3.094132265780275e-06, + "loss": 1.0028, + "num_input_tokens_seen": 30255570, + "step": 4475 + }, + { + "epoch": 0.33638959867728846, + "grad_norm": 1.6572815184059708, + "learning_rate": 3.0937247029366623e-06, + "loss": 1.0052, + "num_input_tokens_seen": 30280270, + "step": 4476 + }, + { + "epoch": 0.33646475274312343, + "grad_norm": 1.587779586682106, + "learning_rate": 3.0933170752856723e-06, + "loss": 0.9454, + "num_input_tokens_seen": 30301935, + "step": 4477 + }, + { + "epoch": 0.33653990680895834, + "grad_norm": 1.7111417744243682, + "learning_rate": 3.0929093828514595e-06, + "loss": 0.9911, + "num_input_tokens_seen": 30325240, + "step": 4478 + }, + { + "epoch": 0.3366150608747933, + "grad_norm": 1.9111784713663091, + "learning_rate": 3.0925016256581805e-06, + "loss": 0.9244, + "num_input_tokens_seen": 30346155, + "step": 4479 + }, + { + "epoch": 0.3366902149406283, + "grad_norm": 1.4334357483817577, + "learning_rate": 3.092093803729997e-06, + "loss": 0.9313, + "num_input_tokens_seen": 30369970, + "step": 4480 + }, + { + "epoch": 0.33676536900646326, + "grad_norm": 1.0495055442252885, + "learning_rate": 3.091685917091073e-06, + "loss": 0.7592, + "num_input_tokens_seen": 30442875, + "step": 4481 + }, + { + "epoch": 0.3368405230722982, + "grad_norm": 1.9648777862088365, + "learning_rate": 3.0912779657655784e-06, + "loss": 0.9521, + "num_input_tokens_seen": 30462755, + "step": 4482 + }, + { + "epoch": 0.3369156771381332, + "grad_norm": 0.9491914972859257, + "learning_rate": 3.0908699497776864e-06, + "loss": 0.8702, + "num_input_tokens_seen": 30544640, + "step": 4483 + }, + { + "epoch": 0.3369908312039681, + "grad_norm": 1.5408286403063944, + "learning_rate": 3.0904618691515714e-06, + "loss": 0.9243, + "num_input_tokens_seen": 30568870, + "step": 4484 + }, + { + "epoch": 0.3370659852698031, + "grad_norm": 1.6374681891925948, + "learning_rate": 3.0900537239114157e-06, + "loss": 1.0029, + "num_input_tokens_seen": 30592440, + "step": 4485 + }, + { + "epoch": 0.33714113933563805, + "grad_norm": 1.5623968471809762, + "learning_rate": 3.089645514081402e-06, + "loss": 1.0765, + "num_input_tokens_seen": 30616960, + "step": 4486 + }, + { + "epoch": 0.337216293401473, + "grad_norm": 1.7670779031485562, + "learning_rate": 3.08923723968572e-06, + "loss": 0.9407, + "num_input_tokens_seen": 30639440, + "step": 4487 + }, + { + "epoch": 0.337291447467308, + "grad_norm": 1.954448099211851, + "learning_rate": 3.0888289007485605e-06, + "loss": 1.0294, + "num_input_tokens_seen": 30660935, + "step": 4488 + }, + { + "epoch": 0.33736660153314296, + "grad_norm": 1.4313424126387886, + "learning_rate": 3.0884204972941187e-06, + "loss": 0.9933, + "num_input_tokens_seen": 30686280, + "step": 4489 + }, + { + "epoch": 0.3374417555989779, + "grad_norm": 1.5346382179495774, + "learning_rate": 3.088012029346595e-06, + "loss": 0.9785, + "num_input_tokens_seen": 30708940, + "step": 4490 + }, + { + "epoch": 0.33751690966481285, + "grad_norm": 1.4688027555040928, + "learning_rate": 3.087603496930192e-06, + "loss": 1.0009, + "num_input_tokens_seen": 30732475, + "step": 4491 + }, + { + "epoch": 0.3375920637306478, + "grad_norm": 1.7861033792297063, + "learning_rate": 3.087194900069117e-06, + "loss": 0.96, + "num_input_tokens_seen": 30754300, + "step": 4492 + }, + { + "epoch": 0.3376672177964828, + "grad_norm": 1.4398977765251397, + "learning_rate": 3.0867862387875815e-06, + "loss": 0.9207, + "num_input_tokens_seen": 30776105, + "step": 4493 + }, + { + "epoch": 0.33774237186231776, + "grad_norm": 1.439228511351044, + "learning_rate": 3.0863775131097995e-06, + "loss": 0.915, + "num_input_tokens_seen": 30800630, + "step": 4494 + }, + { + "epoch": 0.33781752592815273, + "grad_norm": 1.6191405513749606, + "learning_rate": 3.0859687230599897e-06, + "loss": 1.0033, + "num_input_tokens_seen": 30823360, + "step": 4495 + }, + { + "epoch": 0.3378926799939877, + "grad_norm": 1.3244314557401524, + "learning_rate": 3.0855598686623745e-06, + "loss": 1.0977, + "num_input_tokens_seen": 30843830, + "step": 4496 + }, + { + "epoch": 0.3379678340598226, + "grad_norm": 1.2847271698635931, + "learning_rate": 3.085150949941181e-06, + "loss": 1.0232, + "num_input_tokens_seen": 30869395, + "step": 4497 + }, + { + "epoch": 0.3380429881256576, + "grad_norm": 1.3581799921803654, + "learning_rate": 3.084741966920638e-06, + "loss": 0.9318, + "num_input_tokens_seen": 30894105, + "step": 4498 + }, + { + "epoch": 0.33811814219149255, + "grad_norm": 1.6051402254733438, + "learning_rate": 3.0843329196249794e-06, + "loss": 0.9973, + "num_input_tokens_seen": 30914520, + "step": 4499 + }, + { + "epoch": 0.3381932962573275, + "grad_norm": 1.4613355614483918, + "learning_rate": 3.0839238080784435e-06, + "loss": 0.9622, + "num_input_tokens_seen": 30937200, + "step": 4500 + }, + { + "epoch": 0.3382684503231625, + "grad_norm": 1.7431446405605702, + "learning_rate": 3.083514632305271e-06, + "loss": 0.9607, + "num_input_tokens_seen": 30963360, + "step": 4501 + }, + { + "epoch": 0.33834360438899747, + "grad_norm": 1.4294779472824923, + "learning_rate": 3.0831053923297074e-06, + "loss": 0.9078, + "num_input_tokens_seen": 30987740, + "step": 4502 + }, + { + "epoch": 0.3384187584548324, + "grad_norm": 1.877555694871489, + "learning_rate": 3.082696088176002e-06, + "loss": 0.9124, + "num_input_tokens_seen": 31011345, + "step": 4503 + }, + { + "epoch": 0.33849391252066735, + "grad_norm": 1.6305587765634593, + "learning_rate": 3.0822867198684073e-06, + "loss": 0.9576, + "num_input_tokens_seen": 31032220, + "step": 4504 + }, + { + "epoch": 0.3385690665865023, + "grad_norm": 1.3655260726747105, + "learning_rate": 3.0818772874311804e-06, + "loss": 0.9332, + "num_input_tokens_seen": 31054360, + "step": 4505 + }, + { + "epoch": 0.3386442206523373, + "grad_norm": 1.706600248278702, + "learning_rate": 3.081467790888581e-06, + "loss": 0.9684, + "num_input_tokens_seen": 31076280, + "step": 4506 + }, + { + "epoch": 0.33871937471817226, + "grad_norm": 1.4481348472527125, + "learning_rate": 3.0810582302648743e-06, + "loss": 0.945, + "num_input_tokens_seen": 31101700, + "step": 4507 + }, + { + "epoch": 0.33879452878400723, + "grad_norm": 1.3434135548103, + "learning_rate": 3.0806486055843276e-06, + "loss": 0.958, + "num_input_tokens_seen": 31126910, + "step": 4508 + }, + { + "epoch": 0.3388696828498422, + "grad_norm": 1.9512277161757197, + "learning_rate": 3.080238916871213e-06, + "loss": 0.9188, + "num_input_tokens_seen": 31150035, + "step": 4509 + }, + { + "epoch": 0.3389448369156771, + "grad_norm": 1.5908786863666626, + "learning_rate": 3.079829164149806e-06, + "loss": 0.9942, + "num_input_tokens_seen": 31174110, + "step": 4510 + }, + { + "epoch": 0.3390199909815121, + "grad_norm": 1.3956192431264958, + "learning_rate": 3.0794193474443866e-06, + "loss": 0.947, + "num_input_tokens_seen": 31202835, + "step": 4511 + }, + { + "epoch": 0.33909514504734706, + "grad_norm": 1.5386957031636255, + "learning_rate": 3.0790094667792368e-06, + "loss": 0.9861, + "num_input_tokens_seen": 31226535, + "step": 4512 + }, + { + "epoch": 0.339170299113182, + "grad_norm": 2.210923049933455, + "learning_rate": 3.078599522178644e-06, + "loss": 0.9075, + "num_input_tokens_seen": 31250400, + "step": 4513 + }, + { + "epoch": 0.339245453179017, + "grad_norm": 1.7218813721249848, + "learning_rate": 3.0781895136669e-06, + "loss": 0.8848, + "num_input_tokens_seen": 31274370, + "step": 4514 + }, + { + "epoch": 0.33932060724485197, + "grad_norm": 1.841601461519147, + "learning_rate": 3.077779441268299e-06, + "loss": 0.9374, + "num_input_tokens_seen": 31295645, + "step": 4515 + }, + { + "epoch": 0.3393957613106869, + "grad_norm": 1.7309207369035207, + "learning_rate": 3.077369305007138e-06, + "loss": 0.8864, + "num_input_tokens_seen": 31317905, + "step": 4516 + }, + { + "epoch": 0.33947091537652185, + "grad_norm": 1.5416906543961857, + "learning_rate": 3.07695910490772e-06, + "loss": 0.8837, + "num_input_tokens_seen": 31341015, + "step": 4517 + }, + { + "epoch": 0.3395460694423568, + "grad_norm": 0.8008242339716488, + "learning_rate": 3.076548840994352e-06, + "loss": 0.7911, + "num_input_tokens_seen": 31410165, + "step": 4518 + }, + { + "epoch": 0.3396212235081918, + "grad_norm": 2.0594826793159946, + "learning_rate": 3.076138513291342e-06, + "loss": 0.8228, + "num_input_tokens_seen": 31433485, + "step": 4519 + }, + { + "epoch": 0.33969637757402676, + "grad_norm": 1.4492394787727445, + "learning_rate": 3.0757281218230046e-06, + "loss": 0.8827, + "num_input_tokens_seen": 31457610, + "step": 4520 + }, + { + "epoch": 0.33977153163986173, + "grad_norm": 1.3322421267765914, + "learning_rate": 3.0753176666136575e-06, + "loss": 0.9483, + "num_input_tokens_seen": 31481990, + "step": 4521 + }, + { + "epoch": 0.3398466857056967, + "grad_norm": 1.4987361669975812, + "learning_rate": 3.0749071476876203e-06, + "loss": 0.9662, + "num_input_tokens_seen": 31505465, + "step": 4522 + }, + { + "epoch": 0.3399218397715316, + "grad_norm": 1.4743000438697165, + "learning_rate": 3.0744965650692184e-06, + "loss": 0.9804, + "num_input_tokens_seen": 31528130, + "step": 4523 + }, + { + "epoch": 0.3399969938373666, + "grad_norm": 1.6936892283893348, + "learning_rate": 3.0740859187827807e-06, + "loss": 0.9143, + "num_input_tokens_seen": 31549785, + "step": 4524 + }, + { + "epoch": 0.34007214790320156, + "grad_norm": 1.492883889238355, + "learning_rate": 3.0736752088526388e-06, + "loss": 1.0296, + "num_input_tokens_seen": 31572635, + "step": 4525 + }, + { + "epoch": 0.34014730196903653, + "grad_norm": 1.7573353437668835, + "learning_rate": 3.0732644353031304e-06, + "loss": 0.9645, + "num_input_tokens_seen": 31595995, + "step": 4526 + }, + { + "epoch": 0.3402224560348715, + "grad_norm": 1.3163997299078145, + "learning_rate": 3.072853598158594e-06, + "loss": 0.9467, + "num_input_tokens_seen": 31621435, + "step": 4527 + }, + { + "epoch": 0.34029761010070647, + "grad_norm": 1.6829567255042668, + "learning_rate": 3.0724426974433737e-06, + "loss": 0.9902, + "num_input_tokens_seen": 31644915, + "step": 4528 + }, + { + "epoch": 0.3403727641665414, + "grad_norm": 1.3777719214060271, + "learning_rate": 3.0720317331818163e-06, + "loss": 0.9096, + "num_input_tokens_seen": 31668525, + "step": 4529 + }, + { + "epoch": 0.34044791823237636, + "grad_norm": 0.8324626824890209, + "learning_rate": 3.071620705398274e-06, + "loss": 0.8467, + "num_input_tokens_seen": 31744370, + "step": 4530 + }, + { + "epoch": 0.3405230722982113, + "grad_norm": 1.502665456208354, + "learning_rate": 3.0712096141171017e-06, + "loss": 1.0418, + "num_input_tokens_seen": 31765380, + "step": 4531 + }, + { + "epoch": 0.3405982263640463, + "grad_norm": 1.6876543468955179, + "learning_rate": 3.070798459362658e-06, + "loss": 1.0664, + "num_input_tokens_seen": 31784060, + "step": 4532 + }, + { + "epoch": 0.34067338042988127, + "grad_norm": 1.9706085757058984, + "learning_rate": 3.070387241159305e-06, + "loss": 0.8071, + "num_input_tokens_seen": 31810610, + "step": 4533 + }, + { + "epoch": 0.34074853449571624, + "grad_norm": 1.5457981098751319, + "learning_rate": 3.069975959531408e-06, + "loss": 0.9982, + "num_input_tokens_seen": 31833870, + "step": 4534 + }, + { + "epoch": 0.34082368856155115, + "grad_norm": 1.8222512329449083, + "learning_rate": 3.0695646145033404e-06, + "loss": 0.9391, + "num_input_tokens_seen": 31859010, + "step": 4535 + }, + { + "epoch": 0.3408988426273861, + "grad_norm": 1.494863138472848, + "learning_rate": 3.0691532060994722e-06, + "loss": 0.8883, + "num_input_tokens_seen": 31885275, + "step": 4536 + }, + { + "epoch": 0.3409739966932211, + "grad_norm": 1.600920609030078, + "learning_rate": 3.068741734344183e-06, + "loss": 0.9343, + "num_input_tokens_seen": 31909780, + "step": 4537 + }, + { + "epoch": 0.34104915075905606, + "grad_norm": 1.3116090566221819, + "learning_rate": 3.0683301992618538e-06, + "loss": 0.9935, + "num_input_tokens_seen": 31933195, + "step": 4538 + }, + { + "epoch": 0.34112430482489103, + "grad_norm": 1.779866199715178, + "learning_rate": 3.067918600876869e-06, + "loss": 1.0035, + "num_input_tokens_seen": 31956410, + "step": 4539 + }, + { + "epoch": 0.341199458890726, + "grad_norm": 0.8403810423634214, + "learning_rate": 3.067506939213617e-06, + "loss": 0.8474, + "num_input_tokens_seen": 32031480, + "step": 4540 + }, + { + "epoch": 0.341274612956561, + "grad_norm": 1.8942391396501057, + "learning_rate": 3.067095214296492e-06, + "loss": 0.9665, + "num_input_tokens_seen": 32052585, + "step": 4541 + }, + { + "epoch": 0.3413497670223959, + "grad_norm": 2.1757816883134655, + "learning_rate": 3.066683426149889e-06, + "loss": 0.9586, + "num_input_tokens_seen": 32076135, + "step": 4542 + }, + { + "epoch": 0.34142492108823086, + "grad_norm": 1.5580582619033705, + "learning_rate": 3.066271574798209e-06, + "loss": 0.9629, + "num_input_tokens_seen": 32100020, + "step": 4543 + }, + { + "epoch": 0.34150007515406583, + "grad_norm": 1.731316145862398, + "learning_rate": 3.0658596602658548e-06, + "loss": 0.8852, + "num_input_tokens_seen": 32122860, + "step": 4544 + }, + { + "epoch": 0.3415752292199008, + "grad_norm": 1.663897080773659, + "learning_rate": 3.0654476825772338e-06, + "loss": 0.9906, + "num_input_tokens_seen": 32148455, + "step": 4545 + }, + { + "epoch": 0.34165038328573577, + "grad_norm": 1.5667677008062793, + "learning_rate": 3.0650356417567586e-06, + "loss": 0.9638, + "num_input_tokens_seen": 32169690, + "step": 4546 + }, + { + "epoch": 0.34172553735157074, + "grad_norm": 1.6120662224043, + "learning_rate": 3.064623537828843e-06, + "loss": 1.0006, + "num_input_tokens_seen": 32191630, + "step": 4547 + }, + { + "epoch": 0.34180069141740566, + "grad_norm": 1.7650845806182915, + "learning_rate": 3.0642113708179062e-06, + "loss": 0.9708, + "num_input_tokens_seen": 32212140, + "step": 4548 + }, + { + "epoch": 0.3418758454832406, + "grad_norm": 1.5169039963260114, + "learning_rate": 3.0637991407483706e-06, + "loss": 0.9993, + "num_input_tokens_seen": 32236430, + "step": 4549 + }, + { + "epoch": 0.3419509995490756, + "grad_norm": 1.6988833743207499, + "learning_rate": 3.0633868476446615e-06, + "loss": 0.9127, + "num_input_tokens_seen": 32258145, + "step": 4550 + }, + { + "epoch": 0.34202615361491057, + "grad_norm": 1.7204137898768532, + "learning_rate": 3.062974491531211e-06, + "loss": 0.9788, + "num_input_tokens_seen": 32280915, + "step": 4551 + }, + { + "epoch": 0.34210130768074554, + "grad_norm": 1.45904949405484, + "learning_rate": 3.06256207243245e-06, + "loss": 1.07, + "num_input_tokens_seen": 32302505, + "step": 4552 + }, + { + "epoch": 0.3421764617465805, + "grad_norm": 1.6995558130359079, + "learning_rate": 3.0621495903728177e-06, + "loss": 0.9339, + "num_input_tokens_seen": 32326195, + "step": 4553 + }, + { + "epoch": 0.3422516158124155, + "grad_norm": 1.323016450778238, + "learning_rate": 3.061737045376756e-06, + "loss": 0.8725, + "num_input_tokens_seen": 32352405, + "step": 4554 + }, + { + "epoch": 0.3423267698782504, + "grad_norm": 1.4453493680633442, + "learning_rate": 3.061324437468708e-06, + "loss": 0.8592, + "num_input_tokens_seen": 32379460, + "step": 4555 + }, + { + "epoch": 0.34240192394408536, + "grad_norm": 1.3386394549123777, + "learning_rate": 3.060911766673123e-06, + "loss": 1.0712, + "num_input_tokens_seen": 32402115, + "step": 4556 + }, + { + "epoch": 0.34247707800992033, + "grad_norm": 1.686816182929087, + "learning_rate": 3.0604990330144537e-06, + "loss": 0.9801, + "num_input_tokens_seen": 32425765, + "step": 4557 + }, + { + "epoch": 0.3425522320757553, + "grad_norm": 1.4085861298941575, + "learning_rate": 3.0600862365171553e-06, + "loss": 1.0085, + "num_input_tokens_seen": 32450450, + "step": 4558 + }, + { + "epoch": 0.3426273861415903, + "grad_norm": 1.6472724304709951, + "learning_rate": 3.0596733772056884e-06, + "loss": 0.9812, + "num_input_tokens_seen": 32475830, + "step": 4559 + }, + { + "epoch": 0.34270254020742524, + "grad_norm": 1.9823007631185043, + "learning_rate": 3.0592604551045157e-06, + "loss": 0.9449, + "num_input_tokens_seen": 32499585, + "step": 4560 + }, + { + "epoch": 0.34277769427326016, + "grad_norm": 1.747899020457925, + "learning_rate": 3.0588474702381055e-06, + "loss": 0.9469, + "num_input_tokens_seen": 32523115, + "step": 4561 + }, + { + "epoch": 0.34285284833909513, + "grad_norm": 1.7704891618355432, + "learning_rate": 3.0584344226309277e-06, + "loss": 0.9504, + "num_input_tokens_seen": 32544255, + "step": 4562 + }, + { + "epoch": 0.3429280024049301, + "grad_norm": 3.1821312663625783, + "learning_rate": 3.0580213123074573e-06, + "loss": 0.8739, + "num_input_tokens_seen": 32568265, + "step": 4563 + }, + { + "epoch": 0.34300315647076507, + "grad_norm": 0.7715757678016328, + "learning_rate": 3.0576081392921723e-06, + "loss": 0.8305, + "num_input_tokens_seen": 32651880, + "step": 4564 + }, + { + "epoch": 0.34307831053660004, + "grad_norm": 1.2124464908079489, + "learning_rate": 3.057194903609556e-06, + "loss": 0.9628, + "num_input_tokens_seen": 32680005, + "step": 4565 + }, + { + "epoch": 0.343153464602435, + "grad_norm": 1.5036942766096255, + "learning_rate": 3.056781605284093e-06, + "loss": 0.7845, + "num_input_tokens_seen": 32704240, + "step": 4566 + }, + { + "epoch": 0.34322861866827, + "grad_norm": 1.6047730806553673, + "learning_rate": 3.056368244340273e-06, + "loss": 1.0595, + "num_input_tokens_seen": 32726500, + "step": 4567 + }, + { + "epoch": 0.3433037727341049, + "grad_norm": 1.9201414151565033, + "learning_rate": 3.05595482080259e-06, + "loss": 0.9612, + "num_input_tokens_seen": 32748035, + "step": 4568 + }, + { + "epoch": 0.34337892679993987, + "grad_norm": 1.5524475216557618, + "learning_rate": 3.05554133469554e-06, + "loss": 1.0103, + "num_input_tokens_seen": 32769255, + "step": 4569 + }, + { + "epoch": 0.34345408086577484, + "grad_norm": 1.633387633154173, + "learning_rate": 3.055127786043624e-06, + "loss": 0.9419, + "num_input_tokens_seen": 32791945, + "step": 4570 + }, + { + "epoch": 0.3435292349316098, + "grad_norm": 1.8493371781339238, + "learning_rate": 3.0547141748713463e-06, + "loss": 0.9116, + "num_input_tokens_seen": 32813765, + "step": 4571 + }, + { + "epoch": 0.3436043889974448, + "grad_norm": 1.4849189163317416, + "learning_rate": 3.0543005012032152e-06, + "loss": 0.8583, + "num_input_tokens_seen": 32836515, + "step": 4572 + }, + { + "epoch": 0.34367954306327975, + "grad_norm": 1.5540347622011281, + "learning_rate": 3.0538867650637416e-06, + "loss": 1.021, + "num_input_tokens_seen": 32856390, + "step": 4573 + }, + { + "epoch": 0.34375469712911466, + "grad_norm": 0.8122937234045502, + "learning_rate": 3.053472966477442e-06, + "loss": 0.8755, + "num_input_tokens_seen": 32933695, + "step": 4574 + }, + { + "epoch": 0.34382985119494963, + "grad_norm": 1.2752336344114712, + "learning_rate": 3.053059105468835e-06, + "loss": 1.0452, + "num_input_tokens_seen": 32956345, + "step": 4575 + }, + { + "epoch": 0.3439050052607846, + "grad_norm": 2.1073588590772565, + "learning_rate": 3.052645182062444e-06, + "loss": 0.9403, + "num_input_tokens_seen": 32977735, + "step": 4576 + }, + { + "epoch": 0.3439801593266196, + "grad_norm": 1.612010242694658, + "learning_rate": 3.052231196282795e-06, + "loss": 0.9481, + "num_input_tokens_seen": 32999525, + "step": 4577 + }, + { + "epoch": 0.34405531339245454, + "grad_norm": 0.9889516253043309, + "learning_rate": 3.051817148154418e-06, + "loss": 0.8035, + "num_input_tokens_seen": 33081120, + "step": 4578 + }, + { + "epoch": 0.3441304674582895, + "grad_norm": 1.4052517208624284, + "learning_rate": 3.0514030377018473e-06, + "loss": 0.9596, + "num_input_tokens_seen": 33107270, + "step": 4579 + }, + { + "epoch": 0.3442056215241244, + "grad_norm": 1.3544341434509164, + "learning_rate": 3.0509888649496204e-06, + "loss": 0.9018, + "num_input_tokens_seen": 33137575, + "step": 4580 + }, + { + "epoch": 0.3442807755899594, + "grad_norm": 1.9102760074573932, + "learning_rate": 3.05057462992228e-06, + "loss": 1.0824, + "num_input_tokens_seen": 33160000, + "step": 4581 + }, + { + "epoch": 0.34435592965579437, + "grad_norm": 1.5810412664945348, + "learning_rate": 3.050160332644368e-06, + "loss": 0.9032, + "num_input_tokens_seen": 33184305, + "step": 4582 + }, + { + "epoch": 0.34443108372162934, + "grad_norm": 1.2154710137674356, + "learning_rate": 3.0497459731404364e-06, + "loss": 0.9556, + "num_input_tokens_seen": 33213670, + "step": 4583 + }, + { + "epoch": 0.3445062377874643, + "grad_norm": 1.5215598248050635, + "learning_rate": 3.049331551435035e-06, + "loss": 0.9845, + "num_input_tokens_seen": 33236790, + "step": 4584 + }, + { + "epoch": 0.3445813918532993, + "grad_norm": 1.5422319801973399, + "learning_rate": 3.048917067552722e-06, + "loss": 0.9515, + "num_input_tokens_seen": 33259875, + "step": 4585 + }, + { + "epoch": 0.34465654591913425, + "grad_norm": 1.6807483250775461, + "learning_rate": 3.0485025215180554e-06, + "loss": 0.9562, + "num_input_tokens_seen": 33279745, + "step": 4586 + }, + { + "epoch": 0.34473169998496916, + "grad_norm": 1.4184155141299348, + "learning_rate": 3.0480879133556e-06, + "loss": 0.9855, + "num_input_tokens_seen": 33306195, + "step": 4587 + }, + { + "epoch": 0.34480685405080413, + "grad_norm": 1.4224542967790461, + "learning_rate": 3.047673243089922e-06, + "loss": 0.9405, + "num_input_tokens_seen": 33331310, + "step": 4588 + }, + { + "epoch": 0.3448820081166391, + "grad_norm": 1.5565051470952354, + "learning_rate": 3.047258510745593e-06, + "loss": 0.999, + "num_input_tokens_seen": 33353935, + "step": 4589 + }, + { + "epoch": 0.3449571621824741, + "grad_norm": 1.529603656968729, + "learning_rate": 3.046843716347187e-06, + "loss": 0.9532, + "num_input_tokens_seen": 33377660, + "step": 4590 + }, + { + "epoch": 0.34503231624830905, + "grad_norm": 1.473198909596711, + "learning_rate": 3.046428859919281e-06, + "loss": 1.0216, + "num_input_tokens_seen": 33400360, + "step": 4591 + }, + { + "epoch": 0.345107470314144, + "grad_norm": 1.3963655434580986, + "learning_rate": 3.0460139414864593e-06, + "loss": 0.8403, + "num_input_tokens_seen": 33426880, + "step": 4592 + }, + { + "epoch": 0.34518262437997893, + "grad_norm": 2.161054483089641, + "learning_rate": 3.0455989610733057e-06, + "loss": 0.9543, + "num_input_tokens_seen": 33448440, + "step": 4593 + }, + { + "epoch": 0.3452577784458139, + "grad_norm": 1.3987697867186284, + "learning_rate": 3.0451839187044095e-06, + "loss": 0.9697, + "num_input_tokens_seen": 33472225, + "step": 4594 + }, + { + "epoch": 0.34533293251164887, + "grad_norm": 1.3389908547747935, + "learning_rate": 3.0447688144043636e-06, + "loss": 0.9782, + "num_input_tokens_seen": 33496325, + "step": 4595 + }, + { + "epoch": 0.34540808657748384, + "grad_norm": 1.414777880377575, + "learning_rate": 3.0443536481977657e-06, + "loss": 0.9405, + "num_input_tokens_seen": 33521130, + "step": 4596 + }, + { + "epoch": 0.3454832406433188, + "grad_norm": 1.6517226763151662, + "learning_rate": 3.0439384201092145e-06, + "loss": 1.0066, + "num_input_tokens_seen": 33543280, + "step": 4597 + }, + { + "epoch": 0.3455583947091538, + "grad_norm": 1.5066485878751525, + "learning_rate": 3.0435231301633147e-06, + "loss": 0.952, + "num_input_tokens_seen": 33566015, + "step": 4598 + }, + { + "epoch": 0.34563354877498875, + "grad_norm": 1.5038330060289182, + "learning_rate": 3.043107778384673e-06, + "loss": 0.9575, + "num_input_tokens_seen": 33592735, + "step": 4599 + }, + { + "epoch": 0.34570870284082367, + "grad_norm": 1.4438034014713428, + "learning_rate": 3.0426923647979016e-06, + "loss": 1.0171, + "num_input_tokens_seen": 33620285, + "step": 4600 + }, + { + "epoch": 0.34578385690665864, + "grad_norm": 0.8846289253314873, + "learning_rate": 3.042276889427615e-06, + "loss": 0.8609, + "num_input_tokens_seen": 33692870, + "step": 4601 + }, + { + "epoch": 0.3458590109724936, + "grad_norm": 1.637903190071497, + "learning_rate": 3.041861352298431e-06, + "loss": 0.9758, + "num_input_tokens_seen": 33715975, + "step": 4602 + }, + { + "epoch": 0.3459341650383286, + "grad_norm": 1.547103017319972, + "learning_rate": 3.0414457534349727e-06, + "loss": 0.9944, + "num_input_tokens_seen": 33738050, + "step": 4603 + }, + { + "epoch": 0.34600931910416355, + "grad_norm": 2.0084000852051855, + "learning_rate": 3.041030092861866e-06, + "loss": 0.8621, + "num_input_tokens_seen": 33761015, + "step": 4604 + }, + { + "epoch": 0.3460844731699985, + "grad_norm": 1.7315836946339662, + "learning_rate": 3.0406143706037384e-06, + "loss": 0.8681, + "num_input_tokens_seen": 33784745, + "step": 4605 + }, + { + "epoch": 0.34615962723583343, + "grad_norm": 1.5428594236634567, + "learning_rate": 3.040198586685226e-06, + "loss": 0.9634, + "num_input_tokens_seen": 33806150, + "step": 4606 + }, + { + "epoch": 0.3462347813016684, + "grad_norm": 1.4497977280118897, + "learning_rate": 3.0397827411309632e-06, + "loss": 0.8263, + "num_input_tokens_seen": 33829650, + "step": 4607 + }, + { + "epoch": 0.3463099353675034, + "grad_norm": 1.5783659496112852, + "learning_rate": 3.0393668339655917e-06, + "loss": 0.9316, + "num_input_tokens_seen": 33851125, + "step": 4608 + }, + { + "epoch": 0.34638508943333834, + "grad_norm": 1.3859549124568964, + "learning_rate": 3.0389508652137555e-06, + "loss": 0.8862, + "num_input_tokens_seen": 33877310, + "step": 4609 + }, + { + "epoch": 0.3464602434991733, + "grad_norm": 1.9473958942759864, + "learning_rate": 3.0385348349001023e-06, + "loss": 1.0152, + "num_input_tokens_seen": 33899960, + "step": 4610 + }, + { + "epoch": 0.3465353975650083, + "grad_norm": 1.5284801265533932, + "learning_rate": 3.038118743049283e-06, + "loss": 1.0716, + "num_input_tokens_seen": 33923520, + "step": 4611 + }, + { + "epoch": 0.34661055163084326, + "grad_norm": 1.3300877149238992, + "learning_rate": 3.0377025896859532e-06, + "loss": 1.0007, + "num_input_tokens_seen": 33947185, + "step": 4612 + }, + { + "epoch": 0.34668570569667817, + "grad_norm": 0.7271365697293393, + "learning_rate": 3.037286374834771e-06, + "loss": 0.7292, + "num_input_tokens_seen": 34028070, + "step": 4613 + }, + { + "epoch": 0.34676085976251314, + "grad_norm": 1.8522392999682638, + "learning_rate": 3.036870098520399e-06, + "loss": 1.0832, + "num_input_tokens_seen": 34048645, + "step": 4614 + }, + { + "epoch": 0.3468360138283481, + "grad_norm": 1.2098799979391126, + "learning_rate": 3.036453760767504e-06, + "loss": 0.8865, + "num_input_tokens_seen": 34073220, + "step": 4615 + }, + { + "epoch": 0.3469111678941831, + "grad_norm": 1.41397026983622, + "learning_rate": 3.036037361600754e-06, + "loss": 0.9102, + "num_input_tokens_seen": 34098335, + "step": 4616 + }, + { + "epoch": 0.34698632196001805, + "grad_norm": 1.5972494712753484, + "learning_rate": 3.0356209010448234e-06, + "loss": 0.9002, + "num_input_tokens_seen": 34121520, + "step": 4617 + }, + { + "epoch": 0.347061476025853, + "grad_norm": 3.577332521326421, + "learning_rate": 3.0352043791243886e-06, + "loss": 1.0235, + "num_input_tokens_seen": 34144745, + "step": 4618 + }, + { + "epoch": 0.34713663009168794, + "grad_norm": 1.767077835687302, + "learning_rate": 3.0347877958641303e-06, + "loss": 0.9555, + "num_input_tokens_seen": 34170530, + "step": 4619 + }, + { + "epoch": 0.3472117841575229, + "grad_norm": 1.7188439776997453, + "learning_rate": 3.0343711512887325e-06, + "loss": 0.9643, + "num_input_tokens_seen": 34193925, + "step": 4620 + }, + { + "epoch": 0.3472869382233579, + "grad_norm": 1.4268912397238307, + "learning_rate": 3.0339544454228836e-06, + "loss": 1.0529, + "num_input_tokens_seen": 34217475, + "step": 4621 + }, + { + "epoch": 0.34736209228919285, + "grad_norm": 2.2923609173266906, + "learning_rate": 3.0335376782912742e-06, + "loss": 0.9832, + "num_input_tokens_seen": 34241415, + "step": 4622 + }, + { + "epoch": 0.3474372463550278, + "grad_norm": 1.536973261350557, + "learning_rate": 3.0331208499185996e-06, + "loss": 0.9369, + "num_input_tokens_seen": 34265765, + "step": 4623 + }, + { + "epoch": 0.3475124004208628, + "grad_norm": 1.3803703165069001, + "learning_rate": 3.0327039603295587e-06, + "loss": 0.9953, + "num_input_tokens_seen": 34291960, + "step": 4624 + }, + { + "epoch": 0.3475875544866977, + "grad_norm": 1.7390227718562359, + "learning_rate": 3.032287009548853e-06, + "loss": 0.9794, + "num_input_tokens_seen": 34312855, + "step": 4625 + }, + { + "epoch": 0.3476627085525327, + "grad_norm": 1.9257047378785235, + "learning_rate": 3.03186999760119e-06, + "loss": 0.9129, + "num_input_tokens_seen": 34335590, + "step": 4626 + }, + { + "epoch": 0.34773786261836764, + "grad_norm": 1.3475472668795738, + "learning_rate": 3.031452924511279e-06, + "loss": 1.0137, + "num_input_tokens_seen": 34359165, + "step": 4627 + }, + { + "epoch": 0.3478130166842026, + "grad_norm": 1.6627559636911995, + "learning_rate": 3.031035790303831e-06, + "loss": 0.9636, + "num_input_tokens_seen": 34382550, + "step": 4628 + }, + { + "epoch": 0.3478881707500376, + "grad_norm": 1.4670837365651608, + "learning_rate": 3.030618595003565e-06, + "loss": 0.9307, + "num_input_tokens_seen": 34405250, + "step": 4629 + }, + { + "epoch": 0.34796332481587255, + "grad_norm": 1.658263152600731, + "learning_rate": 3.0302013386352004e-06, + "loss": 0.8926, + "num_input_tokens_seen": 34426840, + "step": 4630 + }, + { + "epoch": 0.3480384788817075, + "grad_norm": 1.4331265137669005, + "learning_rate": 3.0297840212234623e-06, + "loss": 0.8848, + "num_input_tokens_seen": 34452845, + "step": 4631 + }, + { + "epoch": 0.34811363294754244, + "grad_norm": 1.3304301284856062, + "learning_rate": 3.029366642793077e-06, + "loss": 0.9778, + "num_input_tokens_seen": 34476410, + "step": 4632 + }, + { + "epoch": 0.3481887870133774, + "grad_norm": 1.4112982101923184, + "learning_rate": 3.0289492033687768e-06, + "loss": 1.0014, + "num_input_tokens_seen": 34498230, + "step": 4633 + }, + { + "epoch": 0.3482639410792124, + "grad_norm": 1.4546535205252868, + "learning_rate": 3.0285317029752957e-06, + "loss": 0.8918, + "num_input_tokens_seen": 34521695, + "step": 4634 + }, + { + "epoch": 0.34833909514504735, + "grad_norm": 1.5722748750213147, + "learning_rate": 3.028114141637373e-06, + "loss": 0.9558, + "num_input_tokens_seen": 34544310, + "step": 4635 + }, + { + "epoch": 0.3484142492108823, + "grad_norm": 1.6717298926012325, + "learning_rate": 3.0276965193797503e-06, + "loss": 1.1067, + "num_input_tokens_seen": 34566730, + "step": 4636 + }, + { + "epoch": 0.3484894032767173, + "grad_norm": 1.4470155322250304, + "learning_rate": 3.0272788362271743e-06, + "loss": 0.9725, + "num_input_tokens_seen": 34590735, + "step": 4637 + }, + { + "epoch": 0.3485645573425522, + "grad_norm": 0.7885112158582985, + "learning_rate": 3.0268610922043925e-06, + "loss": 0.8068, + "num_input_tokens_seen": 34662085, + "step": 4638 + }, + { + "epoch": 0.3486397114083872, + "grad_norm": 1.476314705537741, + "learning_rate": 3.0264432873361594e-06, + "loss": 0.9336, + "num_input_tokens_seen": 34686900, + "step": 4639 + }, + { + "epoch": 0.34871486547422215, + "grad_norm": 1.4977368129658448, + "learning_rate": 3.026025421647231e-06, + "loss": 1.0287, + "num_input_tokens_seen": 34710700, + "step": 4640 + }, + { + "epoch": 0.3487900195400571, + "grad_norm": 1.3999732764282091, + "learning_rate": 3.025607495162367e-06, + "loss": 0.9638, + "num_input_tokens_seen": 34734200, + "step": 4641 + }, + { + "epoch": 0.3488651736058921, + "grad_norm": 2.0041275348398666, + "learning_rate": 3.025189507906332e-06, + "loss": 0.9797, + "num_input_tokens_seen": 34756275, + "step": 4642 + }, + { + "epoch": 0.34894032767172706, + "grad_norm": 1.891172912902459, + "learning_rate": 3.0247714599038936e-06, + "loss": 0.9508, + "num_input_tokens_seen": 34779620, + "step": 4643 + }, + { + "epoch": 0.349015481737562, + "grad_norm": 1.152119087689959, + "learning_rate": 3.0243533511798205e-06, + "loss": 0.9581, + "num_input_tokens_seen": 34807830, + "step": 4644 + }, + { + "epoch": 0.34909063580339694, + "grad_norm": 1.4137728484016612, + "learning_rate": 3.0239351817588903e-06, + "loss": 0.8826, + "num_input_tokens_seen": 34834860, + "step": 4645 + }, + { + "epoch": 0.3491657898692319, + "grad_norm": 1.3021807316273941, + "learning_rate": 3.023516951665879e-06, + "loss": 0.9803, + "num_input_tokens_seen": 34855645, + "step": 4646 + }, + { + "epoch": 0.3492409439350669, + "grad_norm": 1.575113728368114, + "learning_rate": 3.0230986609255687e-06, + "loss": 0.985, + "num_input_tokens_seen": 34877000, + "step": 4647 + }, + { + "epoch": 0.34931609800090185, + "grad_norm": 1.5263867123771209, + "learning_rate": 3.022680309562746e-06, + "loss": 1.0664, + "num_input_tokens_seen": 34899025, + "step": 4648 + }, + { + "epoch": 0.3493912520667368, + "grad_norm": 1.570846163278223, + "learning_rate": 3.022261897602198e-06, + "loss": 0.86, + "num_input_tokens_seen": 34923525, + "step": 4649 + }, + { + "epoch": 0.3494664061325718, + "grad_norm": 1.5013191462592357, + "learning_rate": 3.0218434250687184e-06, + "loss": 0.9328, + "num_input_tokens_seen": 34946850, + "step": 4650 + }, + { + "epoch": 0.3495415601984067, + "grad_norm": 1.5484111364066142, + "learning_rate": 3.021424891987103e-06, + "loss": 0.9272, + "num_input_tokens_seen": 34969055, + "step": 4651 + }, + { + "epoch": 0.3496167142642417, + "grad_norm": 1.4497851475669885, + "learning_rate": 3.0210062983821513e-06, + "loss": 0.9275, + "num_input_tokens_seen": 34991155, + "step": 4652 + }, + { + "epoch": 0.34969186833007665, + "grad_norm": 1.3265916387550125, + "learning_rate": 3.0205876442786666e-06, + "loss": 0.9278, + "num_input_tokens_seen": 35015240, + "step": 4653 + }, + { + "epoch": 0.3497670223959116, + "grad_norm": 1.3951335642736078, + "learning_rate": 3.0201689297014565e-06, + "loss": 0.9314, + "num_input_tokens_seen": 35039265, + "step": 4654 + }, + { + "epoch": 0.3498421764617466, + "grad_norm": 1.6880245806228615, + "learning_rate": 3.01975015467533e-06, + "loss": 0.9859, + "num_input_tokens_seen": 35061225, + "step": 4655 + }, + { + "epoch": 0.34991733052758156, + "grad_norm": 1.4192328585056038, + "learning_rate": 3.019331319225103e-06, + "loss": 0.9923, + "num_input_tokens_seen": 35081940, + "step": 4656 + }, + { + "epoch": 0.34999248459341653, + "grad_norm": 2.3878340043116935, + "learning_rate": 3.018912423375591e-06, + "loss": 0.9326, + "num_input_tokens_seen": 35104460, + "step": 4657 + }, + { + "epoch": 0.35006763865925145, + "grad_norm": 1.783383614137278, + "learning_rate": 3.018493467151616e-06, + "loss": 0.9336, + "num_input_tokens_seen": 35125575, + "step": 4658 + }, + { + "epoch": 0.3501427927250864, + "grad_norm": 1.5773779734525972, + "learning_rate": 3.0180744505780045e-06, + "loss": 0.8437, + "num_input_tokens_seen": 35145695, + "step": 4659 + }, + { + "epoch": 0.3502179467909214, + "grad_norm": 1.6447336973862952, + "learning_rate": 3.0176553736795827e-06, + "loss": 0.9473, + "num_input_tokens_seen": 35169030, + "step": 4660 + }, + { + "epoch": 0.35029310085675636, + "grad_norm": 1.3866020046037517, + "learning_rate": 3.0172362364811827e-06, + "loss": 0.9618, + "num_input_tokens_seen": 35193105, + "step": 4661 + }, + { + "epoch": 0.3503682549225913, + "grad_norm": 1.5746978712185131, + "learning_rate": 3.016817039007641e-06, + "loss": 0.9736, + "num_input_tokens_seen": 35216700, + "step": 4662 + }, + { + "epoch": 0.3504434089884263, + "grad_norm": 1.4772607676148641, + "learning_rate": 3.0163977812837954e-06, + "loss": 0.9894, + "num_input_tokens_seen": 35239355, + "step": 4663 + }, + { + "epoch": 0.3505185630542612, + "grad_norm": 2.088569266754457, + "learning_rate": 3.0159784633344894e-06, + "loss": 0.8744, + "num_input_tokens_seen": 35261610, + "step": 4664 + }, + { + "epoch": 0.3505937171200962, + "grad_norm": 0.7600249017224704, + "learning_rate": 3.0155590851845694e-06, + "loss": 0.7959, + "num_input_tokens_seen": 35336165, + "step": 4665 + }, + { + "epoch": 0.35066887118593115, + "grad_norm": 1.3704519046526236, + "learning_rate": 3.0151396468588844e-06, + "loss": 0.9613, + "num_input_tokens_seen": 35361425, + "step": 4666 + }, + { + "epoch": 0.3507440252517661, + "grad_norm": 1.465346512028582, + "learning_rate": 3.0147201483822884e-06, + "loss": 1.0079, + "num_input_tokens_seen": 35383175, + "step": 4667 + }, + { + "epoch": 0.3508191793176011, + "grad_norm": 1.3690897324010234, + "learning_rate": 3.014300589779638e-06, + "loss": 0.9695, + "num_input_tokens_seen": 35403520, + "step": 4668 + }, + { + "epoch": 0.35089433338343606, + "grad_norm": 1.5329224153358867, + "learning_rate": 3.0138809710757927e-06, + "loss": 0.9768, + "num_input_tokens_seen": 35426665, + "step": 4669 + }, + { + "epoch": 0.350969487449271, + "grad_norm": 2.0774066802180666, + "learning_rate": 3.013461292295619e-06, + "loss": 0.9591, + "num_input_tokens_seen": 35450645, + "step": 4670 + }, + { + "epoch": 0.35104464151510595, + "grad_norm": 1.811166009317952, + "learning_rate": 3.013041553463982e-06, + "loss": 1.0246, + "num_input_tokens_seen": 35475670, + "step": 4671 + }, + { + "epoch": 0.3511197955809409, + "grad_norm": 1.4367924483098677, + "learning_rate": 3.012621754605754e-06, + "loss": 0.8209, + "num_input_tokens_seen": 35500265, + "step": 4672 + }, + { + "epoch": 0.3511949496467759, + "grad_norm": 1.4328080604675804, + "learning_rate": 3.012201895745809e-06, + "loss": 0.8735, + "num_input_tokens_seen": 35526760, + "step": 4673 + }, + { + "epoch": 0.35127010371261086, + "grad_norm": 1.3701319785485475, + "learning_rate": 3.011781976909026e-06, + "loss": 0.9394, + "num_input_tokens_seen": 35552505, + "step": 4674 + }, + { + "epoch": 0.35134525777844583, + "grad_norm": 1.601680025580219, + "learning_rate": 3.011361998120287e-06, + "loss": 0.9169, + "num_input_tokens_seen": 35576670, + "step": 4675 + }, + { + "epoch": 0.3514204118442808, + "grad_norm": 1.3636656823040874, + "learning_rate": 3.0109419594044765e-06, + "loss": 1.0165, + "num_input_tokens_seen": 35597370, + "step": 4676 + }, + { + "epoch": 0.3514955659101157, + "grad_norm": 1.696524261701371, + "learning_rate": 3.0105218607864835e-06, + "loss": 1.0483, + "num_input_tokens_seen": 35618795, + "step": 4677 + }, + { + "epoch": 0.3515707199759507, + "grad_norm": 1.5204759813724187, + "learning_rate": 3.010101702291201e-06, + "loss": 1.0067, + "num_input_tokens_seen": 35639490, + "step": 4678 + }, + { + "epoch": 0.35164587404178566, + "grad_norm": 1.4359899135649774, + "learning_rate": 3.0096814839435244e-06, + "loss": 0.9142, + "num_input_tokens_seen": 35663695, + "step": 4679 + }, + { + "epoch": 0.3517210281076206, + "grad_norm": 1.441157539227306, + "learning_rate": 3.0092612057683532e-06, + "loss": 0.8954, + "num_input_tokens_seen": 35688910, + "step": 4680 + }, + { + "epoch": 0.3517961821734556, + "grad_norm": 1.2870776993091124, + "learning_rate": 3.0088408677905913e-06, + "loss": 1.0614, + "num_input_tokens_seen": 35714440, + "step": 4681 + }, + { + "epoch": 0.35187133623929057, + "grad_norm": 1.7683987730443496, + "learning_rate": 3.0084204700351453e-06, + "loss": 0.848, + "num_input_tokens_seen": 35739740, + "step": 4682 + }, + { + "epoch": 0.3519464903051255, + "grad_norm": 1.2672407409546453, + "learning_rate": 3.0080000125269242e-06, + "loss": 0.9422, + "num_input_tokens_seen": 35764815, + "step": 4683 + }, + { + "epoch": 0.35202164437096045, + "grad_norm": 1.350968776237592, + "learning_rate": 3.0075794952908436e-06, + "loss": 0.8668, + "num_input_tokens_seen": 35788735, + "step": 4684 + }, + { + "epoch": 0.3520967984367954, + "grad_norm": 1.8100839494236673, + "learning_rate": 3.007158918351818e-06, + "loss": 0.9375, + "num_input_tokens_seen": 35812520, + "step": 4685 + }, + { + "epoch": 0.3521719525026304, + "grad_norm": 1.4884305463584815, + "learning_rate": 3.0067382817347712e-06, + "loss": 0.8769, + "num_input_tokens_seen": 35840625, + "step": 4686 + }, + { + "epoch": 0.35224710656846536, + "grad_norm": 1.368532097250159, + "learning_rate": 3.006317585464626e-06, + "loss": 1.0321, + "num_input_tokens_seen": 35864150, + "step": 4687 + }, + { + "epoch": 0.35232226063430033, + "grad_norm": 2.123116612276045, + "learning_rate": 3.0058968295663094e-06, + "loss": 1.0269, + "num_input_tokens_seen": 35887925, + "step": 4688 + }, + { + "epoch": 0.3523974147001353, + "grad_norm": 1.9151380981172688, + "learning_rate": 3.0054760140647547e-06, + "loss": 0.9797, + "num_input_tokens_seen": 35909330, + "step": 4689 + }, + { + "epoch": 0.3524725687659702, + "grad_norm": 1.6046635820041915, + "learning_rate": 3.005055138984896e-06, + "loss": 0.9952, + "num_input_tokens_seen": 35931375, + "step": 4690 + }, + { + "epoch": 0.3525477228318052, + "grad_norm": 1.4173315208461748, + "learning_rate": 3.0046342043516707e-06, + "loss": 0.996, + "num_input_tokens_seen": 35955745, + "step": 4691 + }, + { + "epoch": 0.35262287689764016, + "grad_norm": 2.0496152051461416, + "learning_rate": 3.0042132101900228e-06, + "loss": 0.9907, + "num_input_tokens_seen": 35976900, + "step": 4692 + }, + { + "epoch": 0.35269803096347513, + "grad_norm": 1.4130693182926912, + "learning_rate": 3.003792156524897e-06, + "loss": 0.9283, + "num_input_tokens_seen": 35999205, + "step": 4693 + }, + { + "epoch": 0.3527731850293101, + "grad_norm": 0.840780826576295, + "learning_rate": 3.003371043381241e-06, + "loss": 0.816, + "num_input_tokens_seen": 36076895, + "step": 4694 + }, + { + "epoch": 0.35284833909514507, + "grad_norm": 1.8612531905086065, + "learning_rate": 3.0029498707840094e-06, + "loss": 1.0085, + "num_input_tokens_seen": 36101760, + "step": 4695 + }, + { + "epoch": 0.35292349316098, + "grad_norm": 1.8343578279331594, + "learning_rate": 3.002528638758157e-06, + "loss": 0.9809, + "num_input_tokens_seen": 36120965, + "step": 4696 + }, + { + "epoch": 0.35299864722681495, + "grad_norm": 1.852385004260726, + "learning_rate": 3.0021073473286446e-06, + "loss": 0.8078, + "num_input_tokens_seen": 36144425, + "step": 4697 + }, + { + "epoch": 0.3530738012926499, + "grad_norm": 1.5083145333623633, + "learning_rate": 3.0016859965204336e-06, + "loss": 0.9959, + "num_input_tokens_seen": 36170940, + "step": 4698 + }, + { + "epoch": 0.3531489553584849, + "grad_norm": 0.7492054067941041, + "learning_rate": 3.001264586358492e-06, + "loss": 0.7786, + "num_input_tokens_seen": 36242750, + "step": 4699 + }, + { + "epoch": 0.35322410942431987, + "grad_norm": 1.4328144668349283, + "learning_rate": 3.0008431168677898e-06, + "loss": 0.9996, + "num_input_tokens_seen": 36265955, + "step": 4700 + }, + { + "epoch": 0.35329926349015484, + "grad_norm": 1.7077334288836143, + "learning_rate": 3.0004215880732993e-06, + "loss": 1.0036, + "num_input_tokens_seen": 36284320, + "step": 4701 + }, + { + "epoch": 0.3533744175559898, + "grad_norm": 1.3117834133184245, + "learning_rate": 3e-06, + "loss": 0.9925, + "num_input_tokens_seen": 36307570, + "step": 4702 + }, + { + "epoch": 0.3534495716218247, + "grad_norm": 1.4082844746407888, + "learning_rate": 2.999578352672871e-06, + "loss": 1.022, + "num_input_tokens_seen": 36331985, + "step": 4703 + }, + { + "epoch": 0.3535247256876597, + "grad_norm": 1.272237715680002, + "learning_rate": 2.9991566461168974e-06, + "loss": 0.9404, + "num_input_tokens_seen": 36358275, + "step": 4704 + }, + { + "epoch": 0.35359987975349466, + "grad_norm": 1.722699499992962, + "learning_rate": 2.998734880357066e-06, + "loss": 0.8612, + "num_input_tokens_seen": 36381135, + "step": 4705 + }, + { + "epoch": 0.35367503381932963, + "grad_norm": 1.8244012582058238, + "learning_rate": 2.998313055418369e-06, + "loss": 1.0818, + "num_input_tokens_seen": 36401310, + "step": 4706 + }, + { + "epoch": 0.3537501878851646, + "grad_norm": 2.065055420064908, + "learning_rate": 2.9978911713257998e-06, + "loss": 1.0392, + "num_input_tokens_seen": 36420670, + "step": 4707 + }, + { + "epoch": 0.3538253419509996, + "grad_norm": 2.783874816151832, + "learning_rate": 2.997469228104358e-06, + "loss": 0.9426, + "num_input_tokens_seen": 36442590, + "step": 4708 + }, + { + "epoch": 0.3539004960168345, + "grad_norm": 1.4576067976649674, + "learning_rate": 2.9970472257790454e-06, + "loss": 0.9574, + "num_input_tokens_seen": 36465460, + "step": 4709 + }, + { + "epoch": 0.35397565008266946, + "grad_norm": 1.5753205563536539, + "learning_rate": 2.996625164374866e-06, + "loss": 0.9935, + "num_input_tokens_seen": 36486845, + "step": 4710 + }, + { + "epoch": 0.3540508041485044, + "grad_norm": 0.676413532446107, + "learning_rate": 2.9962030439168297e-06, + "loss": 0.8139, + "num_input_tokens_seen": 36564590, + "step": 4711 + }, + { + "epoch": 0.3541259582143394, + "grad_norm": 1.7382427768521538, + "learning_rate": 2.995780864429948e-06, + "loss": 0.7441, + "num_input_tokens_seen": 36588670, + "step": 4712 + }, + { + "epoch": 0.35420111228017437, + "grad_norm": 1.2664788744821536, + "learning_rate": 2.9953586259392366e-06, + "loss": 0.8925, + "num_input_tokens_seen": 36614955, + "step": 4713 + }, + { + "epoch": 0.35427626634600934, + "grad_norm": 1.3817444033011594, + "learning_rate": 2.994936328469716e-06, + "loss": 1.0042, + "num_input_tokens_seen": 36640205, + "step": 4714 + }, + { + "epoch": 0.35435142041184425, + "grad_norm": 1.7327937573175174, + "learning_rate": 2.9945139720464082e-06, + "loss": 0.9357, + "num_input_tokens_seen": 36663370, + "step": 4715 + }, + { + "epoch": 0.3544265744776792, + "grad_norm": 1.5914617080276503, + "learning_rate": 2.9940915566943384e-06, + "loss": 0.9089, + "num_input_tokens_seen": 36685860, + "step": 4716 + }, + { + "epoch": 0.3545017285435142, + "grad_norm": 1.4145695741384108, + "learning_rate": 2.9936690824385383e-06, + "loss": 1.0558, + "num_input_tokens_seen": 36707865, + "step": 4717 + }, + { + "epoch": 0.35457688260934916, + "grad_norm": 1.5276424964581417, + "learning_rate": 2.9932465493040393e-06, + "loss": 0.9293, + "num_input_tokens_seen": 36730600, + "step": 4718 + }, + { + "epoch": 0.35465203667518413, + "grad_norm": 1.5239964169037117, + "learning_rate": 2.992823957315879e-06, + "loss": 0.9842, + "num_input_tokens_seen": 36749870, + "step": 4719 + }, + { + "epoch": 0.3547271907410191, + "grad_norm": 1.959139360437412, + "learning_rate": 2.9924013064990974e-06, + "loss": 1.0328, + "num_input_tokens_seen": 36772510, + "step": 4720 + }, + { + "epoch": 0.3548023448068541, + "grad_norm": 1.4836401475638825, + "learning_rate": 2.9919785968787384e-06, + "loss": 1.0269, + "num_input_tokens_seen": 36794275, + "step": 4721 + }, + { + "epoch": 0.354877498872689, + "grad_norm": 1.6036159098369998, + "learning_rate": 2.991555828479849e-06, + "loss": 0.9743, + "num_input_tokens_seen": 36817485, + "step": 4722 + }, + { + "epoch": 0.35495265293852396, + "grad_norm": 1.4055887150947062, + "learning_rate": 2.9911330013274792e-06, + "loss": 0.8959, + "num_input_tokens_seen": 36840425, + "step": 4723 + }, + { + "epoch": 0.35502780700435893, + "grad_norm": 0.7764961183602019, + "learning_rate": 2.990710115446684e-06, + "loss": 0.7361, + "num_input_tokens_seen": 36909810, + "step": 4724 + }, + { + "epoch": 0.3551029610701939, + "grad_norm": 1.9998230855896295, + "learning_rate": 2.9902871708625216e-06, + "loss": 1.0613, + "num_input_tokens_seen": 36929180, + "step": 4725 + }, + { + "epoch": 0.35517811513602887, + "grad_norm": 1.609491881042841, + "learning_rate": 2.9898641676000518e-06, + "loss": 0.917, + "num_input_tokens_seen": 36951255, + "step": 4726 + }, + { + "epoch": 0.35525326920186384, + "grad_norm": 1.5059460566765022, + "learning_rate": 2.9894411056843396e-06, + "loss": 0.9634, + "num_input_tokens_seen": 36973895, + "step": 4727 + }, + { + "epoch": 0.35532842326769876, + "grad_norm": 1.4140587906762467, + "learning_rate": 2.9890179851404533e-06, + "loss": 0.9162, + "num_input_tokens_seen": 37002515, + "step": 4728 + }, + { + "epoch": 0.3554035773335337, + "grad_norm": 1.6806503020481858, + "learning_rate": 2.9885948059934635e-06, + "loss": 1.0092, + "num_input_tokens_seen": 37025545, + "step": 4729 + }, + { + "epoch": 0.3554787313993687, + "grad_norm": 1.49543130018051, + "learning_rate": 2.988171568268446e-06, + "loss": 1.0112, + "num_input_tokens_seen": 37049035, + "step": 4730 + }, + { + "epoch": 0.35555388546520367, + "grad_norm": 1.505481479592937, + "learning_rate": 2.98774827199048e-06, + "loss": 0.9845, + "num_input_tokens_seen": 37072715, + "step": 4731 + }, + { + "epoch": 0.35562903953103864, + "grad_norm": 1.4964357786918707, + "learning_rate": 2.9873249171846454e-06, + "loss": 0.9174, + "num_input_tokens_seen": 37095860, + "step": 4732 + }, + { + "epoch": 0.3557041935968736, + "grad_norm": 1.69901563811669, + "learning_rate": 2.9869015038760296e-06, + "loss": 0.9247, + "num_input_tokens_seen": 37117050, + "step": 4733 + }, + { + "epoch": 0.3557793476627086, + "grad_norm": 1.355922216367513, + "learning_rate": 2.98647803208972e-06, + "loss": 1.0845, + "num_input_tokens_seen": 37139795, + "step": 4734 + }, + { + "epoch": 0.3558545017285435, + "grad_norm": 1.4081445012475593, + "learning_rate": 2.98605450185081e-06, + "loss": 0.9783, + "num_input_tokens_seen": 37162725, + "step": 4735 + }, + { + "epoch": 0.35592965579437846, + "grad_norm": 1.6016155699334187, + "learning_rate": 2.9856309131843945e-06, + "loss": 0.9816, + "num_input_tokens_seen": 37187070, + "step": 4736 + }, + { + "epoch": 0.35600480986021343, + "grad_norm": 1.4179357099230472, + "learning_rate": 2.985207266115574e-06, + "loss": 0.9475, + "num_input_tokens_seen": 37210230, + "step": 4737 + }, + { + "epoch": 0.3560799639260484, + "grad_norm": 1.5909082846206009, + "learning_rate": 2.9847835606694494e-06, + "loss": 1.0302, + "num_input_tokens_seen": 37232270, + "step": 4738 + }, + { + "epoch": 0.3561551179918834, + "grad_norm": 1.9866333373472131, + "learning_rate": 2.9843597968711285e-06, + "loss": 0.8795, + "num_input_tokens_seen": 37255520, + "step": 4739 + }, + { + "epoch": 0.35623027205771834, + "grad_norm": 0.8942844766780624, + "learning_rate": 2.9839359747457195e-06, + "loss": 0.7461, + "num_input_tokens_seen": 37333395, + "step": 4740 + }, + { + "epoch": 0.35630542612355326, + "grad_norm": 2.021072124517386, + "learning_rate": 2.9835120943183374e-06, + "loss": 0.989, + "num_input_tokens_seen": 37351775, + "step": 4741 + }, + { + "epoch": 0.35638058018938823, + "grad_norm": 1.5076480439783413, + "learning_rate": 2.9830881556140965e-06, + "loss": 0.918, + "num_input_tokens_seen": 37376865, + "step": 4742 + }, + { + "epoch": 0.3564557342552232, + "grad_norm": 1.4789969745421916, + "learning_rate": 2.9826641586581184e-06, + "loss": 0.9545, + "num_input_tokens_seen": 37400850, + "step": 4743 + }, + { + "epoch": 0.35653088832105817, + "grad_norm": 1.859398721495545, + "learning_rate": 2.9822401034755255e-06, + "loss": 1.0503, + "num_input_tokens_seen": 37420070, + "step": 4744 + }, + { + "epoch": 0.35660604238689314, + "grad_norm": 1.368589456279076, + "learning_rate": 2.981815990091446e-06, + "loss": 0.9402, + "num_input_tokens_seen": 37443875, + "step": 4745 + }, + { + "epoch": 0.3566811964527281, + "grad_norm": 1.5132411811656092, + "learning_rate": 2.9813918185310085e-06, + "loss": 1.0639, + "num_input_tokens_seen": 37466955, + "step": 4746 + }, + { + "epoch": 0.3567563505185631, + "grad_norm": 1.3434574340988301, + "learning_rate": 2.9809675888193486e-06, + "loss": 1.0003, + "num_input_tokens_seen": 37492600, + "step": 4747 + }, + { + "epoch": 0.356831504584398, + "grad_norm": 0.8823488756628434, + "learning_rate": 2.9805433009816024e-06, + "loss": 0.8362, + "num_input_tokens_seen": 37565960, + "step": 4748 + }, + { + "epoch": 0.35690665865023297, + "grad_norm": 1.843781939730964, + "learning_rate": 2.980118955042911e-06, + "loss": 0.9507, + "num_input_tokens_seen": 37591015, + "step": 4749 + }, + { + "epoch": 0.35698181271606794, + "grad_norm": 1.527687521879067, + "learning_rate": 2.9796945510284187e-06, + "loss": 0.8682, + "num_input_tokens_seen": 37617350, + "step": 4750 + }, + { + "epoch": 0.3570569667819029, + "grad_norm": 1.7618494091090213, + "learning_rate": 2.9792700889632716e-06, + "loss": 0.9965, + "num_input_tokens_seen": 37638860, + "step": 4751 + }, + { + "epoch": 0.3571321208477379, + "grad_norm": 1.6045912969142817, + "learning_rate": 2.9788455688726234e-06, + "loss": 0.9527, + "num_input_tokens_seen": 37662375, + "step": 4752 + }, + { + "epoch": 0.35720727491357285, + "grad_norm": 1.6804066382740352, + "learning_rate": 2.978420990781626e-06, + "loss": 1.032, + "num_input_tokens_seen": 37686190, + "step": 4753 + }, + { + "epoch": 0.35728242897940776, + "grad_norm": 1.4719393576383157, + "learning_rate": 2.977996354715438e-06, + "loss": 0.8516, + "num_input_tokens_seen": 37710100, + "step": 4754 + }, + { + "epoch": 0.35735758304524273, + "grad_norm": 1.610290415142515, + "learning_rate": 2.9775716606992217e-06, + "loss": 0.992, + "num_input_tokens_seen": 37737090, + "step": 4755 + }, + { + "epoch": 0.3574327371110777, + "grad_norm": 1.3948845081991517, + "learning_rate": 2.977146908758141e-06, + "loss": 1.0546, + "num_input_tokens_seen": 37758265, + "step": 4756 + }, + { + "epoch": 0.3575078911769127, + "grad_norm": 2.100291608818055, + "learning_rate": 2.9767220989173635e-06, + "loss": 0.9556, + "num_input_tokens_seen": 37778410, + "step": 4757 + }, + { + "epoch": 0.35758304524274764, + "grad_norm": 1.49195148938348, + "learning_rate": 2.9762972312020623e-06, + "loss": 0.9797, + "num_input_tokens_seen": 37799295, + "step": 4758 + }, + { + "epoch": 0.3576581993085826, + "grad_norm": 1.42563032631755, + "learning_rate": 2.975872305637412e-06, + "loss": 0.8845, + "num_input_tokens_seen": 37824130, + "step": 4759 + }, + { + "epoch": 0.35773335337441753, + "grad_norm": 1.7651677847802183, + "learning_rate": 2.97544732224859e-06, + "loss": 0.9031, + "num_input_tokens_seen": 37845830, + "step": 4760 + }, + { + "epoch": 0.3578085074402525, + "grad_norm": 1.6361005612551163, + "learning_rate": 2.975022281060779e-06, + "loss": 0.8924, + "num_input_tokens_seen": 37868215, + "step": 4761 + }, + { + "epoch": 0.35788366150608747, + "grad_norm": 1.3467404999809505, + "learning_rate": 2.9745971820991643e-06, + "loss": 0.9462, + "num_input_tokens_seen": 37894810, + "step": 4762 + }, + { + "epoch": 0.35795881557192244, + "grad_norm": 1.3907635651893235, + "learning_rate": 2.9741720253889346e-06, + "loss": 0.9213, + "num_input_tokens_seen": 37917950, + "step": 4763 + }, + { + "epoch": 0.3580339696377574, + "grad_norm": 1.8610311873752414, + "learning_rate": 2.9737468109552827e-06, + "loss": 0.9746, + "num_input_tokens_seen": 37936480, + "step": 4764 + }, + { + "epoch": 0.3581091237035924, + "grad_norm": 1.6153361462563207, + "learning_rate": 2.973321538823402e-06, + "loss": 0.9902, + "num_input_tokens_seen": 37958270, + "step": 4765 + }, + { + "epoch": 0.35818427776942735, + "grad_norm": 1.4259601637531252, + "learning_rate": 2.9728962090184938e-06, + "loss": 1.1513, + "num_input_tokens_seen": 37981505, + "step": 4766 + }, + { + "epoch": 0.35825943183526227, + "grad_norm": 1.3806952523299696, + "learning_rate": 2.9724708215657603e-06, + "loss": 0.8713, + "num_input_tokens_seen": 38054355, + "step": 4767 + }, + { + "epoch": 0.35833458590109724, + "grad_norm": 1.5734028180721809, + "learning_rate": 2.972045376490406e-06, + "loss": 0.9321, + "num_input_tokens_seen": 38078175, + "step": 4768 + }, + { + "epoch": 0.3584097399669322, + "grad_norm": 1.4767585906758383, + "learning_rate": 2.971619873817642e-06, + "loss": 0.8529, + "num_input_tokens_seen": 38098530, + "step": 4769 + }, + { + "epoch": 0.3584848940327672, + "grad_norm": 1.8546019125443232, + "learning_rate": 2.971194313572679e-06, + "loss": 0.9597, + "num_input_tokens_seen": 38118220, + "step": 4770 + }, + { + "epoch": 0.35856004809860215, + "grad_norm": 1.5884536552020432, + "learning_rate": 2.970768695780734e-06, + "loss": 0.9387, + "num_input_tokens_seen": 38143495, + "step": 4771 + }, + { + "epoch": 0.3586352021644371, + "grad_norm": 1.9013843662953407, + "learning_rate": 2.970343020467027e-06, + "loss": 0.8525, + "num_input_tokens_seen": 38164030, + "step": 4772 + }, + { + "epoch": 0.35871035623027203, + "grad_norm": 0.8133755514712565, + "learning_rate": 2.9699172876567795e-06, + "loss": 0.8107, + "num_input_tokens_seen": 38240485, + "step": 4773 + }, + { + "epoch": 0.358785510296107, + "grad_norm": 1.4505060036504422, + "learning_rate": 2.969491497375219e-06, + "loss": 0.9312, + "num_input_tokens_seen": 38264565, + "step": 4774 + }, + { + "epoch": 0.358860664361942, + "grad_norm": 1.4365634355091819, + "learning_rate": 2.969065649647575e-06, + "loss": 1.0007, + "num_input_tokens_seen": 38288515, + "step": 4775 + }, + { + "epoch": 0.35893581842777694, + "grad_norm": 2.10149306852662, + "learning_rate": 2.9686397444990803e-06, + "loss": 0.9947, + "num_input_tokens_seen": 38311055, + "step": 4776 + }, + { + "epoch": 0.3590109724936119, + "grad_norm": 1.5259037497977552, + "learning_rate": 2.9682137819549718e-06, + "loss": 0.9494, + "num_input_tokens_seen": 38337110, + "step": 4777 + }, + { + "epoch": 0.3590861265594469, + "grad_norm": 1.576531086845974, + "learning_rate": 2.9677877620404887e-06, + "loss": 1.0167, + "num_input_tokens_seen": 38359130, + "step": 4778 + }, + { + "epoch": 0.35916128062528185, + "grad_norm": 1.86193664786781, + "learning_rate": 2.9673616847808755e-06, + "loss": 0.9426, + "num_input_tokens_seen": 38383945, + "step": 4779 + }, + { + "epoch": 0.35923643469111677, + "grad_norm": 1.6459367051637785, + "learning_rate": 2.966935550201378e-06, + "loss": 0.9305, + "num_input_tokens_seen": 38407810, + "step": 4780 + }, + { + "epoch": 0.35931158875695174, + "grad_norm": 1.2608323896009128, + "learning_rate": 2.9665093583272463e-06, + "loss": 0.895, + "num_input_tokens_seen": 38432195, + "step": 4781 + }, + { + "epoch": 0.3593867428227867, + "grad_norm": 0.8649679768428381, + "learning_rate": 2.966083109183734e-06, + "loss": 0.9035, + "num_input_tokens_seen": 38505465, + "step": 4782 + }, + { + "epoch": 0.3594618968886217, + "grad_norm": 1.6043718958112454, + "learning_rate": 2.9656568027960984e-06, + "loss": 0.9433, + "num_input_tokens_seen": 38523920, + "step": 4783 + }, + { + "epoch": 0.35953705095445665, + "grad_norm": 1.6821114726055946, + "learning_rate": 2.9652304391895994e-06, + "loss": 0.9543, + "num_input_tokens_seen": 38547465, + "step": 4784 + }, + { + "epoch": 0.3596122050202916, + "grad_norm": 1.431745538341244, + "learning_rate": 2.9648040183895004e-06, + "loss": 1.0304, + "num_input_tokens_seen": 38570730, + "step": 4785 + }, + { + "epoch": 0.35968735908612653, + "grad_norm": 1.4686964309338377, + "learning_rate": 2.964377540421069e-06, + "loss": 1.0942, + "num_input_tokens_seen": 38595320, + "step": 4786 + }, + { + "epoch": 0.3597625131519615, + "grad_norm": 1.7512966529355476, + "learning_rate": 2.963951005309576e-06, + "loss": 0.9231, + "num_input_tokens_seen": 38621260, + "step": 4787 + }, + { + "epoch": 0.3598376672177965, + "grad_norm": 1.4438607012430653, + "learning_rate": 2.963524413080294e-06, + "loss": 0.9394, + "num_input_tokens_seen": 38645100, + "step": 4788 + }, + { + "epoch": 0.35991282128363145, + "grad_norm": 2.0554768601666775, + "learning_rate": 2.9630977637585016e-06, + "loss": 0.9168, + "num_input_tokens_seen": 38667965, + "step": 4789 + }, + { + "epoch": 0.3599879753494664, + "grad_norm": 1.4892321174781296, + "learning_rate": 2.9626710573694783e-06, + "loss": 0.9452, + "num_input_tokens_seen": 38694660, + "step": 4790 + }, + { + "epoch": 0.3600631294153014, + "grad_norm": 1.5458507036465319, + "learning_rate": 2.9622442939385085e-06, + "loss": 1.0415, + "num_input_tokens_seen": 38716625, + "step": 4791 + }, + { + "epoch": 0.36013828348113636, + "grad_norm": 1.6700375165551402, + "learning_rate": 2.96181747349088e-06, + "loss": 1.0617, + "num_input_tokens_seen": 38739465, + "step": 4792 + }, + { + "epoch": 0.36021343754697127, + "grad_norm": 2.1090042353551817, + "learning_rate": 2.9613905960518832e-06, + "loss": 0.888, + "num_input_tokens_seen": 38762105, + "step": 4793 + }, + { + "epoch": 0.36028859161280624, + "grad_norm": 1.2871452639124663, + "learning_rate": 2.960963661646812e-06, + "loss": 0.9968, + "num_input_tokens_seen": 38788935, + "step": 4794 + }, + { + "epoch": 0.3603637456786412, + "grad_norm": 1.4111260120267581, + "learning_rate": 2.960536670300963e-06, + "loss": 0.9784, + "num_input_tokens_seen": 38813310, + "step": 4795 + }, + { + "epoch": 0.3604388997444762, + "grad_norm": 1.3896330392269907, + "learning_rate": 2.9601096220396392e-06, + "loss": 0.9144, + "num_input_tokens_seen": 38836790, + "step": 4796 + }, + { + "epoch": 0.36051405381031115, + "grad_norm": 3.3553463932417076, + "learning_rate": 2.9596825168881444e-06, + "loss": 1.0732, + "num_input_tokens_seen": 38857970, + "step": 4797 + }, + { + "epoch": 0.3605892078761461, + "grad_norm": 1.483378668245901, + "learning_rate": 2.9592553548717848e-06, + "loss": 1.0948, + "num_input_tokens_seen": 38879055, + "step": 4798 + }, + { + "epoch": 0.36066436194198104, + "grad_norm": 1.4308831890102145, + "learning_rate": 2.958828136015872e-06, + "loss": 1.0066, + "num_input_tokens_seen": 38903305, + "step": 4799 + }, + { + "epoch": 0.360739516007816, + "grad_norm": 1.9880408121060706, + "learning_rate": 2.958400860345721e-06, + "loss": 0.9685, + "num_input_tokens_seen": 38921335, + "step": 4800 + }, + { + "epoch": 0.360814670073651, + "grad_norm": 1.615295999405431, + "learning_rate": 2.9579735278866488e-06, + "loss": 0.9065, + "num_input_tokens_seen": 38945425, + "step": 4801 + }, + { + "epoch": 0.36088982413948595, + "grad_norm": 5.7361006881623275, + "learning_rate": 2.9575461386639768e-06, + "loss": 0.939, + "num_input_tokens_seen": 38968840, + "step": 4802 + }, + { + "epoch": 0.3609649782053209, + "grad_norm": 1.8586591095754783, + "learning_rate": 2.95711869270303e-06, + "loss": 0.8931, + "num_input_tokens_seen": 38992865, + "step": 4803 + }, + { + "epoch": 0.3610401322711559, + "grad_norm": 1.5452381297422262, + "learning_rate": 2.9566911900291346e-06, + "loss": 0.9502, + "num_input_tokens_seen": 39015410, + "step": 4804 + }, + { + "epoch": 0.3611152863369908, + "grad_norm": 1.977550998227869, + "learning_rate": 2.9562636306676237e-06, + "loss": 1.0687, + "num_input_tokens_seen": 39036950, + "step": 4805 + }, + { + "epoch": 0.3611904404028258, + "grad_norm": 1.933041395420044, + "learning_rate": 2.9558360146438303e-06, + "loss": 0.9261, + "num_input_tokens_seen": 39062580, + "step": 4806 + }, + { + "epoch": 0.36126559446866074, + "grad_norm": 1.2950701276282912, + "learning_rate": 2.9554083419830925e-06, + "loss": 0.8349, + "num_input_tokens_seen": 39088845, + "step": 4807 + }, + { + "epoch": 0.3613407485344957, + "grad_norm": 1.4718202196038885, + "learning_rate": 2.954980612710753e-06, + "loss": 0.9594, + "num_input_tokens_seen": 39112795, + "step": 4808 + }, + { + "epoch": 0.3614159026003307, + "grad_norm": 1.7574154553759307, + "learning_rate": 2.9545528268521548e-06, + "loss": 1.0022, + "num_input_tokens_seen": 39137425, + "step": 4809 + }, + { + "epoch": 0.36149105666616566, + "grad_norm": 1.634512495453485, + "learning_rate": 2.954124984432646e-06, + "loss": 0.9395, + "num_input_tokens_seen": 39162125, + "step": 4810 + }, + { + "epoch": 0.3615662107320006, + "grad_norm": 0.9423245901197627, + "learning_rate": 2.953697085477579e-06, + "loss": 0.9095, + "num_input_tokens_seen": 39233150, + "step": 4811 + }, + { + "epoch": 0.36164136479783554, + "grad_norm": 1.7483071586290826, + "learning_rate": 2.953269130012307e-06, + "loss": 0.92, + "num_input_tokens_seen": 39258180, + "step": 4812 + }, + { + "epoch": 0.3617165188636705, + "grad_norm": 1.537667353756366, + "learning_rate": 2.9528411180621894e-06, + "loss": 0.8543, + "num_input_tokens_seen": 39283860, + "step": 4813 + }, + { + "epoch": 0.3617916729295055, + "grad_norm": 1.8808333889284183, + "learning_rate": 2.952413049652587e-06, + "loss": 0.923, + "num_input_tokens_seen": 39308785, + "step": 4814 + }, + { + "epoch": 0.36186682699534045, + "grad_norm": 1.5617089367129309, + "learning_rate": 2.9519849248088633e-06, + "loss": 0.92, + "num_input_tokens_seen": 39331295, + "step": 4815 + }, + { + "epoch": 0.3619419810611754, + "grad_norm": 1.4750071412253059, + "learning_rate": 2.9515567435563886e-06, + "loss": 0.9761, + "num_input_tokens_seen": 39354480, + "step": 4816 + }, + { + "epoch": 0.3620171351270104, + "grad_norm": 1.7678456478102977, + "learning_rate": 2.951128505920532e-06, + "loss": 0.9423, + "num_input_tokens_seen": 39375475, + "step": 4817 + }, + { + "epoch": 0.3620922891928453, + "grad_norm": 1.7438724051004981, + "learning_rate": 2.95070021192667e-06, + "loss": 1.0242, + "num_input_tokens_seen": 39396450, + "step": 4818 + }, + { + "epoch": 0.3621674432586803, + "grad_norm": 2.1232995354055024, + "learning_rate": 2.9502718616001803e-06, + "loss": 0.9648, + "num_input_tokens_seen": 39419175, + "step": 4819 + }, + { + "epoch": 0.36224259732451525, + "grad_norm": 0.9029830385454913, + "learning_rate": 2.9498434549664434e-06, + "loss": 0.7728, + "num_input_tokens_seen": 39493905, + "step": 4820 + }, + { + "epoch": 0.3623177513903502, + "grad_norm": 1.5326978883659177, + "learning_rate": 2.9494149920508443e-06, + "loss": 1.0119, + "num_input_tokens_seen": 39514300, + "step": 4821 + }, + { + "epoch": 0.3623929054561852, + "grad_norm": 1.4993337105094284, + "learning_rate": 2.9489864728787722e-06, + "loss": 0.9507, + "num_input_tokens_seen": 39537350, + "step": 4822 + }, + { + "epoch": 0.36246805952202016, + "grad_norm": 1.417718702395618, + "learning_rate": 2.9485578974756167e-06, + "loss": 0.8873, + "num_input_tokens_seen": 39564165, + "step": 4823 + }, + { + "epoch": 0.36254321358785513, + "grad_norm": 1.3294550799025742, + "learning_rate": 2.9481292658667743e-06, + "loss": 1.0028, + "num_input_tokens_seen": 39593580, + "step": 4824 + }, + { + "epoch": 0.36261836765369004, + "grad_norm": 1.4554439622423747, + "learning_rate": 2.947700578077643e-06, + "loss": 1.0233, + "num_input_tokens_seen": 39616565, + "step": 4825 + }, + { + "epoch": 0.362693521719525, + "grad_norm": 1.5664021403955597, + "learning_rate": 2.947271834133622e-06, + "loss": 0.8808, + "num_input_tokens_seen": 39640595, + "step": 4826 + }, + { + "epoch": 0.36276867578536, + "grad_norm": 1.4851339708771216, + "learning_rate": 2.946843034060118e-06, + "loss": 1.0032, + "num_input_tokens_seen": 39665715, + "step": 4827 + }, + { + "epoch": 0.36284382985119495, + "grad_norm": 1.522585430034765, + "learning_rate": 2.9464141778825384e-06, + "loss": 1.0365, + "num_input_tokens_seen": 39688595, + "step": 4828 + }, + { + "epoch": 0.3629189839170299, + "grad_norm": 1.966059769184668, + "learning_rate": 2.9459852656262945e-06, + "loss": 0.9329, + "num_input_tokens_seen": 39710380, + "step": 4829 + }, + { + "epoch": 0.3629941379828649, + "grad_norm": 1.7003160463396756, + "learning_rate": 2.945556297316802e-06, + "loss": 0.8882, + "num_input_tokens_seen": 39733815, + "step": 4830 + }, + { + "epoch": 0.3630692920486998, + "grad_norm": 1.4468634454583014, + "learning_rate": 2.9451272729794774e-06, + "loss": 1.0061, + "num_input_tokens_seen": 39755640, + "step": 4831 + }, + { + "epoch": 0.3631444461145348, + "grad_norm": 2.012969167292246, + "learning_rate": 2.944698192639743e-06, + "loss": 0.9049, + "num_input_tokens_seen": 39775720, + "step": 4832 + }, + { + "epoch": 0.36321960018036975, + "grad_norm": 1.5207307685164064, + "learning_rate": 2.944269056323023e-06, + "loss": 0.9734, + "num_input_tokens_seen": 39799235, + "step": 4833 + }, + { + "epoch": 0.3632947542462047, + "grad_norm": 1.9358421123610465, + "learning_rate": 2.9438398640547453e-06, + "loss": 0.8427, + "num_input_tokens_seen": 39824675, + "step": 4834 + }, + { + "epoch": 0.3633699083120397, + "grad_norm": 1.5151274346114199, + "learning_rate": 2.943410615860342e-06, + "loss": 0.9493, + "num_input_tokens_seen": 39847115, + "step": 4835 + }, + { + "epoch": 0.36344506237787466, + "grad_norm": 1.5252298619620188, + "learning_rate": 2.9429813117652478e-06, + "loss": 0.9966, + "num_input_tokens_seen": 39869920, + "step": 4836 + }, + { + "epoch": 0.36352021644370963, + "grad_norm": 1.6477410892875919, + "learning_rate": 2.942551951794899e-06, + "loss": 1.0453, + "num_input_tokens_seen": 39890900, + "step": 4837 + }, + { + "epoch": 0.36359537050954455, + "grad_norm": 1.3021633835663309, + "learning_rate": 2.942122535974738e-06, + "loss": 0.9533, + "num_input_tokens_seen": 39915185, + "step": 4838 + }, + { + "epoch": 0.3636705245753795, + "grad_norm": 1.1716471641313788, + "learning_rate": 2.9416930643302086e-06, + "loss": 0.9962, + "num_input_tokens_seen": 39941585, + "step": 4839 + }, + { + "epoch": 0.3637456786412145, + "grad_norm": 2.4535262636155375, + "learning_rate": 2.9412635368867596e-06, + "loss": 0.9985, + "num_input_tokens_seen": 39961480, + "step": 4840 + }, + { + "epoch": 0.36382083270704946, + "grad_norm": 1.3051615327855812, + "learning_rate": 2.9408339536698422e-06, + "loss": 0.8872, + "num_input_tokens_seen": 39984780, + "step": 4841 + }, + { + "epoch": 0.3638959867728844, + "grad_norm": 1.3856347373720057, + "learning_rate": 2.9404043147049097e-06, + "loss": 0.9611, + "num_input_tokens_seen": 40008445, + "step": 4842 + }, + { + "epoch": 0.3639711408387194, + "grad_norm": 1.5918960259572532, + "learning_rate": 2.9399746200174206e-06, + "loss": 0.9749, + "num_input_tokens_seen": 40029705, + "step": 4843 + }, + { + "epoch": 0.3640462949045543, + "grad_norm": 1.2815130475466918, + "learning_rate": 2.939544869632836e-06, + "loss": 0.9339, + "num_input_tokens_seen": 40055410, + "step": 4844 + }, + { + "epoch": 0.3641214489703893, + "grad_norm": 1.5299355450934862, + "learning_rate": 2.9391150635766194e-06, + "loss": 0.9458, + "num_input_tokens_seen": 40078435, + "step": 4845 + }, + { + "epoch": 0.36419660303622425, + "grad_norm": 1.3769646393855894, + "learning_rate": 2.9386852018742404e-06, + "loss": 0.858, + "num_input_tokens_seen": 40102210, + "step": 4846 + }, + { + "epoch": 0.3642717571020592, + "grad_norm": 1.5089672983315716, + "learning_rate": 2.938255284551168e-06, + "loss": 0.9666, + "num_input_tokens_seen": 40123975, + "step": 4847 + }, + { + "epoch": 0.3643469111678942, + "grad_norm": 1.5563476746329294, + "learning_rate": 2.9378253116328777e-06, + "loss": 1.0095, + "num_input_tokens_seen": 40146785, + "step": 4848 + }, + { + "epoch": 0.36442206523372916, + "grad_norm": 1.3188742755348835, + "learning_rate": 2.937395283144846e-06, + "loss": 0.9684, + "num_input_tokens_seen": 40172950, + "step": 4849 + }, + { + "epoch": 0.3644972192995641, + "grad_norm": 1.5254132883339617, + "learning_rate": 2.9369651991125542e-06, + "loss": 0.9464, + "num_input_tokens_seen": 40196690, + "step": 4850 + }, + { + "epoch": 0.36457237336539905, + "grad_norm": 1.4083280256649124, + "learning_rate": 2.9365350595614863e-06, + "loss": 1.0489, + "num_input_tokens_seen": 40221530, + "step": 4851 + }, + { + "epoch": 0.364647527431234, + "grad_norm": 2.1451081714164526, + "learning_rate": 2.936104864517131e-06, + "loss": 0.9775, + "num_input_tokens_seen": 40247155, + "step": 4852 + }, + { + "epoch": 0.364722681497069, + "grad_norm": 1.682872927686748, + "learning_rate": 2.935674614004977e-06, + "loss": 0.9049, + "num_input_tokens_seen": 40271140, + "step": 4853 + }, + { + "epoch": 0.36479783556290396, + "grad_norm": 2.48820555375688, + "learning_rate": 2.9352443080505192e-06, + "loss": 0.9094, + "num_input_tokens_seen": 40296690, + "step": 4854 + }, + { + "epoch": 0.36487298962873893, + "grad_norm": 1.1669189430139608, + "learning_rate": 2.934813946679255e-06, + "loss": 1.0193, + "num_input_tokens_seen": 40323580, + "step": 4855 + }, + { + "epoch": 0.3649481436945739, + "grad_norm": 1.750236767692005, + "learning_rate": 2.9343835299166846e-06, + "loss": 0.927, + "num_input_tokens_seen": 40345270, + "step": 4856 + }, + { + "epoch": 0.3650232977604088, + "grad_norm": 1.6448769104192258, + "learning_rate": 2.9339530577883125e-06, + "loss": 1.0564, + "num_input_tokens_seen": 40367455, + "step": 4857 + }, + { + "epoch": 0.3650984518262438, + "grad_norm": 1.6299864482901374, + "learning_rate": 2.9335225303196454e-06, + "loss": 0.8307, + "num_input_tokens_seen": 40391785, + "step": 4858 + }, + { + "epoch": 0.36517360589207876, + "grad_norm": 1.766532799609262, + "learning_rate": 2.933091947536193e-06, + "loss": 1.1373, + "num_input_tokens_seen": 40414035, + "step": 4859 + }, + { + "epoch": 0.3652487599579137, + "grad_norm": 1.6881605197823333, + "learning_rate": 2.93266130946347e-06, + "loss": 0.9176, + "num_input_tokens_seen": 40436895, + "step": 4860 + }, + { + "epoch": 0.3653239140237487, + "grad_norm": 1.7316139169221418, + "learning_rate": 2.9322306161269933e-06, + "loss": 0.9815, + "num_input_tokens_seen": 40462475, + "step": 4861 + }, + { + "epoch": 0.36539906808958367, + "grad_norm": 1.402027912754486, + "learning_rate": 2.931799867552282e-06, + "loss": 0.989, + "num_input_tokens_seen": 40484670, + "step": 4862 + }, + { + "epoch": 0.3654742221554186, + "grad_norm": 1.2199014945989115, + "learning_rate": 2.931369063764862e-06, + "loss": 0.9781, + "num_input_tokens_seen": 40509825, + "step": 4863 + }, + { + "epoch": 0.36554937622125355, + "grad_norm": 1.4624876885833304, + "learning_rate": 2.9309382047902574e-06, + "loss": 0.8568, + "num_input_tokens_seen": 40532205, + "step": 4864 + }, + { + "epoch": 0.3656245302870885, + "grad_norm": 1.2074764874935378, + "learning_rate": 2.9305072906539993e-06, + "loss": 0.9432, + "num_input_tokens_seen": 40557880, + "step": 4865 + }, + { + "epoch": 0.3656996843529235, + "grad_norm": 0.7043518383479546, + "learning_rate": 2.930076321381622e-06, + "loss": 0.809, + "num_input_tokens_seen": 40642515, + "step": 4866 + }, + { + "epoch": 0.36577483841875846, + "grad_norm": 1.5249361149898757, + "learning_rate": 2.92964529699866e-06, + "loss": 0.9819, + "num_input_tokens_seen": 40665735, + "step": 4867 + }, + { + "epoch": 0.36584999248459343, + "grad_norm": 1.588533803784865, + "learning_rate": 2.9292142175306548e-06, + "loss": 0.9772, + "num_input_tokens_seen": 40688680, + "step": 4868 + }, + { + "epoch": 0.3659251465504284, + "grad_norm": 1.3555131624624708, + "learning_rate": 2.9287830830031492e-06, + "loss": 0.9053, + "num_input_tokens_seen": 40716005, + "step": 4869 + }, + { + "epoch": 0.3660003006162633, + "grad_norm": 1.7784221353974856, + "learning_rate": 2.9283518934416892e-06, + "loss": 1.015, + "num_input_tokens_seen": 40738275, + "step": 4870 + }, + { + "epoch": 0.3660754546820983, + "grad_norm": 1.741446641168622, + "learning_rate": 2.927920648871825e-06, + "loss": 0.8484, + "num_input_tokens_seen": 40763255, + "step": 4871 + }, + { + "epoch": 0.36615060874793326, + "grad_norm": 1.2534719885734606, + "learning_rate": 2.9274893493191084e-06, + "loss": 1.0121, + "num_input_tokens_seen": 40787415, + "step": 4872 + }, + { + "epoch": 0.36622576281376823, + "grad_norm": 1.1594802479642814, + "learning_rate": 2.9270579948090962e-06, + "loss": 0.7957, + "num_input_tokens_seen": 40866170, + "step": 4873 + }, + { + "epoch": 0.3663009168796032, + "grad_norm": 1.5593633066476653, + "learning_rate": 2.9266265853673483e-06, + "loss": 0.9247, + "num_input_tokens_seen": 40888375, + "step": 4874 + }, + { + "epoch": 0.36637607094543817, + "grad_norm": 1.4564158594857852, + "learning_rate": 2.926195121019427e-06, + "loss": 0.8114, + "num_input_tokens_seen": 40915285, + "step": 4875 + }, + { + "epoch": 0.3664512250112731, + "grad_norm": 1.2877440212068751, + "learning_rate": 2.9257636017908984e-06, + "loss": 0.9627, + "num_input_tokens_seen": 40942000, + "step": 4876 + }, + { + "epoch": 0.36652637907710806, + "grad_norm": 2.1831734513815295, + "learning_rate": 2.925332027707331e-06, + "loss": 0.9884, + "num_input_tokens_seen": 40962495, + "step": 4877 + }, + { + "epoch": 0.366601533142943, + "grad_norm": 1.597847626283388, + "learning_rate": 2.9249003987942976e-06, + "loss": 0.9862, + "num_input_tokens_seen": 40987385, + "step": 4878 + }, + { + "epoch": 0.366676687208778, + "grad_norm": 1.8065681731671621, + "learning_rate": 2.924468715077374e-06, + "loss": 1.0228, + "num_input_tokens_seen": 41008630, + "step": 4879 + }, + { + "epoch": 0.36675184127461297, + "grad_norm": 1.6080642102787042, + "learning_rate": 2.9240369765821392e-06, + "loss": 1.0108, + "num_input_tokens_seen": 41033285, + "step": 4880 + }, + { + "epoch": 0.36682699534044794, + "grad_norm": 1.689527000620169, + "learning_rate": 2.9236051833341745e-06, + "loss": 0.9935, + "num_input_tokens_seen": 41057495, + "step": 4881 + }, + { + "epoch": 0.3669021494062829, + "grad_norm": 1.50095154462824, + "learning_rate": 2.9231733353590668e-06, + "loss": 0.9825, + "num_input_tokens_seen": 41079775, + "step": 4882 + }, + { + "epoch": 0.3669773034721178, + "grad_norm": 1.4174789549595774, + "learning_rate": 2.9227414326824027e-06, + "loss": 1.0559, + "num_input_tokens_seen": 41101425, + "step": 4883 + }, + { + "epoch": 0.3670524575379528, + "grad_norm": 1.8605913303085095, + "learning_rate": 2.9223094753297767e-06, + "loss": 0.92, + "num_input_tokens_seen": 41125270, + "step": 4884 + }, + { + "epoch": 0.36712761160378776, + "grad_norm": 1.4941198808813247, + "learning_rate": 2.9218774633267815e-06, + "loss": 0.9108, + "num_input_tokens_seen": 41149220, + "step": 4885 + }, + { + "epoch": 0.36720276566962273, + "grad_norm": 1.6167307038094305, + "learning_rate": 2.9214453966990174e-06, + "loss": 1.0528, + "num_input_tokens_seen": 41169510, + "step": 4886 + }, + { + "epoch": 0.3672779197354577, + "grad_norm": 1.9223134850953025, + "learning_rate": 2.9210132754720845e-06, + "loss": 0.9591, + "num_input_tokens_seen": 41192265, + "step": 4887 + }, + { + "epoch": 0.3673530738012927, + "grad_norm": 1.3162489437667797, + "learning_rate": 2.9205810996715885e-06, + "loss": 0.9667, + "num_input_tokens_seen": 41217180, + "step": 4888 + }, + { + "epoch": 0.3674282278671276, + "grad_norm": 1.3908869035954883, + "learning_rate": 2.9201488693231366e-06, + "loss": 1.0019, + "num_input_tokens_seen": 41244145, + "step": 4889 + }, + { + "epoch": 0.36750338193296256, + "grad_norm": 1.5422761932191829, + "learning_rate": 2.9197165844523416e-06, + "loss": 0.8725, + "num_input_tokens_seen": 41267725, + "step": 4890 + }, + { + "epoch": 0.36757853599879753, + "grad_norm": 1.889075661133007, + "learning_rate": 2.9192842450848164e-06, + "loss": 0.9085, + "num_input_tokens_seen": 41288925, + "step": 4891 + }, + { + "epoch": 0.3676536900646325, + "grad_norm": 1.6230683216280093, + "learning_rate": 2.91885185124618e-06, + "loss": 0.9752, + "num_input_tokens_seen": 41313345, + "step": 4892 + }, + { + "epoch": 0.36772884413046747, + "grad_norm": 1.9545686193186265, + "learning_rate": 2.918419402962053e-06, + "loss": 0.9156, + "num_input_tokens_seen": 41332325, + "step": 4893 + }, + { + "epoch": 0.36780399819630244, + "grad_norm": 1.49090537067592, + "learning_rate": 2.917986900258059e-06, + "loss": 0.9137, + "num_input_tokens_seen": 41356950, + "step": 4894 + }, + { + "epoch": 0.36787915226213735, + "grad_norm": 1.576571993992993, + "learning_rate": 2.9175543431598257e-06, + "loss": 0.8871, + "num_input_tokens_seen": 41379790, + "step": 4895 + }, + { + "epoch": 0.3679543063279723, + "grad_norm": 1.6446986169890543, + "learning_rate": 2.917121731692985e-06, + "loss": 0.8802, + "num_input_tokens_seen": 41407740, + "step": 4896 + }, + { + "epoch": 0.3680294603938073, + "grad_norm": 1.7775077995633506, + "learning_rate": 2.9166890658831695e-06, + "loss": 0.8901, + "num_input_tokens_seen": 41430780, + "step": 4897 + }, + { + "epoch": 0.36810461445964227, + "grad_norm": 1.5265218176833508, + "learning_rate": 2.9162563457560157e-06, + "loss": 0.9768, + "num_input_tokens_seen": 41453275, + "step": 4898 + }, + { + "epoch": 0.36817976852547724, + "grad_norm": 1.3205138735723059, + "learning_rate": 2.915823571337166e-06, + "loss": 1.0045, + "num_input_tokens_seen": 41477415, + "step": 4899 + }, + { + "epoch": 0.3682549225913122, + "grad_norm": 2.0333168664901256, + "learning_rate": 2.915390742652262e-06, + "loss": 0.9271, + "num_input_tokens_seen": 41502550, + "step": 4900 + }, + { + "epoch": 0.3683300766571472, + "grad_norm": 1.6275316471474313, + "learning_rate": 2.914957859726952e-06, + "loss": 1.0137, + "num_input_tokens_seen": 41526780, + "step": 4901 + }, + { + "epoch": 0.3684052307229821, + "grad_norm": 1.6314722428621753, + "learning_rate": 2.9145249225868848e-06, + "loss": 0.9995, + "num_input_tokens_seen": 41550440, + "step": 4902 + }, + { + "epoch": 0.36848038478881706, + "grad_norm": 1.5818116589016091, + "learning_rate": 2.9140919312577134e-06, + "loss": 1.0939, + "num_input_tokens_seen": 41574610, + "step": 4903 + }, + { + "epoch": 0.36855553885465203, + "grad_norm": 1.700376059506807, + "learning_rate": 2.9136588857650956e-06, + "loss": 0.9289, + "num_input_tokens_seen": 41597490, + "step": 4904 + }, + { + "epoch": 0.368630692920487, + "grad_norm": 2.4214102022171136, + "learning_rate": 2.9132257861346897e-06, + "loss": 1.0951, + "num_input_tokens_seen": 41617250, + "step": 4905 + }, + { + "epoch": 0.368705846986322, + "grad_norm": 1.5989862329821984, + "learning_rate": 2.912792632392159e-06, + "loss": 1.013, + "num_input_tokens_seen": 41639040, + "step": 4906 + }, + { + "epoch": 0.36878100105215694, + "grad_norm": 2.2488792595177625, + "learning_rate": 2.9123594245631702e-06, + "loss": 0.9119, + "num_input_tokens_seen": 41660310, + "step": 4907 + }, + { + "epoch": 0.36885615511799186, + "grad_norm": 1.3462658329903079, + "learning_rate": 2.9119261626733915e-06, + "loss": 0.9604, + "num_input_tokens_seen": 41683450, + "step": 4908 + }, + { + "epoch": 0.3689313091838268, + "grad_norm": 1.7938025038209289, + "learning_rate": 2.911492846748495e-06, + "loss": 0.9236, + "num_input_tokens_seen": 41704225, + "step": 4909 + }, + { + "epoch": 0.3690064632496618, + "grad_norm": 1.5123449797831732, + "learning_rate": 2.911059476814158e-06, + "loss": 0.9777, + "num_input_tokens_seen": 41729400, + "step": 4910 + }, + { + "epoch": 0.36908161731549677, + "grad_norm": 1.4268828852231357, + "learning_rate": 2.9106260528960573e-06, + "loss": 0.9847, + "num_input_tokens_seen": 41753025, + "step": 4911 + }, + { + "epoch": 0.36915677138133174, + "grad_norm": 1.71399189639318, + "learning_rate": 2.910192575019877e-06, + "loss": 0.9737, + "num_input_tokens_seen": 41774965, + "step": 4912 + }, + { + "epoch": 0.3692319254471667, + "grad_norm": 1.4911992660295539, + "learning_rate": 2.9097590432113007e-06, + "loss": 0.8922, + "num_input_tokens_seen": 41798215, + "step": 4913 + }, + { + "epoch": 0.3693070795130017, + "grad_norm": 1.681383066690192, + "learning_rate": 2.909325457496017e-06, + "loss": 1.0108, + "num_input_tokens_seen": 41823690, + "step": 4914 + }, + { + "epoch": 0.3693822335788366, + "grad_norm": 1.3883034261586338, + "learning_rate": 2.908891817899718e-06, + "loss": 0.9144, + "num_input_tokens_seen": 41845625, + "step": 4915 + }, + { + "epoch": 0.36945738764467156, + "grad_norm": 1.3192081234205864, + "learning_rate": 2.9084581244480994e-06, + "loss": 1.0347, + "num_input_tokens_seen": 41870680, + "step": 4916 + }, + { + "epoch": 0.36953254171050653, + "grad_norm": 1.2511053919807322, + "learning_rate": 2.908024377166857e-06, + "loss": 0.9579, + "num_input_tokens_seen": 41903955, + "step": 4917 + }, + { + "epoch": 0.3696076957763415, + "grad_norm": 1.347363160922659, + "learning_rate": 2.9075905760816942e-06, + "loss": 0.9999, + "num_input_tokens_seen": 41929130, + "step": 4918 + }, + { + "epoch": 0.3696828498421765, + "grad_norm": 1.3497756436026875, + "learning_rate": 2.9071567212183138e-06, + "loss": 0.8398, + "num_input_tokens_seen": 41952010, + "step": 4919 + }, + { + "epoch": 0.36975800390801145, + "grad_norm": 1.5576839994732454, + "learning_rate": 2.906722812602424e-06, + "loss": 0.9257, + "num_input_tokens_seen": 41976305, + "step": 4920 + }, + { + "epoch": 0.36983315797384636, + "grad_norm": 1.778164449834066, + "learning_rate": 2.906288850259736e-06, + "loss": 0.8913, + "num_input_tokens_seen": 42000165, + "step": 4921 + }, + { + "epoch": 0.36990831203968133, + "grad_norm": 1.403803923225926, + "learning_rate": 2.9058548342159628e-06, + "loss": 0.8195, + "num_input_tokens_seen": 42026295, + "step": 4922 + }, + { + "epoch": 0.3699834661055163, + "grad_norm": 1.8590926909708445, + "learning_rate": 2.9054207644968218e-06, + "loss": 0.9689, + "num_input_tokens_seen": 42048920, + "step": 4923 + }, + { + "epoch": 0.37005862017135127, + "grad_norm": 1.4471688386211874, + "learning_rate": 2.904986641128033e-06, + "loss": 1.0303, + "num_input_tokens_seen": 42073660, + "step": 4924 + }, + { + "epoch": 0.37013377423718624, + "grad_norm": 1.6497397679963846, + "learning_rate": 2.9045524641353208e-06, + "loss": 0.979, + "num_input_tokens_seen": 42095000, + "step": 4925 + }, + { + "epoch": 0.3702089283030212, + "grad_norm": 1.4339549261907363, + "learning_rate": 2.904118233544411e-06, + "loss": 0.9651, + "num_input_tokens_seen": 42119940, + "step": 4926 + }, + { + "epoch": 0.3702840823688562, + "grad_norm": 1.9232236890541459, + "learning_rate": 2.9036839493810348e-06, + "loss": 1.0718, + "num_input_tokens_seen": 42141355, + "step": 4927 + }, + { + "epoch": 0.3703592364346911, + "grad_norm": 1.5577911375488398, + "learning_rate": 2.903249611670923e-06, + "loss": 0.9575, + "num_input_tokens_seen": 42166180, + "step": 4928 + }, + { + "epoch": 0.37043439050052607, + "grad_norm": 1.6875010596377966, + "learning_rate": 2.9028152204398135e-06, + "loss": 0.9071, + "num_input_tokens_seen": 42188900, + "step": 4929 + }, + { + "epoch": 0.37050954456636104, + "grad_norm": 1.3108665201569982, + "learning_rate": 2.9023807757134455e-06, + "loss": 1.0441, + "num_input_tokens_seen": 42214430, + "step": 4930 + }, + { + "epoch": 0.370584698632196, + "grad_norm": 1.5540912193959966, + "learning_rate": 2.90194627751756e-06, + "loss": 1.0073, + "num_input_tokens_seen": 42238070, + "step": 4931 + }, + { + "epoch": 0.370659852698031, + "grad_norm": 1.7720700003616225, + "learning_rate": 2.9015117258779045e-06, + "loss": 0.9333, + "num_input_tokens_seen": 42260240, + "step": 4932 + }, + { + "epoch": 0.37073500676386595, + "grad_norm": 1.45551906791523, + "learning_rate": 2.9010771208202265e-06, + "loss": 1.0028, + "num_input_tokens_seen": 42287310, + "step": 4933 + }, + { + "epoch": 0.37081016082970086, + "grad_norm": 0.9398881694034933, + "learning_rate": 2.900642462370279e-06, + "loss": 0.8543, + "num_input_tokens_seen": 42360580, + "step": 4934 + }, + { + "epoch": 0.37088531489553583, + "grad_norm": 1.5007207251620904, + "learning_rate": 2.900207750553817e-06, + "loss": 0.8955, + "num_input_tokens_seen": 42384980, + "step": 4935 + }, + { + "epoch": 0.3709604689613708, + "grad_norm": 1.5029113649437464, + "learning_rate": 2.899772985396599e-06, + "loss": 0.9704, + "num_input_tokens_seen": 42410520, + "step": 4936 + }, + { + "epoch": 0.3710356230272058, + "grad_norm": 1.2283064477653316, + "learning_rate": 2.8993381669243854e-06, + "loss": 0.9873, + "num_input_tokens_seen": 42435525, + "step": 4937 + }, + { + "epoch": 0.37111077709304074, + "grad_norm": 1.5400163047720057, + "learning_rate": 2.8989032951629417e-06, + "loss": 1.0067, + "num_input_tokens_seen": 42461090, + "step": 4938 + }, + { + "epoch": 0.3711859311588757, + "grad_norm": 1.9954475207786755, + "learning_rate": 2.898468370138036e-06, + "loss": 0.8713, + "num_input_tokens_seen": 42535265, + "step": 4939 + }, + { + "epoch": 0.37126108522471063, + "grad_norm": 1.4864998315990057, + "learning_rate": 2.8980333918754383e-06, + "loss": 0.9427, + "num_input_tokens_seen": 42559695, + "step": 4940 + }, + { + "epoch": 0.3713362392905456, + "grad_norm": 1.4222450194088059, + "learning_rate": 2.8975983604009244e-06, + "loss": 0.9846, + "num_input_tokens_seen": 42581905, + "step": 4941 + }, + { + "epoch": 0.37141139335638057, + "grad_norm": 1.2257039655412287, + "learning_rate": 2.8971632757402694e-06, + "loss": 0.9283, + "num_input_tokens_seen": 42608945, + "step": 4942 + }, + { + "epoch": 0.37148654742221554, + "grad_norm": 1.5234568276768752, + "learning_rate": 2.8967281379192557e-06, + "loss": 1.0162, + "num_input_tokens_seen": 42632525, + "step": 4943 + }, + { + "epoch": 0.3715617014880505, + "grad_norm": 1.402506571803616, + "learning_rate": 2.8962929469636653e-06, + "loss": 0.8737, + "num_input_tokens_seen": 42658750, + "step": 4944 + }, + { + "epoch": 0.3716368555538855, + "grad_norm": 1.447269743425736, + "learning_rate": 2.8958577028992866e-06, + "loss": 0.9204, + "num_input_tokens_seen": 42681960, + "step": 4945 + }, + { + "epoch": 0.37171200961972045, + "grad_norm": 1.7042026435144804, + "learning_rate": 2.895422405751908e-06, + "loss": 1.0052, + "num_input_tokens_seen": 42704310, + "step": 4946 + }, + { + "epoch": 0.37178716368555537, + "grad_norm": 1.453582691592187, + "learning_rate": 2.8949870555473226e-06, + "loss": 0.9594, + "num_input_tokens_seen": 42726665, + "step": 4947 + }, + { + "epoch": 0.37186231775139034, + "grad_norm": 1.3816533807193196, + "learning_rate": 2.8945516523113275e-06, + "loss": 0.9273, + "num_input_tokens_seen": 42747610, + "step": 4948 + }, + { + "epoch": 0.3719374718172253, + "grad_norm": 1.5840554180276687, + "learning_rate": 2.8941161960697217e-06, + "loss": 0.9573, + "num_input_tokens_seen": 42770375, + "step": 4949 + }, + { + "epoch": 0.3720126258830603, + "grad_norm": 1.7723288410659077, + "learning_rate": 2.893680686848307e-06, + "loss": 0.989, + "num_input_tokens_seen": 42793705, + "step": 4950 + }, + { + "epoch": 0.37208777994889525, + "grad_norm": 1.7548747967349625, + "learning_rate": 2.89324512467289e-06, + "loss": 0.9027, + "num_input_tokens_seen": 42816255, + "step": 4951 + }, + { + "epoch": 0.3721629340147302, + "grad_norm": 1.7320477792435518, + "learning_rate": 2.8928095095692783e-06, + "loss": 1.0084, + "num_input_tokens_seen": 42839285, + "step": 4952 + }, + { + "epoch": 0.37223808808056513, + "grad_norm": 1.4197149952453871, + "learning_rate": 2.892373841563285e-06, + "loss": 0.944, + "num_input_tokens_seen": 42862245, + "step": 4953 + }, + { + "epoch": 0.3723132421464001, + "grad_norm": 1.8099207448731471, + "learning_rate": 2.891938120680724e-06, + "loss": 0.9487, + "num_input_tokens_seen": 42883750, + "step": 4954 + }, + { + "epoch": 0.3723883962122351, + "grad_norm": 0.7254593791434507, + "learning_rate": 2.891502346947414e-06, + "loss": 0.7911, + "num_input_tokens_seen": 42957630, + "step": 4955 + }, + { + "epoch": 0.37246355027807004, + "grad_norm": 1.6824832094666886, + "learning_rate": 2.8910665203891763e-06, + "loss": 0.9918, + "num_input_tokens_seen": 42977495, + "step": 4956 + }, + { + "epoch": 0.372538704343905, + "grad_norm": 1.8343255291430267, + "learning_rate": 2.8906306410318353e-06, + "loss": 1.03, + "num_input_tokens_seen": 42999300, + "step": 4957 + }, + { + "epoch": 0.37261385840974, + "grad_norm": 1.406511155356477, + "learning_rate": 2.890194708901218e-06, + "loss": 0.9566, + "num_input_tokens_seen": 43025370, + "step": 4958 + }, + { + "epoch": 0.37268901247557495, + "grad_norm": 1.3688909762866823, + "learning_rate": 2.889758724023155e-06, + "loss": 0.9935, + "num_input_tokens_seen": 43050560, + "step": 4959 + }, + { + "epoch": 0.37276416654140987, + "grad_norm": 1.7641181930251733, + "learning_rate": 2.8893226864234813e-06, + "loss": 0.8147, + "num_input_tokens_seen": 43072905, + "step": 4960 + }, + { + "epoch": 0.37283932060724484, + "grad_norm": 1.524395528970256, + "learning_rate": 2.8888865961280325e-06, + "loss": 0.979, + "num_input_tokens_seen": 43093025, + "step": 4961 + }, + { + "epoch": 0.3729144746730798, + "grad_norm": 1.567074982548948, + "learning_rate": 2.888450453162649e-06, + "loss": 0.9598, + "num_input_tokens_seen": 43119080, + "step": 4962 + }, + { + "epoch": 0.3729896287389148, + "grad_norm": 0.7680633409054574, + "learning_rate": 2.888014257553175e-06, + "loss": 0.7804, + "num_input_tokens_seen": 43202925, + "step": 4963 + }, + { + "epoch": 0.37306478280474975, + "grad_norm": 0.6278010068898119, + "learning_rate": 2.8875780093254545e-06, + "loss": 0.7592, + "num_input_tokens_seen": 43288270, + "step": 4964 + }, + { + "epoch": 0.3731399368705847, + "grad_norm": 1.3957559459736595, + "learning_rate": 2.8871417085053394e-06, + "loss": 0.9579, + "num_input_tokens_seen": 43312530, + "step": 4965 + }, + { + "epoch": 0.37321509093641964, + "grad_norm": 1.5251681156984591, + "learning_rate": 2.88670535511868e-06, + "loss": 1.0501, + "num_input_tokens_seen": 43335925, + "step": 4966 + }, + { + "epoch": 0.3732902450022546, + "grad_norm": 1.9443984737714355, + "learning_rate": 2.886268949191334e-06, + "loss": 0.9038, + "num_input_tokens_seen": 43357345, + "step": 4967 + }, + { + "epoch": 0.3733653990680896, + "grad_norm": 1.3461292406076109, + "learning_rate": 2.885832490749158e-06, + "loss": 0.9922, + "num_input_tokens_seen": 43382920, + "step": 4968 + }, + { + "epoch": 0.37344055313392455, + "grad_norm": 2.0234340829709927, + "learning_rate": 2.885395979818015e-06, + "loss": 0.9186, + "num_input_tokens_seen": 43403685, + "step": 4969 + }, + { + "epoch": 0.3735157071997595, + "grad_norm": 1.41354622030778, + "learning_rate": 2.8849594164237694e-06, + "loss": 0.8762, + "num_input_tokens_seen": 43426970, + "step": 4970 + }, + { + "epoch": 0.3735908612655945, + "grad_norm": 1.3263196904948475, + "learning_rate": 2.8845228005922905e-06, + "loss": 0.9171, + "num_input_tokens_seen": 43454105, + "step": 4971 + }, + { + "epoch": 0.37366601533142946, + "grad_norm": 1.5069727164964768, + "learning_rate": 2.8840861323494482e-06, + "loss": 0.9779, + "num_input_tokens_seen": 43477450, + "step": 4972 + }, + { + "epoch": 0.3737411693972644, + "grad_norm": 1.4310662130769236, + "learning_rate": 2.8836494117211177e-06, + "loss": 0.9464, + "num_input_tokens_seen": 43500140, + "step": 4973 + }, + { + "epoch": 0.37381632346309934, + "grad_norm": 1.607750674765292, + "learning_rate": 2.883212638733175e-06, + "loss": 0.9792, + "num_input_tokens_seen": 43523350, + "step": 4974 + }, + { + "epoch": 0.3738914775289343, + "grad_norm": 1.6021283894500273, + "learning_rate": 2.8827758134115017e-06, + "loss": 0.9495, + "num_input_tokens_seen": 43546220, + "step": 4975 + }, + { + "epoch": 0.3739666315947693, + "grad_norm": 1.6088169204321006, + "learning_rate": 2.8823389357819815e-06, + "loss": 0.9599, + "num_input_tokens_seen": 43568255, + "step": 4976 + }, + { + "epoch": 0.37404178566060425, + "grad_norm": 1.2414982160877281, + "learning_rate": 2.8819020058705003e-06, + "loss": 0.9146, + "num_input_tokens_seen": 43598210, + "step": 4977 + }, + { + "epoch": 0.3741169397264392, + "grad_norm": 1.4427163391258393, + "learning_rate": 2.881465023702948e-06, + "loss": 0.9467, + "num_input_tokens_seen": 43622510, + "step": 4978 + }, + { + "epoch": 0.37419209379227414, + "grad_norm": 1.5925598936929781, + "learning_rate": 2.8810279893052184e-06, + "loss": 0.9143, + "num_input_tokens_seen": 43644940, + "step": 4979 + }, + { + "epoch": 0.3742672478581091, + "grad_norm": 1.5017678651288429, + "learning_rate": 2.880590902703206e-06, + "loss": 1.0298, + "num_input_tokens_seen": 43668835, + "step": 4980 + }, + { + "epoch": 0.3743424019239441, + "grad_norm": 1.6150212176083618, + "learning_rate": 2.8801537639228107e-06, + "loss": 0.9885, + "num_input_tokens_seen": 43691410, + "step": 4981 + }, + { + "epoch": 0.37441755598977905, + "grad_norm": 1.5559026675243302, + "learning_rate": 2.8797165729899347e-06, + "loss": 0.9962, + "num_input_tokens_seen": 43716140, + "step": 4982 + }, + { + "epoch": 0.374492710055614, + "grad_norm": 1.5382483018220665, + "learning_rate": 2.879279329930483e-06, + "loss": 0.8797, + "num_input_tokens_seen": 43738730, + "step": 4983 + }, + { + "epoch": 0.374567864121449, + "grad_norm": 1.483284238295728, + "learning_rate": 2.8788420347703643e-06, + "loss": 0.9763, + "num_input_tokens_seen": 43764420, + "step": 4984 + }, + { + "epoch": 0.3746430181872839, + "grad_norm": 1.2706552078401534, + "learning_rate": 2.87840468753549e-06, + "loss": 1.0098, + "num_input_tokens_seen": 43790355, + "step": 4985 + }, + { + "epoch": 0.3747181722531189, + "grad_norm": 1.8721795803890144, + "learning_rate": 2.8779672882517735e-06, + "loss": 0.9568, + "num_input_tokens_seen": 43811935, + "step": 4986 + }, + { + "epoch": 0.37479332631895385, + "grad_norm": 1.2403602356013248, + "learning_rate": 2.877529836945134e-06, + "loss": 0.9366, + "num_input_tokens_seen": 43837075, + "step": 4987 + }, + { + "epoch": 0.3748684803847888, + "grad_norm": 0.8823355002175493, + "learning_rate": 2.8770923336414906e-06, + "loss": 0.8706, + "num_input_tokens_seen": 43914030, + "step": 4988 + }, + { + "epoch": 0.3749436344506238, + "grad_norm": 1.5307427948142471, + "learning_rate": 2.8766547783667686e-06, + "loss": 0.923, + "num_input_tokens_seen": 43936015, + "step": 4989 + }, + { + "epoch": 0.37501878851645876, + "grad_norm": 1.662561590444909, + "learning_rate": 2.8762171711468935e-06, + "loss": 0.971, + "num_input_tokens_seen": 43958415, + "step": 4990 + }, + { + "epoch": 0.3750939425822937, + "grad_norm": 1.7929999321779577, + "learning_rate": 2.8757795120077955e-06, + "loss": 1.0176, + "num_input_tokens_seen": 43978465, + "step": 4991 + }, + { + "epoch": 0.37516909664812864, + "grad_norm": 1.5129073642678308, + "learning_rate": 2.8753418009754082e-06, + "loss": 0.9563, + "num_input_tokens_seen": 44001090, + "step": 4992 + }, + { + "epoch": 0.3752442507139636, + "grad_norm": 1.6019102881954381, + "learning_rate": 2.874904038075668e-06, + "loss": 1.0035, + "num_input_tokens_seen": 44023835, + "step": 4993 + }, + { + "epoch": 0.3753194047797986, + "grad_norm": 1.6768822954255889, + "learning_rate": 2.874466223334512e-06, + "loss": 0.9405, + "num_input_tokens_seen": 44047175, + "step": 4994 + }, + { + "epoch": 0.37539455884563355, + "grad_norm": 1.7262746993942055, + "learning_rate": 2.8740283567778844e-06, + "loss": 0.9847, + "num_input_tokens_seen": 44068810, + "step": 4995 + }, + { + "epoch": 0.3754697129114685, + "grad_norm": 1.5756329793398893, + "learning_rate": 2.87359043843173e-06, + "loss": 0.8756, + "num_input_tokens_seen": 44096045, + "step": 4996 + }, + { + "epoch": 0.3755448669773035, + "grad_norm": 1.5931826310866226, + "learning_rate": 2.873152468321997e-06, + "loss": 0.8463, + "num_input_tokens_seen": 44121190, + "step": 4997 + }, + { + "epoch": 0.3756200210431384, + "grad_norm": 3.6757045822985512, + "learning_rate": 2.872714446474636e-06, + "loss": 0.9441, + "num_input_tokens_seen": 44145490, + "step": 4998 + }, + { + "epoch": 0.3756951751089734, + "grad_norm": 1.7607458435122192, + "learning_rate": 2.8722763729156027e-06, + "loss": 0.9781, + "num_input_tokens_seen": 44167855, + "step": 4999 + }, + { + "epoch": 0.37577032917480835, + "grad_norm": 1.3473499779356561, + "learning_rate": 2.8718382476708544e-06, + "loss": 1.0163, + "num_input_tokens_seen": 44193705, + "step": 5000 + }, + { + "epoch": 0.3758454832406433, + "grad_norm": 1.7023776453222976, + "learning_rate": 2.8714000707663507e-06, + "loss": 0.9285, + "num_input_tokens_seen": 44215170, + "step": 5001 + }, + { + "epoch": 0.3759206373064783, + "grad_norm": 2.035304556511469, + "learning_rate": 2.8709618422280564e-06, + "loss": 0.8796, + "num_input_tokens_seen": 44239815, + "step": 5002 + }, + { + "epoch": 0.37599579137231326, + "grad_norm": 1.583655533716864, + "learning_rate": 2.8705235620819377e-06, + "loss": 1.0586, + "num_input_tokens_seen": 44262515, + "step": 5003 + }, + { + "epoch": 0.37607094543814823, + "grad_norm": 1.8009770549644837, + "learning_rate": 2.8700852303539647e-06, + "loss": 0.9183, + "num_input_tokens_seen": 44285005, + "step": 5004 + }, + { + "epoch": 0.37614609950398314, + "grad_norm": 2.256893089557875, + "learning_rate": 2.8696468470701096e-06, + "loss": 0.8793, + "num_input_tokens_seen": 44304265, + "step": 5005 + }, + { + "epoch": 0.3762212535698181, + "grad_norm": 1.4693929198489497, + "learning_rate": 2.869208412256349e-06, + "loss": 1.0318, + "num_input_tokens_seen": 44328595, + "step": 5006 + }, + { + "epoch": 0.3762964076356531, + "grad_norm": 1.5156106063316608, + "learning_rate": 2.868769925938662e-06, + "loss": 0.8866, + "num_input_tokens_seen": 44353165, + "step": 5007 + }, + { + "epoch": 0.37637156170148806, + "grad_norm": 1.9563642572058402, + "learning_rate": 2.868331388143029e-06, + "loss": 0.947, + "num_input_tokens_seen": 44375265, + "step": 5008 + }, + { + "epoch": 0.376446715767323, + "grad_norm": 1.5052651347318298, + "learning_rate": 2.867892798895437e-06, + "loss": 1.0043, + "num_input_tokens_seen": 44400195, + "step": 5009 + }, + { + "epoch": 0.376521869833158, + "grad_norm": 1.2667716212877038, + "learning_rate": 2.867454158221873e-06, + "loss": 1.0148, + "num_input_tokens_seen": 44426930, + "step": 5010 + }, + { + "epoch": 0.3765970238989929, + "grad_norm": 1.53164496485653, + "learning_rate": 2.867015466148329e-06, + "loss": 0.9378, + "num_input_tokens_seen": 44453000, + "step": 5011 + }, + { + "epoch": 0.3766721779648279, + "grad_norm": 1.6544562649196075, + "learning_rate": 2.8665767227007985e-06, + "loss": 0.873, + "num_input_tokens_seen": 44477260, + "step": 5012 + }, + { + "epoch": 0.37674733203066285, + "grad_norm": 0.8347091919919912, + "learning_rate": 2.866137927905278e-06, + "loss": 0.8043, + "num_input_tokens_seen": 44556050, + "step": 5013 + }, + { + "epoch": 0.3768224860964978, + "grad_norm": 1.5641815292690129, + "learning_rate": 2.865699081787769e-06, + "loss": 0.9946, + "num_input_tokens_seen": 44577900, + "step": 5014 + }, + { + "epoch": 0.3768976401623328, + "grad_norm": 1.537267577428945, + "learning_rate": 2.865260184374275e-06, + "loss": 0.9838, + "num_input_tokens_seen": 44598595, + "step": 5015 + }, + { + "epoch": 0.37697279422816776, + "grad_norm": 1.374983050502, + "learning_rate": 2.864821235690801e-06, + "loss": 0.9467, + "num_input_tokens_seen": 44622890, + "step": 5016 + }, + { + "epoch": 0.37704794829400273, + "grad_norm": 1.6165240857412222, + "learning_rate": 2.8643822357633576e-06, + "loss": 0.8047, + "num_input_tokens_seen": 44647955, + "step": 5017 + }, + { + "epoch": 0.37712310235983765, + "grad_norm": 1.7690453272695084, + "learning_rate": 2.863943184617957e-06, + "loss": 0.9383, + "num_input_tokens_seen": 44668145, + "step": 5018 + }, + { + "epoch": 0.3771982564256726, + "grad_norm": 2.826396350606014, + "learning_rate": 2.8635040822806135e-06, + "loss": 0.9258, + "num_input_tokens_seen": 44690910, + "step": 5019 + }, + { + "epoch": 0.3772734104915076, + "grad_norm": 1.4551444838580765, + "learning_rate": 2.8630649287773475e-06, + "loss": 1.0255, + "num_input_tokens_seen": 44715720, + "step": 5020 + }, + { + "epoch": 0.37734856455734256, + "grad_norm": 1.393945950474165, + "learning_rate": 2.862625724134179e-06, + "loss": 0.9731, + "num_input_tokens_seen": 44738920, + "step": 5021 + }, + { + "epoch": 0.37742371862317753, + "grad_norm": 1.4279251493022622, + "learning_rate": 2.8621864683771337e-06, + "loss": 0.9856, + "num_input_tokens_seen": 44763010, + "step": 5022 + }, + { + "epoch": 0.3774988726890125, + "grad_norm": 1.3614186239057835, + "learning_rate": 2.8617471615322377e-06, + "loss": 0.8818, + "num_input_tokens_seen": 44786730, + "step": 5023 + }, + { + "epoch": 0.3775740267548474, + "grad_norm": 1.691908975294585, + "learning_rate": 2.8613078036255233e-06, + "loss": 0.8997, + "num_input_tokens_seen": 44811315, + "step": 5024 + }, + { + "epoch": 0.3776491808206824, + "grad_norm": 1.773626645122316, + "learning_rate": 2.8608683946830236e-06, + "loss": 0.8913, + "num_input_tokens_seen": 44834775, + "step": 5025 + }, + { + "epoch": 0.37772433488651735, + "grad_norm": 1.5035496672941215, + "learning_rate": 2.8604289347307746e-06, + "loss": 1.0054, + "num_input_tokens_seen": 44860705, + "step": 5026 + }, + { + "epoch": 0.3777994889523523, + "grad_norm": 1.3601290047692398, + "learning_rate": 2.859989423794816e-06, + "loss": 0.8513, + "num_input_tokens_seen": 44889050, + "step": 5027 + }, + { + "epoch": 0.3778746430181873, + "grad_norm": 1.222188793192418, + "learning_rate": 2.8595498619011916e-06, + "loss": 0.9835, + "num_input_tokens_seen": 44915650, + "step": 5028 + }, + { + "epoch": 0.37794979708402227, + "grad_norm": 1.450010345685159, + "learning_rate": 2.8591102490759468e-06, + "loss": 0.9419, + "num_input_tokens_seen": 44937635, + "step": 5029 + }, + { + "epoch": 0.3780249511498572, + "grad_norm": 1.5516201411946917, + "learning_rate": 2.858670585345129e-06, + "loss": 1.0795, + "num_input_tokens_seen": 44959955, + "step": 5030 + }, + { + "epoch": 0.37810010521569215, + "grad_norm": 1.425401610951258, + "learning_rate": 2.8582308707347913e-06, + "loss": 0.943, + "num_input_tokens_seen": 44983755, + "step": 5031 + }, + { + "epoch": 0.3781752592815271, + "grad_norm": 1.1022542310197885, + "learning_rate": 2.857791105270988e-06, + "loss": 0.832, + "num_input_tokens_seen": 45051710, + "step": 5032 + }, + { + "epoch": 0.3782504133473621, + "grad_norm": 1.614580125702984, + "learning_rate": 2.8573512889797773e-06, + "loss": 1.0016, + "num_input_tokens_seen": 45074160, + "step": 5033 + }, + { + "epoch": 0.37832556741319706, + "grad_norm": 1.5868579467241752, + "learning_rate": 2.8569114218872195e-06, + "loss": 0.981, + "num_input_tokens_seen": 45093505, + "step": 5034 + }, + { + "epoch": 0.37840072147903203, + "grad_norm": 2.188125085076137, + "learning_rate": 2.856471504019379e-06, + "loss": 0.8571, + "num_input_tokens_seen": 45115330, + "step": 5035 + }, + { + "epoch": 0.378475875544867, + "grad_norm": 1.587569876507163, + "learning_rate": 2.856031535402321e-06, + "loss": 0.8765, + "num_input_tokens_seen": 45139295, + "step": 5036 + }, + { + "epoch": 0.3785510296107019, + "grad_norm": 1.839714547903379, + "learning_rate": 2.8555915160621184e-06, + "loss": 1.0165, + "num_input_tokens_seen": 45160620, + "step": 5037 + }, + { + "epoch": 0.3786261836765369, + "grad_norm": 1.360069623595506, + "learning_rate": 2.8551514460248406e-06, + "loss": 1.0669, + "num_input_tokens_seen": 45183330, + "step": 5038 + }, + { + "epoch": 0.37870133774237186, + "grad_norm": 1.7506563454449882, + "learning_rate": 2.8547113253165666e-06, + "loss": 1.0808, + "num_input_tokens_seen": 45205165, + "step": 5039 + }, + { + "epoch": 0.3787764918082068, + "grad_norm": 1.7090215537250553, + "learning_rate": 2.8542711539633723e-06, + "loss": 1.0028, + "num_input_tokens_seen": 45226115, + "step": 5040 + }, + { + "epoch": 0.3788516458740418, + "grad_norm": 1.5976788582169854, + "learning_rate": 2.8538309319913413e-06, + "loss": 0.8837, + "num_input_tokens_seen": 45248720, + "step": 5041 + }, + { + "epoch": 0.37892679993987677, + "grad_norm": 1.4440447221630408, + "learning_rate": 2.8533906594265588e-06, + "loss": 0.9294, + "num_input_tokens_seen": 45275485, + "step": 5042 + }, + { + "epoch": 0.3790019540057117, + "grad_norm": 1.6790063940884117, + "learning_rate": 2.852950336295111e-06, + "loss": 0.8252, + "num_input_tokens_seen": 45299255, + "step": 5043 + }, + { + "epoch": 0.37907710807154665, + "grad_norm": 1.8369186701455702, + "learning_rate": 2.8525099626230894e-06, + "loss": 0.8604, + "num_input_tokens_seen": 45324915, + "step": 5044 + }, + { + "epoch": 0.3791522621373816, + "grad_norm": 1.7724490330824405, + "learning_rate": 2.8520695384365887e-06, + "loss": 0.844, + "num_input_tokens_seen": 45349195, + "step": 5045 + }, + { + "epoch": 0.3792274162032166, + "grad_norm": 1.4993029405293177, + "learning_rate": 2.851629063761705e-06, + "loss": 1.0142, + "num_input_tokens_seen": 45373505, + "step": 5046 + }, + { + "epoch": 0.37930257026905156, + "grad_norm": 1.636700177144211, + "learning_rate": 2.8511885386245373e-06, + "loss": 0.9812, + "num_input_tokens_seen": 45395975, + "step": 5047 + }, + { + "epoch": 0.37937772433488653, + "grad_norm": 1.440752330788073, + "learning_rate": 2.8507479630511905e-06, + "loss": 0.914, + "num_input_tokens_seen": 45420535, + "step": 5048 + }, + { + "epoch": 0.3794528784007215, + "grad_norm": 1.0010585546640256, + "learning_rate": 2.850307337067768e-06, + "loss": 0.9624, + "num_input_tokens_seen": 45498530, + "step": 5049 + }, + { + "epoch": 0.3795280324665564, + "grad_norm": 1.8179895862914777, + "learning_rate": 2.849866660700381e-06, + "loss": 0.8217, + "num_input_tokens_seen": 45521755, + "step": 5050 + }, + { + "epoch": 0.3796031865323914, + "grad_norm": 1.3389107706111825, + "learning_rate": 2.8494259339751396e-06, + "loss": 1.0063, + "num_input_tokens_seen": 45547370, + "step": 5051 + }, + { + "epoch": 0.37967834059822636, + "grad_norm": 1.6711866441243215, + "learning_rate": 2.8489851569181584e-06, + "loss": 0.8815, + "num_input_tokens_seen": 45571480, + "step": 5052 + }, + { + "epoch": 0.37975349466406133, + "grad_norm": 1.2337805366795218, + "learning_rate": 2.848544329555556e-06, + "loss": 0.9919, + "num_input_tokens_seen": 45597640, + "step": 5053 + }, + { + "epoch": 0.3798286487298963, + "grad_norm": 3.9148891284195244, + "learning_rate": 2.8481034519134524e-06, + "loss": 0.9472, + "num_input_tokens_seen": 45626610, + "step": 5054 + }, + { + "epoch": 0.37990380279573127, + "grad_norm": 3.6925438475281465, + "learning_rate": 2.8476625240179726e-06, + "loss": 0.981, + "num_input_tokens_seen": 45645745, + "step": 5055 + }, + { + "epoch": 0.3799789568615662, + "grad_norm": 3.135084284572667, + "learning_rate": 2.847221545895241e-06, + "loss": 0.912, + "num_input_tokens_seen": 45669775, + "step": 5056 + }, + { + "epoch": 0.38005411092740116, + "grad_norm": 1.755078032625401, + "learning_rate": 2.8467805175713897e-06, + "loss": 0.9718, + "num_input_tokens_seen": 45691520, + "step": 5057 + }, + { + "epoch": 0.3801292649932361, + "grad_norm": 1.494999657315116, + "learning_rate": 2.84633943907255e-06, + "loss": 0.9589, + "num_input_tokens_seen": 45714105, + "step": 5058 + }, + { + "epoch": 0.3802044190590711, + "grad_norm": 1.7881728713165297, + "learning_rate": 2.8458983104248575e-06, + "loss": 0.9909, + "num_input_tokens_seen": 45733120, + "step": 5059 + }, + { + "epoch": 0.38027957312490607, + "grad_norm": 1.411512531023551, + "learning_rate": 2.8454571316544504e-06, + "loss": 0.9347, + "num_input_tokens_seen": 45756290, + "step": 5060 + }, + { + "epoch": 0.38035472719074104, + "grad_norm": 1.6744160103839423, + "learning_rate": 2.845015902787472e-06, + "loss": 0.9269, + "num_input_tokens_seen": 45780525, + "step": 5061 + }, + { + "epoch": 0.380429881256576, + "grad_norm": 1.6715315706704577, + "learning_rate": 2.8445746238500647e-06, + "loss": 0.8778, + "num_input_tokens_seen": 45807770, + "step": 5062 + }, + { + "epoch": 0.3805050353224109, + "grad_norm": 1.4812276299312517, + "learning_rate": 2.8441332948683768e-06, + "loss": 0.8425, + "num_input_tokens_seen": 45834625, + "step": 5063 + }, + { + "epoch": 0.3805801893882459, + "grad_norm": 1.2494823814606213, + "learning_rate": 2.8436919158685594e-06, + "loss": 0.9038, + "num_input_tokens_seen": 45909520, + "step": 5064 + }, + { + "epoch": 0.38065534345408086, + "grad_norm": 1.8244880951200035, + "learning_rate": 2.8432504868767648e-06, + "loss": 0.9992, + "num_input_tokens_seen": 45930950, + "step": 5065 + }, + { + "epoch": 0.38073049751991583, + "grad_norm": 1.687583497418868, + "learning_rate": 2.84280900791915e-06, + "loss": 0.972, + "num_input_tokens_seen": 45952390, + "step": 5066 + }, + { + "epoch": 0.3808056515857508, + "grad_norm": 1.6533947210810642, + "learning_rate": 2.8423674790218737e-06, + "loss": 0.8702, + "num_input_tokens_seen": 45977415, + "step": 5067 + }, + { + "epoch": 0.3808808056515858, + "grad_norm": 1.8721905322966441, + "learning_rate": 2.841925900211099e-06, + "loss": 0.8744, + "num_input_tokens_seen": 45999620, + "step": 5068 + }, + { + "epoch": 0.3809559597174207, + "grad_norm": 1.856317843857486, + "learning_rate": 2.841484271512991e-06, + "loss": 0.9124, + "num_input_tokens_seen": 46024335, + "step": 5069 + }, + { + "epoch": 0.38103111378325566, + "grad_norm": 2.4604736859860807, + "learning_rate": 2.8410425929537175e-06, + "loss": 0.914, + "num_input_tokens_seen": 46048630, + "step": 5070 + }, + { + "epoch": 0.38110626784909063, + "grad_norm": 1.4040054209840893, + "learning_rate": 2.8406008645594493e-06, + "loss": 0.9157, + "num_input_tokens_seen": 46072700, + "step": 5071 + }, + { + "epoch": 0.3811814219149256, + "grad_norm": 1.3699399000058325, + "learning_rate": 2.840159086356362e-06, + "loss": 0.9775, + "num_input_tokens_seen": 46097730, + "step": 5072 + }, + { + "epoch": 0.38125657598076057, + "grad_norm": 1.7389654634587821, + "learning_rate": 2.839717258370631e-06, + "loss": 0.9703, + "num_input_tokens_seen": 46122495, + "step": 5073 + }, + { + "epoch": 0.38133173004659554, + "grad_norm": 1.5822239134469323, + "learning_rate": 2.8392753806284367e-06, + "loss": 0.9023, + "num_input_tokens_seen": 46145510, + "step": 5074 + }, + { + "epoch": 0.38140688411243046, + "grad_norm": 1.44622938611268, + "learning_rate": 2.838833453155963e-06, + "loss": 1.0277, + "num_input_tokens_seen": 46167660, + "step": 5075 + }, + { + "epoch": 0.3814820381782654, + "grad_norm": 1.700876413958852, + "learning_rate": 2.8383914759793944e-06, + "loss": 0.9642, + "num_input_tokens_seen": 46189105, + "step": 5076 + }, + { + "epoch": 0.3815571922441004, + "grad_norm": 1.848959336258164, + "learning_rate": 2.8379494491249214e-06, + "loss": 1.0355, + "num_input_tokens_seen": 46210455, + "step": 5077 + }, + { + "epoch": 0.38163234630993537, + "grad_norm": 1.4769116347164661, + "learning_rate": 2.8375073726187334e-06, + "loss": 0.8696, + "num_input_tokens_seen": 46236220, + "step": 5078 + }, + { + "epoch": 0.38170750037577034, + "grad_norm": 1.2640732093989226, + "learning_rate": 2.8370652464870277e-06, + "loss": 0.9986, + "num_input_tokens_seen": 46259930, + "step": 5079 + }, + { + "epoch": 0.3817826544416053, + "grad_norm": 1.1822899413691408, + "learning_rate": 2.836623070756e-06, + "loss": 0.9503, + "num_input_tokens_seen": 46287425, + "step": 5080 + }, + { + "epoch": 0.3818578085074403, + "grad_norm": 1.6111026860524729, + "learning_rate": 2.836180845451852e-06, + "loss": 0.936, + "num_input_tokens_seen": 46309840, + "step": 5081 + }, + { + "epoch": 0.3819329625732752, + "grad_norm": 1.671081211922269, + "learning_rate": 2.835738570600787e-06, + "loss": 0.8994, + "num_input_tokens_seen": 46331770, + "step": 5082 + }, + { + "epoch": 0.38200811663911016, + "grad_norm": 1.5945155978894787, + "learning_rate": 2.835296246229012e-06, + "loss": 1.0026, + "num_input_tokens_seen": 46355795, + "step": 5083 + }, + { + "epoch": 0.38208327070494513, + "grad_norm": 1.745807325522989, + "learning_rate": 2.8348538723627356e-06, + "loss": 0.8973, + "num_input_tokens_seen": 46379720, + "step": 5084 + }, + { + "epoch": 0.3821584247707801, + "grad_norm": 1.6564492879486328, + "learning_rate": 2.83441144902817e-06, + "loss": 1.0246, + "num_input_tokens_seen": 46403705, + "step": 5085 + }, + { + "epoch": 0.3822335788366151, + "grad_norm": 1.783252059415513, + "learning_rate": 2.8339689762515307e-06, + "loss": 0.921, + "num_input_tokens_seen": 46424790, + "step": 5086 + }, + { + "epoch": 0.38230873290245004, + "grad_norm": 1.7419334876483157, + "learning_rate": 2.8335264540590366e-06, + "loss": 0.909, + "num_input_tokens_seen": 46450260, + "step": 5087 + }, + { + "epoch": 0.38238388696828496, + "grad_norm": 1.3026510094730204, + "learning_rate": 2.833083882476908e-06, + "loss": 0.967, + "num_input_tokens_seen": 46473575, + "step": 5088 + }, + { + "epoch": 0.38245904103411993, + "grad_norm": 1.5305867218302964, + "learning_rate": 2.8326412615313695e-06, + "loss": 0.9579, + "num_input_tokens_seen": 46496620, + "step": 5089 + }, + { + "epoch": 0.3825341950999549, + "grad_norm": 1.3723000982239895, + "learning_rate": 2.8321985912486476e-06, + "loss": 0.9482, + "num_input_tokens_seen": 46519095, + "step": 5090 + }, + { + "epoch": 0.38260934916578987, + "grad_norm": 1.333737262496975, + "learning_rate": 2.8317558716549727e-06, + "loss": 0.916, + "num_input_tokens_seen": 46546290, + "step": 5091 + }, + { + "epoch": 0.38268450323162484, + "grad_norm": 1.757032772485897, + "learning_rate": 2.8313131027765774e-06, + "loss": 0.9877, + "num_input_tokens_seen": 46569425, + "step": 5092 + }, + { + "epoch": 0.3827596572974598, + "grad_norm": 1.8716296740528142, + "learning_rate": 2.830870284639697e-06, + "loss": 1.0071, + "num_input_tokens_seen": 46589885, + "step": 5093 + }, + { + "epoch": 0.3828348113632948, + "grad_norm": 0.9939321363028879, + "learning_rate": 2.830427417270571e-06, + "loss": 0.8401, + "num_input_tokens_seen": 46675145, + "step": 5094 + }, + { + "epoch": 0.3829099654291297, + "grad_norm": 1.7837788883855943, + "learning_rate": 2.829984500695441e-06, + "loss": 0.9989, + "num_input_tokens_seen": 46695190, + "step": 5095 + }, + { + "epoch": 0.38298511949496467, + "grad_norm": 1.332862438057921, + "learning_rate": 2.8295415349405508e-06, + "loss": 0.9913, + "num_input_tokens_seen": 46719160, + "step": 5096 + }, + { + "epoch": 0.38306027356079964, + "grad_norm": 1.2634589887061127, + "learning_rate": 2.8290985200321477e-06, + "loss": 0.8846, + "num_input_tokens_seen": 46744555, + "step": 5097 + }, + { + "epoch": 0.3831354276266346, + "grad_norm": 0.8985077871941506, + "learning_rate": 2.8286554559964826e-06, + "loss": 0.7679, + "num_input_tokens_seen": 46819855, + "step": 5098 + }, + { + "epoch": 0.3832105816924696, + "grad_norm": 2.0075874647799172, + "learning_rate": 2.8282123428598096e-06, + "loss": 0.9233, + "num_input_tokens_seen": 46842475, + "step": 5099 + }, + { + "epoch": 0.38328573575830455, + "grad_norm": 1.3620580743964497, + "learning_rate": 2.8277691806483824e-06, + "loss": 0.9971, + "num_input_tokens_seen": 46866965, + "step": 5100 + }, + { + "epoch": 0.38336088982413946, + "grad_norm": 1.4719549072345541, + "learning_rate": 2.8273259693884625e-06, + "loss": 1.0158, + "num_input_tokens_seen": 46888860, + "step": 5101 + }, + { + "epoch": 0.38343604388997443, + "grad_norm": 1.3910763468848677, + "learning_rate": 2.8268827091063105e-06, + "loss": 1.027, + "num_input_tokens_seen": 46912980, + "step": 5102 + }, + { + "epoch": 0.3835111979558094, + "grad_norm": 1.498984310869663, + "learning_rate": 2.8264393998281916e-06, + "loss": 0.8669, + "num_input_tokens_seen": 46937200, + "step": 5103 + }, + { + "epoch": 0.3835863520216444, + "grad_norm": 1.240308960601762, + "learning_rate": 2.825996041580373e-06, + "loss": 0.9423, + "num_input_tokens_seen": 46962500, + "step": 5104 + }, + { + "epoch": 0.38366150608747934, + "grad_norm": 1.6043386821254766, + "learning_rate": 2.825552634389127e-06, + "loss": 0.8935, + "num_input_tokens_seen": 46985310, + "step": 5105 + }, + { + "epoch": 0.3837366601533143, + "grad_norm": 0.8451762683688845, + "learning_rate": 2.8251091782807265e-06, + "loss": 0.8129, + "num_input_tokens_seen": 47066695, + "step": 5106 + }, + { + "epoch": 0.3838118142191493, + "grad_norm": 1.518416753557712, + "learning_rate": 2.8246656732814463e-06, + "loss": 0.8873, + "num_input_tokens_seen": 47093090, + "step": 5107 + }, + { + "epoch": 0.3838869682849842, + "grad_norm": 1.398284946410762, + "learning_rate": 2.8242221194175676e-06, + "loss": 0.9015, + "num_input_tokens_seen": 47119035, + "step": 5108 + }, + { + "epoch": 0.38396212235081917, + "grad_norm": 1.9375920427670876, + "learning_rate": 2.8237785167153726e-06, + "loss": 0.9808, + "num_input_tokens_seen": 47142830, + "step": 5109 + }, + { + "epoch": 0.38403727641665414, + "grad_norm": 1.210132060292503, + "learning_rate": 2.8233348652011456e-06, + "loss": 1.001, + "num_input_tokens_seen": 47169130, + "step": 5110 + }, + { + "epoch": 0.3841124304824891, + "grad_norm": 2.3342820237116046, + "learning_rate": 2.8228911649011755e-06, + "loss": 1.0298, + "num_input_tokens_seen": 47192645, + "step": 5111 + }, + { + "epoch": 0.3841875845483241, + "grad_norm": 1.6289447441401044, + "learning_rate": 2.8224474158417526e-06, + "loss": 0.9916, + "num_input_tokens_seen": 47214140, + "step": 5112 + }, + { + "epoch": 0.38426273861415905, + "grad_norm": 1.731852578741097, + "learning_rate": 2.8220036180491703e-06, + "loss": 0.9489, + "num_input_tokens_seen": 47237455, + "step": 5113 + }, + { + "epoch": 0.38433789267999396, + "grad_norm": 2.0663881355169518, + "learning_rate": 2.8215597715497266e-06, + "loss": 0.9603, + "num_input_tokens_seen": 47259625, + "step": 5114 + }, + { + "epoch": 0.38441304674582893, + "grad_norm": 1.9138805069329996, + "learning_rate": 2.8211158763697205e-06, + "loss": 0.8974, + "num_input_tokens_seen": 47282680, + "step": 5115 + }, + { + "epoch": 0.3844882008116639, + "grad_norm": 2.2712198564274813, + "learning_rate": 2.820671932535455e-06, + "loss": 0.9663, + "num_input_tokens_seen": 47306235, + "step": 5116 + }, + { + "epoch": 0.3845633548774989, + "grad_norm": 1.3951317271725026, + "learning_rate": 2.8202279400732343e-06, + "loss": 0.9045, + "num_input_tokens_seen": 47330165, + "step": 5117 + }, + { + "epoch": 0.38463850894333385, + "grad_norm": 1.4613981285754603, + "learning_rate": 2.819783899009367e-06, + "loss": 0.9138, + "num_input_tokens_seen": 47355940, + "step": 5118 + }, + { + "epoch": 0.3847136630091688, + "grad_norm": 1.5242763459523854, + "learning_rate": 2.819339809370165e-06, + "loss": 0.9083, + "num_input_tokens_seen": 47381470, + "step": 5119 + }, + { + "epoch": 0.38478881707500373, + "grad_norm": 1.6083724964229387, + "learning_rate": 2.8188956711819413e-06, + "loss": 0.8983, + "num_input_tokens_seen": 47408830, + "step": 5120 + }, + { + "epoch": 0.3848639711408387, + "grad_norm": 1.43031789087301, + "learning_rate": 2.818451484471014e-06, + "loss": 1.0106, + "num_input_tokens_seen": 47430385, + "step": 5121 + }, + { + "epoch": 0.38493912520667367, + "grad_norm": 1.47757795021735, + "learning_rate": 2.8180072492637016e-06, + "loss": 0.9131, + "num_input_tokens_seen": 47453670, + "step": 5122 + }, + { + "epoch": 0.38501427927250864, + "grad_norm": 1.9000559371945827, + "learning_rate": 2.817562965586328e-06, + "loss": 0.8284, + "num_input_tokens_seen": 47476715, + "step": 5123 + }, + { + "epoch": 0.3850894333383436, + "grad_norm": 1.4629909343477712, + "learning_rate": 2.8171186334652174e-06, + "loss": 1.0672, + "num_input_tokens_seen": 47501275, + "step": 5124 + }, + { + "epoch": 0.3851645874041786, + "grad_norm": 0.8181275469548713, + "learning_rate": 2.8166742529266988e-06, + "loss": 0.7633, + "num_input_tokens_seen": 47568035, + "step": 5125 + }, + { + "epoch": 0.38523974147001355, + "grad_norm": 1.5257928882004612, + "learning_rate": 2.8162298239971036e-06, + "loss": 0.9905, + "num_input_tokens_seen": 47592115, + "step": 5126 + }, + { + "epoch": 0.38531489553584847, + "grad_norm": 1.7469996208273482, + "learning_rate": 2.8157853467027665e-06, + "loss": 1.0471, + "num_input_tokens_seen": 47616070, + "step": 5127 + }, + { + "epoch": 0.38539004960168344, + "grad_norm": 0.9920473976040749, + "learning_rate": 2.815340821070023e-06, + "loss": 0.8409, + "num_input_tokens_seen": 47690185, + "step": 5128 + }, + { + "epoch": 0.3854652036675184, + "grad_norm": 1.471800051844218, + "learning_rate": 2.8148962471252135e-06, + "loss": 0.9938, + "num_input_tokens_seen": 47712260, + "step": 5129 + }, + { + "epoch": 0.3855403577333534, + "grad_norm": 1.8659356681266588, + "learning_rate": 2.8144516248946813e-06, + "loss": 0.8315, + "num_input_tokens_seen": 47735060, + "step": 5130 + }, + { + "epoch": 0.38561551179918835, + "grad_norm": 1.5285515655963766, + "learning_rate": 2.8140069544047717e-06, + "loss": 0.9566, + "num_input_tokens_seen": 47757860, + "step": 5131 + }, + { + "epoch": 0.3856906658650233, + "grad_norm": 1.6250261891528823, + "learning_rate": 2.813562235681833e-06, + "loss": 0.9596, + "num_input_tokens_seen": 47779100, + "step": 5132 + }, + { + "epoch": 0.38576581993085823, + "grad_norm": 0.8479811577480489, + "learning_rate": 2.813117468752216e-06, + "loss": 0.758, + "num_input_tokens_seen": 47859150, + "step": 5133 + }, + { + "epoch": 0.3858409739966932, + "grad_norm": 1.57389786623272, + "learning_rate": 2.812672653642276e-06, + "loss": 0.9685, + "num_input_tokens_seen": 47883090, + "step": 5134 + }, + { + "epoch": 0.3859161280625282, + "grad_norm": 1.6326999169189895, + "learning_rate": 2.812227790378369e-06, + "loss": 1.0134, + "num_input_tokens_seen": 47905780, + "step": 5135 + }, + { + "epoch": 0.38599128212836314, + "grad_norm": 1.5196049140392838, + "learning_rate": 2.811782878986855e-06, + "loss": 0.9563, + "num_input_tokens_seen": 47930600, + "step": 5136 + }, + { + "epoch": 0.3860664361941981, + "grad_norm": 1.5772479046614039, + "learning_rate": 2.811337919494097e-06, + "loss": 1.0945, + "num_input_tokens_seen": 47953740, + "step": 5137 + }, + { + "epoch": 0.3861415902600331, + "grad_norm": 1.5973075355559807, + "learning_rate": 2.8108929119264608e-06, + "loss": 0.8033, + "num_input_tokens_seen": 47979305, + "step": 5138 + }, + { + "epoch": 0.38621674432586806, + "grad_norm": 2.1303228022653764, + "learning_rate": 2.8104478563103145e-06, + "loss": 0.9486, + "num_input_tokens_seen": 48005815, + "step": 5139 + }, + { + "epoch": 0.38629189839170297, + "grad_norm": 1.5049126762580716, + "learning_rate": 2.8100027526720283e-06, + "loss": 0.9746, + "num_input_tokens_seen": 48026820, + "step": 5140 + }, + { + "epoch": 0.38636705245753794, + "grad_norm": 1.5159081263420173, + "learning_rate": 2.8095576010379784e-06, + "loss": 0.9304, + "num_input_tokens_seen": 48048025, + "step": 5141 + }, + { + "epoch": 0.3864422065233729, + "grad_norm": 1.6125701001920651, + "learning_rate": 2.80911240143454e-06, + "loss": 1.0118, + "num_input_tokens_seen": 48070785, + "step": 5142 + }, + { + "epoch": 0.3865173605892079, + "grad_norm": 1.5016237053915984, + "learning_rate": 2.8086671538880938e-06, + "loss": 1.0109, + "num_input_tokens_seen": 48093000, + "step": 5143 + }, + { + "epoch": 0.38659251465504285, + "grad_norm": 1.853301828834744, + "learning_rate": 2.808221858425022e-06, + "loss": 0.9273, + "num_input_tokens_seen": 48113695, + "step": 5144 + }, + { + "epoch": 0.3866676687208778, + "grad_norm": 1.8908686835508244, + "learning_rate": 2.8077765150717107e-06, + "loss": 1.0438, + "num_input_tokens_seen": 48138070, + "step": 5145 + }, + { + "epoch": 0.38674282278671274, + "grad_norm": 1.4358440480835202, + "learning_rate": 2.807331123854547e-06, + "loss": 0.9973, + "num_input_tokens_seen": 48162520, + "step": 5146 + }, + { + "epoch": 0.3868179768525477, + "grad_norm": 1.4515782810170488, + "learning_rate": 2.806885684799923e-06, + "loss": 0.9524, + "num_input_tokens_seen": 48188390, + "step": 5147 + }, + { + "epoch": 0.3868931309183827, + "grad_norm": 1.7430335757358715, + "learning_rate": 2.8064401979342324e-06, + "loss": 0.9816, + "num_input_tokens_seen": 48209425, + "step": 5148 + }, + { + "epoch": 0.38696828498421765, + "grad_norm": 1.267403988860422, + "learning_rate": 2.805994663283872e-06, + "loss": 0.9089, + "num_input_tokens_seen": 48234965, + "step": 5149 + }, + { + "epoch": 0.3870434390500526, + "grad_norm": 1.4777497055061162, + "learning_rate": 2.805549080875242e-06, + "loss": 0.9875, + "num_input_tokens_seen": 48258410, + "step": 5150 + }, + { + "epoch": 0.3871185931158876, + "grad_norm": 2.5738886388536173, + "learning_rate": 2.8051034507347435e-06, + "loss": 0.9094, + "num_input_tokens_seen": 48283305, + "step": 5151 + }, + { + "epoch": 0.38719374718172256, + "grad_norm": 1.6494609674771303, + "learning_rate": 2.804657772888783e-06, + "loss": 0.9709, + "num_input_tokens_seen": 48305815, + "step": 5152 + }, + { + "epoch": 0.3872689012475575, + "grad_norm": 2.2880243544649512, + "learning_rate": 2.804212047363768e-06, + "loss": 0.8609, + "num_input_tokens_seen": 48325745, + "step": 5153 + }, + { + "epoch": 0.38734405531339244, + "grad_norm": 1.3500107694125656, + "learning_rate": 2.8037662741861097e-06, + "loss": 1.0274, + "num_input_tokens_seen": 48347660, + "step": 5154 + }, + { + "epoch": 0.3874192093792274, + "grad_norm": 1.5583921985761895, + "learning_rate": 2.803320453382222e-06, + "loss": 0.9509, + "num_input_tokens_seen": 48371115, + "step": 5155 + }, + { + "epoch": 0.3874943634450624, + "grad_norm": 5.292287094309853, + "learning_rate": 2.8028745849785213e-06, + "loss": 1.0084, + "num_input_tokens_seen": 48394690, + "step": 5156 + }, + { + "epoch": 0.38756951751089735, + "grad_norm": 1.8204925186141798, + "learning_rate": 2.8024286690014266e-06, + "loss": 0.9925, + "num_input_tokens_seen": 48416705, + "step": 5157 + }, + { + "epoch": 0.3876446715767323, + "grad_norm": 1.6430962977871626, + "learning_rate": 2.801982705477361e-06, + "loss": 1.07, + "num_input_tokens_seen": 48438620, + "step": 5158 + }, + { + "epoch": 0.38771982564256724, + "grad_norm": 2.3329646863819593, + "learning_rate": 2.801536694432749e-06, + "loss": 0.9804, + "num_input_tokens_seen": 48459335, + "step": 5159 + }, + { + "epoch": 0.3877949797084022, + "grad_norm": 1.5879291374913334, + "learning_rate": 2.8010906358940185e-06, + "loss": 0.9314, + "num_input_tokens_seen": 48484105, + "step": 5160 + }, + { + "epoch": 0.3878701337742372, + "grad_norm": 1.6836118055468527, + "learning_rate": 2.8006445298876003e-06, + "loss": 0.9043, + "num_input_tokens_seen": 48507545, + "step": 5161 + }, + { + "epoch": 0.38794528784007215, + "grad_norm": 0.928533783872708, + "learning_rate": 2.800198376439928e-06, + "loss": 0.8526, + "num_input_tokens_seen": 48584175, + "step": 5162 + }, + { + "epoch": 0.3880204419059071, + "grad_norm": 1.8614681867677758, + "learning_rate": 2.7997521755774373e-06, + "loss": 0.9796, + "num_input_tokens_seen": 48608405, + "step": 5163 + }, + { + "epoch": 0.3880955959717421, + "grad_norm": 1.4986683179471794, + "learning_rate": 2.799305927326568e-06, + "loss": 0.9842, + "num_input_tokens_seen": 48632165, + "step": 5164 + }, + { + "epoch": 0.388170750037577, + "grad_norm": 1.3312843891412225, + "learning_rate": 2.7988596317137623e-06, + "loss": 0.966, + "num_input_tokens_seen": 48659355, + "step": 5165 + }, + { + "epoch": 0.388245904103412, + "grad_norm": 1.4187400548645186, + "learning_rate": 2.7984132887654633e-06, + "loss": 0.9663, + "num_input_tokens_seen": 48682545, + "step": 5166 + }, + { + "epoch": 0.38832105816924695, + "grad_norm": 1.7175160226532615, + "learning_rate": 2.7979668985081204e-06, + "loss": 0.9255, + "num_input_tokens_seen": 48705885, + "step": 5167 + }, + { + "epoch": 0.3883962122350819, + "grad_norm": 1.577809443887345, + "learning_rate": 2.797520460968183e-06, + "loss": 0.9298, + "num_input_tokens_seen": 48730595, + "step": 5168 + }, + { + "epoch": 0.3884713663009169, + "grad_norm": 1.61975203912309, + "learning_rate": 2.797073976172104e-06, + "loss": 0.9062, + "num_input_tokens_seen": 48756725, + "step": 5169 + }, + { + "epoch": 0.38854652036675186, + "grad_norm": 1.4548737862499401, + "learning_rate": 2.79662744414634e-06, + "loss": 1.0209, + "num_input_tokens_seen": 48777995, + "step": 5170 + }, + { + "epoch": 0.3886216744325868, + "grad_norm": 1.7362372907773478, + "learning_rate": 2.79618086491735e-06, + "loss": 1.0259, + "num_input_tokens_seen": 48801200, + "step": 5171 + }, + { + "epoch": 0.38869682849842174, + "grad_norm": 1.5440327898852553, + "learning_rate": 2.7957342385115944e-06, + "loss": 0.9524, + "num_input_tokens_seen": 48824120, + "step": 5172 + }, + { + "epoch": 0.3887719825642567, + "grad_norm": 1.7327606661227082, + "learning_rate": 2.795287564955538e-06, + "loss": 1.0181, + "num_input_tokens_seen": 48847590, + "step": 5173 + }, + { + "epoch": 0.3888471366300917, + "grad_norm": 1.739034219599274, + "learning_rate": 2.7948408442756477e-06, + "loss": 0.9836, + "num_input_tokens_seen": 48870525, + "step": 5174 + }, + { + "epoch": 0.38892229069592665, + "grad_norm": 1.5892689108534304, + "learning_rate": 2.794394076498394e-06, + "loss": 1.0963, + "num_input_tokens_seen": 48893390, + "step": 5175 + }, + { + "epoch": 0.3889974447617616, + "grad_norm": 1.3600286030566424, + "learning_rate": 2.79394726165025e-06, + "loss": 0.9544, + "num_input_tokens_seen": 48915025, + "step": 5176 + }, + { + "epoch": 0.3890725988275966, + "grad_norm": 0.7745572471295864, + "learning_rate": 2.79350039975769e-06, + "loss": 0.798, + "num_input_tokens_seen": 48990365, + "step": 5177 + }, + { + "epoch": 0.3891477528934315, + "grad_norm": 0.7505895363897344, + "learning_rate": 2.7930534908471927e-06, + "loss": 0.8215, + "num_input_tokens_seen": 49062470, + "step": 5178 + }, + { + "epoch": 0.3892229069592665, + "grad_norm": 1.5365640722409875, + "learning_rate": 2.792606534945239e-06, + "loss": 0.9602, + "num_input_tokens_seen": 49084355, + "step": 5179 + }, + { + "epoch": 0.38929806102510145, + "grad_norm": 1.752689474749438, + "learning_rate": 2.7921595320783136e-06, + "loss": 0.9173, + "num_input_tokens_seen": 49105540, + "step": 5180 + }, + { + "epoch": 0.3893732150909364, + "grad_norm": 1.5116333456828728, + "learning_rate": 2.7917124822729022e-06, + "loss": 1.0181, + "num_input_tokens_seen": 49129675, + "step": 5181 + }, + { + "epoch": 0.3894483691567714, + "grad_norm": 1.525492998075217, + "learning_rate": 2.791265385555495e-06, + "loss": 0.9437, + "num_input_tokens_seen": 49151380, + "step": 5182 + }, + { + "epoch": 0.38952352322260636, + "grad_norm": 1.3923687235729023, + "learning_rate": 2.7908182419525834e-06, + "loss": 0.9172, + "num_input_tokens_seen": 49176225, + "step": 5183 + }, + { + "epoch": 0.38959867728844133, + "grad_norm": 0.967385407692139, + "learning_rate": 2.7903710514906626e-06, + "loss": 0.8425, + "num_input_tokens_seen": 49252315, + "step": 5184 + }, + { + "epoch": 0.38967383135427625, + "grad_norm": 1.594407787337285, + "learning_rate": 2.7899238141962304e-06, + "loss": 0.9373, + "num_input_tokens_seen": 49274545, + "step": 5185 + }, + { + "epoch": 0.3897489854201112, + "grad_norm": 1.6661267439176064, + "learning_rate": 2.7894765300957875e-06, + "loss": 1.0161, + "num_input_tokens_seen": 49294870, + "step": 5186 + }, + { + "epoch": 0.3898241394859462, + "grad_norm": 1.3125481369637169, + "learning_rate": 2.7890291992158376e-06, + "loss": 1.0128, + "num_input_tokens_seen": 49322150, + "step": 5187 + }, + { + "epoch": 0.38989929355178116, + "grad_norm": 1.6547084327937034, + "learning_rate": 2.7885818215828856e-06, + "loss": 1.0156, + "num_input_tokens_seen": 49344835, + "step": 5188 + }, + { + "epoch": 0.3899744476176161, + "grad_norm": 2.1889770970372804, + "learning_rate": 2.7881343972234416e-06, + "loss": 0.9217, + "num_input_tokens_seen": 49369355, + "step": 5189 + }, + { + "epoch": 0.3900496016834511, + "grad_norm": 1.8477929611114927, + "learning_rate": 2.787686926164016e-06, + "loss": 0.7494, + "num_input_tokens_seen": 49392670, + "step": 5190 + }, + { + "epoch": 0.390124755749286, + "grad_norm": 1.5515680503169702, + "learning_rate": 2.787239408431124e-06, + "loss": 0.9319, + "num_input_tokens_seen": 49416395, + "step": 5191 + }, + { + "epoch": 0.390199909815121, + "grad_norm": 1.7933076691947822, + "learning_rate": 2.786791844051282e-06, + "loss": 0.8582, + "num_input_tokens_seen": 49436565, + "step": 5192 + }, + { + "epoch": 0.39027506388095595, + "grad_norm": 1.6222240506050107, + "learning_rate": 2.7863442330510115e-06, + "loss": 0.9472, + "num_input_tokens_seen": 49459310, + "step": 5193 + }, + { + "epoch": 0.3903502179467909, + "grad_norm": 1.6681957939050105, + "learning_rate": 2.7858965754568335e-06, + "loss": 0.9174, + "num_input_tokens_seen": 49482230, + "step": 5194 + }, + { + "epoch": 0.3904253720126259, + "grad_norm": 1.5969635224114676, + "learning_rate": 2.7854488712952735e-06, + "loss": 0.9085, + "num_input_tokens_seen": 49506375, + "step": 5195 + }, + { + "epoch": 0.39050052607846086, + "grad_norm": 2.0599814556731535, + "learning_rate": 2.7850011205928607e-06, + "loss": 1.0016, + "num_input_tokens_seen": 49531770, + "step": 5196 + }, + { + "epoch": 0.39057568014429583, + "grad_norm": 1.780793784093162, + "learning_rate": 2.7845533233761256e-06, + "loss": 0.9386, + "num_input_tokens_seen": 49554800, + "step": 5197 + }, + { + "epoch": 0.39065083421013075, + "grad_norm": 1.2999398199237067, + "learning_rate": 2.784105479671602e-06, + "loss": 0.9819, + "num_input_tokens_seen": 49578865, + "step": 5198 + }, + { + "epoch": 0.3907259882759657, + "grad_norm": 2.2136485593829045, + "learning_rate": 2.783657589505826e-06, + "loss": 0.8711, + "num_input_tokens_seen": 49604935, + "step": 5199 + }, + { + "epoch": 0.3908011423418007, + "grad_norm": 1.363285928188047, + "learning_rate": 2.783209652905337e-06, + "loss": 0.9269, + "num_input_tokens_seen": 49630115, + "step": 5200 + }, + { + "epoch": 0.39087629640763566, + "grad_norm": 0.9596445643050673, + "learning_rate": 2.7827616698966763e-06, + "loss": 0.8572, + "num_input_tokens_seen": 49709515, + "step": 5201 + }, + { + "epoch": 0.39095145047347063, + "grad_norm": 1.5642889843960677, + "learning_rate": 2.78231364050639e-06, + "loss": 1.0485, + "num_input_tokens_seen": 49729865, + "step": 5202 + }, + { + "epoch": 0.3910266045393056, + "grad_norm": 1.4339846875744517, + "learning_rate": 2.781865564761025e-06, + "loss": 0.9033, + "num_input_tokens_seen": 49757070, + "step": 5203 + }, + { + "epoch": 0.3911017586051405, + "grad_norm": 1.5681281388434951, + "learning_rate": 2.781417442687131e-06, + "loss": 0.9557, + "num_input_tokens_seen": 49780495, + "step": 5204 + }, + { + "epoch": 0.3911769126709755, + "grad_norm": 1.8917836193728568, + "learning_rate": 2.7809692743112616e-06, + "loss": 0.9205, + "num_input_tokens_seen": 49800930, + "step": 5205 + }, + { + "epoch": 0.39125206673681046, + "grad_norm": 1.8433045964590737, + "learning_rate": 2.780521059659972e-06, + "loss": 1.0717, + "num_input_tokens_seen": 49822945, + "step": 5206 + }, + { + "epoch": 0.3913272208026454, + "grad_norm": 0.8700548485753177, + "learning_rate": 2.78007279875982e-06, + "loss": 0.732, + "num_input_tokens_seen": 49905100, + "step": 5207 + }, + { + "epoch": 0.3914023748684804, + "grad_norm": 0.7518861418776085, + "learning_rate": 2.7796244916373686e-06, + "loss": 0.7876, + "num_input_tokens_seen": 49982320, + "step": 5208 + }, + { + "epoch": 0.39147752893431537, + "grad_norm": 1.6322941299481561, + "learning_rate": 2.7791761383191807e-06, + "loss": 0.965, + "num_input_tokens_seen": 50003470, + "step": 5209 + }, + { + "epoch": 0.3915526830001503, + "grad_norm": 1.5914618578387119, + "learning_rate": 2.778727738831822e-06, + "loss": 0.9132, + "num_input_tokens_seen": 50028870, + "step": 5210 + }, + { + "epoch": 0.39162783706598525, + "grad_norm": 1.3640671353081144, + "learning_rate": 2.7782792932018635e-06, + "loss": 0.8306, + "num_input_tokens_seen": 50105415, + "step": 5211 + }, + { + "epoch": 0.3917029911318202, + "grad_norm": 1.4909199228971686, + "learning_rate": 2.7778308014558767e-06, + "loss": 0.9346, + "num_input_tokens_seen": 50128030, + "step": 5212 + }, + { + "epoch": 0.3917781451976552, + "grad_norm": 1.5895116960707778, + "learning_rate": 2.777382263620436e-06, + "loss": 0.9215, + "num_input_tokens_seen": 50155430, + "step": 5213 + }, + { + "epoch": 0.39185329926349016, + "grad_norm": 1.752949817084979, + "learning_rate": 2.7769336797221197e-06, + "loss": 1.0212, + "num_input_tokens_seen": 50175085, + "step": 5214 + }, + { + "epoch": 0.39192845332932513, + "grad_norm": 1.579032485951465, + "learning_rate": 2.7764850497875076e-06, + "loss": 0.8717, + "num_input_tokens_seen": 50198450, + "step": 5215 + }, + { + "epoch": 0.3920036073951601, + "grad_norm": 1.5235515845203864, + "learning_rate": 2.776036373843183e-06, + "loss": 0.9151, + "num_input_tokens_seen": 50223565, + "step": 5216 + }, + { + "epoch": 0.392078761460995, + "grad_norm": 1.8030927473386862, + "learning_rate": 2.775587651915732e-06, + "loss": 1.0478, + "num_input_tokens_seen": 50243110, + "step": 5217 + }, + { + "epoch": 0.39215391552683, + "grad_norm": 1.511840499982619, + "learning_rate": 2.775138884031742e-06, + "loss": 0.9045, + "num_input_tokens_seen": 50268710, + "step": 5218 + }, + { + "epoch": 0.39222906959266496, + "grad_norm": 1.8409358385499792, + "learning_rate": 2.7746900702178053e-06, + "loss": 1.1112, + "num_input_tokens_seen": 50289285, + "step": 5219 + }, + { + "epoch": 0.39230422365849993, + "grad_norm": 1.4192027037714132, + "learning_rate": 2.7742412105005154e-06, + "loss": 0.96, + "num_input_tokens_seen": 50315395, + "step": 5220 + }, + { + "epoch": 0.3923793777243349, + "grad_norm": 1.6123419514167396, + "learning_rate": 2.773792304906469e-06, + "loss": 0.9945, + "num_input_tokens_seen": 50338130, + "step": 5221 + }, + { + "epoch": 0.39245453179016987, + "grad_norm": 1.7385696686199446, + "learning_rate": 2.7733433534622655e-06, + "loss": 0.9497, + "num_input_tokens_seen": 50360695, + "step": 5222 + }, + { + "epoch": 0.3925296858560048, + "grad_norm": 2.222901235262916, + "learning_rate": 2.772894356194507e-06, + "loss": 1.026, + "num_input_tokens_seen": 50381680, + "step": 5223 + }, + { + "epoch": 0.39260483992183975, + "grad_norm": 0.988170361234703, + "learning_rate": 2.7724453131297988e-06, + "loss": 0.7981, + "num_input_tokens_seen": 50448200, + "step": 5224 + }, + { + "epoch": 0.3926799939876747, + "grad_norm": 1.7625466577292663, + "learning_rate": 2.771996224294747e-06, + "loss": 0.9625, + "num_input_tokens_seen": 50469840, + "step": 5225 + }, + { + "epoch": 0.3927551480535097, + "grad_norm": 1.8199694396167525, + "learning_rate": 2.7715470897159636e-06, + "loss": 1.067, + "num_input_tokens_seen": 50489055, + "step": 5226 + }, + { + "epoch": 0.39283030211934467, + "grad_norm": 1.6811632637424896, + "learning_rate": 2.7710979094200593e-06, + "loss": 1.0159, + "num_input_tokens_seen": 50510410, + "step": 5227 + }, + { + "epoch": 0.39290545618517964, + "grad_norm": 1.4248359786552824, + "learning_rate": 2.7706486834336524e-06, + "loss": 0.9363, + "num_input_tokens_seen": 50532330, + "step": 5228 + }, + { + "epoch": 0.3929806102510146, + "grad_norm": 1.4433944759824007, + "learning_rate": 2.7701994117833596e-06, + "loss": 1.0062, + "num_input_tokens_seen": 50556700, + "step": 5229 + }, + { + "epoch": 0.3930557643168495, + "grad_norm": 1.571837107389356, + "learning_rate": 2.7697500944958024e-06, + "loss": 0.9704, + "num_input_tokens_seen": 50577660, + "step": 5230 + }, + { + "epoch": 0.3931309183826845, + "grad_norm": 1.569824750315812, + "learning_rate": 2.7693007315976047e-06, + "loss": 1.0222, + "num_input_tokens_seen": 50601715, + "step": 5231 + }, + { + "epoch": 0.39320607244851946, + "grad_norm": 1.3239468095559528, + "learning_rate": 2.7688513231153926e-06, + "loss": 0.8899, + "num_input_tokens_seen": 50625380, + "step": 5232 + }, + { + "epoch": 0.39328122651435443, + "grad_norm": 2.0042061683531696, + "learning_rate": 2.7684018690757954e-06, + "loss": 0.8901, + "num_input_tokens_seen": 50647120, + "step": 5233 + }, + { + "epoch": 0.3933563805801894, + "grad_norm": 2.844327092564614, + "learning_rate": 2.767952369505445e-06, + "loss": 1.0014, + "num_input_tokens_seen": 50667415, + "step": 5234 + }, + { + "epoch": 0.3934315346460244, + "grad_norm": 1.9066066720944819, + "learning_rate": 2.7675028244309766e-06, + "loss": 0.9381, + "num_input_tokens_seen": 50688730, + "step": 5235 + }, + { + "epoch": 0.3935066887118593, + "grad_norm": 1.2574461409332895, + "learning_rate": 2.767053233879026e-06, + "loss": 0.9584, + "num_input_tokens_seen": 50714060, + "step": 5236 + }, + { + "epoch": 0.39358184277769426, + "grad_norm": 1.7267384849375367, + "learning_rate": 2.766603597876235e-06, + "loss": 0.9941, + "num_input_tokens_seen": 50734910, + "step": 5237 + }, + { + "epoch": 0.3936569968435292, + "grad_norm": 1.2604234500839087, + "learning_rate": 2.7661539164492442e-06, + "loss": 0.9075, + "num_input_tokens_seen": 50761940, + "step": 5238 + }, + { + "epoch": 0.3937321509093642, + "grad_norm": 1.420836268372548, + "learning_rate": 2.765704189624701e-06, + "loss": 0.8976, + "num_input_tokens_seen": 50786625, + "step": 5239 + }, + { + "epoch": 0.39380730497519917, + "grad_norm": 1.5859384771635305, + "learning_rate": 2.765254417429252e-06, + "loss": 0.9122, + "num_input_tokens_seen": 50807300, + "step": 5240 + }, + { + "epoch": 0.39388245904103414, + "grad_norm": 1.1052115852291338, + "learning_rate": 2.764804599889549e-06, + "loss": 0.823, + "num_input_tokens_seen": 50876200, + "step": 5241 + }, + { + "epoch": 0.3939576131068691, + "grad_norm": 1.6213820603231428, + "learning_rate": 2.7643547370322446e-06, + "loss": 0.9696, + "num_input_tokens_seen": 50897665, + "step": 5242 + }, + { + "epoch": 0.394032767172704, + "grad_norm": 1.5717529219134079, + "learning_rate": 2.763904828883995e-06, + "loss": 0.9283, + "num_input_tokens_seen": 50922185, + "step": 5243 + }, + { + "epoch": 0.394107921238539, + "grad_norm": 1.7302000642318986, + "learning_rate": 2.763454875471459e-06, + "loss": 1.0314, + "num_input_tokens_seen": 50945470, + "step": 5244 + }, + { + "epoch": 0.39418307530437396, + "grad_norm": 1.594229906419427, + "learning_rate": 2.7630048768212975e-06, + "loss": 1.049, + "num_input_tokens_seen": 50969135, + "step": 5245 + }, + { + "epoch": 0.39425822937020893, + "grad_norm": 1.9676639649488248, + "learning_rate": 2.7625548329601763e-06, + "loss": 0.9294, + "num_input_tokens_seen": 50992160, + "step": 5246 + }, + { + "epoch": 0.3943333834360439, + "grad_norm": 1.4639414538782967, + "learning_rate": 2.7621047439147606e-06, + "loss": 0.9352, + "num_input_tokens_seen": 51016875, + "step": 5247 + }, + { + "epoch": 0.3944085375018789, + "grad_norm": 1.4181003982340021, + "learning_rate": 2.7616546097117213e-06, + "loss": 0.8956, + "num_input_tokens_seen": 51038365, + "step": 5248 + }, + { + "epoch": 0.3944836915677138, + "grad_norm": 1.5956962519933073, + "learning_rate": 2.761204430377729e-06, + "loss": 0.9813, + "num_input_tokens_seen": 51065040, + "step": 5249 + }, + { + "epoch": 0.39455884563354876, + "grad_norm": 1.4125544056499926, + "learning_rate": 2.7607542059394604e-06, + "loss": 0.9776, + "num_input_tokens_seen": 51089550, + "step": 5250 + }, + { + "epoch": 0.39463399969938373, + "grad_norm": 1.983734387899403, + "learning_rate": 2.760303936423591e-06, + "loss": 0.9168, + "num_input_tokens_seen": 51110755, + "step": 5251 + }, + { + "epoch": 0.3947091537652187, + "grad_norm": 1.5001411371589288, + "learning_rate": 2.759853621856802e-06, + "loss": 0.8763, + "num_input_tokens_seen": 51136010, + "step": 5252 + }, + { + "epoch": 0.39478430783105367, + "grad_norm": 1.6495710335498022, + "learning_rate": 2.759403262265777e-06, + "loss": 0.9137, + "num_input_tokens_seen": 51160430, + "step": 5253 + }, + { + "epoch": 0.39485946189688864, + "grad_norm": 1.406435297308026, + "learning_rate": 2.7589528576772e-06, + "loss": 0.991, + "num_input_tokens_seen": 51184465, + "step": 5254 + }, + { + "epoch": 0.39493461596272356, + "grad_norm": 1.4788008582300736, + "learning_rate": 2.7585024081177602e-06, + "loss": 0.8651, + "num_input_tokens_seen": 51207015, + "step": 5255 + }, + { + "epoch": 0.3950097700285585, + "grad_norm": 1.388093594569821, + "learning_rate": 2.7580519136141483e-06, + "loss": 0.9624, + "num_input_tokens_seen": 51230835, + "step": 5256 + }, + { + "epoch": 0.3950849240943935, + "grad_norm": 1.427361853306746, + "learning_rate": 2.7576013741930576e-06, + "loss": 0.9769, + "num_input_tokens_seen": 51252040, + "step": 5257 + }, + { + "epoch": 0.39516007816022847, + "grad_norm": 2.006264174937147, + "learning_rate": 2.7571507898811846e-06, + "loss": 1.0665, + "num_input_tokens_seen": 51270910, + "step": 5258 + }, + { + "epoch": 0.39523523222606344, + "grad_norm": 2.217472729606629, + "learning_rate": 2.756700160705228e-06, + "loss": 0.8422, + "num_input_tokens_seen": 51297815, + "step": 5259 + }, + { + "epoch": 0.3953103862918984, + "grad_norm": 2.801703010094914, + "learning_rate": 2.756249486691889e-06, + "loss": 1.0512, + "num_input_tokens_seen": 51318205, + "step": 5260 + }, + { + "epoch": 0.3953855403577334, + "grad_norm": 1.395010131334274, + "learning_rate": 2.7557987678678723e-06, + "loss": 0.8936, + "num_input_tokens_seen": 51343965, + "step": 5261 + }, + { + "epoch": 0.3954606944235683, + "grad_norm": 1.7721641775502581, + "learning_rate": 2.755348004259884e-06, + "loss": 0.9525, + "num_input_tokens_seen": 51368215, + "step": 5262 + }, + { + "epoch": 0.39553584848940326, + "grad_norm": 1.4101151225255792, + "learning_rate": 2.7548971958946347e-06, + "loss": 1.0314, + "num_input_tokens_seen": 51392580, + "step": 5263 + }, + { + "epoch": 0.39561100255523823, + "grad_norm": 1.7098546891070387, + "learning_rate": 2.7544463427988355e-06, + "loss": 1.0793, + "num_input_tokens_seen": 51416980, + "step": 5264 + }, + { + "epoch": 0.3956861566210732, + "grad_norm": 2.1300250831882206, + "learning_rate": 2.7539954449992014e-06, + "loss": 0.9603, + "num_input_tokens_seen": 51438475, + "step": 5265 + }, + { + "epoch": 0.3957613106869082, + "grad_norm": 0.9839713464883741, + "learning_rate": 2.7535445025224506e-06, + "loss": 0.8195, + "num_input_tokens_seen": 51506300, + "step": 5266 + }, + { + "epoch": 0.39583646475274314, + "grad_norm": 0.764704092053868, + "learning_rate": 2.7530935153953016e-06, + "loss": 0.7999, + "num_input_tokens_seen": 51578390, + "step": 5267 + }, + { + "epoch": 0.39591161881857806, + "grad_norm": 1.5625097655944826, + "learning_rate": 2.752642483644478e-06, + "loss": 0.9775, + "num_input_tokens_seen": 51602900, + "step": 5268 + }, + { + "epoch": 0.39598677288441303, + "grad_norm": 1.1709559078922844, + "learning_rate": 2.752191407296706e-06, + "loss": 0.856, + "num_input_tokens_seen": 51632325, + "step": 5269 + }, + { + "epoch": 0.396061926950248, + "grad_norm": 1.5046426134554591, + "learning_rate": 2.7517402863787123e-06, + "loss": 0.8134, + "num_input_tokens_seen": 51658665, + "step": 5270 + }, + { + "epoch": 0.39613708101608297, + "grad_norm": 1.3522318246628129, + "learning_rate": 2.751289120917228e-06, + "loss": 1.0324, + "num_input_tokens_seen": 51684150, + "step": 5271 + }, + { + "epoch": 0.39621223508191794, + "grad_norm": 1.4708326467597086, + "learning_rate": 2.750837910938987e-06, + "loss": 0.9368, + "num_input_tokens_seen": 51707200, + "step": 5272 + }, + { + "epoch": 0.3962873891477529, + "grad_norm": 1.7925133885265334, + "learning_rate": 2.7503866564707236e-06, + "loss": 0.981, + "num_input_tokens_seen": 51728025, + "step": 5273 + }, + { + "epoch": 0.3963625432135879, + "grad_norm": 1.606164266472818, + "learning_rate": 2.7499353575391784e-06, + "loss": 1.0344, + "num_input_tokens_seen": 51750840, + "step": 5274 + }, + { + "epoch": 0.3964376972794228, + "grad_norm": 1.6792800054170918, + "learning_rate": 2.749484014171091e-06, + "loss": 0.8611, + "num_input_tokens_seen": 51772970, + "step": 5275 + }, + { + "epoch": 0.39651285134525777, + "grad_norm": 1.7581036813950741, + "learning_rate": 2.749032626393206e-06, + "loss": 1.0177, + "num_input_tokens_seen": 51794115, + "step": 5276 + }, + { + "epoch": 0.39658800541109274, + "grad_norm": 1.6212665511722466, + "learning_rate": 2.74858119423227e-06, + "loss": 0.8871, + "num_input_tokens_seen": 51817950, + "step": 5277 + }, + { + "epoch": 0.3966631594769277, + "grad_norm": 1.9093986522252253, + "learning_rate": 2.748129717715031e-06, + "loss": 1.023, + "num_input_tokens_seen": 51838615, + "step": 5278 + }, + { + "epoch": 0.3967383135427627, + "grad_norm": 1.3733701149095061, + "learning_rate": 2.747678196868241e-06, + "loss": 0.9084, + "num_input_tokens_seen": 51860400, + "step": 5279 + }, + { + "epoch": 0.39681346760859765, + "grad_norm": 1.5952315830255042, + "learning_rate": 2.747226631718656e-06, + "loss": 1.0483, + "num_input_tokens_seen": 51883800, + "step": 5280 + }, + { + "epoch": 0.39688862167443256, + "grad_norm": 1.5718738138789712, + "learning_rate": 2.746775022293032e-06, + "loss": 1.0174, + "num_input_tokens_seen": 51907955, + "step": 5281 + }, + { + "epoch": 0.39696377574026753, + "grad_norm": 1.3988100546591389, + "learning_rate": 2.746323368618127e-06, + "loss": 0.9024, + "num_input_tokens_seen": 51932745, + "step": 5282 + }, + { + "epoch": 0.3970389298061025, + "grad_norm": 1.6818778018084621, + "learning_rate": 2.7458716707207054e-06, + "loss": 1.0017, + "num_input_tokens_seen": 51958170, + "step": 5283 + }, + { + "epoch": 0.3971140838719375, + "grad_norm": 1.8798561153745539, + "learning_rate": 2.74541992862753e-06, + "loss": 0.9826, + "num_input_tokens_seen": 51977080, + "step": 5284 + }, + { + "epoch": 0.39718923793777244, + "grad_norm": 1.4970337626256263, + "learning_rate": 2.744968142365371e-06, + "loss": 0.9919, + "num_input_tokens_seen": 51999505, + "step": 5285 + }, + { + "epoch": 0.3972643920036074, + "grad_norm": 1.3349350408401675, + "learning_rate": 2.744516311960996e-06, + "loss": 0.9529, + "num_input_tokens_seen": 52025735, + "step": 5286 + }, + { + "epoch": 0.3973395460694424, + "grad_norm": 1.516600226250015, + "learning_rate": 2.744064437441179e-06, + "loss": 0.9753, + "num_input_tokens_seen": 52045330, + "step": 5287 + }, + { + "epoch": 0.3974147001352773, + "grad_norm": 1.7846851353269797, + "learning_rate": 2.743612518832695e-06, + "loss": 1.0186, + "num_input_tokens_seen": 52067915, + "step": 5288 + }, + { + "epoch": 0.39748985420111227, + "grad_norm": 1.6870102348265905, + "learning_rate": 2.743160556162321e-06, + "loss": 1.0058, + "num_input_tokens_seen": 52093955, + "step": 5289 + }, + { + "epoch": 0.39756500826694724, + "grad_norm": 2.8635255467384484, + "learning_rate": 2.7427085494568383e-06, + "loss": 0.919, + "num_input_tokens_seen": 52121940, + "step": 5290 + }, + { + "epoch": 0.3976401623327822, + "grad_norm": 1.392352713255424, + "learning_rate": 2.742256498743031e-06, + "loss": 0.9186, + "num_input_tokens_seen": 52145050, + "step": 5291 + }, + { + "epoch": 0.3977153163986172, + "grad_norm": 1.7181811691877453, + "learning_rate": 2.7418044040476838e-06, + "loss": 0.8575, + "num_input_tokens_seen": 52166635, + "step": 5292 + }, + { + "epoch": 0.39779047046445215, + "grad_norm": 1.4795166013610672, + "learning_rate": 2.7413522653975842e-06, + "loss": 0.8978, + "num_input_tokens_seen": 52192170, + "step": 5293 + }, + { + "epoch": 0.39786562453028707, + "grad_norm": 1.5822267011306415, + "learning_rate": 2.7409000828195247e-06, + "loss": 0.9176, + "num_input_tokens_seen": 52218315, + "step": 5294 + }, + { + "epoch": 0.39794077859612204, + "grad_norm": 1.468502591986442, + "learning_rate": 2.7404478563402976e-06, + "loss": 0.9287, + "num_input_tokens_seen": 52242360, + "step": 5295 + }, + { + "epoch": 0.398015932661957, + "grad_norm": 1.6192034995614182, + "learning_rate": 2.7399955859867e-06, + "loss": 0.9456, + "num_input_tokens_seen": 52267020, + "step": 5296 + }, + { + "epoch": 0.398091086727792, + "grad_norm": 2.03247678830264, + "learning_rate": 2.739543271785531e-06, + "loss": 1.0323, + "num_input_tokens_seen": 52287655, + "step": 5297 + }, + { + "epoch": 0.39816624079362695, + "grad_norm": 1.922311314623357, + "learning_rate": 2.7390909137635906e-06, + "loss": 0.9129, + "num_input_tokens_seen": 52310125, + "step": 5298 + }, + { + "epoch": 0.3982413948594619, + "grad_norm": 1.4843658446983286, + "learning_rate": 2.7386385119476833e-06, + "loss": 0.9425, + "num_input_tokens_seen": 52333050, + "step": 5299 + }, + { + "epoch": 0.39831654892529683, + "grad_norm": 3.337460029734175, + "learning_rate": 2.738186066364616e-06, + "loss": 0.9317, + "num_input_tokens_seen": 52357265, + "step": 5300 + }, + { + "epoch": 0.3983917029911318, + "grad_norm": 1.5990618279809603, + "learning_rate": 2.7377335770411965e-06, + "loss": 1.0037, + "num_input_tokens_seen": 52380340, + "step": 5301 + }, + { + "epoch": 0.3984668570569668, + "grad_norm": 1.5058768700598903, + "learning_rate": 2.737281044004239e-06, + "loss": 0.903, + "num_input_tokens_seen": 52405875, + "step": 5302 + }, + { + "epoch": 0.39854201112280174, + "grad_norm": 1.5073458092628886, + "learning_rate": 2.7368284672805558e-06, + "loss": 0.974, + "num_input_tokens_seen": 52430705, + "step": 5303 + }, + { + "epoch": 0.3986171651886367, + "grad_norm": 1.6591596416948637, + "learning_rate": 2.7363758468969643e-06, + "loss": 1.0482, + "num_input_tokens_seen": 52451580, + "step": 5304 + }, + { + "epoch": 0.3986923192544717, + "grad_norm": 1.774654760500384, + "learning_rate": 2.735923182880285e-06, + "loss": 1.0519, + "num_input_tokens_seen": 52469485, + "step": 5305 + }, + { + "epoch": 0.39876747332030665, + "grad_norm": 1.498099712847509, + "learning_rate": 2.7354704752573376e-06, + "loss": 0.9868, + "num_input_tokens_seen": 52492640, + "step": 5306 + }, + { + "epoch": 0.39884262738614157, + "grad_norm": 3.24027484410947, + "learning_rate": 2.735017724054949e-06, + "loss": 1.0007, + "num_input_tokens_seen": 52518690, + "step": 5307 + }, + { + "epoch": 0.39891778145197654, + "grad_norm": 1.5595777268970985, + "learning_rate": 2.7345649292999456e-06, + "loss": 0.9342, + "num_input_tokens_seen": 52544235, + "step": 5308 + }, + { + "epoch": 0.3989929355178115, + "grad_norm": 1.729146413416389, + "learning_rate": 2.7341120910191575e-06, + "loss": 0.8403, + "num_input_tokens_seen": 52568855, + "step": 5309 + }, + { + "epoch": 0.3990680895836465, + "grad_norm": 1.4698292439733294, + "learning_rate": 2.733659209239417e-06, + "loss": 1.0355, + "num_input_tokens_seen": 52589520, + "step": 5310 + }, + { + "epoch": 0.39914324364948145, + "grad_norm": 1.7575138092583757, + "learning_rate": 2.7332062839875586e-06, + "loss": 1.0341, + "num_input_tokens_seen": 52611475, + "step": 5311 + }, + { + "epoch": 0.3992183977153164, + "grad_norm": 0.8035183533291634, + "learning_rate": 2.73275331529042e-06, + "loss": 0.7876, + "num_input_tokens_seen": 52689830, + "step": 5312 + }, + { + "epoch": 0.39929355178115133, + "grad_norm": 1.919388379690378, + "learning_rate": 2.7323003031748424e-06, + "loss": 0.9835, + "num_input_tokens_seen": 52713930, + "step": 5313 + }, + { + "epoch": 0.3993687058469863, + "grad_norm": 1.5692795729707663, + "learning_rate": 2.731847247667667e-06, + "loss": 0.9671, + "num_input_tokens_seen": 52739320, + "step": 5314 + }, + { + "epoch": 0.3994438599128213, + "grad_norm": 1.8335875710577079, + "learning_rate": 2.7313941487957398e-06, + "loss": 0.9174, + "num_input_tokens_seen": 52762975, + "step": 5315 + }, + { + "epoch": 0.39951901397865625, + "grad_norm": 1.7855848469824434, + "learning_rate": 2.730941006585909e-06, + "loss": 1.0591, + "num_input_tokens_seen": 52784690, + "step": 5316 + }, + { + "epoch": 0.3995941680444912, + "grad_norm": 2.0316507971094007, + "learning_rate": 2.7304878210650243e-06, + "loss": 0.9064, + "num_input_tokens_seen": 52809425, + "step": 5317 + }, + { + "epoch": 0.3996693221103262, + "grad_norm": 1.250867637877167, + "learning_rate": 2.7300345922599394e-06, + "loss": 0.8995, + "num_input_tokens_seen": 52834375, + "step": 5318 + }, + { + "epoch": 0.39974447617616116, + "grad_norm": 1.3383931573598256, + "learning_rate": 2.7295813201975087e-06, + "loss": 0.9114, + "num_input_tokens_seen": 52859580, + "step": 5319 + }, + { + "epoch": 0.39981963024199607, + "grad_norm": 1.6195076042248804, + "learning_rate": 2.7291280049045916e-06, + "loss": 0.9616, + "num_input_tokens_seen": 52883805, + "step": 5320 + }, + { + "epoch": 0.39989478430783104, + "grad_norm": 3.6630812512466693, + "learning_rate": 2.728674646408048e-06, + "loss": 0.929, + "num_input_tokens_seen": 52910075, + "step": 5321 + }, + { + "epoch": 0.399969938373666, + "grad_norm": 1.5443336341100662, + "learning_rate": 2.7282212447347413e-06, + "loss": 0.9244, + "num_input_tokens_seen": 52933310, + "step": 5322 + }, + { + "epoch": 0.400045092439501, + "grad_norm": 1.699021391533535, + "learning_rate": 2.7277677999115368e-06, + "loss": 0.9273, + "num_input_tokens_seen": 52955630, + "step": 5323 + }, + { + "epoch": 0.40012024650533595, + "grad_norm": 1.4626777606104078, + "learning_rate": 2.7273143119653042e-06, + "loss": 0.948, + "num_input_tokens_seen": 52979560, + "step": 5324 + }, + { + "epoch": 0.4001954005711709, + "grad_norm": 1.588082802792534, + "learning_rate": 2.7268607809229137e-06, + "loss": 0.9013, + "num_input_tokens_seen": 53002400, + "step": 5325 + }, + { + "epoch": 0.40027055463700584, + "grad_norm": 1.838002054564082, + "learning_rate": 2.7264072068112377e-06, + "loss": 1.0373, + "num_input_tokens_seen": 53025860, + "step": 5326 + }, + { + "epoch": 0.4003457087028408, + "grad_norm": 1.6186250940391047, + "learning_rate": 2.725953589657154e-06, + "loss": 0.9786, + "num_input_tokens_seen": 53048330, + "step": 5327 + }, + { + "epoch": 0.4004208627686758, + "grad_norm": 1.2797924217649543, + "learning_rate": 2.7254999294875395e-06, + "loss": 0.9625, + "num_input_tokens_seen": 53072265, + "step": 5328 + }, + { + "epoch": 0.40049601683451075, + "grad_norm": 1.7361509149015113, + "learning_rate": 2.725046226329276e-06, + "loss": 0.9254, + "num_input_tokens_seen": 53093270, + "step": 5329 + }, + { + "epoch": 0.4005711709003457, + "grad_norm": 1.5834349633108644, + "learning_rate": 2.7245924802092476e-06, + "loss": 0.9298, + "num_input_tokens_seen": 53114985, + "step": 5330 + }, + { + "epoch": 0.4006463249661807, + "grad_norm": 0.8353672399826866, + "learning_rate": 2.7241386911543397e-06, + "loss": 0.8018, + "num_input_tokens_seen": 53190045, + "step": 5331 + }, + { + "epoch": 0.40072147903201566, + "grad_norm": 1.318936641195188, + "learning_rate": 2.7236848591914422e-06, + "loss": 0.9097, + "num_input_tokens_seen": 53214975, + "step": 5332 + }, + { + "epoch": 0.4007966330978506, + "grad_norm": 1.7214142697027934, + "learning_rate": 2.7232309843474446e-06, + "loss": 0.8625, + "num_input_tokens_seen": 53238065, + "step": 5333 + }, + { + "epoch": 0.40087178716368554, + "grad_norm": 1.3392451870553896, + "learning_rate": 2.7227770666492423e-06, + "loss": 0.8902, + "num_input_tokens_seen": 53262360, + "step": 5334 + }, + { + "epoch": 0.4009469412295205, + "grad_norm": 0.9608376613948467, + "learning_rate": 2.722323106123731e-06, + "loss": 0.8288, + "num_input_tokens_seen": 53338330, + "step": 5335 + }, + { + "epoch": 0.4010220952953555, + "grad_norm": 1.9160087188689565, + "learning_rate": 2.7218691027978103e-06, + "loss": 0.951, + "num_input_tokens_seen": 53358725, + "step": 5336 + }, + { + "epoch": 0.40109724936119046, + "grad_norm": 1.9620218508453418, + "learning_rate": 2.7214150566983807e-06, + "loss": 0.9482, + "num_input_tokens_seen": 53380800, + "step": 5337 + }, + { + "epoch": 0.4011724034270254, + "grad_norm": 1.6980031543383376, + "learning_rate": 2.7209609678523462e-06, + "loss": 1.0063, + "num_input_tokens_seen": 53401575, + "step": 5338 + }, + { + "epoch": 0.40124755749286034, + "grad_norm": 1.4660044229812828, + "learning_rate": 2.7205068362866134e-06, + "loss": 0.9998, + "num_input_tokens_seen": 53427495, + "step": 5339 + }, + { + "epoch": 0.4013227115586953, + "grad_norm": 1.2917569754419387, + "learning_rate": 2.7200526620280923e-06, + "loss": 0.9516, + "num_input_tokens_seen": 53454330, + "step": 5340 + }, + { + "epoch": 0.4013978656245303, + "grad_norm": 1.3989450396798295, + "learning_rate": 2.719598445103693e-06, + "loss": 0.8657, + "num_input_tokens_seen": 53480190, + "step": 5341 + }, + { + "epoch": 0.40147301969036525, + "grad_norm": 1.8618773603728496, + "learning_rate": 2.7191441855403304e-06, + "loss": 1.0522, + "num_input_tokens_seen": 53502225, + "step": 5342 + }, + { + "epoch": 0.4015481737562002, + "grad_norm": 1.5079548032347099, + "learning_rate": 2.718689883364922e-06, + "loss": 0.9784, + "num_input_tokens_seen": 53525630, + "step": 5343 + }, + { + "epoch": 0.4016233278220352, + "grad_norm": 2.0668477558454823, + "learning_rate": 2.7182355386043847e-06, + "loss": 0.8456, + "num_input_tokens_seen": 53548730, + "step": 5344 + }, + { + "epoch": 0.4016984818878701, + "grad_norm": 2.8246602284503113, + "learning_rate": 2.7177811512856415e-06, + "loss": 0.9443, + "num_input_tokens_seen": 53569310, + "step": 5345 + }, + { + "epoch": 0.4017736359537051, + "grad_norm": 1.7691787469672766, + "learning_rate": 2.7173267214356173e-06, + "loss": 0.939, + "num_input_tokens_seen": 53592785, + "step": 5346 + }, + { + "epoch": 0.40184879001954005, + "grad_norm": 1.5128684391174796, + "learning_rate": 2.716872249081238e-06, + "loss": 0.9653, + "num_input_tokens_seen": 53620080, + "step": 5347 + }, + { + "epoch": 0.401923944085375, + "grad_norm": 1.3376525221519717, + "learning_rate": 2.7164177342494323e-06, + "loss": 0.9816, + "num_input_tokens_seen": 53642400, + "step": 5348 + }, + { + "epoch": 0.40199909815121, + "grad_norm": 1.4492069048538831, + "learning_rate": 2.7159631769671326e-06, + "loss": 1.057, + "num_input_tokens_seen": 53662905, + "step": 5349 + }, + { + "epoch": 0.40207425221704496, + "grad_norm": 1.5266107619903897, + "learning_rate": 2.715508577261273e-06, + "loss": 0.9491, + "num_input_tokens_seen": 53687755, + "step": 5350 + }, + { + "epoch": 0.40214940628287993, + "grad_norm": 1.8056434642551458, + "learning_rate": 2.715053935158791e-06, + "loss": 0.8608, + "num_input_tokens_seen": 53710330, + "step": 5351 + }, + { + "epoch": 0.40222456034871484, + "grad_norm": 1.208135007608627, + "learning_rate": 2.7145992506866242e-06, + "loss": 0.9028, + "num_input_tokens_seen": 53777595, + "step": 5352 + }, + { + "epoch": 0.4022997144145498, + "grad_norm": 1.923889902739858, + "learning_rate": 2.714144523871716e-06, + "loss": 0.9853, + "num_input_tokens_seen": 53799220, + "step": 5353 + }, + { + "epoch": 0.4023748684803848, + "grad_norm": 1.5602016329671322, + "learning_rate": 2.7136897547410105e-06, + "loss": 1.0237, + "num_input_tokens_seen": 53819280, + "step": 5354 + }, + { + "epoch": 0.40245002254621975, + "grad_norm": 1.7196566184624629, + "learning_rate": 2.7132349433214536e-06, + "loss": 0.939, + "num_input_tokens_seen": 53844730, + "step": 5355 + }, + { + "epoch": 0.4025251766120547, + "grad_norm": 1.459121227867697, + "learning_rate": 2.712780089639995e-06, + "loss": 0.9209, + "num_input_tokens_seen": 53867660, + "step": 5356 + }, + { + "epoch": 0.4026003306778897, + "grad_norm": 1.5160815937459513, + "learning_rate": 2.7123251937235873e-06, + "loss": 0.9082, + "num_input_tokens_seen": 53891275, + "step": 5357 + }, + { + "epoch": 0.4026754847437246, + "grad_norm": 1.895429260733103, + "learning_rate": 2.7118702555991835e-06, + "loss": 0.99, + "num_input_tokens_seen": 53910715, + "step": 5358 + }, + { + "epoch": 0.4027506388095596, + "grad_norm": 1.5851358727301168, + "learning_rate": 2.7114152752937417e-06, + "loss": 0.9146, + "num_input_tokens_seen": 53936190, + "step": 5359 + }, + { + "epoch": 0.40282579287539455, + "grad_norm": 1.580494376713501, + "learning_rate": 2.71096025283422e-06, + "loss": 0.9582, + "num_input_tokens_seen": 53959045, + "step": 5360 + }, + { + "epoch": 0.4029009469412295, + "grad_norm": 1.4368086271815006, + "learning_rate": 2.7105051882475813e-06, + "loss": 0.952, + "num_input_tokens_seen": 53986530, + "step": 5361 + }, + { + "epoch": 0.4029761010070645, + "grad_norm": 1.66064567980885, + "learning_rate": 2.7100500815607898e-06, + "loss": 1.0289, + "num_input_tokens_seen": 54009015, + "step": 5362 + }, + { + "epoch": 0.40305125507289946, + "grad_norm": 1.404196362884162, + "learning_rate": 2.7095949328008113e-06, + "loss": 0.9745, + "num_input_tokens_seen": 54033905, + "step": 5363 + }, + { + "epoch": 0.40312640913873443, + "grad_norm": 1.4645715079049462, + "learning_rate": 2.7091397419946162e-06, + "loss": 0.9829, + "num_input_tokens_seen": 54057700, + "step": 5364 + }, + { + "epoch": 0.40320156320456935, + "grad_norm": 1.485109809899907, + "learning_rate": 2.708684509169176e-06, + "loss": 0.9484, + "num_input_tokens_seen": 54081085, + "step": 5365 + }, + { + "epoch": 0.4032767172704043, + "grad_norm": 1.3915525621855596, + "learning_rate": 2.7082292343514646e-06, + "loss": 0.9398, + "num_input_tokens_seen": 54104145, + "step": 5366 + }, + { + "epoch": 0.4033518713362393, + "grad_norm": 1.719137806923673, + "learning_rate": 2.707773917568459e-06, + "loss": 0.9596, + "num_input_tokens_seen": 54126980, + "step": 5367 + }, + { + "epoch": 0.40342702540207426, + "grad_norm": 1.4484545397039106, + "learning_rate": 2.707318558847139e-06, + "loss": 1.043, + "num_input_tokens_seen": 54148100, + "step": 5368 + }, + { + "epoch": 0.4035021794679092, + "grad_norm": 1.596494667462464, + "learning_rate": 2.706863158214486e-06, + "loss": 1.0089, + "num_input_tokens_seen": 54170965, + "step": 5369 + }, + { + "epoch": 0.4035773335337442, + "grad_norm": 1.6328741180620787, + "learning_rate": 2.7064077156974835e-06, + "loss": 0.8987, + "num_input_tokens_seen": 54196390, + "step": 5370 + }, + { + "epoch": 0.4036524875995791, + "grad_norm": 1.4519988587745407, + "learning_rate": 2.705952231323119e-06, + "loss": 0.8572, + "num_input_tokens_seen": 54225575, + "step": 5371 + }, + { + "epoch": 0.4037276416654141, + "grad_norm": 1.4092182622059182, + "learning_rate": 2.7054967051183813e-06, + "loss": 0.9178, + "num_input_tokens_seen": 54251600, + "step": 5372 + }, + { + "epoch": 0.40380279573124905, + "grad_norm": 1.4225053333256237, + "learning_rate": 2.705041137110263e-06, + "loss": 0.9959, + "num_input_tokens_seen": 54272855, + "step": 5373 + }, + { + "epoch": 0.403877949797084, + "grad_norm": 1.8836612925054819, + "learning_rate": 2.704585527325757e-06, + "loss": 0.9487, + "num_input_tokens_seen": 54298175, + "step": 5374 + }, + { + "epoch": 0.403953103862919, + "grad_norm": 1.9809800791463323, + "learning_rate": 2.704129875791861e-06, + "loss": 1.0391, + "num_input_tokens_seen": 54319950, + "step": 5375 + }, + { + "epoch": 0.40402825792875396, + "grad_norm": 1.5474555727446686, + "learning_rate": 2.7036741825355728e-06, + "loss": 0.9517, + "num_input_tokens_seen": 54344545, + "step": 5376 + }, + { + "epoch": 0.40410341199458893, + "grad_norm": 1.41475032706492, + "learning_rate": 2.7032184475838953e-06, + "loss": 0.9856, + "num_input_tokens_seen": 54370345, + "step": 5377 + }, + { + "epoch": 0.40417856606042385, + "grad_norm": 1.5668181450978091, + "learning_rate": 2.7027626709638317e-06, + "loss": 1.0012, + "num_input_tokens_seen": 54391410, + "step": 5378 + }, + { + "epoch": 0.4042537201262588, + "grad_norm": 2.0989974580688497, + "learning_rate": 2.702306852702389e-06, + "loss": 0.8995, + "num_input_tokens_seen": 54411700, + "step": 5379 + }, + { + "epoch": 0.4043288741920938, + "grad_norm": 1.464816243231323, + "learning_rate": 2.7018509928265763e-06, + "loss": 1.0366, + "num_input_tokens_seen": 54434385, + "step": 5380 + }, + { + "epoch": 0.40440402825792876, + "grad_norm": 1.6325373805675003, + "learning_rate": 2.7013950913634036e-06, + "loss": 0.972, + "num_input_tokens_seen": 54463160, + "step": 5381 + }, + { + "epoch": 0.40447918232376373, + "grad_norm": 1.3853670221291947, + "learning_rate": 2.7009391483398868e-06, + "loss": 1.0282, + "num_input_tokens_seen": 54485735, + "step": 5382 + }, + { + "epoch": 0.4045543363895987, + "grad_norm": 1.571450575451559, + "learning_rate": 2.7004831637830416e-06, + "loss": 0.9121, + "num_input_tokens_seen": 54508420, + "step": 5383 + }, + { + "epoch": 0.4046294904554336, + "grad_norm": 1.9259792017224717, + "learning_rate": 2.700027137719886e-06, + "loss": 0.9361, + "num_input_tokens_seen": 54529900, + "step": 5384 + }, + { + "epoch": 0.4047046445212686, + "grad_norm": 1.383254632472905, + "learning_rate": 2.699571070177442e-06, + "loss": 0.9798, + "num_input_tokens_seen": 54554160, + "step": 5385 + }, + { + "epoch": 0.40477979858710356, + "grad_norm": 2.023174136192351, + "learning_rate": 2.6991149611827335e-06, + "loss": 1.0153, + "num_input_tokens_seen": 54574175, + "step": 5386 + }, + { + "epoch": 0.4048549526529385, + "grad_norm": 1.4956739349585375, + "learning_rate": 2.6986588107627858e-06, + "loss": 1.0178, + "num_input_tokens_seen": 54599575, + "step": 5387 + }, + { + "epoch": 0.4049301067187735, + "grad_norm": 1.53082401812022, + "learning_rate": 2.698202618944629e-06, + "loss": 0.9882, + "num_input_tokens_seen": 54623065, + "step": 5388 + }, + { + "epoch": 0.40500526078460847, + "grad_norm": 1.4006358507885917, + "learning_rate": 2.697746385755293e-06, + "loss": 0.9552, + "num_input_tokens_seen": 54647040, + "step": 5389 + }, + { + "epoch": 0.4050804148504434, + "grad_norm": 1.5896313874600132, + "learning_rate": 2.6972901112218123e-06, + "loss": 1.0018, + "num_input_tokens_seen": 54668510, + "step": 5390 + }, + { + "epoch": 0.40515556891627835, + "grad_norm": 1.7595910463097089, + "learning_rate": 2.696833795371222e-06, + "loss": 0.9184, + "num_input_tokens_seen": 54689315, + "step": 5391 + }, + { + "epoch": 0.4052307229821133, + "grad_norm": 2.1824946221878547, + "learning_rate": 2.696377438230561e-06, + "loss": 0.9337, + "num_input_tokens_seen": 54712465, + "step": 5392 + }, + { + "epoch": 0.4053058770479483, + "grad_norm": 1.695344775195906, + "learning_rate": 2.6959210398268703e-06, + "loss": 0.9444, + "num_input_tokens_seen": 54736405, + "step": 5393 + }, + { + "epoch": 0.40538103111378326, + "grad_norm": 1.5532617213287232, + "learning_rate": 2.6954646001871928e-06, + "loss": 0.903, + "num_input_tokens_seen": 54762430, + "step": 5394 + }, + { + "epoch": 0.40545618517961823, + "grad_norm": 1.5184670769542372, + "learning_rate": 2.695008119338575e-06, + "loss": 1.0052, + "num_input_tokens_seen": 54784205, + "step": 5395 + }, + { + "epoch": 0.4055313392454532, + "grad_norm": 1.502871308360975, + "learning_rate": 2.6945515973080643e-06, + "loss": 0.9686, + "num_input_tokens_seen": 54807210, + "step": 5396 + }, + { + "epoch": 0.4056064933112881, + "grad_norm": 1.5404417567350426, + "learning_rate": 2.6940950341227124e-06, + "loss": 0.9849, + "num_input_tokens_seen": 54829790, + "step": 5397 + }, + { + "epoch": 0.4056816473771231, + "grad_norm": 1.4358666303909935, + "learning_rate": 2.693638429809572e-06, + "loss": 0.8818, + "num_input_tokens_seen": 54853445, + "step": 5398 + }, + { + "epoch": 0.40575680144295806, + "grad_norm": 1.793822108291418, + "learning_rate": 2.6931817843956977e-06, + "loss": 1.0359, + "num_input_tokens_seen": 54875330, + "step": 5399 + }, + { + "epoch": 0.40583195550879303, + "grad_norm": 1.467687100468278, + "learning_rate": 2.692725097908149e-06, + "loss": 0.8417, + "num_input_tokens_seen": 54903315, + "step": 5400 + }, + { + "epoch": 0.405907109574628, + "grad_norm": 1.5131822544817, + "learning_rate": 2.692268370373985e-06, + "loss": 0.9062, + "num_input_tokens_seen": 54927760, + "step": 5401 + }, + { + "epoch": 0.40598226364046297, + "grad_norm": 1.3634858510731456, + "learning_rate": 2.69181160182027e-06, + "loss": 1.0718, + "num_input_tokens_seen": 54953110, + "step": 5402 + }, + { + "epoch": 0.4060574177062979, + "grad_norm": 1.3443381774342764, + "learning_rate": 2.691354792274068e-06, + "loss": 0.9646, + "num_input_tokens_seen": 54977885, + "step": 5403 + }, + { + "epoch": 0.40613257177213286, + "grad_norm": 1.4125143607430757, + "learning_rate": 2.690897941762447e-06, + "loss": 0.967, + "num_input_tokens_seen": 55001830, + "step": 5404 + }, + { + "epoch": 0.4062077258379678, + "grad_norm": 1.279228942465187, + "learning_rate": 2.6904410503124774e-06, + "loss": 0.9439, + "num_input_tokens_seen": 55030440, + "step": 5405 + }, + { + "epoch": 0.4062828799038028, + "grad_norm": 1.3825354864384483, + "learning_rate": 2.6899841179512324e-06, + "loss": 0.8879, + "num_input_tokens_seen": 55058390, + "step": 5406 + }, + { + "epoch": 0.40635803396963777, + "grad_norm": 1.6245640389972253, + "learning_rate": 2.689527144705785e-06, + "loss": 0.9096, + "num_input_tokens_seen": 55082875, + "step": 5407 + }, + { + "epoch": 0.40643318803547274, + "grad_norm": 1.3593940733525194, + "learning_rate": 2.6890701306032154e-06, + "loss": 0.8751, + "num_input_tokens_seen": 55109850, + "step": 5408 + }, + { + "epoch": 0.4065083421013077, + "grad_norm": 1.6307193901608, + "learning_rate": 2.6886130756706003e-06, + "loss": 0.9989, + "num_input_tokens_seen": 55131625, + "step": 5409 + }, + { + "epoch": 0.4065834961671426, + "grad_norm": 1.9351933192004933, + "learning_rate": 2.688155979935025e-06, + "loss": 0.9235, + "num_input_tokens_seen": 55153895, + "step": 5410 + }, + { + "epoch": 0.4066586502329776, + "grad_norm": 1.344213405812677, + "learning_rate": 2.687698843423572e-06, + "loss": 0.9002, + "num_input_tokens_seen": 55175635, + "step": 5411 + }, + { + "epoch": 0.40673380429881256, + "grad_norm": 1.5392788187971926, + "learning_rate": 2.6872416661633296e-06, + "loss": 0.9258, + "num_input_tokens_seen": 55199300, + "step": 5412 + }, + { + "epoch": 0.40680895836464753, + "grad_norm": 1.4240826247172194, + "learning_rate": 2.6867844481813868e-06, + "loss": 0.9879, + "num_input_tokens_seen": 55221000, + "step": 5413 + }, + { + "epoch": 0.4068841124304825, + "grad_norm": 1.761407795125987, + "learning_rate": 2.6863271895048353e-06, + "loss": 0.9332, + "num_input_tokens_seen": 55244835, + "step": 5414 + }, + { + "epoch": 0.4069592664963175, + "grad_norm": 1.6543061707650943, + "learning_rate": 2.6858698901607696e-06, + "loss": 0.9852, + "num_input_tokens_seen": 55267520, + "step": 5415 + }, + { + "epoch": 0.4070344205621524, + "grad_norm": 1.497581757875341, + "learning_rate": 2.6854125501762863e-06, + "loss": 0.9132, + "num_input_tokens_seen": 55291545, + "step": 5416 + }, + { + "epoch": 0.40710957462798736, + "grad_norm": 1.86416202097237, + "learning_rate": 2.684955169578486e-06, + "loss": 0.9309, + "num_input_tokens_seen": 55313110, + "step": 5417 + }, + { + "epoch": 0.40718472869382233, + "grad_norm": 1.587513408454078, + "learning_rate": 2.684497748394468e-06, + "loss": 0.9814, + "num_input_tokens_seen": 55338700, + "step": 5418 + }, + { + "epoch": 0.4072598827596573, + "grad_norm": 1.9572572254062646, + "learning_rate": 2.6840402866513377e-06, + "loss": 0.9066, + "num_input_tokens_seen": 55364355, + "step": 5419 + }, + { + "epoch": 0.40733503682549227, + "grad_norm": 1.3427085389146236, + "learning_rate": 2.6835827843762006e-06, + "loss": 1.0304, + "num_input_tokens_seen": 55386590, + "step": 5420 + }, + { + "epoch": 0.40741019089132724, + "grad_norm": 1.6255894838858422, + "learning_rate": 2.6831252415961665e-06, + "loss": 0.9731, + "num_input_tokens_seen": 55408010, + "step": 5421 + }, + { + "epoch": 0.4074853449571622, + "grad_norm": 1.656801168033551, + "learning_rate": 2.682667658338345e-06, + "loss": 0.9611, + "num_input_tokens_seen": 55431015, + "step": 5422 + }, + { + "epoch": 0.4075604990229971, + "grad_norm": 1.658832767494269, + "learning_rate": 2.6822100346298517e-06, + "loss": 0.928, + "num_input_tokens_seen": 55457675, + "step": 5423 + }, + { + "epoch": 0.4076356530888321, + "grad_norm": 1.4299160300456693, + "learning_rate": 2.6817523704978014e-06, + "loss": 1.0548, + "num_input_tokens_seen": 55481190, + "step": 5424 + }, + { + "epoch": 0.40771080715466707, + "grad_norm": 1.9012975302830113, + "learning_rate": 2.681294665969312e-06, + "loss": 0.8997, + "num_input_tokens_seen": 55504375, + "step": 5425 + }, + { + "epoch": 0.40778596122050204, + "grad_norm": 1.3916467063675422, + "learning_rate": 2.6808369210715055e-06, + "loss": 0.9515, + "num_input_tokens_seen": 55531035, + "step": 5426 + }, + { + "epoch": 0.407861115286337, + "grad_norm": 1.7331147985367925, + "learning_rate": 2.6803791358315035e-06, + "loss": 0.9339, + "num_input_tokens_seen": 55551435, + "step": 5427 + }, + { + "epoch": 0.407936269352172, + "grad_norm": 1.3150645133609025, + "learning_rate": 2.679921310276432e-06, + "loss": 1.0303, + "num_input_tokens_seen": 55575690, + "step": 5428 + }, + { + "epoch": 0.4080114234180069, + "grad_norm": 1.4422833015140166, + "learning_rate": 2.6794634444334203e-06, + "loss": 0.8031, + "num_input_tokens_seen": 55602055, + "step": 5429 + }, + { + "epoch": 0.40808657748384186, + "grad_norm": 1.72243486223123, + "learning_rate": 2.679005538329598e-06, + "loss": 0.9808, + "num_input_tokens_seen": 55624805, + "step": 5430 + }, + { + "epoch": 0.40816173154967683, + "grad_norm": 2.319233447571045, + "learning_rate": 2.678547591992096e-06, + "loss": 0.9973, + "num_input_tokens_seen": 55647950, + "step": 5431 + }, + { + "epoch": 0.4082368856155118, + "grad_norm": 1.4309958219753776, + "learning_rate": 2.6780896054480526e-06, + "loss": 0.8084, + "num_input_tokens_seen": 55671285, + "step": 5432 + }, + { + "epoch": 0.4083120396813468, + "grad_norm": 1.503399255678846, + "learning_rate": 2.6776315787246024e-06, + "loss": 1.0191, + "num_input_tokens_seen": 55696270, + "step": 5433 + }, + { + "epoch": 0.40838719374718174, + "grad_norm": 0.9705457041639008, + "learning_rate": 2.6771735118488864e-06, + "loss": 0.8765, + "num_input_tokens_seen": 55778190, + "step": 5434 + }, + { + "epoch": 0.40846234781301666, + "grad_norm": 2.4923417094482194, + "learning_rate": 2.676715404848047e-06, + "loss": 1.0407, + "num_input_tokens_seen": 55801440, + "step": 5435 + }, + { + "epoch": 0.4085375018788516, + "grad_norm": 2.1447764883543363, + "learning_rate": 2.676257257749228e-06, + "loss": 0.9866, + "num_input_tokens_seen": 55818355, + "step": 5436 + }, + { + "epoch": 0.4086126559446866, + "grad_norm": 1.7996437965381746, + "learning_rate": 2.6757990705795777e-06, + "loss": 0.9669, + "num_input_tokens_seen": 55841215, + "step": 5437 + }, + { + "epoch": 0.40868781001052157, + "grad_norm": 1.5277248989934, + "learning_rate": 2.675340843366244e-06, + "loss": 1.0266, + "num_input_tokens_seen": 55862925, + "step": 5438 + }, + { + "epoch": 0.40876296407635654, + "grad_norm": 2.073678213834946, + "learning_rate": 2.6748825761363794e-06, + "loss": 0.8512, + "num_input_tokens_seen": 55884170, + "step": 5439 + }, + { + "epoch": 0.4088381181421915, + "grad_norm": 1.7763106812669502, + "learning_rate": 2.674424268917138e-06, + "loss": 1.0073, + "num_input_tokens_seen": 55906330, + "step": 5440 + }, + { + "epoch": 0.4089132722080265, + "grad_norm": 1.5217434585403118, + "learning_rate": 2.6739659217356766e-06, + "loss": 0.985, + "num_input_tokens_seen": 55928110, + "step": 5441 + }, + { + "epoch": 0.4089884262738614, + "grad_norm": 1.4350799423971323, + "learning_rate": 2.6735075346191526e-06, + "loss": 0.939, + "num_input_tokens_seen": 55952275, + "step": 5442 + }, + { + "epoch": 0.40906358033969636, + "grad_norm": 0.8202894661712183, + "learning_rate": 2.6730491075947294e-06, + "loss": 0.8069, + "num_input_tokens_seen": 56032510, + "step": 5443 + }, + { + "epoch": 0.40913873440553133, + "grad_norm": 1.7277928452211262, + "learning_rate": 2.672590640689568e-06, + "loss": 0.9369, + "num_input_tokens_seen": 56056210, + "step": 5444 + }, + { + "epoch": 0.4092138884713663, + "grad_norm": 1.5309773416468273, + "learning_rate": 2.6721321339308365e-06, + "loss": 0.9427, + "num_input_tokens_seen": 56080825, + "step": 5445 + }, + { + "epoch": 0.4092890425372013, + "grad_norm": 0.7916918842582462, + "learning_rate": 2.671673587345702e-06, + "loss": 0.7649, + "num_input_tokens_seen": 56164515, + "step": 5446 + }, + { + "epoch": 0.40936419660303625, + "grad_norm": 1.459558499090758, + "learning_rate": 2.671215000961335e-06, + "loss": 0.9907, + "num_input_tokens_seen": 56186755, + "step": 5447 + }, + { + "epoch": 0.40943935066887116, + "grad_norm": 1.7436346364570994, + "learning_rate": 2.6707563748049094e-06, + "loss": 1.0136, + "num_input_tokens_seen": 56208400, + "step": 5448 + }, + { + "epoch": 0.40951450473470613, + "grad_norm": 0.8614611879706605, + "learning_rate": 2.6702977089036e-06, + "loss": 0.7876, + "num_input_tokens_seen": 56282105, + "step": 5449 + }, + { + "epoch": 0.4095896588005411, + "grad_norm": 1.6915517836900256, + "learning_rate": 2.6698390032845844e-06, + "loss": 1.0585, + "num_input_tokens_seen": 56304615, + "step": 5450 + }, + { + "epoch": 0.40966481286637607, + "grad_norm": 1.788259734244713, + "learning_rate": 2.6693802579750434e-06, + "loss": 0.8865, + "num_input_tokens_seen": 56327745, + "step": 5451 + }, + { + "epoch": 0.40973996693221104, + "grad_norm": 1.7066557796448916, + "learning_rate": 2.668921473002159e-06, + "loss": 0.9729, + "num_input_tokens_seen": 56350605, + "step": 5452 + }, + { + "epoch": 0.409815120998046, + "grad_norm": 1.5730431950016708, + "learning_rate": 2.668462648393115e-06, + "loss": 0.9311, + "num_input_tokens_seen": 56374665, + "step": 5453 + }, + { + "epoch": 0.409890275063881, + "grad_norm": 1.7800877443396486, + "learning_rate": 2.6680037841751e-06, + "loss": 1.0276, + "num_input_tokens_seen": 56394420, + "step": 5454 + }, + { + "epoch": 0.4099654291297159, + "grad_norm": 1.4247326482976959, + "learning_rate": 2.6675448803753026e-06, + "loss": 0.9035, + "num_input_tokens_seen": 56423280, + "step": 5455 + }, + { + "epoch": 0.41004058319555087, + "grad_norm": 1.7399665121166423, + "learning_rate": 2.667085937020915e-06, + "loss": 0.979, + "num_input_tokens_seen": 56446590, + "step": 5456 + }, + { + "epoch": 0.41011573726138584, + "grad_norm": 1.040313025554915, + "learning_rate": 2.6666269541391313e-06, + "loss": 0.9203, + "num_input_tokens_seen": 56517745, + "step": 5457 + }, + { + "epoch": 0.4101908913272208, + "grad_norm": 1.5044714720110897, + "learning_rate": 2.6661679317571473e-06, + "loss": 0.9781, + "num_input_tokens_seen": 56540400, + "step": 5458 + }, + { + "epoch": 0.4102660453930558, + "grad_norm": 1.5516084631558993, + "learning_rate": 2.665708869902163e-06, + "loss": 1.0117, + "num_input_tokens_seen": 56561315, + "step": 5459 + }, + { + "epoch": 0.41034119945889075, + "grad_norm": 1.6827201258277755, + "learning_rate": 2.6652497686013786e-06, + "loss": 0.9501, + "num_input_tokens_seen": 56584815, + "step": 5460 + }, + { + "epoch": 0.41041635352472566, + "grad_norm": 1.3400309796097185, + "learning_rate": 2.664790627881998e-06, + "loss": 0.9388, + "num_input_tokens_seen": 56608780, + "step": 5461 + }, + { + "epoch": 0.41049150759056063, + "grad_norm": 1.4967795610364767, + "learning_rate": 2.664331447771227e-06, + "loss": 1.0215, + "num_input_tokens_seen": 56631060, + "step": 5462 + }, + { + "epoch": 0.4105666616563956, + "grad_norm": 1.7677577817883428, + "learning_rate": 2.663872228296275e-06, + "loss": 0.8912, + "num_input_tokens_seen": 56653900, + "step": 5463 + }, + { + "epoch": 0.4106418157222306, + "grad_norm": 0.8086337019776006, + "learning_rate": 2.6634129694843497e-06, + "loss": 0.7909, + "num_input_tokens_seen": 56723500, + "step": 5464 + }, + { + "epoch": 0.41071696978806554, + "grad_norm": 1.6305580454685855, + "learning_rate": 2.6629536713626664e-06, + "loss": 1.0338, + "num_input_tokens_seen": 56744365, + "step": 5465 + }, + { + "epoch": 0.4107921238539005, + "grad_norm": 1.3880052213763, + "learning_rate": 2.662494333958439e-06, + "loss": 0.9796, + "num_input_tokens_seen": 56771135, + "step": 5466 + }, + { + "epoch": 0.4108672779197355, + "grad_norm": 1.3924512123176875, + "learning_rate": 2.662034957298886e-06, + "loss": 0.949, + "num_input_tokens_seen": 56795990, + "step": 5467 + }, + { + "epoch": 0.4109424319855704, + "grad_norm": 1.6739921355934402, + "learning_rate": 2.6615755414112266e-06, + "loss": 0.874, + "num_input_tokens_seen": 56820705, + "step": 5468 + }, + { + "epoch": 0.41101758605140537, + "grad_norm": 1.6225969446077078, + "learning_rate": 2.6611160863226826e-06, + "loss": 0.8803, + "num_input_tokens_seen": 56841215, + "step": 5469 + }, + { + "epoch": 0.41109274011724034, + "grad_norm": 1.379333602774956, + "learning_rate": 2.6606565920604793e-06, + "loss": 0.9154, + "num_input_tokens_seen": 56863920, + "step": 5470 + }, + { + "epoch": 0.4111678941830753, + "grad_norm": 1.4901334995910245, + "learning_rate": 2.6601970586518428e-06, + "loss": 0.8697, + "num_input_tokens_seen": 56886470, + "step": 5471 + }, + { + "epoch": 0.4112430482489103, + "grad_norm": 1.5069594267741522, + "learning_rate": 2.6597374861240026e-06, + "loss": 0.9386, + "num_input_tokens_seen": 56912025, + "step": 5472 + }, + { + "epoch": 0.41131820231474525, + "grad_norm": 1.8713365369419843, + "learning_rate": 2.65927787450419e-06, + "loss": 0.9838, + "num_input_tokens_seen": 56932605, + "step": 5473 + }, + { + "epoch": 0.41139335638058017, + "grad_norm": 0.9604343865029421, + "learning_rate": 2.6588182238196395e-06, + "loss": 0.8394, + "num_input_tokens_seen": 57008830, + "step": 5474 + }, + { + "epoch": 0.41146851044641514, + "grad_norm": 1.4569544222215005, + "learning_rate": 2.6583585340975854e-06, + "loss": 0.9103, + "num_input_tokens_seen": 57032870, + "step": 5475 + }, + { + "epoch": 0.4115436645122501, + "grad_norm": 1.3833399914194455, + "learning_rate": 2.657898805365268e-06, + "loss": 0.9905, + "num_input_tokens_seen": 57059480, + "step": 5476 + }, + { + "epoch": 0.4116188185780851, + "grad_norm": 1.5188537499425676, + "learning_rate": 2.6574390376499265e-06, + "loss": 1.0199, + "num_input_tokens_seen": 57082205, + "step": 5477 + }, + { + "epoch": 0.41169397264392005, + "grad_norm": 2.7043576820124073, + "learning_rate": 2.6569792309788046e-06, + "loss": 0.8132, + "num_input_tokens_seen": 57110715, + "step": 5478 + }, + { + "epoch": 0.411769126709755, + "grad_norm": 1.6432399437052707, + "learning_rate": 2.656519385379148e-06, + "loss": 0.9302, + "num_input_tokens_seen": 57131650, + "step": 5479 + }, + { + "epoch": 0.41184428077558993, + "grad_norm": 1.5643282303503694, + "learning_rate": 2.6560595008782032e-06, + "loss": 0.9373, + "num_input_tokens_seen": 57154360, + "step": 5480 + }, + { + "epoch": 0.4119194348414249, + "grad_norm": 2.3050379599798876, + "learning_rate": 2.655599577503221e-06, + "loss": 0.9695, + "num_input_tokens_seen": 57181820, + "step": 5481 + }, + { + "epoch": 0.4119945889072599, + "grad_norm": 1.6432298598814488, + "learning_rate": 2.6551396152814534e-06, + "loss": 0.9865, + "num_input_tokens_seen": 57205070, + "step": 5482 + }, + { + "epoch": 0.41206974297309484, + "grad_norm": 1.9714360155040078, + "learning_rate": 2.6546796142401547e-06, + "loss": 0.9773, + "num_input_tokens_seen": 57221290, + "step": 5483 + }, + { + "epoch": 0.4121448970389298, + "grad_norm": 0.799783485140882, + "learning_rate": 2.6542195744065826e-06, + "loss": 0.7662, + "num_input_tokens_seen": 57295930, + "step": 5484 + }, + { + "epoch": 0.4122200511047648, + "grad_norm": 1.4679625815058415, + "learning_rate": 2.653759495807995e-06, + "loss": 1.0035, + "num_input_tokens_seen": 57318435, + "step": 5485 + }, + { + "epoch": 0.41229520517059975, + "grad_norm": 0.8009867334731657, + "learning_rate": 2.6532993784716535e-06, + "loss": 0.8476, + "num_input_tokens_seen": 57392600, + "step": 5486 + }, + { + "epoch": 0.41237035923643467, + "grad_norm": 1.674959563720897, + "learning_rate": 2.652839222424823e-06, + "loss": 1.0223, + "num_input_tokens_seen": 57415475, + "step": 5487 + }, + { + "epoch": 0.41244551330226964, + "grad_norm": 1.270689263020475, + "learning_rate": 2.652379027694768e-06, + "loss": 0.9472, + "num_input_tokens_seen": 57441715, + "step": 5488 + }, + { + "epoch": 0.4125206673681046, + "grad_norm": 1.3686360559981152, + "learning_rate": 2.651918794308758e-06, + "loss": 0.8748, + "num_input_tokens_seen": 57467615, + "step": 5489 + }, + { + "epoch": 0.4125958214339396, + "grad_norm": 1.728042175625985, + "learning_rate": 2.651458522294063e-06, + "loss": 1.0175, + "num_input_tokens_seen": 57489120, + "step": 5490 + }, + { + "epoch": 0.41267097549977455, + "grad_norm": 2.8243950120048527, + "learning_rate": 2.650998211677956e-06, + "loss": 0.9759, + "num_input_tokens_seen": 57512340, + "step": 5491 + }, + { + "epoch": 0.4127461295656095, + "grad_norm": 1.5390629647346221, + "learning_rate": 2.6505378624877116e-06, + "loss": 0.9993, + "num_input_tokens_seen": 57535465, + "step": 5492 + }, + { + "epoch": 0.41282128363144444, + "grad_norm": 1.8452897835156121, + "learning_rate": 2.650077474750608e-06, + "loss": 0.9333, + "num_input_tokens_seen": 57559210, + "step": 5493 + }, + { + "epoch": 0.4128964376972794, + "grad_norm": 0.9212181207945935, + "learning_rate": 2.649617048493925e-06, + "loss": 0.8899, + "num_input_tokens_seen": 57642335, + "step": 5494 + }, + { + "epoch": 0.4129715917631144, + "grad_norm": 1.621213609791442, + "learning_rate": 2.649156583744944e-06, + "loss": 0.9462, + "num_input_tokens_seen": 57665020, + "step": 5495 + }, + { + "epoch": 0.41304674582894935, + "grad_norm": 1.6262228839323034, + "learning_rate": 2.64869608053095e-06, + "loss": 0.8976, + "num_input_tokens_seen": 57687810, + "step": 5496 + }, + { + "epoch": 0.4131218998947843, + "grad_norm": 1.7214975763814078, + "learning_rate": 2.648235538879229e-06, + "loss": 0.8929, + "num_input_tokens_seen": 57710700, + "step": 5497 + }, + { + "epoch": 0.4131970539606193, + "grad_norm": 1.3611298995548584, + "learning_rate": 2.6477749588170703e-06, + "loss": 0.9786, + "num_input_tokens_seen": 57735925, + "step": 5498 + }, + { + "epoch": 0.41327220802645426, + "grad_norm": 1.4843745181434251, + "learning_rate": 2.647314340371764e-06, + "loss": 0.9949, + "num_input_tokens_seen": 57757535, + "step": 5499 + }, + { + "epoch": 0.4133473620922892, + "grad_norm": 1.6472265487208093, + "learning_rate": 2.646853683570605e-06, + "loss": 0.854, + "num_input_tokens_seen": 57782155, + "step": 5500 + }, + { + "epoch": 0.41342251615812414, + "grad_norm": 1.808766828457384, + "learning_rate": 2.646392988440888e-06, + "loss": 1.0034, + "num_input_tokens_seen": 57804655, + "step": 5501 + }, + { + "epoch": 0.4134976702239591, + "grad_norm": 1.7389741695212044, + "learning_rate": 2.6459322550099113e-06, + "loss": 0.9314, + "num_input_tokens_seen": 57825930, + "step": 5502 + }, + { + "epoch": 0.4135728242897941, + "grad_norm": 1.5201486936419246, + "learning_rate": 2.645471483304975e-06, + "loss": 0.9679, + "num_input_tokens_seen": 57851390, + "step": 5503 + }, + { + "epoch": 0.41364797835562905, + "grad_norm": 1.6313249300413046, + "learning_rate": 2.645010673353382e-06, + "loss": 1.03, + "num_input_tokens_seen": 57873520, + "step": 5504 + }, + { + "epoch": 0.413723132421464, + "grad_norm": 1.7331436184840165, + "learning_rate": 2.644549825182436e-06, + "loss": 0.986, + "num_input_tokens_seen": 57894135, + "step": 5505 + }, + { + "epoch": 0.41379828648729894, + "grad_norm": 1.6606343377553952, + "learning_rate": 2.644088938819445e-06, + "loss": 0.9632, + "num_input_tokens_seen": 57916315, + "step": 5506 + }, + { + "epoch": 0.4138734405531339, + "grad_norm": 1.7171981741770437, + "learning_rate": 2.6436280142917183e-06, + "loss": 0.9607, + "num_input_tokens_seen": 57940655, + "step": 5507 + }, + { + "epoch": 0.4139485946189689, + "grad_norm": 1.6750790563117937, + "learning_rate": 2.6431670516265668e-06, + "loss": 0.9172, + "num_input_tokens_seen": 57960335, + "step": 5508 + }, + { + "epoch": 0.41402374868480385, + "grad_norm": 1.5288657485673274, + "learning_rate": 2.6427060508513052e-06, + "loss": 0.8547, + "num_input_tokens_seen": 57984680, + "step": 5509 + }, + { + "epoch": 0.4140989027506388, + "grad_norm": 1.4542155121885796, + "learning_rate": 2.6422450119932484e-06, + "loss": 0.9835, + "num_input_tokens_seen": 58007985, + "step": 5510 + }, + { + "epoch": 0.4141740568164738, + "grad_norm": 1.9033831942697497, + "learning_rate": 2.641783935079716e-06, + "loss": 0.9275, + "num_input_tokens_seen": 58032025, + "step": 5511 + }, + { + "epoch": 0.41424921088230876, + "grad_norm": 1.338347196986188, + "learning_rate": 2.641322820138027e-06, + "loss": 0.9027, + "num_input_tokens_seen": 58056415, + "step": 5512 + }, + { + "epoch": 0.4143243649481437, + "grad_norm": 1.6111077915172067, + "learning_rate": 2.6408616671955053e-06, + "loss": 0.9234, + "num_input_tokens_seen": 58081470, + "step": 5513 + }, + { + "epoch": 0.41439951901397865, + "grad_norm": 2.6853061859472147, + "learning_rate": 2.6404004762794766e-06, + "loss": 0.9472, + "num_input_tokens_seen": 58104650, + "step": 5514 + }, + { + "epoch": 0.4144746730798136, + "grad_norm": 0.8102129512651011, + "learning_rate": 2.6399392474172667e-06, + "loss": 0.7733, + "num_input_tokens_seen": 58184975, + "step": 5515 + }, + { + "epoch": 0.4145498271456486, + "grad_norm": 0.8899036966683669, + "learning_rate": 2.6394779806362057e-06, + "loss": 0.8112, + "num_input_tokens_seen": 58265245, + "step": 5516 + }, + { + "epoch": 0.41462498121148356, + "grad_norm": 1.373324413656385, + "learning_rate": 2.6390166759636263e-06, + "loss": 0.9263, + "num_input_tokens_seen": 58288530, + "step": 5517 + }, + { + "epoch": 0.4147001352773185, + "grad_norm": 0.9184369455642818, + "learning_rate": 2.638555333426862e-06, + "loss": 0.9252, + "num_input_tokens_seen": 58357460, + "step": 5518 + }, + { + "epoch": 0.41477528934315344, + "grad_norm": 1.5378510795394253, + "learning_rate": 2.638093953053248e-06, + "loss": 0.9689, + "num_input_tokens_seen": 58378430, + "step": 5519 + }, + { + "epoch": 0.4148504434089884, + "grad_norm": 1.4839901073773858, + "learning_rate": 2.6376325348701244e-06, + "loss": 0.9976, + "num_input_tokens_seen": 58403715, + "step": 5520 + }, + { + "epoch": 0.4149255974748234, + "grad_norm": 1.597211482062603, + "learning_rate": 2.6371710789048313e-06, + "loss": 1.0163, + "num_input_tokens_seen": 58427255, + "step": 5521 + }, + { + "epoch": 0.41500075154065835, + "grad_norm": 1.4966451161765184, + "learning_rate": 2.6367095851847125e-06, + "loss": 0.899, + "num_input_tokens_seen": 58452810, + "step": 5522 + }, + { + "epoch": 0.4150759056064933, + "grad_norm": 1.3396311777607257, + "learning_rate": 2.636248053737112e-06, + "loss": 0.9267, + "num_input_tokens_seen": 58475620, + "step": 5523 + }, + { + "epoch": 0.4151510596723283, + "grad_norm": 1.5808804317624463, + "learning_rate": 2.635786484589378e-06, + "loss": 0.9967, + "num_input_tokens_seen": 58498200, + "step": 5524 + }, + { + "epoch": 0.4152262137381632, + "grad_norm": 2.4442589778573347, + "learning_rate": 2.6353248777688606e-06, + "loss": 0.9604, + "num_input_tokens_seen": 58521815, + "step": 5525 + }, + { + "epoch": 0.4153013678039982, + "grad_norm": 1.431102448566222, + "learning_rate": 2.634863233302911e-06, + "loss": 0.8586, + "num_input_tokens_seen": 58549965, + "step": 5526 + }, + { + "epoch": 0.41537652186983315, + "grad_norm": 1.6611671152671457, + "learning_rate": 2.634401551218884e-06, + "loss": 0.9058, + "num_input_tokens_seen": 58570370, + "step": 5527 + }, + { + "epoch": 0.4154516759356681, + "grad_norm": 1.8300045816958246, + "learning_rate": 2.6339398315441353e-06, + "loss": 1.0365, + "num_input_tokens_seen": 58592145, + "step": 5528 + }, + { + "epoch": 0.4155268300015031, + "grad_norm": 1.3306109331712457, + "learning_rate": 2.633478074306025e-06, + "loss": 0.9627, + "num_input_tokens_seen": 58615890, + "step": 5529 + }, + { + "epoch": 0.41560198406733806, + "grad_norm": 1.535708107413097, + "learning_rate": 2.633016279531912e-06, + "loss": 0.9178, + "num_input_tokens_seen": 58637130, + "step": 5530 + }, + { + "epoch": 0.41567713813317303, + "grad_norm": 1.8227391256432603, + "learning_rate": 2.6325544472491616e-06, + "loss": 0.864, + "num_input_tokens_seen": 58661010, + "step": 5531 + }, + { + "epoch": 0.41575229219900794, + "grad_norm": 0.7329332118058146, + "learning_rate": 2.632092577485137e-06, + "loss": 0.7907, + "num_input_tokens_seen": 58748325, + "step": 5532 + }, + { + "epoch": 0.4158274462648429, + "grad_norm": 2.2770938376288825, + "learning_rate": 2.631630670267207e-06, + "loss": 0.9052, + "num_input_tokens_seen": 58773130, + "step": 5533 + }, + { + "epoch": 0.4159026003306779, + "grad_norm": 1.62539660675701, + "learning_rate": 2.631168725622742e-06, + "loss": 0.9834, + "num_input_tokens_seen": 58794980, + "step": 5534 + }, + { + "epoch": 0.41597775439651286, + "grad_norm": 1.8410167801738233, + "learning_rate": 2.630706743579112e-06, + "loss": 1.0418, + "num_input_tokens_seen": 58817450, + "step": 5535 + }, + { + "epoch": 0.4160529084623478, + "grad_norm": 1.7609991054370173, + "learning_rate": 2.6302447241636924e-06, + "loss": 0.8745, + "num_input_tokens_seen": 58839405, + "step": 5536 + }, + { + "epoch": 0.4161280625281828, + "grad_norm": 1.795825220371457, + "learning_rate": 2.6297826674038595e-06, + "loss": 1.0496, + "num_input_tokens_seen": 58860735, + "step": 5537 + }, + { + "epoch": 0.4162032165940177, + "grad_norm": 1.5162279954840403, + "learning_rate": 2.6293205733269924e-06, + "loss": 0.8352, + "num_input_tokens_seen": 58885235, + "step": 5538 + }, + { + "epoch": 0.4162783706598527, + "grad_norm": 1.5095130146497893, + "learning_rate": 2.6288584419604713e-06, + "loss": 1.0232, + "num_input_tokens_seen": 58905520, + "step": 5539 + }, + { + "epoch": 0.41635352472568765, + "grad_norm": 1.2957704735297109, + "learning_rate": 2.62839627333168e-06, + "loss": 0.99, + "num_input_tokens_seen": 58929535, + "step": 5540 + }, + { + "epoch": 0.4164286787915226, + "grad_norm": 0.8364818975059736, + "learning_rate": 2.6279340674680025e-06, + "loss": 0.8315, + "num_input_tokens_seen": 59014405, + "step": 5541 + }, + { + "epoch": 0.4165038328573576, + "grad_norm": 1.4537284992494401, + "learning_rate": 2.627471824396827e-06, + "loss": 0.9674, + "num_input_tokens_seen": 59037510, + "step": 5542 + }, + { + "epoch": 0.41657898692319256, + "grad_norm": 1.4962614677582495, + "learning_rate": 2.6270095441455435e-06, + "loss": 0.9271, + "num_input_tokens_seen": 59064440, + "step": 5543 + }, + { + "epoch": 0.41665414098902753, + "grad_norm": 0.7896606897257019, + "learning_rate": 2.6265472267415432e-06, + "loss": 0.7981, + "num_input_tokens_seen": 59145990, + "step": 5544 + }, + { + "epoch": 0.41672929505486245, + "grad_norm": 1.6116469559093713, + "learning_rate": 2.626084872212221e-06, + "loss": 0.9767, + "num_input_tokens_seen": 59168275, + "step": 5545 + }, + { + "epoch": 0.4168044491206974, + "grad_norm": 1.5166686878959406, + "learning_rate": 2.625622480584972e-06, + "loss": 0.9937, + "num_input_tokens_seen": 59191935, + "step": 5546 + }, + { + "epoch": 0.4168796031865324, + "grad_norm": 1.7620129392497366, + "learning_rate": 2.6251600518871953e-06, + "loss": 1.0541, + "num_input_tokens_seen": 59213785, + "step": 5547 + }, + { + "epoch": 0.41695475725236736, + "grad_norm": 1.569991501220248, + "learning_rate": 2.6246975861462927e-06, + "loss": 0.9292, + "num_input_tokens_seen": 59236465, + "step": 5548 + }, + { + "epoch": 0.41702991131820233, + "grad_norm": 1.1805264350939748, + "learning_rate": 2.6242350833896645e-06, + "loss": 0.9767, + "num_input_tokens_seen": 59262405, + "step": 5549 + }, + { + "epoch": 0.4171050653840373, + "grad_norm": 1.7970870846518228, + "learning_rate": 2.623772543644718e-06, + "loss": 0.837, + "num_input_tokens_seen": 59283260, + "step": 5550 + }, + { + "epoch": 0.4171802194498722, + "grad_norm": 1.1735603928501097, + "learning_rate": 2.6233099669388605e-06, + "loss": 0.8946, + "num_input_tokens_seen": 59313045, + "step": 5551 + }, + { + "epoch": 0.4172553735157072, + "grad_norm": 1.5720782624435976, + "learning_rate": 2.6228473532995e-06, + "loss": 1.0, + "num_input_tokens_seen": 59334245, + "step": 5552 + }, + { + "epoch": 0.41733052758154215, + "grad_norm": 1.509090297219402, + "learning_rate": 2.6223847027540485e-06, + "loss": 0.8507, + "num_input_tokens_seen": 59359805, + "step": 5553 + }, + { + "epoch": 0.4174056816473771, + "grad_norm": 1.4674312271301888, + "learning_rate": 2.62192201532992e-06, + "loss": 0.9439, + "num_input_tokens_seen": 59383575, + "step": 5554 + }, + { + "epoch": 0.4174808357132121, + "grad_norm": 1.4183766852782593, + "learning_rate": 2.621459291054531e-06, + "loss": 0.8285, + "num_input_tokens_seen": 59404990, + "step": 5555 + }, + { + "epoch": 0.41755598977904707, + "grad_norm": 1.8550320723812266, + "learning_rate": 2.6209965299552994e-06, + "loss": 1.0041, + "num_input_tokens_seen": 59425945, + "step": 5556 + }, + { + "epoch": 0.41763114384488204, + "grad_norm": 1.313010389000727, + "learning_rate": 2.6205337320596452e-06, + "loss": 0.9895, + "num_input_tokens_seen": 59448565, + "step": 5557 + }, + { + "epoch": 0.41770629791071695, + "grad_norm": 1.6366967538889645, + "learning_rate": 2.620070897394991e-06, + "loss": 0.9993, + "num_input_tokens_seen": 59469780, + "step": 5558 + }, + { + "epoch": 0.4177814519765519, + "grad_norm": 1.6640049024473011, + "learning_rate": 2.619608025988762e-06, + "loss": 0.8849, + "num_input_tokens_seen": 59493910, + "step": 5559 + }, + { + "epoch": 0.4178566060423869, + "grad_norm": 1.2898264615190123, + "learning_rate": 2.6191451178683842e-06, + "loss": 0.8894, + "num_input_tokens_seen": 59519930, + "step": 5560 + }, + { + "epoch": 0.41793176010822186, + "grad_norm": 0.8126735135184562, + "learning_rate": 2.6186821730612884e-06, + "loss": 0.7854, + "num_input_tokens_seen": 59593940, + "step": 5561 + }, + { + "epoch": 0.41800691417405683, + "grad_norm": 1.5642182629868104, + "learning_rate": 2.6182191915949043e-06, + "loss": 1.0756, + "num_input_tokens_seen": 59614740, + "step": 5562 + }, + { + "epoch": 0.4180820682398918, + "grad_norm": 1.650232720580335, + "learning_rate": 2.6177561734966653e-06, + "loss": 0.9185, + "num_input_tokens_seen": 59637930, + "step": 5563 + }, + { + "epoch": 0.4181572223057267, + "grad_norm": 1.4018502389776404, + "learning_rate": 2.6172931187940084e-06, + "loss": 0.8088, + "num_input_tokens_seen": 59661040, + "step": 5564 + }, + { + "epoch": 0.4182323763715617, + "grad_norm": 1.3300564353843687, + "learning_rate": 2.6168300275143695e-06, + "loss": 0.9318, + "num_input_tokens_seen": 59685430, + "step": 5565 + }, + { + "epoch": 0.41830753043739666, + "grad_norm": 1.7422363548100266, + "learning_rate": 2.61636689968519e-06, + "loss": 0.9346, + "num_input_tokens_seen": 59710245, + "step": 5566 + }, + { + "epoch": 0.4183826845032316, + "grad_norm": 1.6177450499714574, + "learning_rate": 2.6159037353339113e-06, + "loss": 0.9263, + "num_input_tokens_seen": 59733575, + "step": 5567 + }, + { + "epoch": 0.4184578385690666, + "grad_norm": 0.7949293731410801, + "learning_rate": 2.6154405344879776e-06, + "loss": 0.756, + "num_input_tokens_seen": 59811355, + "step": 5568 + }, + { + "epoch": 0.41853299263490157, + "grad_norm": 3.0087936584576225, + "learning_rate": 2.6149772971748357e-06, + "loss": 0.8389, + "num_input_tokens_seen": 59834155, + "step": 5569 + }, + { + "epoch": 0.4186081467007365, + "grad_norm": 1.6121751443264118, + "learning_rate": 2.614514023421934e-06, + "loss": 1.0153, + "num_input_tokens_seen": 59858680, + "step": 5570 + }, + { + "epoch": 0.41868330076657145, + "grad_norm": 1.6492715325666558, + "learning_rate": 2.6140507132567238e-06, + "loss": 0.9381, + "num_input_tokens_seen": 59882905, + "step": 5571 + }, + { + "epoch": 0.4187584548324064, + "grad_norm": 2.2251527840953167, + "learning_rate": 2.6135873667066567e-06, + "loss": 1.0142, + "num_input_tokens_seen": 59904100, + "step": 5572 + }, + { + "epoch": 0.4188336088982414, + "grad_norm": 1.267847582092158, + "learning_rate": 2.6131239837991894e-06, + "loss": 1.0484, + "num_input_tokens_seen": 59929290, + "step": 5573 + }, + { + "epoch": 0.41890876296407636, + "grad_norm": 1.3858174578471723, + "learning_rate": 2.6126605645617777e-06, + "loss": 0.8732, + "num_input_tokens_seen": 59953620, + "step": 5574 + }, + { + "epoch": 0.41898391702991133, + "grad_norm": 1.4480342497068135, + "learning_rate": 2.6121971090218816e-06, + "loss": 0.9189, + "num_input_tokens_seen": 59975815, + "step": 5575 + }, + { + "epoch": 0.4190590710957463, + "grad_norm": 0.9157653011071579, + "learning_rate": 2.6117336172069625e-06, + "loss": 0.8765, + "num_input_tokens_seen": 60058610, + "step": 5576 + }, + { + "epoch": 0.4191342251615812, + "grad_norm": 1.5214389309444076, + "learning_rate": 2.6112700891444845e-06, + "loss": 0.9486, + "num_input_tokens_seen": 60080395, + "step": 5577 + }, + { + "epoch": 0.4192093792274162, + "grad_norm": 1.6275990315674203, + "learning_rate": 2.6108065248619128e-06, + "loss": 0.9727, + "num_input_tokens_seen": 60100625, + "step": 5578 + }, + { + "epoch": 0.41928453329325116, + "grad_norm": 1.6693169105903134, + "learning_rate": 2.6103429243867147e-06, + "loss": 0.9345, + "num_input_tokens_seen": 60125040, + "step": 5579 + }, + { + "epoch": 0.41935968735908613, + "grad_norm": 2.085088206925059, + "learning_rate": 2.609879287746362e-06, + "loss": 0.953, + "num_input_tokens_seen": 60148670, + "step": 5580 + }, + { + "epoch": 0.4194348414249211, + "grad_norm": 1.422221214986153, + "learning_rate": 2.609415614968326e-06, + "loss": 0.9176, + "num_input_tokens_seen": 60172395, + "step": 5581 + }, + { + "epoch": 0.41950999549075607, + "grad_norm": 0.8022716144349709, + "learning_rate": 2.608951906080081e-06, + "loss": 0.8321, + "num_input_tokens_seen": 60251705, + "step": 5582 + }, + { + "epoch": 0.419585149556591, + "grad_norm": 1.7134168245082855, + "learning_rate": 2.608488161109104e-06, + "loss": 0.9077, + "num_input_tokens_seen": 60272575, + "step": 5583 + }, + { + "epoch": 0.41966030362242596, + "grad_norm": 1.460852880628204, + "learning_rate": 2.608024380082874e-06, + "loss": 1.0031, + "num_input_tokens_seen": 60295755, + "step": 5584 + }, + { + "epoch": 0.4197354576882609, + "grad_norm": 1.507754072727477, + "learning_rate": 2.60756056302887e-06, + "loss": 0.8567, + "num_input_tokens_seen": 60316560, + "step": 5585 + }, + { + "epoch": 0.4198106117540959, + "grad_norm": 1.4209637918550428, + "learning_rate": 2.6070967099745773e-06, + "loss": 0.9912, + "num_input_tokens_seen": 60338855, + "step": 5586 + }, + { + "epoch": 0.41988576581993087, + "grad_norm": 1.6598678517883256, + "learning_rate": 2.6066328209474786e-06, + "loss": 0.9617, + "num_input_tokens_seen": 60364425, + "step": 5587 + }, + { + "epoch": 0.41996091988576584, + "grad_norm": 2.532776454737297, + "learning_rate": 2.6061688959750633e-06, + "loss": 0.7976, + "num_input_tokens_seen": 60386835, + "step": 5588 + }, + { + "epoch": 0.4200360739516008, + "grad_norm": 1.5431328408593261, + "learning_rate": 2.6057049350848194e-06, + "loss": 1.0077, + "num_input_tokens_seen": 60410300, + "step": 5589 + }, + { + "epoch": 0.4201112280174357, + "grad_norm": 6.994718193829112, + "learning_rate": 2.6052409383042383e-06, + "loss": 0.8629, + "num_input_tokens_seen": 60429480, + "step": 5590 + }, + { + "epoch": 0.4201863820832707, + "grad_norm": 1.4557458356836717, + "learning_rate": 2.604776905660814e-06, + "loss": 0.9723, + "num_input_tokens_seen": 60453845, + "step": 5591 + }, + { + "epoch": 0.42026153614910566, + "grad_norm": 1.521566640958324, + "learning_rate": 2.6043128371820427e-06, + "loss": 0.9832, + "num_input_tokens_seen": 60475035, + "step": 5592 + }, + { + "epoch": 0.42033669021494063, + "grad_norm": 1.5605806387637355, + "learning_rate": 2.603848732895421e-06, + "loss": 0.8289, + "num_input_tokens_seen": 60496520, + "step": 5593 + }, + { + "epoch": 0.4204118442807756, + "grad_norm": 1.6299078992555198, + "learning_rate": 2.6033845928284503e-06, + "loss": 0.9192, + "num_input_tokens_seen": 60518585, + "step": 5594 + }, + { + "epoch": 0.4204869983466106, + "grad_norm": 3.7845275797746796, + "learning_rate": 2.602920417008632e-06, + "loss": 0.9931, + "num_input_tokens_seen": 60541240, + "step": 5595 + }, + { + "epoch": 0.4205621524124455, + "grad_norm": 1.2455433075219748, + "learning_rate": 2.60245620546347e-06, + "loss": 0.9197, + "num_input_tokens_seen": 60566670, + "step": 5596 + }, + { + "epoch": 0.42063730647828046, + "grad_norm": 1.7276001310794937, + "learning_rate": 2.6019919582204713e-06, + "loss": 0.9915, + "num_input_tokens_seen": 60589400, + "step": 5597 + }, + { + "epoch": 0.42071246054411543, + "grad_norm": 1.6456122752913578, + "learning_rate": 2.601527675307143e-06, + "loss": 0.9457, + "num_input_tokens_seen": 60613375, + "step": 5598 + }, + { + "epoch": 0.4207876146099504, + "grad_norm": 1.7083405595331955, + "learning_rate": 2.601063356750997e-06, + "loss": 0.9543, + "num_input_tokens_seen": 60636490, + "step": 5599 + }, + { + "epoch": 0.42086276867578537, + "grad_norm": 1.489840994293753, + "learning_rate": 2.600599002579546e-06, + "loss": 0.9106, + "num_input_tokens_seen": 60658805, + "step": 5600 + }, + { + "epoch": 0.42093792274162034, + "grad_norm": 1.5218576700528075, + "learning_rate": 2.6001346128203036e-06, + "loss": 0.9579, + "num_input_tokens_seen": 60685650, + "step": 5601 + }, + { + "epoch": 0.4210130768074553, + "grad_norm": 1.5025489924816893, + "learning_rate": 2.5996701875007873e-06, + "loss": 0.953, + "num_input_tokens_seen": 60708655, + "step": 5602 + }, + { + "epoch": 0.4210882308732902, + "grad_norm": 1.4060633641476148, + "learning_rate": 2.5992057266485162e-06, + "loss": 0.9658, + "num_input_tokens_seen": 60734155, + "step": 5603 + }, + { + "epoch": 0.4211633849391252, + "grad_norm": 1.3431569498485927, + "learning_rate": 2.5987412302910114e-06, + "loss": 0.9234, + "num_input_tokens_seen": 60759200, + "step": 5604 + }, + { + "epoch": 0.42123853900496017, + "grad_norm": 1.4582753941970064, + "learning_rate": 2.598276698455796e-06, + "loss": 0.9471, + "num_input_tokens_seen": 60783400, + "step": 5605 + }, + { + "epoch": 0.42131369307079514, + "grad_norm": 0.7395446704555119, + "learning_rate": 2.5978121311703955e-06, + "loss": 0.766, + "num_input_tokens_seen": 60859155, + "step": 5606 + }, + { + "epoch": 0.4213888471366301, + "grad_norm": 1.52338397592273, + "learning_rate": 2.5973475284623366e-06, + "loss": 0.9503, + "num_input_tokens_seen": 60881540, + "step": 5607 + }, + { + "epoch": 0.4214640012024651, + "grad_norm": 1.4833709533435622, + "learning_rate": 2.5968828903591492e-06, + "loss": 1.0327, + "num_input_tokens_seen": 60902625, + "step": 5608 + }, + { + "epoch": 0.4215391552683, + "grad_norm": 1.7417222843081264, + "learning_rate": 2.5964182168883654e-06, + "loss": 0.977, + "num_input_tokens_seen": 60924155, + "step": 5609 + }, + { + "epoch": 0.42161430933413496, + "grad_norm": 1.7955390275980319, + "learning_rate": 2.5959535080775176e-06, + "loss": 0.7876, + "num_input_tokens_seen": 60947420, + "step": 5610 + }, + { + "epoch": 0.42168946339996993, + "grad_norm": 1.4943289843741403, + "learning_rate": 2.595488763954143e-06, + "loss": 0.9355, + "num_input_tokens_seen": 60971870, + "step": 5611 + }, + { + "epoch": 0.4217646174658049, + "grad_norm": 2.2320719767624526, + "learning_rate": 2.5950239845457792e-06, + "loss": 0.9289, + "num_input_tokens_seen": 60993725, + "step": 5612 + }, + { + "epoch": 0.4218397715316399, + "grad_norm": 1.5172167412781368, + "learning_rate": 2.594559169879965e-06, + "loss": 0.9193, + "num_input_tokens_seen": 61016960, + "step": 5613 + }, + { + "epoch": 0.42191492559747484, + "grad_norm": 1.2293830562573882, + "learning_rate": 2.594094319984244e-06, + "loss": 1.0205, + "num_input_tokens_seen": 61042010, + "step": 5614 + }, + { + "epoch": 0.42199007966330976, + "grad_norm": 0.8613139387074017, + "learning_rate": 2.593629434886159e-06, + "loss": 0.7967, + "num_input_tokens_seen": 61118290, + "step": 5615 + }, + { + "epoch": 0.42206523372914473, + "grad_norm": 1.6063814933181955, + "learning_rate": 2.5931645146132576e-06, + "loss": 0.8777, + "num_input_tokens_seen": 61143920, + "step": 5616 + }, + { + "epoch": 0.4221403877949797, + "grad_norm": 1.546029987148014, + "learning_rate": 2.592699559193086e-06, + "loss": 1.0366, + "num_input_tokens_seen": 61167015, + "step": 5617 + }, + { + "epoch": 0.42221554186081467, + "grad_norm": 1.4737061822927613, + "learning_rate": 2.592234568653197e-06, + "loss": 0.9514, + "num_input_tokens_seen": 61192000, + "step": 5618 + }, + { + "epoch": 0.42229069592664964, + "grad_norm": 2.4774471592200094, + "learning_rate": 2.5917695430211416e-06, + "loss": 0.9131, + "num_input_tokens_seen": 61216000, + "step": 5619 + }, + { + "epoch": 0.4223658499924846, + "grad_norm": 1.5083682760473032, + "learning_rate": 2.591304482324475e-06, + "loss": 1.0175, + "num_input_tokens_seen": 61241020, + "step": 5620 + }, + { + "epoch": 0.4224410040583196, + "grad_norm": 1.3354714456274166, + "learning_rate": 2.590839386590754e-06, + "loss": 0.9168, + "num_input_tokens_seen": 61265530, + "step": 5621 + }, + { + "epoch": 0.4225161581241545, + "grad_norm": 1.703571156150154, + "learning_rate": 2.5903742558475358e-06, + "loss": 0.9503, + "num_input_tokens_seen": 61285140, + "step": 5622 + }, + { + "epoch": 0.42259131218998947, + "grad_norm": 1.493332588260896, + "learning_rate": 2.589909090122383e-06, + "loss": 0.9916, + "num_input_tokens_seen": 61309690, + "step": 5623 + }, + { + "epoch": 0.42266646625582444, + "grad_norm": 1.568358462892792, + "learning_rate": 2.589443889442857e-06, + "loss": 0.8696, + "num_input_tokens_seen": 61334780, + "step": 5624 + }, + { + "epoch": 0.4227416203216594, + "grad_norm": 2.0404412880610936, + "learning_rate": 2.5889786538365243e-06, + "loss": 0.9615, + "num_input_tokens_seen": 61359485, + "step": 5625 + }, + { + "epoch": 0.4228167743874944, + "grad_norm": 1.527597313635882, + "learning_rate": 2.588513383330951e-06, + "loss": 0.9005, + "num_input_tokens_seen": 61382300, + "step": 5626 + }, + { + "epoch": 0.42289192845332935, + "grad_norm": 1.578592212370035, + "learning_rate": 2.588048077953705e-06, + "loss": 0.9535, + "num_input_tokens_seen": 61406235, + "step": 5627 + }, + { + "epoch": 0.42296708251916426, + "grad_norm": 1.7152739313299559, + "learning_rate": 2.58758273773236e-06, + "loss": 1.0049, + "num_input_tokens_seen": 61428905, + "step": 5628 + }, + { + "epoch": 0.42304223658499923, + "grad_norm": 1.7446036925523754, + "learning_rate": 2.5871173626944864e-06, + "loss": 0.9641, + "num_input_tokens_seen": 61450830, + "step": 5629 + }, + { + "epoch": 0.4231173906508342, + "grad_norm": 1.6786458178678112, + "learning_rate": 2.586651952867662e-06, + "loss": 0.9623, + "num_input_tokens_seen": 61474810, + "step": 5630 + }, + { + "epoch": 0.4231925447166692, + "grad_norm": 1.6830641052811868, + "learning_rate": 2.5861865082794625e-06, + "loss": 1.0183, + "num_input_tokens_seen": 61498425, + "step": 5631 + }, + { + "epoch": 0.42326769878250414, + "grad_norm": 1.4319240398825812, + "learning_rate": 2.5857210289574675e-06, + "loss": 0.8837, + "num_input_tokens_seen": 61523095, + "step": 5632 + }, + { + "epoch": 0.4233428528483391, + "grad_norm": 1.8357676326004118, + "learning_rate": 2.5852555149292593e-06, + "loss": 0.9618, + "num_input_tokens_seen": 61542080, + "step": 5633 + }, + { + "epoch": 0.4234180069141741, + "grad_norm": 1.5183791473098383, + "learning_rate": 2.5847899662224195e-06, + "loss": 1.0266, + "num_input_tokens_seen": 61565665, + "step": 5634 + }, + { + "epoch": 0.423493160980009, + "grad_norm": 1.5826990295889232, + "learning_rate": 2.584324382864536e-06, + "loss": 0.8182, + "num_input_tokens_seen": 61588890, + "step": 5635 + }, + { + "epoch": 0.42356831504584397, + "grad_norm": 0.8375811238640378, + "learning_rate": 2.583858764883195e-06, + "loss": 0.7696, + "num_input_tokens_seen": 61670275, + "step": 5636 + }, + { + "epoch": 0.42364346911167894, + "grad_norm": 1.6197855255239502, + "learning_rate": 2.5833931123059865e-06, + "loss": 0.9062, + "num_input_tokens_seen": 61694335, + "step": 5637 + }, + { + "epoch": 0.4237186231775139, + "grad_norm": 1.8520241978590795, + "learning_rate": 2.5829274251605023e-06, + "loss": 1.0194, + "num_input_tokens_seen": 61716905, + "step": 5638 + }, + { + "epoch": 0.4237937772433489, + "grad_norm": 1.4302820522408866, + "learning_rate": 2.5824617034743354e-06, + "loss": 0.9082, + "num_input_tokens_seen": 61741060, + "step": 5639 + }, + { + "epoch": 0.42386893130918385, + "grad_norm": 1.4672373022822505, + "learning_rate": 2.5819959472750827e-06, + "loss": 0.9552, + "num_input_tokens_seen": 61763115, + "step": 5640 + }, + { + "epoch": 0.42394408537501876, + "grad_norm": 1.4201066475431054, + "learning_rate": 2.581530156590341e-06, + "loss": 0.9732, + "num_input_tokens_seen": 61787060, + "step": 5641 + }, + { + "epoch": 0.42401923944085373, + "grad_norm": 0.7879026972801382, + "learning_rate": 2.5810643314477116e-06, + "loss": 0.7861, + "num_input_tokens_seen": 61857015, + "step": 5642 + }, + { + "epoch": 0.4240943935066887, + "grad_norm": 1.8450327791232948, + "learning_rate": 2.5805984718747953e-06, + "loss": 0.9281, + "num_input_tokens_seen": 61879995, + "step": 5643 + }, + { + "epoch": 0.4241695475725237, + "grad_norm": 1.5546049930014816, + "learning_rate": 2.5801325778991958e-06, + "loss": 0.9324, + "num_input_tokens_seen": 61902875, + "step": 5644 + }, + { + "epoch": 0.42424470163835865, + "grad_norm": 1.562217534282651, + "learning_rate": 2.5796666495485196e-06, + "loss": 0.9618, + "num_input_tokens_seen": 61927530, + "step": 5645 + }, + { + "epoch": 0.4243198557041936, + "grad_norm": 1.4651820898325119, + "learning_rate": 2.579200686850375e-06, + "loss": 0.964, + "num_input_tokens_seen": 61950850, + "step": 5646 + }, + { + "epoch": 0.4243950097700286, + "grad_norm": 1.6363778499024506, + "learning_rate": 2.5787346898323716e-06, + "loss": 0.9596, + "num_input_tokens_seen": 61973945, + "step": 5647 + }, + { + "epoch": 0.4244701638358635, + "grad_norm": 1.5902520482602278, + "learning_rate": 2.578268658522122e-06, + "loss": 0.9194, + "num_input_tokens_seen": 61995415, + "step": 5648 + }, + { + "epoch": 0.42454531790169847, + "grad_norm": 1.2463427447337525, + "learning_rate": 2.5778025929472397e-06, + "loss": 0.8553, + "num_input_tokens_seen": 62023975, + "step": 5649 + }, + { + "epoch": 0.42462047196753344, + "grad_norm": 1.6380874628323947, + "learning_rate": 2.577336493135341e-06, + "loss": 0.9804, + "num_input_tokens_seen": 62046245, + "step": 5650 + }, + { + "epoch": 0.4246956260333684, + "grad_norm": 1.469150001761696, + "learning_rate": 2.5768703591140445e-06, + "loss": 0.9838, + "num_input_tokens_seen": 62070090, + "step": 5651 + }, + { + "epoch": 0.4247707800992034, + "grad_norm": 1.40611715748957, + "learning_rate": 2.5764041909109706e-06, + "loss": 0.9179, + "num_input_tokens_seen": 62095240, + "step": 5652 + }, + { + "epoch": 0.42484593416503835, + "grad_norm": 1.4908210929309826, + "learning_rate": 2.5759379885537414e-06, + "loss": 1.0673, + "num_input_tokens_seen": 62119430, + "step": 5653 + }, + { + "epoch": 0.42492108823087327, + "grad_norm": 1.3619847294515741, + "learning_rate": 2.57547175206998e-06, + "loss": 0.9546, + "num_input_tokens_seen": 62144720, + "step": 5654 + }, + { + "epoch": 0.42499624229670824, + "grad_norm": 2.009734425056961, + "learning_rate": 2.5750054814873144e-06, + "loss": 1.0191, + "num_input_tokens_seen": 62168120, + "step": 5655 + }, + { + "epoch": 0.4250713963625432, + "grad_norm": 1.9726885198087667, + "learning_rate": 2.5745391768333715e-06, + "loss": 1.0379, + "num_input_tokens_seen": 62192565, + "step": 5656 + }, + { + "epoch": 0.4251465504283782, + "grad_norm": 1.743986082332093, + "learning_rate": 2.574072838135783e-06, + "loss": 1.0514, + "num_input_tokens_seen": 62211965, + "step": 5657 + }, + { + "epoch": 0.42522170449421315, + "grad_norm": 1.3869160686392996, + "learning_rate": 2.5736064654221806e-06, + "loss": 0.978, + "num_input_tokens_seen": 62236115, + "step": 5658 + }, + { + "epoch": 0.4252968585600481, + "grad_norm": 0.9846564904659728, + "learning_rate": 2.573140058720198e-06, + "loss": 0.8856, + "num_input_tokens_seen": 62305720, + "step": 5659 + }, + { + "epoch": 0.42537201262588303, + "grad_norm": 1.2233733003221965, + "learning_rate": 2.572673618057473e-06, + "loss": 0.9034, + "num_input_tokens_seen": 62332530, + "step": 5660 + }, + { + "epoch": 0.425447166691718, + "grad_norm": 1.365356658130641, + "learning_rate": 2.5722071434616426e-06, + "loss": 0.8898, + "num_input_tokens_seen": 62358320, + "step": 5661 + }, + { + "epoch": 0.425522320757553, + "grad_norm": 1.5014856451811107, + "learning_rate": 2.5717406349603483e-06, + "loss": 0.9624, + "num_input_tokens_seen": 62380495, + "step": 5662 + }, + { + "epoch": 0.42559747482338794, + "grad_norm": 1.8788995881468111, + "learning_rate": 2.5712740925812314e-06, + "loss": 0.9756, + "num_input_tokens_seen": 62403450, + "step": 5663 + }, + { + "epoch": 0.4256726288892229, + "grad_norm": 0.8576484676208844, + "learning_rate": 2.5708075163519373e-06, + "loss": 0.825, + "num_input_tokens_seen": 62472455, + "step": 5664 + }, + { + "epoch": 0.4257477829550579, + "grad_norm": 1.5640193418784882, + "learning_rate": 2.5703409063001124e-06, + "loss": 0.9067, + "num_input_tokens_seen": 62501145, + "step": 5665 + }, + { + "epoch": 0.42582293702089286, + "grad_norm": 1.5515513009611253, + "learning_rate": 2.5698742624534046e-06, + "loss": 0.9297, + "num_input_tokens_seen": 62526920, + "step": 5666 + }, + { + "epoch": 0.42589809108672777, + "grad_norm": 1.5822135914401658, + "learning_rate": 2.5694075848394646e-06, + "loss": 0.9991, + "num_input_tokens_seen": 62549630, + "step": 5667 + }, + { + "epoch": 0.42597324515256274, + "grad_norm": 2.1687651400078867, + "learning_rate": 2.5689408734859445e-06, + "loss": 0.9298, + "num_input_tokens_seen": 62571275, + "step": 5668 + }, + { + "epoch": 0.4260483992183977, + "grad_norm": 1.607211538391317, + "learning_rate": 2.568474128420499e-06, + "loss": 0.8111, + "num_input_tokens_seen": 62599965, + "step": 5669 + }, + { + "epoch": 0.4261235532842327, + "grad_norm": 1.873076214923523, + "learning_rate": 2.5680073496707854e-06, + "loss": 0.8895, + "num_input_tokens_seen": 62624620, + "step": 5670 + }, + { + "epoch": 0.42619870735006765, + "grad_norm": 1.3068490767313714, + "learning_rate": 2.5675405372644606e-06, + "loss": 0.9552, + "num_input_tokens_seen": 62649735, + "step": 5671 + }, + { + "epoch": 0.4262738614159026, + "grad_norm": 1.4235712350775889, + "learning_rate": 2.567073691229186e-06, + "loss": 0.8096, + "num_input_tokens_seen": 62675480, + "step": 5672 + }, + { + "epoch": 0.42634901548173754, + "grad_norm": 1.447439494737906, + "learning_rate": 2.5666068115926223e-06, + "loss": 0.9564, + "num_input_tokens_seen": 62699570, + "step": 5673 + }, + { + "epoch": 0.4264241695475725, + "grad_norm": 1.5606956745196532, + "learning_rate": 2.5661398983824375e-06, + "loss": 0.9335, + "num_input_tokens_seen": 62719520, + "step": 5674 + }, + { + "epoch": 0.4264993236134075, + "grad_norm": 1.4581506433042937, + "learning_rate": 2.565672951626295e-06, + "loss": 0.9361, + "num_input_tokens_seen": 62743925, + "step": 5675 + }, + { + "epoch": 0.42657447767924245, + "grad_norm": 1.377960615669358, + "learning_rate": 2.5652059713518636e-06, + "loss": 0.9536, + "num_input_tokens_seen": 62766075, + "step": 5676 + }, + { + "epoch": 0.4266496317450774, + "grad_norm": 1.610786855704344, + "learning_rate": 2.5647389575868142e-06, + "loss": 0.8614, + "num_input_tokens_seen": 62788895, + "step": 5677 + }, + { + "epoch": 0.4267247858109124, + "grad_norm": 1.6959327231234982, + "learning_rate": 2.564271910358819e-06, + "loss": 0.9695, + "num_input_tokens_seen": 62811125, + "step": 5678 + }, + { + "epoch": 0.42679993987674736, + "grad_norm": 1.626547736578203, + "learning_rate": 2.563804829695553e-06, + "loss": 0.9419, + "num_input_tokens_seen": 62835790, + "step": 5679 + }, + { + "epoch": 0.4268750939425823, + "grad_norm": 1.6665913882739074, + "learning_rate": 2.5633377156246917e-06, + "loss": 0.9728, + "num_input_tokens_seen": 62857950, + "step": 5680 + }, + { + "epoch": 0.42695024800841724, + "grad_norm": 0.7935623427708838, + "learning_rate": 2.5628705681739124e-06, + "loss": 0.7813, + "num_input_tokens_seen": 62935215, + "step": 5681 + }, + { + "epoch": 0.4270254020742522, + "grad_norm": 1.6494495485146228, + "learning_rate": 2.5624033873708983e-06, + "loss": 0.9323, + "num_input_tokens_seen": 62956835, + "step": 5682 + }, + { + "epoch": 0.4271005561400872, + "grad_norm": 1.3781000217241486, + "learning_rate": 2.5619361732433287e-06, + "loss": 1.0591, + "num_input_tokens_seen": 62979880, + "step": 5683 + }, + { + "epoch": 0.42717571020592215, + "grad_norm": 1.5069148895287507, + "learning_rate": 2.5614689258188896e-06, + "loss": 0.9528, + "num_input_tokens_seen": 63003880, + "step": 5684 + }, + { + "epoch": 0.4272508642717571, + "grad_norm": 1.4617507221552222, + "learning_rate": 2.561001645125266e-06, + "loss": 0.9283, + "num_input_tokens_seen": 63030975, + "step": 5685 + }, + { + "epoch": 0.42732601833759204, + "grad_norm": 1.6566020663360785, + "learning_rate": 2.560534331190148e-06, + "loss": 1.0451, + "num_input_tokens_seen": 63054480, + "step": 5686 + }, + { + "epoch": 0.427401172403427, + "grad_norm": 1.219542123671545, + "learning_rate": 2.5600669840412233e-06, + "loss": 0.972, + "num_input_tokens_seen": 63083650, + "step": 5687 + }, + { + "epoch": 0.427476326469262, + "grad_norm": 0.7751530588426899, + "learning_rate": 2.5595996037061853e-06, + "loss": 0.7978, + "num_input_tokens_seen": 63170080, + "step": 5688 + }, + { + "epoch": 0.42755148053509695, + "grad_norm": 1.9618381696332539, + "learning_rate": 2.559132190212728e-06, + "loss": 1.0557, + "num_input_tokens_seen": 63194380, + "step": 5689 + }, + { + "epoch": 0.4276266346009319, + "grad_norm": 1.3894610302056492, + "learning_rate": 2.558664743588547e-06, + "loss": 0.9754, + "num_input_tokens_seen": 63217075, + "step": 5690 + }, + { + "epoch": 0.4277017886667669, + "grad_norm": 1.6613763611601169, + "learning_rate": 2.5581972638613417e-06, + "loss": 1.0034, + "num_input_tokens_seen": 63240440, + "step": 5691 + }, + { + "epoch": 0.42777694273260186, + "grad_norm": 0.798747200819194, + "learning_rate": 2.557729751058811e-06, + "loss": 0.8331, + "num_input_tokens_seen": 63313485, + "step": 5692 + }, + { + "epoch": 0.4278520967984368, + "grad_norm": 2.024161073193518, + "learning_rate": 2.557262205208656e-06, + "loss": 0.9146, + "num_input_tokens_seen": 63335780, + "step": 5693 + }, + { + "epoch": 0.42792725086427175, + "grad_norm": 1.5023892765219031, + "learning_rate": 2.556794626338582e-06, + "loss": 0.9363, + "num_input_tokens_seen": 63356455, + "step": 5694 + }, + { + "epoch": 0.4280024049301067, + "grad_norm": 1.7743357933472177, + "learning_rate": 2.5563270144762933e-06, + "loss": 1.0702, + "num_input_tokens_seen": 63378020, + "step": 5695 + }, + { + "epoch": 0.4280775589959417, + "grad_norm": 1.5896703827080567, + "learning_rate": 2.5558593696495e-06, + "loss": 0.9851, + "num_input_tokens_seen": 63400775, + "step": 5696 + }, + { + "epoch": 0.42815271306177666, + "grad_norm": 1.5429603286706501, + "learning_rate": 2.5553916918859102e-06, + "loss": 0.998, + "num_input_tokens_seen": 63421455, + "step": 5697 + }, + { + "epoch": 0.4282278671276116, + "grad_norm": 3.4471970159603154, + "learning_rate": 2.554923981213235e-06, + "loss": 0.8824, + "num_input_tokens_seen": 63445370, + "step": 5698 + }, + { + "epoch": 0.42830302119344654, + "grad_norm": 1.8448682077369258, + "learning_rate": 2.55445623765919e-06, + "loss": 1.1054, + "num_input_tokens_seen": 63466065, + "step": 5699 + }, + { + "epoch": 0.4283781752592815, + "grad_norm": 1.6277519545533858, + "learning_rate": 2.553988461251489e-06, + "loss": 0.912, + "num_input_tokens_seen": 63485295, + "step": 5700 + }, + { + "epoch": 0.4284533293251165, + "grad_norm": 1.0179784649892574, + "learning_rate": 2.553520652017851e-06, + "loss": 0.8189, + "num_input_tokens_seen": 63551650, + "step": 5701 + }, + { + "epoch": 0.42852848339095145, + "grad_norm": 1.8381126991231345, + "learning_rate": 2.5530528099859946e-06, + "loss": 0.9335, + "num_input_tokens_seen": 63569960, + "step": 5702 + }, + { + "epoch": 0.4286036374567864, + "grad_norm": 1.2395743949418276, + "learning_rate": 2.5525849351836414e-06, + "loss": 0.9791, + "num_input_tokens_seen": 63597385, + "step": 5703 + }, + { + "epoch": 0.4286787915226214, + "grad_norm": 1.4467562501098987, + "learning_rate": 2.5521170276385147e-06, + "loss": 0.902, + "num_input_tokens_seen": 63620775, + "step": 5704 + }, + { + "epoch": 0.4287539455884563, + "grad_norm": 0.963151846034383, + "learning_rate": 2.5516490873783397e-06, + "loss": 0.7789, + "num_input_tokens_seen": 63688915, + "step": 5705 + }, + { + "epoch": 0.4288290996542913, + "grad_norm": 1.5343695024753141, + "learning_rate": 2.5511811144308447e-06, + "loss": 0.9232, + "num_input_tokens_seen": 63714935, + "step": 5706 + }, + { + "epoch": 0.42890425372012625, + "grad_norm": 1.731075206429678, + "learning_rate": 2.550713108823757e-06, + "loss": 0.9503, + "num_input_tokens_seen": 63736110, + "step": 5707 + }, + { + "epoch": 0.4289794077859612, + "grad_norm": 1.414774762751385, + "learning_rate": 2.5502450705848097e-06, + "loss": 0.9114, + "num_input_tokens_seen": 63758670, + "step": 5708 + }, + { + "epoch": 0.4290545618517962, + "grad_norm": 1.6698600534517896, + "learning_rate": 2.5497769997417347e-06, + "loss": 0.9757, + "num_input_tokens_seen": 63781020, + "step": 5709 + }, + { + "epoch": 0.42912971591763116, + "grad_norm": 1.382547256117361, + "learning_rate": 2.5493088963222668e-06, + "loss": 0.9363, + "num_input_tokens_seen": 63806250, + "step": 5710 + }, + { + "epoch": 0.42920486998346613, + "grad_norm": 0.9511591227027116, + "learning_rate": 2.5488407603541437e-06, + "loss": 0.8314, + "num_input_tokens_seen": 63875190, + "step": 5711 + }, + { + "epoch": 0.42928002404930105, + "grad_norm": 2.1455843759452895, + "learning_rate": 2.5483725918651034e-06, + "loss": 1.0099, + "num_input_tokens_seen": 63898285, + "step": 5712 + }, + { + "epoch": 0.429355178115136, + "grad_norm": 1.7635806916619403, + "learning_rate": 2.5479043908828877e-06, + "loss": 1.0007, + "num_input_tokens_seen": 63917960, + "step": 5713 + }, + { + "epoch": 0.429430332180971, + "grad_norm": 1.4846431289237896, + "learning_rate": 2.547436157435239e-06, + "loss": 1.0141, + "num_input_tokens_seen": 63943360, + "step": 5714 + }, + { + "epoch": 0.42950548624680596, + "grad_norm": 2.040556729213064, + "learning_rate": 2.546967891549901e-06, + "loss": 0.9252, + "num_input_tokens_seen": 63965810, + "step": 5715 + }, + { + "epoch": 0.4295806403126409, + "grad_norm": 1.4563093672672396, + "learning_rate": 2.5464995932546217e-06, + "loss": 0.919, + "num_input_tokens_seen": 63991165, + "step": 5716 + }, + { + "epoch": 0.4296557943784759, + "grad_norm": 1.7707791338835328, + "learning_rate": 2.5460312625771475e-06, + "loss": 0.9452, + "num_input_tokens_seen": 64011870, + "step": 5717 + }, + { + "epoch": 0.4297309484443108, + "grad_norm": 1.4739819132757574, + "learning_rate": 2.5455628995452313e-06, + "loss": 0.9823, + "num_input_tokens_seen": 64038465, + "step": 5718 + }, + { + "epoch": 0.4298061025101458, + "grad_norm": 1.5408262419275252, + "learning_rate": 2.5450945041866246e-06, + "loss": 0.9606, + "num_input_tokens_seen": 64060825, + "step": 5719 + }, + { + "epoch": 0.42988125657598075, + "grad_norm": 1.2772798668054968, + "learning_rate": 2.54462607652908e-06, + "loss": 0.9609, + "num_input_tokens_seen": 64087880, + "step": 5720 + }, + { + "epoch": 0.4299564106418157, + "grad_norm": 1.3650571519813206, + "learning_rate": 2.5441576166003555e-06, + "loss": 1.0198, + "num_input_tokens_seen": 64112590, + "step": 5721 + }, + { + "epoch": 0.4300315647076507, + "grad_norm": 1.5891931500734442, + "learning_rate": 2.5436891244282076e-06, + "loss": 1.009, + "num_input_tokens_seen": 64135385, + "step": 5722 + }, + { + "epoch": 0.43010671877348566, + "grad_norm": 1.8879364064950606, + "learning_rate": 2.5432206000403982e-06, + "loss": 0.9682, + "num_input_tokens_seen": 64158785, + "step": 5723 + }, + { + "epoch": 0.43018187283932063, + "grad_norm": 1.691136292373002, + "learning_rate": 2.5427520434646884e-06, + "loss": 0.9302, + "num_input_tokens_seen": 64181850, + "step": 5724 + }, + { + "epoch": 0.43025702690515555, + "grad_norm": 5.671789583766268, + "learning_rate": 2.5422834547288406e-06, + "loss": 0.9765, + "num_input_tokens_seen": 64204745, + "step": 5725 + }, + { + "epoch": 0.4303321809709905, + "grad_norm": 2.553954039576102, + "learning_rate": 2.5418148338606226e-06, + "loss": 0.965, + "num_input_tokens_seen": 64227190, + "step": 5726 + }, + { + "epoch": 0.4304073350368255, + "grad_norm": 1.5257219450520076, + "learning_rate": 2.5413461808878e-06, + "loss": 0.9076, + "num_input_tokens_seen": 64252330, + "step": 5727 + }, + { + "epoch": 0.43048248910266046, + "grad_norm": 1.3197584174628978, + "learning_rate": 2.5408774958381436e-06, + "loss": 0.948, + "num_input_tokens_seen": 64275625, + "step": 5728 + }, + { + "epoch": 0.43055764316849543, + "grad_norm": 1.378557025896307, + "learning_rate": 2.5404087787394248e-06, + "loss": 0.8972, + "num_input_tokens_seen": 64299840, + "step": 5729 + }, + { + "epoch": 0.4306327972343304, + "grad_norm": 1.9604266591751498, + "learning_rate": 2.5399400296194164e-06, + "loss": 0.9504, + "num_input_tokens_seen": 64323705, + "step": 5730 + }, + { + "epoch": 0.4307079513001653, + "grad_norm": 1.8155784115091291, + "learning_rate": 2.5394712485058933e-06, + "loss": 0.8848, + "num_input_tokens_seen": 64348180, + "step": 5731 + }, + { + "epoch": 0.4307831053660003, + "grad_norm": 1.8717782634008997, + "learning_rate": 2.539002435426633e-06, + "loss": 0.9397, + "num_input_tokens_seen": 64373535, + "step": 5732 + }, + { + "epoch": 0.43085825943183526, + "grad_norm": 1.6603857981238728, + "learning_rate": 2.5385335904094147e-06, + "loss": 0.9757, + "num_input_tokens_seen": 64395640, + "step": 5733 + }, + { + "epoch": 0.4309334134976702, + "grad_norm": 1.5542160162705614, + "learning_rate": 2.5380647134820186e-06, + "loss": 0.9339, + "num_input_tokens_seen": 64419570, + "step": 5734 + }, + { + "epoch": 0.4310085675635052, + "grad_norm": 1.5349377192682674, + "learning_rate": 2.5375958046722283e-06, + "loss": 0.9856, + "num_input_tokens_seen": 64440910, + "step": 5735 + }, + { + "epoch": 0.43108372162934017, + "grad_norm": 1.4911597742123424, + "learning_rate": 2.5371268640078277e-06, + "loss": 1.0676, + "num_input_tokens_seen": 64463165, + "step": 5736 + }, + { + "epoch": 0.43115887569517514, + "grad_norm": 1.4842246531589893, + "learning_rate": 2.5366578915166033e-06, + "loss": 0.9594, + "num_input_tokens_seen": 64485270, + "step": 5737 + }, + { + "epoch": 0.43123402976101005, + "grad_norm": 1.7454270833241055, + "learning_rate": 2.536188887226345e-06, + "loss": 0.9384, + "num_input_tokens_seen": 64507360, + "step": 5738 + }, + { + "epoch": 0.431309183826845, + "grad_norm": 2.0388067677053803, + "learning_rate": 2.53571985116484e-06, + "loss": 0.8812, + "num_input_tokens_seen": 64529570, + "step": 5739 + }, + { + "epoch": 0.43138433789268, + "grad_norm": 1.530470747116367, + "learning_rate": 2.535250783359884e-06, + "loss": 0.9716, + "num_input_tokens_seen": 64550420, + "step": 5740 + }, + { + "epoch": 0.43145949195851496, + "grad_norm": 1.9546927301838501, + "learning_rate": 2.5347816838392695e-06, + "loss": 0.9062, + "num_input_tokens_seen": 64573100, + "step": 5741 + }, + { + "epoch": 0.43153464602434993, + "grad_norm": 1.8888290324108157, + "learning_rate": 2.534312552630791e-06, + "loss": 0.9504, + "num_input_tokens_seen": 64594585, + "step": 5742 + }, + { + "epoch": 0.4316098000901849, + "grad_norm": 1.510529358266316, + "learning_rate": 2.533843389762249e-06, + "loss": 1.0571, + "num_input_tokens_seen": 64619500, + "step": 5743 + }, + { + "epoch": 0.4316849541560198, + "grad_norm": 1.7125126803408988, + "learning_rate": 2.5333741952614412e-06, + "loss": 0.9684, + "num_input_tokens_seen": 64641170, + "step": 5744 + }, + { + "epoch": 0.4317601082218548, + "grad_norm": 1.593923297510824, + "learning_rate": 2.5329049691561705e-06, + "loss": 0.9406, + "num_input_tokens_seen": 64660820, + "step": 5745 + }, + { + "epoch": 0.43183526228768976, + "grad_norm": 1.3429688689457617, + "learning_rate": 2.53243571147424e-06, + "loss": 1.0035, + "num_input_tokens_seen": 64681785, + "step": 5746 + }, + { + "epoch": 0.43191041635352473, + "grad_norm": 1.640454310666009, + "learning_rate": 2.5319664222434534e-06, + "loss": 0.9622, + "num_input_tokens_seen": 64705375, + "step": 5747 + }, + { + "epoch": 0.4319855704193597, + "grad_norm": 1.2364751116742527, + "learning_rate": 2.5314971014916207e-06, + "loss": 0.9499, + "num_input_tokens_seen": 64731060, + "step": 5748 + }, + { + "epoch": 0.43206072448519467, + "grad_norm": 1.7988387282532228, + "learning_rate": 2.5310277492465486e-06, + "loss": 1.0106, + "num_input_tokens_seen": 64751380, + "step": 5749 + }, + { + "epoch": 0.4321358785510296, + "grad_norm": 1.6002899324464315, + "learning_rate": 2.5305583655360495e-06, + "loss": 1.0125, + "num_input_tokens_seen": 64773520, + "step": 5750 + }, + { + "epoch": 0.43221103261686455, + "grad_norm": 1.6990015351466472, + "learning_rate": 2.530088950387935e-06, + "loss": 0.9521, + "num_input_tokens_seen": 64795400, + "step": 5751 + }, + { + "epoch": 0.4322861866826995, + "grad_norm": 1.2730144429882089, + "learning_rate": 2.529619503830021e-06, + "loss": 0.9639, + "num_input_tokens_seen": 64820620, + "step": 5752 + }, + { + "epoch": 0.4323613407485345, + "grad_norm": 1.9372795964056553, + "learning_rate": 2.5291500258901234e-06, + "loss": 0.9795, + "num_input_tokens_seen": 64841485, + "step": 5753 + }, + { + "epoch": 0.43243649481436947, + "grad_norm": 1.3751973097290566, + "learning_rate": 2.5286805165960597e-06, + "loss": 0.9447, + "num_input_tokens_seen": 64867035, + "step": 5754 + }, + { + "epoch": 0.43251164888020444, + "grad_norm": 1.546855502535543, + "learning_rate": 2.528210975975652e-06, + "loss": 0.9943, + "num_input_tokens_seen": 64891660, + "step": 5755 + }, + { + "epoch": 0.4325868029460394, + "grad_norm": 1.6613274246669991, + "learning_rate": 2.52774140405672e-06, + "loss": 0.9826, + "num_input_tokens_seen": 64916395, + "step": 5756 + }, + { + "epoch": 0.4326619570118743, + "grad_norm": 1.3895276057990094, + "learning_rate": 2.5272718008670895e-06, + "loss": 0.9969, + "num_input_tokens_seen": 64940485, + "step": 5757 + }, + { + "epoch": 0.4327371110777093, + "grad_norm": 1.47512472886134, + "learning_rate": 2.5268021664345865e-06, + "loss": 0.9662, + "num_input_tokens_seen": 64962015, + "step": 5758 + }, + { + "epoch": 0.43281226514354426, + "grad_norm": 0.8142613612861921, + "learning_rate": 2.526332500787037e-06, + "loss": 0.8165, + "num_input_tokens_seen": 65037525, + "step": 5759 + }, + { + "epoch": 0.43288741920937923, + "grad_norm": 1.4251110435185894, + "learning_rate": 2.525862803952272e-06, + "loss": 0.9917, + "num_input_tokens_seen": 65063135, + "step": 5760 + }, + { + "epoch": 0.4329625732752142, + "grad_norm": 1.4445085918821958, + "learning_rate": 2.5253930759581213e-06, + "loss": 1.065, + "num_input_tokens_seen": 65086715, + "step": 5761 + }, + { + "epoch": 0.4330377273410492, + "grad_norm": 1.4129237020563572, + "learning_rate": 2.5249233168324196e-06, + "loss": 0.9519, + "num_input_tokens_seen": 65110330, + "step": 5762 + }, + { + "epoch": 0.4331128814068841, + "grad_norm": 1.4675637994574222, + "learning_rate": 2.5244535266030014e-06, + "loss": 1.0333, + "num_input_tokens_seen": 65131335, + "step": 5763 + }, + { + "epoch": 0.43318803547271906, + "grad_norm": 1.7776895375340178, + "learning_rate": 2.5239837052977032e-06, + "loss": 1.0972, + "num_input_tokens_seen": 65152370, + "step": 5764 + }, + { + "epoch": 0.433263189538554, + "grad_norm": 1.7421897580256176, + "learning_rate": 2.523513852944364e-06, + "loss": 0.9417, + "num_input_tokens_seen": 65174930, + "step": 5765 + }, + { + "epoch": 0.433338343604389, + "grad_norm": 2.0722802365693043, + "learning_rate": 2.5230439695708244e-06, + "loss": 0.8631, + "num_input_tokens_seen": 65196845, + "step": 5766 + }, + { + "epoch": 0.43341349767022397, + "grad_norm": 1.7550256635957666, + "learning_rate": 2.5225740552049267e-06, + "loss": 0.9804, + "num_input_tokens_seen": 65218595, + "step": 5767 + }, + { + "epoch": 0.43348865173605894, + "grad_norm": 1.590362988909425, + "learning_rate": 2.5221041098745157e-06, + "loss": 1.0166, + "num_input_tokens_seen": 65238835, + "step": 5768 + }, + { + "epoch": 0.4335638058018939, + "grad_norm": 0.76469271204037, + "learning_rate": 2.5216341336074363e-06, + "loss": 0.8161, + "num_input_tokens_seen": 65316365, + "step": 5769 + }, + { + "epoch": 0.4336389598677288, + "grad_norm": 1.3073172649091174, + "learning_rate": 2.5211641264315372e-06, + "loss": 0.9407, + "num_input_tokens_seen": 65341175, + "step": 5770 + }, + { + "epoch": 0.4337141139335638, + "grad_norm": 1.8599180822894186, + "learning_rate": 2.520694088374668e-06, + "loss": 1.0347, + "num_input_tokens_seen": 65361050, + "step": 5771 + }, + { + "epoch": 0.43378926799939876, + "grad_norm": 1.3039620489817298, + "learning_rate": 2.52022401946468e-06, + "loss": 0.9911, + "num_input_tokens_seen": 65388075, + "step": 5772 + }, + { + "epoch": 0.43386442206523373, + "grad_norm": 1.5713128846178175, + "learning_rate": 2.519753919729427e-06, + "loss": 1.0058, + "num_input_tokens_seen": 65407495, + "step": 5773 + }, + { + "epoch": 0.4339395761310687, + "grad_norm": 3.2585480589168094, + "learning_rate": 2.519283789196764e-06, + "loss": 1.0361, + "num_input_tokens_seen": 65428505, + "step": 5774 + }, + { + "epoch": 0.4340147301969037, + "grad_norm": 1.4674738757847923, + "learning_rate": 2.518813627894548e-06, + "loss": 1.01, + "num_input_tokens_seen": 65451260, + "step": 5775 + }, + { + "epoch": 0.4340898842627386, + "grad_norm": 1.2205748956087887, + "learning_rate": 2.5183434358506373e-06, + "loss": 1.008, + "num_input_tokens_seen": 65476795, + "step": 5776 + }, + { + "epoch": 0.43416503832857356, + "grad_norm": 0.7762809273644754, + "learning_rate": 2.5178732130928943e-06, + "loss": 0.8362, + "num_input_tokens_seen": 65551685, + "step": 5777 + }, + { + "epoch": 0.43424019239440853, + "grad_norm": 1.8166770158992036, + "learning_rate": 2.5174029596491792e-06, + "loss": 0.9944, + "num_input_tokens_seen": 65575095, + "step": 5778 + }, + { + "epoch": 0.4343153464602435, + "grad_norm": 5.131817261056643, + "learning_rate": 2.5169326755473582e-06, + "loss": 0.9187, + "num_input_tokens_seen": 65596870, + "step": 5779 + }, + { + "epoch": 0.43439050052607847, + "grad_norm": 2.291192323548761, + "learning_rate": 2.516462360815297e-06, + "loss": 0.9209, + "num_input_tokens_seen": 65621585, + "step": 5780 + }, + { + "epoch": 0.43446565459191344, + "grad_norm": 1.5044852591261832, + "learning_rate": 2.5159920154808615e-06, + "loss": 0.9581, + "num_input_tokens_seen": 65643335, + "step": 5781 + }, + { + "epoch": 0.4345408086577484, + "grad_norm": 1.841245210619341, + "learning_rate": 2.5155216395719253e-06, + "loss": 0.9707, + "num_input_tokens_seen": 65664435, + "step": 5782 + }, + { + "epoch": 0.4346159627235833, + "grad_norm": 0.8755027484709049, + "learning_rate": 2.5150512331163564e-06, + "loss": 0.8913, + "num_input_tokens_seen": 65737005, + "step": 5783 + }, + { + "epoch": 0.4346911167894183, + "grad_norm": 1.6263714651594625, + "learning_rate": 2.5145807961420303e-06, + "loss": 0.8111, + "num_input_tokens_seen": 65759575, + "step": 5784 + }, + { + "epoch": 0.43476627085525327, + "grad_norm": 1.4384981919039599, + "learning_rate": 2.514110328676822e-06, + "loss": 0.9986, + "num_input_tokens_seen": 65781820, + "step": 5785 + }, + { + "epoch": 0.43484142492108824, + "grad_norm": 1.4304744868265034, + "learning_rate": 2.5136398307486075e-06, + "loss": 0.9752, + "num_input_tokens_seen": 65806725, + "step": 5786 + }, + { + "epoch": 0.4349165789869232, + "grad_norm": 2.3092373694677084, + "learning_rate": 2.5131693023852663e-06, + "loss": 0.8111, + "num_input_tokens_seen": 65831715, + "step": 5787 + }, + { + "epoch": 0.4349917330527582, + "grad_norm": 1.783200517800839, + "learning_rate": 2.5126987436146786e-06, + "loss": 0.9555, + "num_input_tokens_seen": 65852765, + "step": 5788 + }, + { + "epoch": 0.4350668871185931, + "grad_norm": 1.421725338615481, + "learning_rate": 2.5122281544647273e-06, + "loss": 0.8815, + "num_input_tokens_seen": 65876900, + "step": 5789 + }, + { + "epoch": 0.43514204118442806, + "grad_norm": 1.5127294350624052, + "learning_rate": 2.511757534963297e-06, + "loss": 0.9068, + "num_input_tokens_seen": 65899875, + "step": 5790 + }, + { + "epoch": 0.43521719525026303, + "grad_norm": 1.607423432396914, + "learning_rate": 2.5112868851382724e-06, + "loss": 0.9302, + "num_input_tokens_seen": 65921455, + "step": 5791 + }, + { + "epoch": 0.435292349316098, + "grad_norm": 1.5206043994290521, + "learning_rate": 2.5108162050175425e-06, + "loss": 0.9527, + "num_input_tokens_seen": 65943565, + "step": 5792 + }, + { + "epoch": 0.435367503381933, + "grad_norm": 1.5916894045632002, + "learning_rate": 2.510345494628996e-06, + "loss": 1.0199, + "num_input_tokens_seen": 65966645, + "step": 5793 + }, + { + "epoch": 0.43544265744776794, + "grad_norm": 1.2965823843041058, + "learning_rate": 2.509874754000525e-06, + "loss": 0.9963, + "num_input_tokens_seen": 65991300, + "step": 5794 + }, + { + "epoch": 0.43551781151360286, + "grad_norm": 1.5260548336081614, + "learning_rate": 2.5094039831600217e-06, + "loss": 0.9128, + "num_input_tokens_seen": 66014565, + "step": 5795 + }, + { + "epoch": 0.43559296557943783, + "grad_norm": 1.7115315142089114, + "learning_rate": 2.5089331821353827e-06, + "loss": 0.9305, + "num_input_tokens_seen": 66038835, + "step": 5796 + }, + { + "epoch": 0.4356681196452728, + "grad_norm": 1.848797822531058, + "learning_rate": 2.5084623509545034e-06, + "loss": 1.0286, + "num_input_tokens_seen": 66058295, + "step": 5797 + }, + { + "epoch": 0.43574327371110777, + "grad_norm": 1.8333374442430126, + "learning_rate": 2.5079914896452823e-06, + "loss": 0.9529, + "num_input_tokens_seen": 66082380, + "step": 5798 + }, + { + "epoch": 0.43581842777694274, + "grad_norm": 1.7834300704221109, + "learning_rate": 2.507520598235621e-06, + "loss": 1.0098, + "num_input_tokens_seen": 66102385, + "step": 5799 + }, + { + "epoch": 0.4358935818427777, + "grad_norm": 1.748938988159624, + "learning_rate": 2.5070496767534202e-06, + "loss": 0.9462, + "num_input_tokens_seen": 66124935, + "step": 5800 + }, + { + "epoch": 0.4359687359086127, + "grad_norm": 1.369609366156848, + "learning_rate": 2.5065787252265848e-06, + "loss": 0.9916, + "num_input_tokens_seen": 66148925, + "step": 5801 + }, + { + "epoch": 0.4360438899744476, + "grad_norm": 1.3075432503660285, + "learning_rate": 2.50610774368302e-06, + "loss": 0.9465, + "num_input_tokens_seen": 66174290, + "step": 5802 + }, + { + "epoch": 0.43611904404028257, + "grad_norm": 1.4993624921931339, + "learning_rate": 2.505636732150633e-06, + "loss": 0.9805, + "num_input_tokens_seen": 66195695, + "step": 5803 + }, + { + "epoch": 0.43619419810611754, + "grad_norm": 1.753492752643246, + "learning_rate": 2.505165690657334e-06, + "loss": 0.9877, + "num_input_tokens_seen": 66222275, + "step": 5804 + }, + { + "epoch": 0.4362693521719525, + "grad_norm": 1.5578147069366342, + "learning_rate": 2.504694619231033e-06, + "loss": 1.0018, + "num_input_tokens_seen": 66244515, + "step": 5805 + }, + { + "epoch": 0.4363445062377875, + "grad_norm": 1.5734452460514086, + "learning_rate": 2.5042235178996436e-06, + "loss": 1.0358, + "num_input_tokens_seen": 66266175, + "step": 5806 + }, + { + "epoch": 0.43641966030362245, + "grad_norm": 1.782311240496931, + "learning_rate": 2.5037523866910797e-06, + "loss": 0.9346, + "num_input_tokens_seen": 66288895, + "step": 5807 + }, + { + "epoch": 0.43649481436945736, + "grad_norm": 1.5672224016192102, + "learning_rate": 2.503281225633258e-06, + "loss": 0.9654, + "num_input_tokens_seen": 66312315, + "step": 5808 + }, + { + "epoch": 0.43656996843529233, + "grad_norm": 1.2970920518989961, + "learning_rate": 2.5028100347540967e-06, + "loss": 0.9561, + "num_input_tokens_seen": 66337315, + "step": 5809 + }, + { + "epoch": 0.4366451225011273, + "grad_norm": 1.4264368509567664, + "learning_rate": 2.5023388140815148e-06, + "loss": 0.9435, + "num_input_tokens_seen": 66360360, + "step": 5810 + }, + { + "epoch": 0.4367202765669623, + "grad_norm": 1.6107041139012628, + "learning_rate": 2.5018675636434353e-06, + "loss": 1.0298, + "num_input_tokens_seen": 66383305, + "step": 5811 + }, + { + "epoch": 0.43679543063279724, + "grad_norm": 2.3313621187222893, + "learning_rate": 2.5013962834677804e-06, + "loss": 1.0198, + "num_input_tokens_seen": 66404170, + "step": 5812 + }, + { + "epoch": 0.4368705846986322, + "grad_norm": 0.8528352727536839, + "learning_rate": 2.5009249735824757e-06, + "loss": 0.8464, + "num_input_tokens_seen": 66481490, + "step": 5813 + }, + { + "epoch": 0.4369457387644672, + "grad_norm": 1.5016072722080565, + "learning_rate": 2.500453634015449e-06, + "loss": 0.9752, + "num_input_tokens_seen": 66505240, + "step": 5814 + }, + { + "epoch": 0.4370208928303021, + "grad_norm": 2.1079440314085147, + "learning_rate": 2.4999822647946273e-06, + "loss": 0.9912, + "num_input_tokens_seen": 66527360, + "step": 5815 + }, + { + "epoch": 0.43709604689613707, + "grad_norm": 1.59763147759294, + "learning_rate": 2.499510865947942e-06, + "loss": 0.937, + "num_input_tokens_seen": 66552570, + "step": 5816 + }, + { + "epoch": 0.43717120096197204, + "grad_norm": 1.8489299360073546, + "learning_rate": 2.4990394375033247e-06, + "loss": 0.9884, + "num_input_tokens_seen": 66573595, + "step": 5817 + }, + { + "epoch": 0.437246355027807, + "grad_norm": 2.073876189197445, + "learning_rate": 2.4985679794887106e-06, + "loss": 0.9338, + "num_input_tokens_seen": 66596540, + "step": 5818 + }, + { + "epoch": 0.437321509093642, + "grad_norm": 1.4265469102122554, + "learning_rate": 2.4980964919320343e-06, + "loss": 0.9724, + "num_input_tokens_seen": 66617935, + "step": 5819 + }, + { + "epoch": 0.43739666315947695, + "grad_norm": 1.7836101355783767, + "learning_rate": 2.4976249748612332e-06, + "loss": 0.9966, + "num_input_tokens_seen": 66640835, + "step": 5820 + }, + { + "epoch": 0.43747181722531187, + "grad_norm": 1.4458186732418334, + "learning_rate": 2.497153428304247e-06, + "loss": 0.9283, + "num_input_tokens_seen": 66664715, + "step": 5821 + }, + { + "epoch": 0.43754697129114684, + "grad_norm": 1.4748062911774489, + "learning_rate": 2.496681852289016e-06, + "loss": 0.9921, + "num_input_tokens_seen": 66688515, + "step": 5822 + }, + { + "epoch": 0.4376221253569818, + "grad_norm": 1.6668842729926308, + "learning_rate": 2.4962102468434843e-06, + "loss": 0.896, + "num_input_tokens_seen": 66713000, + "step": 5823 + }, + { + "epoch": 0.4376972794228168, + "grad_norm": 1.6706086826633608, + "learning_rate": 2.4957386119955954e-06, + "loss": 1.041, + "num_input_tokens_seen": 66736205, + "step": 5824 + }, + { + "epoch": 0.43777243348865175, + "grad_norm": 1.5973647021815378, + "learning_rate": 2.4952669477732938e-06, + "loss": 0.9689, + "num_input_tokens_seen": 66760355, + "step": 5825 + }, + { + "epoch": 0.4378475875544867, + "grad_norm": 1.8380977826014973, + "learning_rate": 2.4947952542045307e-06, + "loss": 1.0536, + "num_input_tokens_seen": 66782270, + "step": 5826 + }, + { + "epoch": 0.43792274162032163, + "grad_norm": 1.4636983640506955, + "learning_rate": 2.494323531317253e-06, + "loss": 0.9871, + "num_input_tokens_seen": 66805230, + "step": 5827 + }, + { + "epoch": 0.4379978956861566, + "grad_norm": 1.3549927417096206, + "learning_rate": 2.493851779139414e-06, + "loss": 0.9486, + "num_input_tokens_seen": 66833995, + "step": 5828 + }, + { + "epoch": 0.4380730497519916, + "grad_norm": 1.528508203431422, + "learning_rate": 2.493379997698966e-06, + "loss": 0.9342, + "num_input_tokens_seen": 66858090, + "step": 5829 + }, + { + "epoch": 0.43814820381782654, + "grad_norm": 1.3069669261912624, + "learning_rate": 2.4929081870238635e-06, + "loss": 0.9406, + "num_input_tokens_seen": 66881570, + "step": 5830 + }, + { + "epoch": 0.4382233578836615, + "grad_norm": 0.6581877301959844, + "learning_rate": 2.4924363471420634e-06, + "loss": 0.7381, + "num_input_tokens_seen": 66960960, + "step": 5831 + }, + { + "epoch": 0.4382985119494965, + "grad_norm": 1.3765413141947678, + "learning_rate": 2.491964478081524e-06, + "loss": 1.0111, + "num_input_tokens_seen": 66983480, + "step": 5832 + }, + { + "epoch": 0.43837366601533145, + "grad_norm": 1.5001176152212434, + "learning_rate": 2.4914925798702057e-06, + "loss": 0.9511, + "num_input_tokens_seen": 67007655, + "step": 5833 + }, + { + "epoch": 0.43844882008116637, + "grad_norm": 0.7753975433117841, + "learning_rate": 2.49102065253607e-06, + "loss": 0.7704, + "num_input_tokens_seen": 67085095, + "step": 5834 + }, + { + "epoch": 0.43852397414700134, + "grad_norm": 1.409989788343097, + "learning_rate": 2.49054869610708e-06, + "loss": 0.7529, + "num_input_tokens_seen": 67113220, + "step": 5835 + }, + { + "epoch": 0.4385991282128363, + "grad_norm": 1.578312173451125, + "learning_rate": 2.490076710611202e-06, + "loss": 0.9952, + "num_input_tokens_seen": 67136330, + "step": 5836 + }, + { + "epoch": 0.4386742822786713, + "grad_norm": 1.8941646889864925, + "learning_rate": 2.4896046960764015e-06, + "loss": 0.8516, + "num_input_tokens_seen": 67160400, + "step": 5837 + }, + { + "epoch": 0.43874943634450625, + "grad_norm": 1.7350632959962242, + "learning_rate": 2.4891326525306487e-06, + "loss": 0.8368, + "num_input_tokens_seen": 67182415, + "step": 5838 + }, + { + "epoch": 0.4388245904103412, + "grad_norm": 1.8612044494465863, + "learning_rate": 2.4886605800019123e-06, + "loss": 1.0152, + "num_input_tokens_seen": 67202540, + "step": 5839 + }, + { + "epoch": 0.43889974447617613, + "grad_norm": 1.5628862285099294, + "learning_rate": 2.488188478518166e-06, + "loss": 0.9753, + "num_input_tokens_seen": 67226210, + "step": 5840 + }, + { + "epoch": 0.4389748985420111, + "grad_norm": 1.826357647427562, + "learning_rate": 2.487716348107383e-06, + "loss": 0.9296, + "num_input_tokens_seen": 67249470, + "step": 5841 + }, + { + "epoch": 0.4390500526078461, + "grad_norm": 1.4674877667932589, + "learning_rate": 2.4872441887975386e-06, + "loss": 0.925, + "num_input_tokens_seen": 67280225, + "step": 5842 + }, + { + "epoch": 0.43912520667368105, + "grad_norm": 1.2931906979273202, + "learning_rate": 2.48677200061661e-06, + "loss": 0.9531, + "num_input_tokens_seen": 67309125, + "step": 5843 + }, + { + "epoch": 0.439200360739516, + "grad_norm": 0.7431913714209792, + "learning_rate": 2.486299783592576e-06, + "loss": 0.8387, + "num_input_tokens_seen": 67385850, + "step": 5844 + }, + { + "epoch": 0.439275514805351, + "grad_norm": 1.6580990959758122, + "learning_rate": 2.485827537753419e-06, + "loss": 0.9589, + "num_input_tokens_seen": 67407370, + "step": 5845 + }, + { + "epoch": 0.43935066887118596, + "grad_norm": 1.537990835973575, + "learning_rate": 2.4853552631271193e-06, + "loss": 0.9767, + "num_input_tokens_seen": 67429640, + "step": 5846 + }, + { + "epoch": 0.43942582293702087, + "grad_norm": 1.3808948560847787, + "learning_rate": 2.4848829597416615e-06, + "loss": 0.929, + "num_input_tokens_seen": 67459990, + "step": 5847 + }, + { + "epoch": 0.43950097700285584, + "grad_norm": 1.4605662644296276, + "learning_rate": 2.484410627625032e-06, + "loss": 1.007, + "num_input_tokens_seen": 67483640, + "step": 5848 + }, + { + "epoch": 0.4395761310686908, + "grad_norm": 1.1523143699909317, + "learning_rate": 2.483938266805217e-06, + "loss": 0.9043, + "num_input_tokens_seen": 67510665, + "step": 5849 + }, + { + "epoch": 0.4396512851345258, + "grad_norm": 2.015819096645731, + "learning_rate": 2.483465877310208e-06, + "loss": 0.9186, + "num_input_tokens_seen": 67531070, + "step": 5850 + }, + { + "epoch": 0.43972643920036075, + "grad_norm": 1.530460076066538, + "learning_rate": 2.482993459167993e-06, + "loss": 0.9832, + "num_input_tokens_seen": 67554660, + "step": 5851 + }, + { + "epoch": 0.4398015932661957, + "grad_norm": 0.8526401175856227, + "learning_rate": 2.482521012406567e-06, + "loss": 0.8037, + "num_input_tokens_seen": 67636805, + "step": 5852 + }, + { + "epoch": 0.43987674733203064, + "grad_norm": 1.29171969192118, + "learning_rate": 2.4820485370539233e-06, + "loss": 0.9852, + "num_input_tokens_seen": 67662840, + "step": 5853 + }, + { + "epoch": 0.4399519013978656, + "grad_norm": 1.4474968152842957, + "learning_rate": 2.481576033138057e-06, + "loss": 1.0028, + "num_input_tokens_seen": 67687285, + "step": 5854 + }, + { + "epoch": 0.4400270554637006, + "grad_norm": 1.447931010589172, + "learning_rate": 2.4811035006869677e-06, + "loss": 1.0175, + "num_input_tokens_seen": 67711725, + "step": 5855 + }, + { + "epoch": 0.44010220952953555, + "grad_norm": 0.9612236527303296, + "learning_rate": 2.4806309397286534e-06, + "loss": 0.8411, + "num_input_tokens_seen": 67786725, + "step": 5856 + }, + { + "epoch": 0.4401773635953705, + "grad_norm": 1.30350070358392, + "learning_rate": 2.4801583502911154e-06, + "loss": 1.0052, + "num_input_tokens_seen": 67812935, + "step": 5857 + }, + { + "epoch": 0.4402525176612055, + "grad_norm": 1.896063556595779, + "learning_rate": 2.4796857324023564e-06, + "loss": 0.84, + "num_input_tokens_seen": 67837090, + "step": 5858 + }, + { + "epoch": 0.44032767172704046, + "grad_norm": 7.397793992605583, + "learning_rate": 2.479213086090381e-06, + "loss": 0.9728, + "num_input_tokens_seen": 67857220, + "step": 5859 + }, + { + "epoch": 0.4404028257928754, + "grad_norm": 1.6153031580162789, + "learning_rate": 2.478740411383195e-06, + "loss": 0.9965, + "num_input_tokens_seen": 67880650, + "step": 5860 + }, + { + "epoch": 0.44047797985871034, + "grad_norm": 1.3970465108901682, + "learning_rate": 2.478267708308807e-06, + "loss": 0.8778, + "num_input_tokens_seen": 67906570, + "step": 5861 + }, + { + "epoch": 0.4405531339245453, + "grad_norm": 1.2551193786166965, + "learning_rate": 2.4777949768952255e-06, + "loss": 0.9788, + "num_input_tokens_seen": 67932460, + "step": 5862 + }, + { + "epoch": 0.4406282879903803, + "grad_norm": 1.7892947650046367, + "learning_rate": 2.477322217170462e-06, + "loss": 0.9924, + "num_input_tokens_seen": 67957620, + "step": 5863 + }, + { + "epoch": 0.44070344205621526, + "grad_norm": 1.9399681982402999, + "learning_rate": 2.476849429162529e-06, + "loss": 0.9122, + "num_input_tokens_seen": 67981330, + "step": 5864 + }, + { + "epoch": 0.4407785961220502, + "grad_norm": 1.7487767576470554, + "learning_rate": 2.4763766128994423e-06, + "loss": 1.0046, + "num_input_tokens_seen": 68003320, + "step": 5865 + }, + { + "epoch": 0.44085375018788514, + "grad_norm": 1.640032924810066, + "learning_rate": 2.475903768409216e-06, + "loss": 0.9533, + "num_input_tokens_seen": 68026030, + "step": 5866 + }, + { + "epoch": 0.4409289042537201, + "grad_norm": 1.4530452583337221, + "learning_rate": 2.47543089571987e-06, + "loss": 0.8637, + "num_input_tokens_seen": 68052480, + "step": 5867 + }, + { + "epoch": 0.4410040583195551, + "grad_norm": 1.7758550760318577, + "learning_rate": 2.4749579948594224e-06, + "loss": 0.8475, + "num_input_tokens_seen": 68073500, + "step": 5868 + }, + { + "epoch": 0.44107921238539005, + "grad_norm": 1.4298710105981072, + "learning_rate": 2.4744850658558943e-06, + "loss": 0.9148, + "num_input_tokens_seen": 68097550, + "step": 5869 + }, + { + "epoch": 0.441154366451225, + "grad_norm": 1.6024550765193304, + "learning_rate": 2.47401210873731e-06, + "loss": 0.8958, + "num_input_tokens_seen": 68121270, + "step": 5870 + }, + { + "epoch": 0.44122952051706, + "grad_norm": 1.3887841545457957, + "learning_rate": 2.473539123531693e-06, + "loss": 0.9357, + "num_input_tokens_seen": 68146630, + "step": 5871 + }, + { + "epoch": 0.4413046745828949, + "grad_norm": 1.5399727235583067, + "learning_rate": 2.4730661102670692e-06, + "loss": 1.0443, + "num_input_tokens_seen": 68170445, + "step": 5872 + }, + { + "epoch": 0.4413798286487299, + "grad_norm": 2.0911941871104047, + "learning_rate": 2.4725930689714673e-06, + "loss": 0.9272, + "num_input_tokens_seen": 68193370, + "step": 5873 + }, + { + "epoch": 0.44145498271456485, + "grad_norm": 1.5229742666363564, + "learning_rate": 2.4721199996729167e-06, + "loss": 0.9409, + "num_input_tokens_seen": 68220010, + "step": 5874 + }, + { + "epoch": 0.4415301367803998, + "grad_norm": 1.3624657757861658, + "learning_rate": 2.471646902399448e-06, + "loss": 0.9068, + "num_input_tokens_seen": 68243700, + "step": 5875 + }, + { + "epoch": 0.4416052908462348, + "grad_norm": 1.6035049939935733, + "learning_rate": 2.471173777179094e-06, + "loss": 0.9598, + "num_input_tokens_seen": 68267965, + "step": 5876 + }, + { + "epoch": 0.44168044491206976, + "grad_norm": 1.6129921568687546, + "learning_rate": 2.4707006240398894e-06, + "loss": 0.9642, + "num_input_tokens_seen": 68291975, + "step": 5877 + }, + { + "epoch": 0.44175559897790473, + "grad_norm": 1.8468380945329812, + "learning_rate": 2.4702274430098707e-06, + "loss": 0.9444, + "num_input_tokens_seen": 68316380, + "step": 5878 + }, + { + "epoch": 0.44183075304373964, + "grad_norm": 1.4647630999933652, + "learning_rate": 2.469754234117075e-06, + "loss": 0.9767, + "num_input_tokens_seen": 68342280, + "step": 5879 + }, + { + "epoch": 0.4419059071095746, + "grad_norm": 1.494584639652994, + "learning_rate": 2.4692809973895426e-06, + "loss": 0.9532, + "num_input_tokens_seen": 68366315, + "step": 5880 + }, + { + "epoch": 0.4419810611754096, + "grad_norm": 1.560041171410932, + "learning_rate": 2.4688077328553136e-06, + "loss": 0.959, + "num_input_tokens_seen": 68389985, + "step": 5881 + }, + { + "epoch": 0.44205621524124455, + "grad_norm": 1.6666982488818807, + "learning_rate": 2.4683344405424316e-06, + "loss": 0.9134, + "num_input_tokens_seen": 68411305, + "step": 5882 + }, + { + "epoch": 0.4421313693070795, + "grad_norm": 4.6020242300709535, + "learning_rate": 2.4678611204789405e-06, + "loss": 0.9432, + "num_input_tokens_seen": 68435425, + "step": 5883 + }, + { + "epoch": 0.4422065233729145, + "grad_norm": 2.1647486511289706, + "learning_rate": 2.4673877726928865e-06, + "loss": 1.0265, + "num_input_tokens_seen": 68453290, + "step": 5884 + }, + { + "epoch": 0.4422816774387494, + "grad_norm": 1.7868196513145507, + "learning_rate": 2.4669143972123178e-06, + "loss": 0.7954, + "num_input_tokens_seen": 68474525, + "step": 5885 + }, + { + "epoch": 0.4423568315045844, + "grad_norm": 1.3200257571192742, + "learning_rate": 2.4664409940652817e-06, + "loss": 0.9587, + "num_input_tokens_seen": 68499485, + "step": 5886 + }, + { + "epoch": 0.44243198557041935, + "grad_norm": 1.5585748580514265, + "learning_rate": 2.465967563279832e-06, + "loss": 1.0159, + "num_input_tokens_seen": 68522300, + "step": 5887 + }, + { + "epoch": 0.4425071396362543, + "grad_norm": 1.4281097210957852, + "learning_rate": 2.4654941048840184e-06, + "loss": 0.9878, + "num_input_tokens_seen": 68550375, + "step": 5888 + }, + { + "epoch": 0.4425822937020893, + "grad_norm": 1.5578782965903675, + "learning_rate": 2.465020618905898e-06, + "loss": 1.0354, + "num_input_tokens_seen": 68576570, + "step": 5889 + }, + { + "epoch": 0.44265744776792426, + "grad_norm": 1.609446181417545, + "learning_rate": 2.464547105373525e-06, + "loss": 1.0305, + "num_input_tokens_seen": 68597750, + "step": 5890 + }, + { + "epoch": 0.44273260183375923, + "grad_norm": 1.5000538021611265, + "learning_rate": 2.4640735643149566e-06, + "loss": 0.9446, + "num_input_tokens_seen": 68621155, + "step": 5891 + }, + { + "epoch": 0.44280775589959415, + "grad_norm": 0.8594249017272142, + "learning_rate": 2.4635999957582526e-06, + "loss": 0.8297, + "num_input_tokens_seen": 68697975, + "step": 5892 + }, + { + "epoch": 0.4428829099654291, + "grad_norm": 1.3520707955192994, + "learning_rate": 2.4631263997314734e-06, + "loss": 0.9555, + "num_input_tokens_seen": 68722145, + "step": 5893 + }, + { + "epoch": 0.4429580640312641, + "grad_norm": 1.631362490228314, + "learning_rate": 2.4626527762626822e-06, + "loss": 1.0332, + "num_input_tokens_seen": 68744700, + "step": 5894 + }, + { + "epoch": 0.44303321809709906, + "grad_norm": 1.8229302832685266, + "learning_rate": 2.462179125379942e-06, + "loss": 0.9432, + "num_input_tokens_seen": 68765290, + "step": 5895 + }, + { + "epoch": 0.443108372162934, + "grad_norm": 1.7127201774115557, + "learning_rate": 2.461705447111319e-06, + "loss": 0.9457, + "num_input_tokens_seen": 68787765, + "step": 5896 + }, + { + "epoch": 0.443183526228769, + "grad_norm": 1.678594828213556, + "learning_rate": 2.4612317414848803e-06, + "loss": 0.9647, + "num_input_tokens_seen": 68808970, + "step": 5897 + }, + { + "epoch": 0.4432586802946039, + "grad_norm": 1.8500521652496444, + "learning_rate": 2.460758008528694e-06, + "loss": 1.0009, + "num_input_tokens_seen": 68831720, + "step": 5898 + }, + { + "epoch": 0.4433338343604389, + "grad_norm": 1.226523234716853, + "learning_rate": 2.460284248270833e-06, + "loss": 0.9339, + "num_input_tokens_seen": 68865215, + "step": 5899 + }, + { + "epoch": 0.44340898842627385, + "grad_norm": 1.3715990227760044, + "learning_rate": 2.4598104607393666e-06, + "loss": 0.9828, + "num_input_tokens_seen": 68890020, + "step": 5900 + }, + { + "epoch": 0.4434841424921088, + "grad_norm": 1.8496990242554114, + "learning_rate": 2.4593366459623698e-06, + "loss": 1.0309, + "num_input_tokens_seen": 68911700, + "step": 5901 + }, + { + "epoch": 0.4435592965579438, + "grad_norm": 1.5826113543873128, + "learning_rate": 2.458862803967918e-06, + "loss": 0.8652, + "num_input_tokens_seen": 68937400, + "step": 5902 + }, + { + "epoch": 0.44363445062377876, + "grad_norm": 0.8766585707079136, + "learning_rate": 2.4583889347840873e-06, + "loss": 0.8454, + "num_input_tokens_seen": 69009575, + "step": 5903 + }, + { + "epoch": 0.44370960468961373, + "grad_norm": 3.4686255905248196, + "learning_rate": 2.4579150384389574e-06, + "loss": 0.9614, + "num_input_tokens_seen": 69029165, + "step": 5904 + }, + { + "epoch": 0.44378475875544865, + "grad_norm": 1.8636637361085318, + "learning_rate": 2.4574411149606076e-06, + "loss": 0.8881, + "num_input_tokens_seen": 69050710, + "step": 5905 + }, + { + "epoch": 0.4438599128212836, + "grad_norm": 1.4828062099568577, + "learning_rate": 2.456967164377121e-06, + "loss": 1.0436, + "num_input_tokens_seen": 69072740, + "step": 5906 + }, + { + "epoch": 0.4439350668871186, + "grad_norm": 1.3552243674774036, + "learning_rate": 2.4564931867165795e-06, + "loss": 0.905, + "num_input_tokens_seen": 69101865, + "step": 5907 + }, + { + "epoch": 0.44401022095295356, + "grad_norm": 1.3215220401728678, + "learning_rate": 2.4560191820070683e-06, + "loss": 0.9414, + "num_input_tokens_seen": 69127120, + "step": 5908 + }, + { + "epoch": 0.44408537501878853, + "grad_norm": 1.3829067219676128, + "learning_rate": 2.4555451502766754e-06, + "loss": 0.831, + "num_input_tokens_seen": 69154210, + "step": 5909 + }, + { + "epoch": 0.4441605290846235, + "grad_norm": 0.8137499291556192, + "learning_rate": 2.4550710915534863e-06, + "loss": 0.8409, + "num_input_tokens_seen": 69229940, + "step": 5910 + }, + { + "epoch": 0.4442356831504584, + "grad_norm": 1.4063133225489222, + "learning_rate": 2.4545970058655938e-06, + "loss": 1.027, + "num_input_tokens_seen": 69253045, + "step": 5911 + }, + { + "epoch": 0.4443108372162934, + "grad_norm": 3.01271208970207, + "learning_rate": 2.454122893241088e-06, + "loss": 0.9188, + "num_input_tokens_seen": 69275275, + "step": 5912 + }, + { + "epoch": 0.44438599128212836, + "grad_norm": 1.5718842037833958, + "learning_rate": 2.453648753708061e-06, + "loss": 0.9423, + "num_input_tokens_seen": 69300390, + "step": 5913 + }, + { + "epoch": 0.4444611453479633, + "grad_norm": 1.3309142499739386, + "learning_rate": 2.4531745872946085e-06, + "loss": 0.9462, + "num_input_tokens_seen": 69326050, + "step": 5914 + }, + { + "epoch": 0.4445362994137983, + "grad_norm": 1.56389547974888, + "learning_rate": 2.4527003940288264e-06, + "loss": 1.0179, + "num_input_tokens_seen": 69348220, + "step": 5915 + }, + { + "epoch": 0.44461145347963327, + "grad_norm": 1.3961661022801872, + "learning_rate": 2.4522261739388127e-06, + "loss": 0.9238, + "num_input_tokens_seen": 69373025, + "step": 5916 + }, + { + "epoch": 0.4446866075454682, + "grad_norm": 1.5290870181483538, + "learning_rate": 2.451751927052666e-06, + "loss": 0.9822, + "num_input_tokens_seen": 69395585, + "step": 5917 + }, + { + "epoch": 0.44476176161130315, + "grad_norm": 1.5568446237486633, + "learning_rate": 2.4512776533984882e-06, + "loss": 0.8348, + "num_input_tokens_seen": 69420575, + "step": 5918 + }, + { + "epoch": 0.4448369156771381, + "grad_norm": 1.6013097796273525, + "learning_rate": 2.450803353004382e-06, + "loss": 0.8896, + "num_input_tokens_seen": 69446065, + "step": 5919 + }, + { + "epoch": 0.4449120697429731, + "grad_norm": 1.4824439890221042, + "learning_rate": 2.4503290258984493e-06, + "loss": 1.0194, + "num_input_tokens_seen": 69470110, + "step": 5920 + }, + { + "epoch": 0.44498722380880806, + "grad_norm": 1.6316731689656068, + "learning_rate": 2.4498546721087984e-06, + "loss": 0.9754, + "num_input_tokens_seen": 69492225, + "step": 5921 + }, + { + "epoch": 0.44506237787464303, + "grad_norm": 1.6450865616676948, + "learning_rate": 2.4493802916635355e-06, + "loss": 1.0119, + "num_input_tokens_seen": 69516015, + "step": 5922 + }, + { + "epoch": 0.445137531940478, + "grad_norm": 1.3753964979527942, + "learning_rate": 2.448905884590769e-06, + "loss": 1.0213, + "num_input_tokens_seen": 69539485, + "step": 5923 + }, + { + "epoch": 0.4452126860063129, + "grad_norm": 1.3767118201959732, + "learning_rate": 2.448431450918611e-06, + "loss": 0.948, + "num_input_tokens_seen": 69566365, + "step": 5924 + }, + { + "epoch": 0.4452878400721479, + "grad_norm": 1.621671569061449, + "learning_rate": 2.4479569906751714e-06, + "loss": 0.8758, + "num_input_tokens_seen": 69590525, + "step": 5925 + }, + { + "epoch": 0.44536299413798286, + "grad_norm": 1.7139685967594303, + "learning_rate": 2.4474825038885655e-06, + "loss": 0.8216, + "num_input_tokens_seen": 69610725, + "step": 5926 + }, + { + "epoch": 0.44543814820381783, + "grad_norm": 1.63431994197732, + "learning_rate": 2.4470079905869066e-06, + "loss": 0.8953, + "num_input_tokens_seen": 69634035, + "step": 5927 + }, + { + "epoch": 0.4455133022696528, + "grad_norm": 1.8455900934718246, + "learning_rate": 2.446533450798314e-06, + "loss": 0.8894, + "num_input_tokens_seen": 69656535, + "step": 5928 + }, + { + "epoch": 0.44558845633548777, + "grad_norm": 1.6146967037965565, + "learning_rate": 2.4460588845509036e-06, + "loss": 0.9726, + "num_input_tokens_seen": 69678440, + "step": 5929 + }, + { + "epoch": 0.4456636104013227, + "grad_norm": 1.4701789235845735, + "learning_rate": 2.4455842918727957e-06, + "loss": 0.902, + "num_input_tokens_seen": 69701195, + "step": 5930 + }, + { + "epoch": 0.44573876446715766, + "grad_norm": 1.6072535189116728, + "learning_rate": 2.4451096727921135e-06, + "loss": 0.9497, + "num_input_tokens_seen": 69724465, + "step": 5931 + }, + { + "epoch": 0.4458139185329926, + "grad_norm": 1.2217823354408561, + "learning_rate": 2.444635027336977e-06, + "loss": 1.0153, + "num_input_tokens_seen": 69749800, + "step": 5932 + }, + { + "epoch": 0.4458890725988276, + "grad_norm": 1.5992276175159472, + "learning_rate": 2.4441603555355142e-06, + "loss": 0.9673, + "num_input_tokens_seen": 69770965, + "step": 5933 + }, + { + "epoch": 0.44596422666466257, + "grad_norm": 1.384725895622446, + "learning_rate": 2.443685657415849e-06, + "loss": 1.0081, + "num_input_tokens_seen": 69794760, + "step": 5934 + }, + { + "epoch": 0.44603938073049754, + "grad_norm": 1.8808127265762458, + "learning_rate": 2.4432109330061096e-06, + "loss": 0.8983, + "num_input_tokens_seen": 69821280, + "step": 5935 + }, + { + "epoch": 0.4461145347963325, + "grad_norm": 1.5198231394670234, + "learning_rate": 2.4427361823344256e-06, + "loss": 0.9781, + "num_input_tokens_seen": 69844735, + "step": 5936 + }, + { + "epoch": 0.4461896888621674, + "grad_norm": 1.937110861800574, + "learning_rate": 2.4422614054289264e-06, + "loss": 0.8916, + "num_input_tokens_seen": 69868695, + "step": 5937 + }, + { + "epoch": 0.4462648429280024, + "grad_norm": 1.3319427818833363, + "learning_rate": 2.4417866023177466e-06, + "loss": 0.9347, + "num_input_tokens_seen": 69894530, + "step": 5938 + }, + { + "epoch": 0.44633999699383736, + "grad_norm": 1.6796232699156568, + "learning_rate": 2.4413117730290186e-06, + "loss": 1.0098, + "num_input_tokens_seen": 69916680, + "step": 5939 + }, + { + "epoch": 0.44641515105967233, + "grad_norm": 2.0149297180413615, + "learning_rate": 2.440836917590878e-06, + "loss": 1.0125, + "num_input_tokens_seen": 69936890, + "step": 5940 + }, + { + "epoch": 0.4464903051255073, + "grad_norm": 1.310113872281731, + "learning_rate": 2.440362036031462e-06, + "loss": 0.8774, + "num_input_tokens_seen": 69962125, + "step": 5941 + }, + { + "epoch": 0.4465654591913423, + "grad_norm": 1.364165579450326, + "learning_rate": 2.4398871283789088e-06, + "loss": 0.9581, + "num_input_tokens_seen": 69986285, + "step": 5942 + }, + { + "epoch": 0.4466406132571772, + "grad_norm": 1.486202359988028, + "learning_rate": 2.439412194661359e-06, + "loss": 0.931, + "num_input_tokens_seen": 70011040, + "step": 5943 + }, + { + "epoch": 0.44671576732301216, + "grad_norm": 1.5873257428745713, + "learning_rate": 2.4389372349069544e-06, + "loss": 0.8674, + "num_input_tokens_seen": 70038915, + "step": 5944 + }, + { + "epoch": 0.44679092138884713, + "grad_norm": 1.4478463720735237, + "learning_rate": 2.4384622491438374e-06, + "loss": 0.8966, + "num_input_tokens_seen": 70061695, + "step": 5945 + }, + { + "epoch": 0.4468660754546821, + "grad_norm": 1.7119774292589947, + "learning_rate": 2.437987237400153e-06, + "loss": 0.8885, + "num_input_tokens_seen": 70083410, + "step": 5946 + }, + { + "epoch": 0.44694122952051707, + "grad_norm": 1.4994251421247868, + "learning_rate": 2.4375121997040477e-06, + "loss": 0.8913, + "num_input_tokens_seen": 70103555, + "step": 5947 + }, + { + "epoch": 0.44701638358635204, + "grad_norm": 1.7018954537150561, + "learning_rate": 2.4370371360836697e-06, + "loss": 1.0291, + "num_input_tokens_seen": 70124740, + "step": 5948 + }, + { + "epoch": 0.447091537652187, + "grad_norm": 1.3852615221773592, + "learning_rate": 2.436562046567167e-06, + "loss": 0.9641, + "num_input_tokens_seen": 70151025, + "step": 5949 + }, + { + "epoch": 0.4471666917180219, + "grad_norm": 1.7638733538580136, + "learning_rate": 2.4360869311826927e-06, + "loss": 0.9566, + "num_input_tokens_seen": 70173110, + "step": 5950 + }, + { + "epoch": 0.4472418457838569, + "grad_norm": 1.2136849030973187, + "learning_rate": 2.435611789958397e-06, + "loss": 0.9479, + "num_input_tokens_seen": 70199860, + "step": 5951 + }, + { + "epoch": 0.44731699984969187, + "grad_norm": 1.604239557104733, + "learning_rate": 2.435136622922434e-06, + "loss": 0.9389, + "num_input_tokens_seen": 70223495, + "step": 5952 + }, + { + "epoch": 0.44739215391552684, + "grad_norm": 1.5648426042572565, + "learning_rate": 2.4346614301029613e-06, + "loss": 0.9816, + "num_input_tokens_seen": 70246045, + "step": 5953 + }, + { + "epoch": 0.4474673079813618, + "grad_norm": 1.529540840264605, + "learning_rate": 2.434186211528133e-06, + "loss": 1.0593, + "num_input_tokens_seen": 70267130, + "step": 5954 + }, + { + "epoch": 0.4475424620471968, + "grad_norm": 2.242044796539521, + "learning_rate": 2.4337109672261097e-06, + "loss": 1.0007, + "num_input_tokens_seen": 70286725, + "step": 5955 + }, + { + "epoch": 0.4476176161130317, + "grad_norm": 1.3716638580716851, + "learning_rate": 2.433235697225051e-06, + "loss": 0.8318, + "num_input_tokens_seen": 70311420, + "step": 5956 + }, + { + "epoch": 0.44769277017886666, + "grad_norm": 1.4910413724351697, + "learning_rate": 2.4327604015531177e-06, + "loss": 1.0436, + "num_input_tokens_seen": 70334420, + "step": 5957 + }, + { + "epoch": 0.44776792424470163, + "grad_norm": 1.5006045871475882, + "learning_rate": 2.432285080238474e-06, + "loss": 0.9867, + "num_input_tokens_seen": 70357750, + "step": 5958 + }, + { + "epoch": 0.4478430783105366, + "grad_norm": 0.7626824957909984, + "learning_rate": 2.4318097333092837e-06, + "loss": 0.8122, + "num_input_tokens_seen": 70435465, + "step": 5959 + }, + { + "epoch": 0.4479182323763716, + "grad_norm": 1.5871156722893773, + "learning_rate": 2.4313343607937135e-06, + "loss": 0.9438, + "num_input_tokens_seen": 70458290, + "step": 5960 + }, + { + "epoch": 0.44799338644220654, + "grad_norm": 1.469252885801043, + "learning_rate": 2.430858962719931e-06, + "loss": 0.8481, + "num_input_tokens_seen": 70479565, + "step": 5961 + }, + { + "epoch": 0.44806854050804146, + "grad_norm": 1.2923194661036403, + "learning_rate": 2.4303835391161047e-06, + "loss": 0.9841, + "num_input_tokens_seen": 70502315, + "step": 5962 + }, + { + "epoch": 0.4481436945738764, + "grad_norm": 5.383549832142916, + "learning_rate": 2.4299080900104055e-06, + "loss": 1.0203, + "num_input_tokens_seen": 70528465, + "step": 5963 + }, + { + "epoch": 0.4482188486397114, + "grad_norm": 1.247022229061519, + "learning_rate": 2.4294326154310058e-06, + "loss": 0.8941, + "num_input_tokens_seen": 70557950, + "step": 5964 + }, + { + "epoch": 0.44829400270554637, + "grad_norm": 1.3796777917125078, + "learning_rate": 2.4289571154060794e-06, + "loss": 1.039, + "num_input_tokens_seen": 70581370, + "step": 5965 + }, + { + "epoch": 0.44836915677138134, + "grad_norm": 0.7706895041095927, + "learning_rate": 2.4284815899638012e-06, + "loss": 0.8433, + "num_input_tokens_seen": 70656625, + "step": 5966 + }, + { + "epoch": 0.4484443108372163, + "grad_norm": 1.5162243788552248, + "learning_rate": 2.428006039132348e-06, + "loss": 0.8969, + "num_input_tokens_seen": 70681570, + "step": 5967 + }, + { + "epoch": 0.4485194649030513, + "grad_norm": 1.4399600333389553, + "learning_rate": 2.4275304629398985e-06, + "loss": 0.9652, + "num_input_tokens_seen": 70708295, + "step": 5968 + }, + { + "epoch": 0.4485946189688862, + "grad_norm": 1.3721576269204077, + "learning_rate": 2.427054861414631e-06, + "loss": 0.9701, + "num_input_tokens_seen": 70730625, + "step": 5969 + }, + { + "epoch": 0.44866977303472116, + "grad_norm": 1.5940171560132441, + "learning_rate": 2.426579234584728e-06, + "loss": 1.0155, + "num_input_tokens_seen": 70753700, + "step": 5970 + }, + { + "epoch": 0.44874492710055613, + "grad_norm": 1.7075843875077463, + "learning_rate": 2.426103582478372e-06, + "loss": 0.9665, + "num_input_tokens_seen": 70775540, + "step": 5971 + }, + { + "epoch": 0.4488200811663911, + "grad_norm": 1.4577460696017435, + "learning_rate": 2.4256279051237473e-06, + "loss": 0.9314, + "num_input_tokens_seen": 70800660, + "step": 5972 + }, + { + "epoch": 0.4488952352322261, + "grad_norm": 1.5904261016898973, + "learning_rate": 2.4251522025490393e-06, + "loss": 0.8741, + "num_input_tokens_seen": 70823610, + "step": 5973 + }, + { + "epoch": 0.44897038929806105, + "grad_norm": 1.7837869747529322, + "learning_rate": 2.4246764747824347e-06, + "loss": 1.0409, + "num_input_tokens_seen": 70841780, + "step": 5974 + }, + { + "epoch": 0.44904554336389596, + "grad_norm": 1.536775390325886, + "learning_rate": 2.4242007218521236e-06, + "loss": 0.9446, + "num_input_tokens_seen": 70863210, + "step": 5975 + }, + { + "epoch": 0.44912069742973093, + "grad_norm": 1.4832711381767476, + "learning_rate": 2.423724943786295e-06, + "loss": 0.9414, + "num_input_tokens_seen": 70887380, + "step": 5976 + }, + { + "epoch": 0.4491958514955659, + "grad_norm": 1.4201582721028898, + "learning_rate": 2.4232491406131408e-06, + "loss": 0.866, + "num_input_tokens_seen": 70911105, + "step": 5977 + }, + { + "epoch": 0.44927100556140087, + "grad_norm": 1.5236892883372906, + "learning_rate": 2.4227733123608548e-06, + "loss": 0.9303, + "num_input_tokens_seen": 70933565, + "step": 5978 + }, + { + "epoch": 0.44934615962723584, + "grad_norm": 1.4432648871727942, + "learning_rate": 2.4222974590576303e-06, + "loss": 1.0033, + "num_input_tokens_seen": 70956635, + "step": 5979 + }, + { + "epoch": 0.4494213136930708, + "grad_norm": 1.2848567439724294, + "learning_rate": 2.4218215807316647e-06, + "loss": 0.9714, + "num_input_tokens_seen": 70980190, + "step": 5980 + }, + { + "epoch": 0.4494964677589058, + "grad_norm": 1.7111011583622284, + "learning_rate": 2.4213456774111553e-06, + "loss": 1.0181, + "num_input_tokens_seen": 71001905, + "step": 5981 + }, + { + "epoch": 0.4495716218247407, + "grad_norm": 1.539642689360385, + "learning_rate": 2.420869749124301e-06, + "loss": 0.8319, + "num_input_tokens_seen": 71025100, + "step": 5982 + }, + { + "epoch": 0.44964677589057567, + "grad_norm": 1.2346832759614028, + "learning_rate": 2.4203937958993027e-06, + "loss": 0.9456, + "num_input_tokens_seen": 71054790, + "step": 5983 + }, + { + "epoch": 0.44972192995641064, + "grad_norm": 1.5380860150049585, + "learning_rate": 2.4199178177643617e-06, + "loss": 0.9444, + "num_input_tokens_seen": 71077830, + "step": 5984 + }, + { + "epoch": 0.4497970840222456, + "grad_norm": 1.9208323433874177, + "learning_rate": 2.4194418147476827e-06, + "loss": 0.9635, + "num_input_tokens_seen": 71097485, + "step": 5985 + }, + { + "epoch": 0.4498722380880806, + "grad_norm": 2.0426619173489398, + "learning_rate": 2.4189657868774688e-06, + "loss": 1.0059, + "num_input_tokens_seen": 71115145, + "step": 5986 + }, + { + "epoch": 0.44994739215391555, + "grad_norm": 1.3438588918877783, + "learning_rate": 2.418489734181929e-06, + "loss": 1.0139, + "num_input_tokens_seen": 71137775, + "step": 5987 + }, + { + "epoch": 0.45002254621975046, + "grad_norm": 1.5195474893442575, + "learning_rate": 2.4180136566892696e-06, + "loss": 1.0083, + "num_input_tokens_seen": 71163570, + "step": 5988 + }, + { + "epoch": 0.45009770028558543, + "grad_norm": 2.419399448311651, + "learning_rate": 2.4175375544276998e-06, + "loss": 0.9427, + "num_input_tokens_seen": 71191375, + "step": 5989 + }, + { + "epoch": 0.4501728543514204, + "grad_norm": 1.4200038125430223, + "learning_rate": 2.4170614274254317e-06, + "loss": 0.9703, + "num_input_tokens_seen": 71215080, + "step": 5990 + }, + { + "epoch": 0.4502480084172554, + "grad_norm": 1.482388743556806, + "learning_rate": 2.4165852757106762e-06, + "loss": 0.8778, + "num_input_tokens_seen": 71237895, + "step": 5991 + }, + { + "epoch": 0.45032316248309034, + "grad_norm": 1.4752640439353943, + "learning_rate": 2.416109099311649e-06, + "loss": 0.9715, + "num_input_tokens_seen": 71260435, + "step": 5992 + }, + { + "epoch": 0.4503983165489253, + "grad_norm": 1.7041417158145462, + "learning_rate": 2.4156328982565636e-06, + "loss": 0.977, + "num_input_tokens_seen": 71282235, + "step": 5993 + }, + { + "epoch": 0.4504734706147603, + "grad_norm": 1.459714081476662, + "learning_rate": 2.4151566725736375e-06, + "loss": 1.043, + "num_input_tokens_seen": 71307905, + "step": 5994 + }, + { + "epoch": 0.4505486246805952, + "grad_norm": 1.628320529201512, + "learning_rate": 2.414680422291089e-06, + "loss": 0.848, + "num_input_tokens_seen": 71329435, + "step": 5995 + }, + { + "epoch": 0.45062377874643017, + "grad_norm": 1.2725267956803452, + "learning_rate": 2.4142041474371368e-06, + "loss": 0.9342, + "num_input_tokens_seen": 71352135, + "step": 5996 + }, + { + "epoch": 0.45069893281226514, + "grad_norm": 0.7770286189667008, + "learning_rate": 2.4137278480400038e-06, + "loss": 0.8853, + "num_input_tokens_seen": 71435760, + "step": 5997 + }, + { + "epoch": 0.4507740868781001, + "grad_norm": 1.8534164482571547, + "learning_rate": 2.4132515241279106e-06, + "loss": 0.8666, + "num_input_tokens_seen": 71460475, + "step": 5998 + }, + { + "epoch": 0.4508492409439351, + "grad_norm": 1.7455638789844485, + "learning_rate": 2.4127751757290826e-06, + "loss": 0.9113, + "num_input_tokens_seen": 71482340, + "step": 5999 + }, + { + "epoch": 0.45092439500977005, + "grad_norm": 1.6535399956262382, + "learning_rate": 2.4122988028717454e-06, + "loss": 0.9055, + "num_input_tokens_seen": 71506225, + "step": 6000 + }, + { + "epoch": 0.45099954907560497, + "grad_norm": 2.6039519463073773, + "learning_rate": 2.4118224055841243e-06, + "loss": 0.9362, + "num_input_tokens_seen": 71525475, + "step": 6001 + }, + { + "epoch": 0.45107470314143994, + "grad_norm": 1.9023022656940451, + "learning_rate": 2.4113459838944496e-06, + "loss": 0.8866, + "num_input_tokens_seen": 71550190, + "step": 6002 + }, + { + "epoch": 0.4511498572072749, + "grad_norm": 1.2678030605025872, + "learning_rate": 2.4108695378309495e-06, + "loss": 0.9272, + "num_input_tokens_seen": 71574895, + "step": 6003 + }, + { + "epoch": 0.4512250112731099, + "grad_norm": 1.6557517111860278, + "learning_rate": 2.4103930674218565e-06, + "loss": 0.892, + "num_input_tokens_seen": 71598255, + "step": 6004 + }, + { + "epoch": 0.45130016533894485, + "grad_norm": 1.5275530659499894, + "learning_rate": 2.4099165726954026e-06, + "loss": 0.9499, + "num_input_tokens_seen": 71619025, + "step": 6005 + }, + { + "epoch": 0.4513753194047798, + "grad_norm": 1.5182318537466395, + "learning_rate": 2.409440053679822e-06, + "loss": 0.9415, + "num_input_tokens_seen": 71642955, + "step": 6006 + }, + { + "epoch": 0.45145047347061473, + "grad_norm": 2.5998688187884493, + "learning_rate": 2.40896351040335e-06, + "loss": 0.9696, + "num_input_tokens_seen": 71667400, + "step": 6007 + }, + { + "epoch": 0.4515256275364497, + "grad_norm": 1.5099509299433516, + "learning_rate": 2.4084869428942243e-06, + "loss": 0.97, + "num_input_tokens_seen": 71692935, + "step": 6008 + }, + { + "epoch": 0.4516007816022847, + "grad_norm": 1.366352502061317, + "learning_rate": 2.4080103511806836e-06, + "loss": 1.0115, + "num_input_tokens_seen": 71715695, + "step": 6009 + }, + { + "epoch": 0.45167593566811964, + "grad_norm": 1.5375257846562007, + "learning_rate": 2.4075337352909667e-06, + "loss": 1.0262, + "num_input_tokens_seen": 71739695, + "step": 6010 + }, + { + "epoch": 0.4517510897339546, + "grad_norm": 1.493399881486593, + "learning_rate": 2.4070570952533155e-06, + "loss": 0.9615, + "num_input_tokens_seen": 71765535, + "step": 6011 + }, + { + "epoch": 0.4518262437997896, + "grad_norm": 1.3918202439197698, + "learning_rate": 2.4065804310959725e-06, + "loss": 0.9631, + "num_input_tokens_seen": 71790070, + "step": 6012 + }, + { + "epoch": 0.45190139786562455, + "grad_norm": 2.570414625185484, + "learning_rate": 2.406103742847182e-06, + "loss": 1.1404, + "num_input_tokens_seen": 71804670, + "step": 6013 + }, + { + "epoch": 0.45197655193145947, + "grad_norm": 1.1968904937458418, + "learning_rate": 2.4056270305351896e-06, + "loss": 0.8814, + "num_input_tokens_seen": 71831275, + "step": 6014 + }, + { + "epoch": 0.45205170599729444, + "grad_norm": 0.8567327280276575, + "learning_rate": 2.4051502941882422e-06, + "loss": 0.8067, + "num_input_tokens_seen": 71900790, + "step": 6015 + }, + { + "epoch": 0.4521268600631294, + "grad_norm": 1.7778581817975108, + "learning_rate": 2.4046735338345897e-06, + "loss": 0.938, + "num_input_tokens_seen": 71923450, + "step": 6016 + }, + { + "epoch": 0.4522020141289644, + "grad_norm": 1.5642790012875, + "learning_rate": 2.4041967495024796e-06, + "loss": 0.9312, + "num_input_tokens_seen": 71946440, + "step": 6017 + }, + { + "epoch": 0.45227716819479935, + "grad_norm": 1.7106850494448607, + "learning_rate": 2.403719941220164e-06, + "loss": 0.9706, + "num_input_tokens_seen": 71967000, + "step": 6018 + }, + { + "epoch": 0.4523523222606343, + "grad_norm": 2.2431141071793803, + "learning_rate": 2.403243109015897e-06, + "loss": 0.7607, + "num_input_tokens_seen": 71989765, + "step": 6019 + }, + { + "epoch": 0.45242747632646924, + "grad_norm": 1.7221301739544288, + "learning_rate": 2.402766252917931e-06, + "loss": 1.0093, + "num_input_tokens_seen": 72012175, + "step": 6020 + }, + { + "epoch": 0.4525026303923042, + "grad_norm": 1.5972967139670549, + "learning_rate": 2.402289372954523e-06, + "loss": 1.015, + "num_input_tokens_seen": 72035420, + "step": 6021 + }, + { + "epoch": 0.4525777844581392, + "grad_norm": 1.577332401470985, + "learning_rate": 2.4018124691539286e-06, + "loss": 0.9406, + "num_input_tokens_seen": 72058815, + "step": 6022 + }, + { + "epoch": 0.45265293852397415, + "grad_norm": 1.614346279667113, + "learning_rate": 2.401335541544406e-06, + "loss": 0.9153, + "num_input_tokens_seen": 72085525, + "step": 6023 + }, + { + "epoch": 0.4527280925898091, + "grad_norm": 1.9661419260015673, + "learning_rate": 2.400858590154217e-06, + "loss": 0.8873, + "num_input_tokens_seen": 72110235, + "step": 6024 + }, + { + "epoch": 0.4528032466556441, + "grad_norm": 1.622427444599074, + "learning_rate": 2.400381615011621e-06, + "loss": 1.0175, + "num_input_tokens_seen": 72131500, + "step": 6025 + }, + { + "epoch": 0.45287840072147906, + "grad_norm": 1.5231746348987854, + "learning_rate": 2.399904616144881e-06, + "loss": 0.9831, + "num_input_tokens_seen": 72153900, + "step": 6026 + }, + { + "epoch": 0.452953554787314, + "grad_norm": 1.4481953505944136, + "learning_rate": 2.3994275935822618e-06, + "loss": 0.9368, + "num_input_tokens_seen": 72180265, + "step": 6027 + }, + { + "epoch": 0.45302870885314894, + "grad_norm": 2.167457424040453, + "learning_rate": 2.398950547352028e-06, + "loss": 0.9035, + "num_input_tokens_seen": 72200505, + "step": 6028 + }, + { + "epoch": 0.4531038629189839, + "grad_norm": 0.878044350066612, + "learning_rate": 2.398473477482446e-06, + "loss": 0.8285, + "num_input_tokens_seen": 72271180, + "step": 6029 + }, + { + "epoch": 0.4531790169848189, + "grad_norm": 1.8762745022428684, + "learning_rate": 2.397996384001785e-06, + "loss": 0.9136, + "num_input_tokens_seen": 72292200, + "step": 6030 + }, + { + "epoch": 0.45325417105065385, + "grad_norm": 1.4436976303576836, + "learning_rate": 2.397519266938314e-06, + "loss": 0.866, + "num_input_tokens_seen": 72317770, + "step": 6031 + }, + { + "epoch": 0.4533293251164888, + "grad_norm": 1.4522361908431944, + "learning_rate": 2.3970421263203045e-06, + "loss": 0.9338, + "num_input_tokens_seen": 72342015, + "step": 6032 + }, + { + "epoch": 0.45340447918232374, + "grad_norm": 1.6515208691102172, + "learning_rate": 2.396564962176028e-06, + "loss": 0.9768, + "num_input_tokens_seen": 72364235, + "step": 6033 + }, + { + "epoch": 0.4534796332481587, + "grad_norm": 1.499563074055466, + "learning_rate": 2.39608777453376e-06, + "loss": 0.9592, + "num_input_tokens_seen": 72385100, + "step": 6034 + }, + { + "epoch": 0.4535547873139937, + "grad_norm": 1.976349830494677, + "learning_rate": 2.395610563421774e-06, + "loss": 0.9647, + "num_input_tokens_seen": 72407980, + "step": 6035 + }, + { + "epoch": 0.45362994137982865, + "grad_norm": 1.5876085469539416, + "learning_rate": 2.3951333288683476e-06, + "loss": 1.0217, + "num_input_tokens_seen": 72430920, + "step": 6036 + }, + { + "epoch": 0.4537050954456636, + "grad_norm": 1.6283887594597348, + "learning_rate": 2.394656070901757e-06, + "loss": 0.9305, + "num_input_tokens_seen": 72458290, + "step": 6037 + }, + { + "epoch": 0.4537802495114986, + "grad_norm": 1.4232000527676192, + "learning_rate": 2.394178789550285e-06, + "loss": 0.9849, + "num_input_tokens_seen": 72482690, + "step": 6038 + }, + { + "epoch": 0.45385540357733356, + "grad_norm": 1.7348931715559393, + "learning_rate": 2.3937014848422094e-06, + "loss": 0.9679, + "num_input_tokens_seen": 72504685, + "step": 6039 + }, + { + "epoch": 0.4539305576431685, + "grad_norm": 1.4624205217202857, + "learning_rate": 2.3932241568058127e-06, + "loss": 1.0432, + "num_input_tokens_seen": 72528895, + "step": 6040 + }, + { + "epoch": 0.45400571170900345, + "grad_norm": 1.5471639652269022, + "learning_rate": 2.3927468054693797e-06, + "loss": 0.8679, + "num_input_tokens_seen": 72550790, + "step": 6041 + }, + { + "epoch": 0.4540808657748384, + "grad_norm": 1.4261007708646989, + "learning_rate": 2.392269430861194e-06, + "loss": 0.9432, + "num_input_tokens_seen": 72572745, + "step": 6042 + }, + { + "epoch": 0.4541560198406734, + "grad_norm": 0.7866270797946895, + "learning_rate": 2.391792033009543e-06, + "loss": 0.8408, + "num_input_tokens_seen": 72647590, + "step": 6043 + }, + { + "epoch": 0.45423117390650836, + "grad_norm": 1.4208611866895444, + "learning_rate": 2.391314611942714e-06, + "loss": 0.8145, + "num_input_tokens_seen": 72674220, + "step": 6044 + }, + { + "epoch": 0.4543063279723433, + "grad_norm": 1.435774804529561, + "learning_rate": 2.390837167688995e-06, + "loss": 0.9516, + "num_input_tokens_seen": 72698055, + "step": 6045 + }, + { + "epoch": 0.45438148203817824, + "grad_norm": 1.5015073196481699, + "learning_rate": 2.3903597002766777e-06, + "loss": 0.8602, + "num_input_tokens_seen": 72726955, + "step": 6046 + }, + { + "epoch": 0.4544566361040132, + "grad_norm": 1.2374992910055576, + "learning_rate": 2.3898822097340527e-06, + "loss": 0.9751, + "num_input_tokens_seen": 72754060, + "step": 6047 + }, + { + "epoch": 0.4545317901698482, + "grad_norm": 1.3363162836633495, + "learning_rate": 2.389404696089415e-06, + "loss": 0.9686, + "num_input_tokens_seen": 72777345, + "step": 6048 + }, + { + "epoch": 0.45460694423568315, + "grad_norm": 1.6075695243501933, + "learning_rate": 2.388927159371057e-06, + "loss": 1.1267, + "num_input_tokens_seen": 72796640, + "step": 6049 + }, + { + "epoch": 0.4546820983015181, + "grad_norm": 2.0847170113093125, + "learning_rate": 2.3884495996072755e-06, + "loss": 0.9097, + "num_input_tokens_seen": 72817755, + "step": 6050 + }, + { + "epoch": 0.4547572523673531, + "grad_norm": 1.310021330608813, + "learning_rate": 2.3879720168263683e-06, + "loss": 0.9241, + "num_input_tokens_seen": 72841720, + "step": 6051 + }, + { + "epoch": 0.454832406433188, + "grad_norm": 1.5424450722038665, + "learning_rate": 2.387494411056633e-06, + "loss": 1.028, + "num_input_tokens_seen": 72864085, + "step": 6052 + }, + { + "epoch": 0.454907560499023, + "grad_norm": 1.6442062527969081, + "learning_rate": 2.38701678232637e-06, + "loss": 0.9902, + "num_input_tokens_seen": 72886900, + "step": 6053 + }, + { + "epoch": 0.45498271456485795, + "grad_norm": 1.762046495895818, + "learning_rate": 2.386539130663881e-06, + "loss": 1.0562, + "num_input_tokens_seen": 72909335, + "step": 6054 + }, + { + "epoch": 0.4550578686306929, + "grad_norm": 1.7427616969312383, + "learning_rate": 2.386061456097468e-06, + "loss": 1.0073, + "num_input_tokens_seen": 72933155, + "step": 6055 + }, + { + "epoch": 0.4551330226965279, + "grad_norm": 1.4027502379321188, + "learning_rate": 2.3855837586554356e-06, + "loss": 0.962, + "num_input_tokens_seen": 72956795, + "step": 6056 + }, + { + "epoch": 0.45520817676236286, + "grad_norm": 1.5547004584150987, + "learning_rate": 2.3851060383660893e-06, + "loss": 0.9936, + "num_input_tokens_seen": 72980475, + "step": 6057 + }, + { + "epoch": 0.45528333082819783, + "grad_norm": 1.682401655477307, + "learning_rate": 2.3846282952577354e-06, + "loss": 0.8117, + "num_input_tokens_seen": 73004700, + "step": 6058 + }, + { + "epoch": 0.45535848489403274, + "grad_norm": 1.6202144582105427, + "learning_rate": 2.384150529358681e-06, + "loss": 1.0087, + "num_input_tokens_seen": 73027465, + "step": 6059 + }, + { + "epoch": 0.4554336389598677, + "grad_norm": 2.3973220070170735, + "learning_rate": 2.383672740697238e-06, + "loss": 1.0696, + "num_input_tokens_seen": 73050995, + "step": 6060 + }, + { + "epoch": 0.4555087930257027, + "grad_norm": 1.6764889803193466, + "learning_rate": 2.3831949293017166e-06, + "loss": 0.8879, + "num_input_tokens_seen": 73073965, + "step": 6061 + }, + { + "epoch": 0.45558394709153766, + "grad_norm": 1.6149496913904302, + "learning_rate": 2.3827170952004266e-06, + "loss": 0.9478, + "num_input_tokens_seen": 73096205, + "step": 6062 + }, + { + "epoch": 0.4556591011573726, + "grad_norm": 2.3511490062254876, + "learning_rate": 2.382239238421684e-06, + "loss": 1.0002, + "num_input_tokens_seen": 73120455, + "step": 6063 + }, + { + "epoch": 0.4557342552232076, + "grad_norm": 1.570510092682518, + "learning_rate": 2.3817613589938026e-06, + "loss": 0.883, + "num_input_tokens_seen": 73145085, + "step": 6064 + }, + { + "epoch": 0.4558094092890425, + "grad_norm": 1.6669511393324132, + "learning_rate": 2.381283456945099e-06, + "loss": 0.843, + "num_input_tokens_seen": 73171560, + "step": 6065 + }, + { + "epoch": 0.4558845633548775, + "grad_norm": 1.6017999240459415, + "learning_rate": 2.3808055323038907e-06, + "loss": 0.8633, + "num_input_tokens_seen": 73195320, + "step": 6066 + }, + { + "epoch": 0.45595971742071245, + "grad_norm": 2.353720653673104, + "learning_rate": 2.3803275850984963e-06, + "loss": 0.9173, + "num_input_tokens_seen": 73220385, + "step": 6067 + }, + { + "epoch": 0.4560348714865474, + "grad_norm": 1.2596680596000627, + "learning_rate": 2.3798496153572363e-06, + "loss": 0.9737, + "num_input_tokens_seen": 73245155, + "step": 6068 + }, + { + "epoch": 0.4561100255523824, + "grad_norm": 1.6247400662796012, + "learning_rate": 2.3793716231084313e-06, + "loss": 0.903, + "num_input_tokens_seen": 73270010, + "step": 6069 + }, + { + "epoch": 0.45618517961821736, + "grad_norm": 1.6439422494996414, + "learning_rate": 2.3788936083804058e-06, + "loss": 0.9057, + "num_input_tokens_seen": 73291735, + "step": 6070 + }, + { + "epoch": 0.45626033368405233, + "grad_norm": 1.703301446856344, + "learning_rate": 2.3784155712014827e-06, + "loss": 0.8618, + "num_input_tokens_seen": 73314775, + "step": 6071 + }, + { + "epoch": 0.45633548774988725, + "grad_norm": 1.9423610250671293, + "learning_rate": 2.3779375115999877e-06, + "loss": 0.9863, + "num_input_tokens_seen": 73338260, + "step": 6072 + }, + { + "epoch": 0.4564106418157222, + "grad_norm": 1.4730690294135846, + "learning_rate": 2.3774594296042485e-06, + "loss": 0.9544, + "num_input_tokens_seen": 73361220, + "step": 6073 + }, + { + "epoch": 0.4564857958815572, + "grad_norm": 1.3641186959313165, + "learning_rate": 2.376981325242592e-06, + "loss": 0.9048, + "num_input_tokens_seen": 73385490, + "step": 6074 + }, + { + "epoch": 0.45656094994739216, + "grad_norm": 1.3829590455760874, + "learning_rate": 2.376503198543349e-06, + "loss": 0.9799, + "num_input_tokens_seen": 73411175, + "step": 6075 + }, + { + "epoch": 0.45663610401322713, + "grad_norm": 1.5407633411746564, + "learning_rate": 2.3760250495348495e-06, + "loss": 0.9712, + "num_input_tokens_seen": 73434290, + "step": 6076 + }, + { + "epoch": 0.4567112580790621, + "grad_norm": 1.960771652442594, + "learning_rate": 2.3755468782454265e-06, + "loss": 0.9669, + "num_input_tokens_seen": 73456525, + "step": 6077 + }, + { + "epoch": 0.456786412144897, + "grad_norm": 1.231199501667331, + "learning_rate": 2.375068684703413e-06, + "loss": 0.9475, + "num_input_tokens_seen": 73480555, + "step": 6078 + }, + { + "epoch": 0.456861566210732, + "grad_norm": 1.8057646736268143, + "learning_rate": 2.3745904689371423e-06, + "loss": 1.0165, + "num_input_tokens_seen": 73501710, + "step": 6079 + }, + { + "epoch": 0.45693672027656695, + "grad_norm": 1.4353363505548946, + "learning_rate": 2.374112230974953e-06, + "loss": 0.8748, + "num_input_tokens_seen": 73522640, + "step": 6080 + }, + { + "epoch": 0.4570118743424019, + "grad_norm": 1.4573989690505562, + "learning_rate": 2.3736339708451803e-06, + "loss": 0.9525, + "num_input_tokens_seen": 73545520, + "step": 6081 + }, + { + "epoch": 0.4570870284082369, + "grad_norm": 2.059466587744894, + "learning_rate": 2.3731556885761656e-06, + "loss": 0.9606, + "num_input_tokens_seen": 73568500, + "step": 6082 + }, + { + "epoch": 0.45716218247407187, + "grad_norm": 1.335370350519042, + "learning_rate": 2.3726773841962472e-06, + "loss": 1.0373, + "num_input_tokens_seen": 73594500, + "step": 6083 + }, + { + "epoch": 0.45723733653990684, + "grad_norm": 1.4447016069341112, + "learning_rate": 2.372199057733766e-06, + "loss": 0.9024, + "num_input_tokens_seen": 73618365, + "step": 6084 + }, + { + "epoch": 0.45731249060574175, + "grad_norm": 1.5089052816030195, + "learning_rate": 2.371720709217066e-06, + "loss": 0.8812, + "num_input_tokens_seen": 73644760, + "step": 6085 + }, + { + "epoch": 0.4573876446715767, + "grad_norm": 0.9680876467192934, + "learning_rate": 2.3712423386744897e-06, + "loss": 0.8794, + "num_input_tokens_seen": 73717035, + "step": 6086 + }, + { + "epoch": 0.4574627987374117, + "grad_norm": 1.01549499853453, + "learning_rate": 2.370763946134384e-06, + "loss": 0.9415, + "num_input_tokens_seen": 73782705, + "step": 6087 + }, + { + "epoch": 0.45753795280324666, + "grad_norm": 1.408494175899924, + "learning_rate": 2.3702855316250943e-06, + "loss": 0.9296, + "num_input_tokens_seen": 73808275, + "step": 6088 + }, + { + "epoch": 0.45761310686908163, + "grad_norm": 1.7123978187538231, + "learning_rate": 2.3698070951749692e-06, + "loss": 0.9113, + "num_input_tokens_seen": 73834085, + "step": 6089 + }, + { + "epoch": 0.4576882609349166, + "grad_norm": 1.4844312155267994, + "learning_rate": 2.3693286368123576e-06, + "loss": 0.9461, + "num_input_tokens_seen": 73858230, + "step": 6090 + }, + { + "epoch": 0.4577634150007515, + "grad_norm": 1.8031983279651178, + "learning_rate": 2.3688501565656104e-06, + "loss": 1.1202, + "num_input_tokens_seen": 73877300, + "step": 6091 + }, + { + "epoch": 0.4578385690665865, + "grad_norm": 1.9192706193928144, + "learning_rate": 2.3683716544630784e-06, + "loss": 0.8923, + "num_input_tokens_seen": 73898355, + "step": 6092 + }, + { + "epoch": 0.45791372313242146, + "grad_norm": 1.9669630024062688, + "learning_rate": 2.367893130533116e-06, + "loss": 0.9313, + "num_input_tokens_seen": 73921900, + "step": 6093 + }, + { + "epoch": 0.4579888771982564, + "grad_norm": 2.1366740052944726, + "learning_rate": 2.367414584804076e-06, + "loss": 1.0409, + "num_input_tokens_seen": 73944580, + "step": 6094 + }, + { + "epoch": 0.4580640312640914, + "grad_norm": 1.4149572587045063, + "learning_rate": 2.3669360173043155e-06, + "loss": 1.0267, + "num_input_tokens_seen": 73969780, + "step": 6095 + }, + { + "epoch": 0.45813918532992637, + "grad_norm": 2.3351244069516217, + "learning_rate": 2.3664574280621907e-06, + "loss": 0.8866, + "num_input_tokens_seen": 73992250, + "step": 6096 + }, + { + "epoch": 0.4582143393957613, + "grad_norm": 1.6492472463194194, + "learning_rate": 2.36597881710606e-06, + "loss": 0.9624, + "num_input_tokens_seen": 74015000, + "step": 6097 + }, + { + "epoch": 0.45828949346159625, + "grad_norm": 1.381761442407281, + "learning_rate": 2.3655001844642828e-06, + "loss": 0.9874, + "num_input_tokens_seen": 74040290, + "step": 6098 + }, + { + "epoch": 0.4583646475274312, + "grad_norm": 1.9009167216449527, + "learning_rate": 2.3650215301652207e-06, + "loss": 1.0262, + "num_input_tokens_seen": 74060255, + "step": 6099 + }, + { + "epoch": 0.4584398015932662, + "grad_norm": 1.695085360872942, + "learning_rate": 2.3645428542372347e-06, + "loss": 0.9849, + "num_input_tokens_seen": 74083005, + "step": 6100 + }, + { + "epoch": 0.45851495565910116, + "grad_norm": 1.3456636154509367, + "learning_rate": 2.3640641567086887e-06, + "loss": 0.9009, + "num_input_tokens_seen": 74107740, + "step": 6101 + }, + { + "epoch": 0.45859010972493613, + "grad_norm": 1.7406467073412204, + "learning_rate": 2.363585437607947e-06, + "loss": 0.8988, + "num_input_tokens_seen": 74129315, + "step": 6102 + }, + { + "epoch": 0.4586652637907711, + "grad_norm": 1.9047107422860263, + "learning_rate": 2.3631066969633755e-06, + "loss": 1.0511, + "num_input_tokens_seen": 74150905, + "step": 6103 + }, + { + "epoch": 0.458740417856606, + "grad_norm": 1.9404972028193652, + "learning_rate": 2.362627934803343e-06, + "loss": 0.9778, + "num_input_tokens_seen": 74171020, + "step": 6104 + }, + { + "epoch": 0.458815571922441, + "grad_norm": 1.6761383885399295, + "learning_rate": 2.362149151156216e-06, + "loss": 0.9835, + "num_input_tokens_seen": 74195185, + "step": 6105 + }, + { + "epoch": 0.45889072598827596, + "grad_norm": 1.5721828273035747, + "learning_rate": 2.3616703460503654e-06, + "loss": 1.0781, + "num_input_tokens_seen": 74215955, + "step": 6106 + }, + { + "epoch": 0.45896588005411093, + "grad_norm": 1.6458741275544286, + "learning_rate": 2.3611915195141615e-06, + "loss": 0.8291, + "num_input_tokens_seen": 74243025, + "step": 6107 + }, + { + "epoch": 0.4590410341199459, + "grad_norm": 1.507541691470862, + "learning_rate": 2.3607126715759773e-06, + "loss": 0.924, + "num_input_tokens_seen": 74266000, + "step": 6108 + }, + { + "epoch": 0.45911618818578087, + "grad_norm": 1.517480559876558, + "learning_rate": 2.360233802264186e-06, + "loss": 1.0148, + "num_input_tokens_seen": 74290800, + "step": 6109 + }, + { + "epoch": 0.4591913422516158, + "grad_norm": 3.048392363057208, + "learning_rate": 2.359754911607163e-06, + "loss": 0.9343, + "num_input_tokens_seen": 74315825, + "step": 6110 + }, + { + "epoch": 0.45926649631745076, + "grad_norm": 1.281165655197658, + "learning_rate": 2.3592759996332824e-06, + "loss": 1.0145, + "num_input_tokens_seen": 74341250, + "step": 6111 + }, + { + "epoch": 0.4593416503832857, + "grad_norm": 1.344751162501363, + "learning_rate": 2.358797066370924e-06, + "loss": 0.8932, + "num_input_tokens_seen": 74367805, + "step": 6112 + }, + { + "epoch": 0.4594168044491207, + "grad_norm": 1.7147204239626752, + "learning_rate": 2.358318111848466e-06, + "loss": 0.9621, + "num_input_tokens_seen": 74390455, + "step": 6113 + }, + { + "epoch": 0.45949195851495567, + "grad_norm": 1.5479057807933203, + "learning_rate": 2.3578391360942872e-06, + "loss": 1.0113, + "num_input_tokens_seen": 74413180, + "step": 6114 + }, + { + "epoch": 0.45956711258079064, + "grad_norm": 1.43287894491135, + "learning_rate": 2.3573601391367696e-06, + "loss": 1.0118, + "num_input_tokens_seen": 74438770, + "step": 6115 + }, + { + "epoch": 0.4596422666466256, + "grad_norm": 1.4327078844522463, + "learning_rate": 2.3568811210042947e-06, + "loss": 0.9195, + "num_input_tokens_seen": 74462690, + "step": 6116 + }, + { + "epoch": 0.4597174207124605, + "grad_norm": 1.725824880329075, + "learning_rate": 2.3564020817252476e-06, + "loss": 0.8691, + "num_input_tokens_seen": 74485225, + "step": 6117 + }, + { + "epoch": 0.4597925747782955, + "grad_norm": 1.704028388695799, + "learning_rate": 2.3559230213280115e-06, + "loss": 0.9323, + "num_input_tokens_seen": 74513000, + "step": 6118 + }, + { + "epoch": 0.45986772884413046, + "grad_norm": 0.7714883538638496, + "learning_rate": 2.3554439398409743e-06, + "loss": 0.8118, + "num_input_tokens_seen": 74591575, + "step": 6119 + }, + { + "epoch": 0.45994288290996543, + "grad_norm": 1.4314993961718023, + "learning_rate": 2.354964837292522e-06, + "loss": 1.002, + "num_input_tokens_seen": 74615260, + "step": 6120 + }, + { + "epoch": 0.4600180369758004, + "grad_norm": 1.7698997130781868, + "learning_rate": 2.354485713711044e-06, + "loss": 0.8637, + "num_input_tokens_seen": 74639100, + "step": 6121 + }, + { + "epoch": 0.4600931910416354, + "grad_norm": 2.3596065988616837, + "learning_rate": 2.354006569124931e-06, + "loss": 0.8998, + "num_input_tokens_seen": 74660360, + "step": 6122 + }, + { + "epoch": 0.4601683451074703, + "grad_norm": 1.6402019863651092, + "learning_rate": 2.3535274035625713e-06, + "loss": 0.8494, + "num_input_tokens_seen": 74685445, + "step": 6123 + }, + { + "epoch": 0.46024349917330526, + "grad_norm": 1.5200563126723523, + "learning_rate": 2.353048217052361e-06, + "loss": 0.9823, + "num_input_tokens_seen": 74707815, + "step": 6124 + }, + { + "epoch": 0.46031865323914023, + "grad_norm": 1.6092265532834338, + "learning_rate": 2.3525690096226906e-06, + "loss": 0.9653, + "num_input_tokens_seen": 74731150, + "step": 6125 + }, + { + "epoch": 0.4603938073049752, + "grad_norm": 1.5907109025643564, + "learning_rate": 2.3520897813019566e-06, + "loss": 1.0432, + "num_input_tokens_seen": 74757900, + "step": 6126 + }, + { + "epoch": 0.46046896137081017, + "grad_norm": 1.4472092837591028, + "learning_rate": 2.351610532118555e-06, + "loss": 0.9681, + "num_input_tokens_seen": 74781325, + "step": 6127 + }, + { + "epoch": 0.46054411543664514, + "grad_norm": 1.599835315812139, + "learning_rate": 2.3511312621008832e-06, + "loss": 0.9983, + "num_input_tokens_seen": 74803695, + "step": 6128 + }, + { + "epoch": 0.4606192695024801, + "grad_norm": 1.6252169831000975, + "learning_rate": 2.35065197127734e-06, + "loss": 0.9404, + "num_input_tokens_seen": 74826890, + "step": 6129 + }, + { + "epoch": 0.460694423568315, + "grad_norm": 1.796925817683243, + "learning_rate": 2.350172659676323e-06, + "loss": 0.9834, + "num_input_tokens_seen": 74849920, + "step": 6130 + }, + { + "epoch": 0.46076957763415, + "grad_norm": 1.8727339720879745, + "learning_rate": 2.349693327326237e-06, + "loss": 0.9516, + "num_input_tokens_seen": 74872530, + "step": 6131 + }, + { + "epoch": 0.46084473169998497, + "grad_norm": 1.6549169826142618, + "learning_rate": 2.3492139742554816e-06, + "loss": 0.955, + "num_input_tokens_seen": 74895105, + "step": 6132 + }, + { + "epoch": 0.46091988576581994, + "grad_norm": 1.6369561726477082, + "learning_rate": 2.3487346004924605e-06, + "loss": 0.9311, + "num_input_tokens_seen": 74919730, + "step": 6133 + }, + { + "epoch": 0.4609950398316549, + "grad_norm": 1.6021492232260661, + "learning_rate": 2.34825520606558e-06, + "loss": 0.8434, + "num_input_tokens_seen": 74942535, + "step": 6134 + }, + { + "epoch": 0.4610701938974899, + "grad_norm": 1.765430811102917, + "learning_rate": 2.3477757910032434e-06, + "loss": 0.8818, + "num_input_tokens_seen": 74966125, + "step": 6135 + }, + { + "epoch": 0.4611453479633248, + "grad_norm": 2.06447934050142, + "learning_rate": 2.347296355333861e-06, + "loss": 0.892, + "num_input_tokens_seen": 74985195, + "step": 6136 + }, + { + "epoch": 0.46122050202915976, + "grad_norm": 1.7113332073081347, + "learning_rate": 2.346816899085839e-06, + "loss": 1.0432, + "num_input_tokens_seen": 75008740, + "step": 6137 + }, + { + "epoch": 0.46129565609499473, + "grad_norm": 1.8362419362991516, + "learning_rate": 2.346337422287587e-06, + "loss": 0.9188, + "num_input_tokens_seen": 75035285, + "step": 6138 + }, + { + "epoch": 0.4613708101608297, + "grad_norm": 1.6285973857541507, + "learning_rate": 2.3458579249675176e-06, + "loss": 0.9424, + "num_input_tokens_seen": 75056515, + "step": 6139 + }, + { + "epoch": 0.4614459642266647, + "grad_norm": 1.6508843219603937, + "learning_rate": 2.345378407154041e-06, + "loss": 0.9987, + "num_input_tokens_seen": 75077990, + "step": 6140 + }, + { + "epoch": 0.46152111829249964, + "grad_norm": 1.50023394985478, + "learning_rate": 2.344898868875572e-06, + "loss": 0.9439, + "num_input_tokens_seen": 75101680, + "step": 6141 + }, + { + "epoch": 0.46159627235833456, + "grad_norm": 1.3223004695746967, + "learning_rate": 2.3444193101605237e-06, + "loss": 0.9332, + "num_input_tokens_seen": 75127500, + "step": 6142 + }, + { + "epoch": 0.46167142642416953, + "grad_norm": 1.5876650116232613, + "learning_rate": 2.3439397310373126e-06, + "loss": 0.9515, + "num_input_tokens_seen": 75151295, + "step": 6143 + }, + { + "epoch": 0.4617465804900045, + "grad_norm": 0.7594125903209346, + "learning_rate": 2.343460131534356e-06, + "loss": 0.7514, + "num_input_tokens_seen": 75223375, + "step": 6144 + }, + { + "epoch": 0.46182173455583947, + "grad_norm": 1.4471511280725622, + "learning_rate": 2.34298051168007e-06, + "loss": 0.9549, + "num_input_tokens_seen": 75245980, + "step": 6145 + }, + { + "epoch": 0.46189688862167444, + "grad_norm": 1.4509425026920137, + "learning_rate": 2.3425008715028766e-06, + "loss": 1.0233, + "num_input_tokens_seen": 75272530, + "step": 6146 + }, + { + "epoch": 0.4619720426875094, + "grad_norm": 1.5412792371960762, + "learning_rate": 2.3420212110311943e-06, + "loss": 0.8536, + "num_input_tokens_seen": 75295280, + "step": 6147 + }, + { + "epoch": 0.4620471967533444, + "grad_norm": 1.4902270155658712, + "learning_rate": 2.3415415302934457e-06, + "loss": 1.0201, + "num_input_tokens_seen": 75318425, + "step": 6148 + }, + { + "epoch": 0.4621223508191793, + "grad_norm": 1.7674819498610777, + "learning_rate": 2.341061829318054e-06, + "loss": 0.9414, + "num_input_tokens_seen": 75340150, + "step": 6149 + }, + { + "epoch": 0.46219750488501427, + "grad_norm": 1.5828693195172707, + "learning_rate": 2.340582108133442e-06, + "loss": 0.9511, + "num_input_tokens_seen": 75363555, + "step": 6150 + }, + { + "epoch": 0.46227265895084924, + "grad_norm": 1.548396739557963, + "learning_rate": 2.340102366768037e-06, + "loss": 0.9322, + "num_input_tokens_seen": 75385170, + "step": 6151 + }, + { + "epoch": 0.4623478130166842, + "grad_norm": 1.6880695653384297, + "learning_rate": 2.339622605250264e-06, + "loss": 0.9764, + "num_input_tokens_seen": 75408785, + "step": 6152 + }, + { + "epoch": 0.4624229670825192, + "grad_norm": 1.5418718905933635, + "learning_rate": 2.339142823608551e-06, + "loss": 0.9841, + "num_input_tokens_seen": 75430005, + "step": 6153 + }, + { + "epoch": 0.46249812114835415, + "grad_norm": 1.629950172856294, + "learning_rate": 2.3386630218713273e-06, + "loss": 0.9361, + "num_input_tokens_seen": 75455120, + "step": 6154 + }, + { + "epoch": 0.46257327521418906, + "grad_norm": 1.6781266168057345, + "learning_rate": 2.3381832000670223e-06, + "loss": 0.9789, + "num_input_tokens_seen": 75477370, + "step": 6155 + }, + { + "epoch": 0.46264842928002403, + "grad_norm": 1.6780670866394387, + "learning_rate": 2.3377033582240684e-06, + "loss": 0.9685, + "num_input_tokens_seen": 75499605, + "step": 6156 + }, + { + "epoch": 0.462723583345859, + "grad_norm": 1.6610309764722677, + "learning_rate": 2.3372234963708966e-06, + "loss": 0.9118, + "num_input_tokens_seen": 75524125, + "step": 6157 + }, + { + "epoch": 0.462798737411694, + "grad_norm": 1.8438858370123385, + "learning_rate": 2.336743614535942e-06, + "loss": 0.9292, + "num_input_tokens_seen": 75544380, + "step": 6158 + }, + { + "epoch": 0.46287389147752894, + "grad_norm": 1.7975623764866213, + "learning_rate": 2.3362637127476383e-06, + "loss": 0.9682, + "num_input_tokens_seen": 75568615, + "step": 6159 + }, + { + "epoch": 0.4629490455433639, + "grad_norm": 1.7078279429973882, + "learning_rate": 2.335783791034422e-06, + "loss": 0.9037, + "num_input_tokens_seen": 75590235, + "step": 6160 + }, + { + "epoch": 0.4630241996091989, + "grad_norm": 2.339115596313851, + "learning_rate": 2.3353038494247305e-06, + "loss": 0.8916, + "num_input_tokens_seen": 75611000, + "step": 6161 + }, + { + "epoch": 0.4630993536750338, + "grad_norm": 1.8655491109010547, + "learning_rate": 2.3348238879470015e-06, + "loss": 0.8656, + "num_input_tokens_seen": 75635215, + "step": 6162 + }, + { + "epoch": 0.46317450774086877, + "grad_norm": 3.231528388598553, + "learning_rate": 2.334343906629676e-06, + "loss": 0.9115, + "num_input_tokens_seen": 75658115, + "step": 6163 + }, + { + "epoch": 0.46324966180670374, + "grad_norm": 1.5111876350704743, + "learning_rate": 2.3338639055011924e-06, + "loss": 0.89, + "num_input_tokens_seen": 75683645, + "step": 6164 + }, + { + "epoch": 0.4633248158725387, + "grad_norm": 1.370088213983344, + "learning_rate": 2.333383884589995e-06, + "loss": 0.9887, + "num_input_tokens_seen": 75706945, + "step": 6165 + }, + { + "epoch": 0.4633999699383737, + "grad_norm": 1.4723855333897935, + "learning_rate": 2.3329038439245257e-06, + "loss": 0.9357, + "num_input_tokens_seen": 75728330, + "step": 6166 + }, + { + "epoch": 0.46347512400420865, + "grad_norm": 1.99941835290259, + "learning_rate": 2.332423783533228e-06, + "loss": 1.0303, + "num_input_tokens_seen": 75746115, + "step": 6167 + }, + { + "epoch": 0.46355027807004356, + "grad_norm": 1.5235151222212133, + "learning_rate": 2.331943703444549e-06, + "loss": 0.8975, + "num_input_tokens_seen": 75770300, + "step": 6168 + }, + { + "epoch": 0.46362543213587853, + "grad_norm": 1.8052802495467806, + "learning_rate": 2.331463603686934e-06, + "loss": 1.0362, + "num_input_tokens_seen": 75793875, + "step": 6169 + }, + { + "epoch": 0.4637005862017135, + "grad_norm": 1.4216243816465957, + "learning_rate": 2.330983484288832e-06, + "loss": 0.8261, + "num_input_tokens_seen": 75822165, + "step": 6170 + }, + { + "epoch": 0.4637757402675485, + "grad_norm": 1.826967051976797, + "learning_rate": 2.3305033452786905e-06, + "loss": 1.0292, + "num_input_tokens_seen": 75840990, + "step": 6171 + }, + { + "epoch": 0.46385089433338345, + "grad_norm": 1.5872576250694355, + "learning_rate": 2.3300231866849606e-06, + "loss": 0.9781, + "num_input_tokens_seen": 75864555, + "step": 6172 + }, + { + "epoch": 0.4639260483992184, + "grad_norm": 1.4455354260880504, + "learning_rate": 2.3295430085360927e-06, + "loss": 0.8969, + "num_input_tokens_seen": 75888555, + "step": 6173 + }, + { + "epoch": 0.4640012024650534, + "grad_norm": 1.1530083243209344, + "learning_rate": 2.32906281086054e-06, + "loss": 1.0186, + "num_input_tokens_seen": 75914330, + "step": 6174 + }, + { + "epoch": 0.4640763565308883, + "grad_norm": 1.5698963581791616, + "learning_rate": 2.3285825936867556e-06, + "loss": 1.0135, + "num_input_tokens_seen": 75936665, + "step": 6175 + }, + { + "epoch": 0.46415151059672327, + "grad_norm": 1.5238269797103008, + "learning_rate": 2.328102357043194e-06, + "loss": 0.9949, + "num_input_tokens_seen": 75962075, + "step": 6176 + }, + { + "epoch": 0.46422666466255824, + "grad_norm": 1.5622634708668013, + "learning_rate": 2.3276221009583116e-06, + "loss": 0.9058, + "num_input_tokens_seen": 75988910, + "step": 6177 + }, + { + "epoch": 0.4643018187283932, + "grad_norm": 3.145016309395668, + "learning_rate": 2.327141825460566e-06, + "loss": 1.0022, + "num_input_tokens_seen": 76012355, + "step": 6178 + }, + { + "epoch": 0.4643769727942282, + "grad_norm": 2.0309934013856785, + "learning_rate": 2.3266615305784126e-06, + "loss": 0.923, + "num_input_tokens_seen": 76030830, + "step": 6179 + }, + { + "epoch": 0.46445212686006315, + "grad_norm": 1.6391288111158109, + "learning_rate": 2.3261812163403144e-06, + "loss": 1.0908, + "num_input_tokens_seen": 76054845, + "step": 6180 + }, + { + "epoch": 0.46452728092589807, + "grad_norm": 1.4627086490568642, + "learning_rate": 2.3257008827747294e-06, + "loss": 0.9785, + "num_input_tokens_seen": 76078035, + "step": 6181 + }, + { + "epoch": 0.46460243499173304, + "grad_norm": 1.7100163597305917, + "learning_rate": 2.32522052991012e-06, + "loss": 0.9754, + "num_input_tokens_seen": 76098090, + "step": 6182 + }, + { + "epoch": 0.464677589057568, + "grad_norm": 1.8252659146699644, + "learning_rate": 2.324740157774949e-06, + "loss": 0.9673, + "num_input_tokens_seen": 76119985, + "step": 6183 + }, + { + "epoch": 0.464752743123403, + "grad_norm": 1.7624667793551991, + "learning_rate": 2.3242597663976793e-06, + "loss": 0.8797, + "num_input_tokens_seen": 76147295, + "step": 6184 + }, + { + "epoch": 0.46482789718923795, + "grad_norm": 0.7617327566570786, + "learning_rate": 2.3237793558067776e-06, + "loss": 0.8512, + "num_input_tokens_seen": 76228625, + "step": 6185 + }, + { + "epoch": 0.4649030512550729, + "grad_norm": 1.8503408143850886, + "learning_rate": 2.3232989260307087e-06, + "loss": 1.0062, + "num_input_tokens_seen": 76247615, + "step": 6186 + }, + { + "epoch": 0.46497820532090783, + "grad_norm": 1.7077616999016938, + "learning_rate": 2.322818477097941e-06, + "loss": 1.0165, + "num_input_tokens_seen": 76267480, + "step": 6187 + }, + { + "epoch": 0.4650533593867428, + "grad_norm": 1.6468629956034113, + "learning_rate": 2.322338009036943e-06, + "loss": 0.9798, + "num_input_tokens_seen": 76290805, + "step": 6188 + }, + { + "epoch": 0.4651285134525778, + "grad_norm": 1.9856644893726303, + "learning_rate": 2.3218575218761816e-06, + "loss": 0.963, + "num_input_tokens_seen": 76317430, + "step": 6189 + }, + { + "epoch": 0.46520366751841274, + "grad_norm": 1.5006779092519802, + "learning_rate": 2.3213770156441314e-06, + "loss": 0.9396, + "num_input_tokens_seen": 76340345, + "step": 6190 + }, + { + "epoch": 0.4652788215842477, + "grad_norm": 1.3730180935400282, + "learning_rate": 2.3208964903692613e-06, + "loss": 0.8798, + "num_input_tokens_seen": 76368040, + "step": 6191 + }, + { + "epoch": 0.4653539756500827, + "grad_norm": 1.3762676291241822, + "learning_rate": 2.3204159460800458e-06, + "loss": 0.9319, + "num_input_tokens_seen": 76392250, + "step": 6192 + }, + { + "epoch": 0.46542912971591766, + "grad_norm": 1.5359616096288546, + "learning_rate": 2.319935382804959e-06, + "loss": 1.0084, + "num_input_tokens_seen": 76412500, + "step": 6193 + }, + { + "epoch": 0.46550428378175257, + "grad_norm": 1.8076691540173335, + "learning_rate": 2.3194548005724748e-06, + "loss": 0.958, + "num_input_tokens_seen": 76434640, + "step": 6194 + }, + { + "epoch": 0.46557943784758754, + "grad_norm": 1.6952193274142848, + "learning_rate": 2.318974199411071e-06, + "loss": 0.9601, + "num_input_tokens_seen": 76456645, + "step": 6195 + }, + { + "epoch": 0.4656545919134225, + "grad_norm": 1.873936033372439, + "learning_rate": 2.318493579349224e-06, + "loss": 0.9258, + "num_input_tokens_seen": 76484025, + "step": 6196 + }, + { + "epoch": 0.4657297459792575, + "grad_norm": 1.4068243761076118, + "learning_rate": 2.3180129404154133e-06, + "loss": 0.9234, + "num_input_tokens_seen": 76507925, + "step": 6197 + }, + { + "epoch": 0.46580490004509245, + "grad_norm": 1.7247000613189738, + "learning_rate": 2.317532282638118e-06, + "loss": 0.9645, + "num_input_tokens_seen": 76529845, + "step": 6198 + }, + { + "epoch": 0.4658800541109274, + "grad_norm": 1.6430703966565585, + "learning_rate": 2.3170516060458188e-06, + "loss": 0.9159, + "num_input_tokens_seen": 76553000, + "step": 6199 + }, + { + "epoch": 0.46595520817676234, + "grad_norm": 1.761876135179926, + "learning_rate": 2.3165709106669983e-06, + "loss": 0.9353, + "num_input_tokens_seen": 76573625, + "step": 6200 + }, + { + "epoch": 0.4660303622425973, + "grad_norm": 1.767344287558248, + "learning_rate": 2.3160901965301386e-06, + "loss": 0.8866, + "num_input_tokens_seen": 76598500, + "step": 6201 + }, + { + "epoch": 0.4661055163084323, + "grad_norm": 0.887209181082288, + "learning_rate": 2.315609463663725e-06, + "loss": 0.7703, + "num_input_tokens_seen": 76667565, + "step": 6202 + }, + { + "epoch": 0.46618067037426725, + "grad_norm": 1.438630447391966, + "learning_rate": 2.315128712096242e-06, + "loss": 0.7888, + "num_input_tokens_seen": 76692015, + "step": 6203 + }, + { + "epoch": 0.4662558244401022, + "grad_norm": 1.4551262969165348, + "learning_rate": 2.314647941856175e-06, + "loss": 0.9167, + "num_input_tokens_seen": 76711215, + "step": 6204 + }, + { + "epoch": 0.4663309785059372, + "grad_norm": 1.640220228853249, + "learning_rate": 2.314167152972014e-06, + "loss": 0.9697, + "num_input_tokens_seen": 76733540, + "step": 6205 + }, + { + "epoch": 0.46640613257177216, + "grad_norm": 1.457713113322355, + "learning_rate": 2.313686345472245e-06, + "loss": 0.9814, + "num_input_tokens_seen": 76758825, + "step": 6206 + }, + { + "epoch": 0.4664812866376071, + "grad_norm": 1.374816665564718, + "learning_rate": 2.3132055193853597e-06, + "loss": 0.9137, + "num_input_tokens_seen": 76785740, + "step": 6207 + }, + { + "epoch": 0.46655644070344204, + "grad_norm": 1.9586733326760624, + "learning_rate": 2.312724674739847e-06, + "loss": 0.9211, + "num_input_tokens_seen": 76810705, + "step": 6208 + }, + { + "epoch": 0.466631594769277, + "grad_norm": 2.487608335043252, + "learning_rate": 2.3122438115642013e-06, + "loss": 1.0502, + "num_input_tokens_seen": 76834260, + "step": 6209 + }, + { + "epoch": 0.466706748835112, + "grad_norm": 1.6162459724063358, + "learning_rate": 2.3117629298869135e-06, + "loss": 0.9717, + "num_input_tokens_seen": 76855355, + "step": 6210 + }, + { + "epoch": 0.46678190290094695, + "grad_norm": 1.6365934701829614, + "learning_rate": 2.3112820297364775e-06, + "loss": 0.9575, + "num_input_tokens_seen": 76877910, + "step": 6211 + }, + { + "epoch": 0.4668570569667819, + "grad_norm": 0.7994553917612525, + "learning_rate": 2.3108011111413904e-06, + "loss": 0.7576, + "num_input_tokens_seen": 76949565, + "step": 6212 + }, + { + "epoch": 0.46693221103261684, + "grad_norm": 2.5758495114658766, + "learning_rate": 2.3103201741301465e-06, + "loss": 1.0217, + "num_input_tokens_seen": 76973215, + "step": 6213 + }, + { + "epoch": 0.4670073650984518, + "grad_norm": 1.7549453749673414, + "learning_rate": 2.3098392187312445e-06, + "loss": 0.9902, + "num_input_tokens_seen": 76997495, + "step": 6214 + }, + { + "epoch": 0.4670825191642868, + "grad_norm": 1.8394745863340762, + "learning_rate": 2.309358244973182e-06, + "loss": 0.9709, + "num_input_tokens_seen": 77019515, + "step": 6215 + }, + { + "epoch": 0.46715767323012175, + "grad_norm": 1.5648797795246874, + "learning_rate": 2.3088772528844588e-06, + "loss": 0.8928, + "num_input_tokens_seen": 77043045, + "step": 6216 + }, + { + "epoch": 0.4672328272959567, + "grad_norm": 1.4920442522405026, + "learning_rate": 2.308396242493576e-06, + "loss": 0.9694, + "num_input_tokens_seen": 77065285, + "step": 6217 + }, + { + "epoch": 0.4673079813617917, + "grad_norm": 1.3177932088705022, + "learning_rate": 2.3079152138290347e-06, + "loss": 0.945, + "num_input_tokens_seen": 77090875, + "step": 6218 + }, + { + "epoch": 0.46738313542762666, + "grad_norm": 1.6925662250043247, + "learning_rate": 2.307434166919338e-06, + "loss": 0.904, + "num_input_tokens_seen": 77112450, + "step": 6219 + }, + { + "epoch": 0.4674582894934616, + "grad_norm": 1.7645152157273123, + "learning_rate": 2.30695310179299e-06, + "loss": 0.9439, + "num_input_tokens_seen": 77134220, + "step": 6220 + }, + { + "epoch": 0.46753344355929655, + "grad_norm": 1.4275926762308306, + "learning_rate": 2.3064720184784946e-06, + "loss": 0.9615, + "num_input_tokens_seen": 77157995, + "step": 6221 + }, + { + "epoch": 0.4676085976251315, + "grad_norm": 1.48473418056207, + "learning_rate": 2.305990917004359e-06, + "loss": 0.93, + "num_input_tokens_seen": 77182170, + "step": 6222 + }, + { + "epoch": 0.4676837516909665, + "grad_norm": 1.7048503990208002, + "learning_rate": 2.3055097973990894e-06, + "loss": 1.0404, + "num_input_tokens_seen": 77203845, + "step": 6223 + }, + { + "epoch": 0.46775890575680146, + "grad_norm": 1.2437271077718646, + "learning_rate": 2.305028659691195e-06, + "loss": 0.9669, + "num_input_tokens_seen": 77229475, + "step": 6224 + }, + { + "epoch": 0.4678340598226364, + "grad_norm": 1.5809102926120582, + "learning_rate": 2.3045475039091846e-06, + "loss": 0.858, + "num_input_tokens_seen": 77253720, + "step": 6225 + }, + { + "epoch": 0.46790921388847134, + "grad_norm": 1.5353155077767566, + "learning_rate": 2.3040663300815673e-06, + "loss": 0.843, + "num_input_tokens_seen": 77277170, + "step": 6226 + }, + { + "epoch": 0.4679843679543063, + "grad_norm": 1.5294168359469575, + "learning_rate": 2.303585138236857e-06, + "loss": 0.9619, + "num_input_tokens_seen": 77300260, + "step": 6227 + }, + { + "epoch": 0.4680595220201413, + "grad_norm": 1.4556444538105533, + "learning_rate": 2.3031039284035636e-06, + "loss": 0.9061, + "num_input_tokens_seen": 77324380, + "step": 6228 + }, + { + "epoch": 0.46813467608597625, + "grad_norm": 1.4175870187143509, + "learning_rate": 2.3026227006102025e-06, + "loss": 0.8988, + "num_input_tokens_seen": 77350440, + "step": 6229 + }, + { + "epoch": 0.4682098301518112, + "grad_norm": 1.7139187970772867, + "learning_rate": 2.3021414548852864e-06, + "loss": 0.9246, + "num_input_tokens_seen": 77370140, + "step": 6230 + }, + { + "epoch": 0.4682849842176462, + "grad_norm": 1.300075717701668, + "learning_rate": 2.3016601912573333e-06, + "loss": 1.0247, + "num_input_tokens_seen": 77393870, + "step": 6231 + }, + { + "epoch": 0.4683601382834811, + "grad_norm": 1.3186479718844157, + "learning_rate": 2.301178909754859e-06, + "loss": 0.9896, + "num_input_tokens_seen": 77419845, + "step": 6232 + }, + { + "epoch": 0.4684352923493161, + "grad_norm": 1.4827902113884417, + "learning_rate": 2.30069761040638e-06, + "loss": 1.0126, + "num_input_tokens_seen": 77444775, + "step": 6233 + }, + { + "epoch": 0.46851044641515105, + "grad_norm": 1.464270721252456, + "learning_rate": 2.300216293240417e-06, + "loss": 0.8716, + "num_input_tokens_seen": 77471760, + "step": 6234 + }, + { + "epoch": 0.468585600480986, + "grad_norm": 1.566847437040599, + "learning_rate": 2.299734958285488e-06, + "loss": 1.0058, + "num_input_tokens_seen": 77495935, + "step": 6235 + }, + { + "epoch": 0.468660754546821, + "grad_norm": 1.8836626847958111, + "learning_rate": 2.2992536055701157e-06, + "loss": 1.0593, + "num_input_tokens_seen": 77511730, + "step": 6236 + }, + { + "epoch": 0.46873590861265596, + "grad_norm": 1.7547574770689731, + "learning_rate": 2.2987722351228216e-06, + "loss": 0.9529, + "num_input_tokens_seen": 77535180, + "step": 6237 + }, + { + "epoch": 0.46881106267849093, + "grad_norm": 1.8697339813203189, + "learning_rate": 2.298290846972128e-06, + "loss": 0.8966, + "num_input_tokens_seen": 77559690, + "step": 6238 + }, + { + "epoch": 0.46888621674432585, + "grad_norm": 1.5460343051178775, + "learning_rate": 2.29780944114656e-06, + "loss": 1.0586, + "num_input_tokens_seen": 77583560, + "step": 6239 + }, + { + "epoch": 0.4689613708101608, + "grad_norm": 1.6572390787959141, + "learning_rate": 2.2973280176746413e-06, + "loss": 0.9199, + "num_input_tokens_seen": 77607145, + "step": 6240 + }, + { + "epoch": 0.4690365248759958, + "grad_norm": 1.7600782923190585, + "learning_rate": 2.2968465765849e-06, + "loss": 0.8385, + "num_input_tokens_seen": 77630645, + "step": 6241 + }, + { + "epoch": 0.46911167894183076, + "grad_norm": 1.5231012217747875, + "learning_rate": 2.296365117905862e-06, + "loss": 0.988, + "num_input_tokens_seen": 77655755, + "step": 6242 + }, + { + "epoch": 0.4691868330076657, + "grad_norm": 1.5015127977606415, + "learning_rate": 2.2958836416660556e-06, + "loss": 0.9564, + "num_input_tokens_seen": 77681230, + "step": 6243 + }, + { + "epoch": 0.4692619870735007, + "grad_norm": 1.4778884504858503, + "learning_rate": 2.295402147894011e-06, + "loss": 1.0317, + "num_input_tokens_seen": 77705165, + "step": 6244 + }, + { + "epoch": 0.4693371411393356, + "grad_norm": 2.0045570670054746, + "learning_rate": 2.294920636618257e-06, + "loss": 0.9629, + "num_input_tokens_seen": 77728900, + "step": 6245 + }, + { + "epoch": 0.4694122952051706, + "grad_norm": 1.7106212166858297, + "learning_rate": 2.2944391078673267e-06, + "loss": 0.9571, + "num_input_tokens_seen": 77752615, + "step": 6246 + }, + { + "epoch": 0.46948744927100555, + "grad_norm": 1.536908263800179, + "learning_rate": 2.2939575616697516e-06, + "loss": 0.8917, + "num_input_tokens_seen": 77776545, + "step": 6247 + }, + { + "epoch": 0.4695626033368405, + "grad_norm": 1.3341678253079843, + "learning_rate": 2.2934759980540654e-06, + "loss": 0.9477, + "num_input_tokens_seen": 77799295, + "step": 6248 + }, + { + "epoch": 0.4696377574026755, + "grad_norm": 1.6439537792298233, + "learning_rate": 2.2929944170488025e-06, + "loss": 1.0035, + "num_input_tokens_seen": 77820105, + "step": 6249 + }, + { + "epoch": 0.46971291146851046, + "grad_norm": 1.2253475610901854, + "learning_rate": 2.2925128186824983e-06, + "loss": 0.7985, + "num_input_tokens_seen": 77899100, + "step": 6250 + }, + { + "epoch": 0.46978806553434543, + "grad_norm": 1.9729255719035927, + "learning_rate": 2.29203120298369e-06, + "loss": 0.8853, + "num_input_tokens_seen": 77921475, + "step": 6251 + }, + { + "epoch": 0.46986321960018035, + "grad_norm": 1.531439866731446, + "learning_rate": 2.2915495699809134e-06, + "loss": 0.918, + "num_input_tokens_seen": 77940970, + "step": 6252 + }, + { + "epoch": 0.4699383736660153, + "grad_norm": 0.9392934810269663, + "learning_rate": 2.2910679197027093e-06, + "loss": 0.9144, + "num_input_tokens_seen": 78017370, + "step": 6253 + }, + { + "epoch": 0.4700135277318503, + "grad_norm": 1.5367165903406914, + "learning_rate": 2.290586252177617e-06, + "loss": 0.965, + "num_input_tokens_seen": 78040000, + "step": 6254 + }, + { + "epoch": 0.47008868179768526, + "grad_norm": 1.6537006839656436, + "learning_rate": 2.290104567434175e-06, + "loss": 1.0144, + "num_input_tokens_seen": 78061740, + "step": 6255 + }, + { + "epoch": 0.47016383586352023, + "grad_norm": 1.39928768463124, + "learning_rate": 2.2896228655009276e-06, + "loss": 0.8518, + "num_input_tokens_seen": 78086720, + "step": 6256 + }, + { + "epoch": 0.4702389899293552, + "grad_norm": 1.4889652152369675, + "learning_rate": 2.2891411464064155e-06, + "loss": 0.926, + "num_input_tokens_seen": 78110725, + "step": 6257 + }, + { + "epoch": 0.4703141439951901, + "grad_norm": 1.4897427329116466, + "learning_rate": 2.2886594101791845e-06, + "loss": 0.948, + "num_input_tokens_seen": 78133555, + "step": 6258 + }, + { + "epoch": 0.4703892980610251, + "grad_norm": 1.5551914950883907, + "learning_rate": 2.2881776568477777e-06, + "loss": 0.9234, + "num_input_tokens_seen": 78156030, + "step": 6259 + }, + { + "epoch": 0.47046445212686006, + "grad_norm": 1.6829180501069725, + "learning_rate": 2.2876958864407407e-06, + "loss": 0.964, + "num_input_tokens_seen": 78175900, + "step": 6260 + }, + { + "epoch": 0.470539606192695, + "grad_norm": 2.170099973880688, + "learning_rate": 2.287214098986621e-06, + "loss": 0.9905, + "num_input_tokens_seen": 78195150, + "step": 6261 + }, + { + "epoch": 0.47061476025853, + "grad_norm": 1.5305853199050319, + "learning_rate": 2.286732294513966e-06, + "loss": 0.9568, + "num_input_tokens_seen": 78217445, + "step": 6262 + }, + { + "epoch": 0.47068991432436497, + "grad_norm": 0.8012123994143, + "learning_rate": 2.286250473051325e-06, + "loss": 0.7974, + "num_input_tokens_seen": 78291850, + "step": 6263 + }, + { + "epoch": 0.47076506839019994, + "grad_norm": 1.799906023539284, + "learning_rate": 2.2857686346272475e-06, + "loss": 0.9737, + "num_input_tokens_seen": 78311710, + "step": 6264 + }, + { + "epoch": 0.47084022245603485, + "grad_norm": 1.499675795169024, + "learning_rate": 2.2852867792702835e-06, + "loss": 0.9134, + "num_input_tokens_seen": 78334820, + "step": 6265 + }, + { + "epoch": 0.4709153765218698, + "grad_norm": 1.3286552716972568, + "learning_rate": 2.284804907008986e-06, + "loss": 0.9712, + "num_input_tokens_seen": 78359790, + "step": 6266 + }, + { + "epoch": 0.4709905305877048, + "grad_norm": 1.657912373747744, + "learning_rate": 2.2843230178719063e-06, + "loss": 0.9272, + "num_input_tokens_seen": 78383330, + "step": 6267 + }, + { + "epoch": 0.47106568465353976, + "grad_norm": 0.8677678271379943, + "learning_rate": 2.2838411118875997e-06, + "loss": 0.7818, + "num_input_tokens_seen": 78461510, + "step": 6268 + }, + { + "epoch": 0.47114083871937473, + "grad_norm": 2.153992507751526, + "learning_rate": 2.2833591890846204e-06, + "loss": 0.8536, + "num_input_tokens_seen": 78486570, + "step": 6269 + }, + { + "epoch": 0.4712159927852097, + "grad_norm": 1.5579962098855376, + "learning_rate": 2.282877249491523e-06, + "loss": 0.8653, + "num_input_tokens_seen": 78512310, + "step": 6270 + }, + { + "epoch": 0.4712911468510446, + "grad_norm": 1.0357265412963221, + "learning_rate": 2.2823952931368667e-06, + "loss": 0.8429, + "num_input_tokens_seen": 78571195, + "step": 6271 + }, + { + "epoch": 0.4713663009168796, + "grad_norm": 1.4728899297306397, + "learning_rate": 2.2819133200492073e-06, + "loss": 0.886, + "num_input_tokens_seen": 78594890, + "step": 6272 + }, + { + "epoch": 0.47144145498271456, + "grad_norm": 3.5473049088130084, + "learning_rate": 2.281431330257105e-06, + "loss": 0.9831, + "num_input_tokens_seen": 78618330, + "step": 6273 + }, + { + "epoch": 0.47151660904854953, + "grad_norm": 1.2685715554564398, + "learning_rate": 2.280949323789117e-06, + "loss": 0.9428, + "num_input_tokens_seen": 78643595, + "step": 6274 + }, + { + "epoch": 0.4715917631143845, + "grad_norm": 1.2825692990562787, + "learning_rate": 2.280467300673807e-06, + "loss": 0.9139, + "num_input_tokens_seen": 78669005, + "step": 6275 + }, + { + "epoch": 0.47166691718021947, + "grad_norm": 0.9088438665613381, + "learning_rate": 2.2799852609397353e-06, + "loss": 0.7908, + "num_input_tokens_seen": 78743925, + "step": 6276 + }, + { + "epoch": 0.4717420712460544, + "grad_norm": 1.633704858811513, + "learning_rate": 2.2795032046154644e-06, + "loss": 0.9383, + "num_input_tokens_seen": 78768655, + "step": 6277 + }, + { + "epoch": 0.47181722531188935, + "grad_norm": 1.5058902484979575, + "learning_rate": 2.279021131729559e-06, + "loss": 0.9128, + "num_input_tokens_seen": 78792735, + "step": 6278 + }, + { + "epoch": 0.4718923793777243, + "grad_norm": 1.4351552000753167, + "learning_rate": 2.2785390423105822e-06, + "loss": 1.0411, + "num_input_tokens_seen": 78815475, + "step": 6279 + }, + { + "epoch": 0.4719675334435593, + "grad_norm": 1.7443033550424167, + "learning_rate": 2.2780569363871016e-06, + "loss": 1.0097, + "num_input_tokens_seen": 78840685, + "step": 6280 + }, + { + "epoch": 0.47204268750939427, + "grad_norm": 1.058903159781117, + "learning_rate": 2.277574813987682e-06, + "loss": 0.8288, + "num_input_tokens_seen": 78914505, + "step": 6281 + }, + { + "epoch": 0.47211784157522924, + "grad_norm": 1.9760999175960512, + "learning_rate": 2.2770926751408916e-06, + "loss": 0.9562, + "num_input_tokens_seen": 78935850, + "step": 6282 + }, + { + "epoch": 0.4721929956410642, + "grad_norm": 1.4835170474922745, + "learning_rate": 2.2766105198753e-06, + "loss": 1.0267, + "num_input_tokens_seen": 78959960, + "step": 6283 + }, + { + "epoch": 0.4722681497068991, + "grad_norm": 1.6432276835084763, + "learning_rate": 2.2761283482194747e-06, + "loss": 0.9491, + "num_input_tokens_seen": 78980305, + "step": 6284 + }, + { + "epoch": 0.4723433037727341, + "grad_norm": 1.5836071313067783, + "learning_rate": 2.2756461602019886e-06, + "loss": 0.8573, + "num_input_tokens_seen": 79004120, + "step": 6285 + }, + { + "epoch": 0.47241845783856906, + "grad_norm": 1.165274993464756, + "learning_rate": 2.2751639558514117e-06, + "loss": 0.8987, + "num_input_tokens_seen": 79028890, + "step": 6286 + }, + { + "epoch": 0.47249361190440403, + "grad_norm": 1.530029121570568, + "learning_rate": 2.2746817351963163e-06, + "loss": 0.9145, + "num_input_tokens_seen": 79054105, + "step": 6287 + }, + { + "epoch": 0.472568765970239, + "grad_norm": 1.4689675028032418, + "learning_rate": 2.274199498265276e-06, + "loss": 0.8843, + "num_input_tokens_seen": 79077075, + "step": 6288 + }, + { + "epoch": 0.472643920036074, + "grad_norm": 2.07640520796074, + "learning_rate": 2.2737172450868663e-06, + "loss": 0.9702, + "num_input_tokens_seen": 79098220, + "step": 6289 + }, + { + "epoch": 0.4727190741019089, + "grad_norm": 1.5365712097430058, + "learning_rate": 2.2732349756896615e-06, + "loss": 1.0202, + "num_input_tokens_seen": 79122160, + "step": 6290 + }, + { + "epoch": 0.47279422816774386, + "grad_norm": 1.4371538989507733, + "learning_rate": 2.272752690102238e-06, + "loss": 0.9484, + "num_input_tokens_seen": 79146315, + "step": 6291 + }, + { + "epoch": 0.4728693822335788, + "grad_norm": 1.7180544746386617, + "learning_rate": 2.272270388353173e-06, + "loss": 0.9287, + "num_input_tokens_seen": 79170695, + "step": 6292 + }, + { + "epoch": 0.4729445362994138, + "grad_norm": 4.95267154771995, + "learning_rate": 2.2717880704710453e-06, + "loss": 0.9671, + "num_input_tokens_seen": 79194580, + "step": 6293 + }, + { + "epoch": 0.47301969036524877, + "grad_norm": 1.5208140942475792, + "learning_rate": 2.2713057364844323e-06, + "loss": 0.843, + "num_input_tokens_seen": 79220055, + "step": 6294 + }, + { + "epoch": 0.47309484443108374, + "grad_norm": 1.5518367836064788, + "learning_rate": 2.2708233864219175e-06, + "loss": 0.9315, + "num_input_tokens_seen": 79242830, + "step": 6295 + }, + { + "epoch": 0.4731699984969187, + "grad_norm": 1.4137233517105863, + "learning_rate": 2.270341020312078e-06, + "loss": 0.9553, + "num_input_tokens_seen": 79268340, + "step": 6296 + }, + { + "epoch": 0.4732451525627536, + "grad_norm": 1.7084231081688366, + "learning_rate": 2.2698586381834993e-06, + "loss": 0.894, + "num_input_tokens_seen": 79291360, + "step": 6297 + }, + { + "epoch": 0.4733203066285886, + "grad_norm": 1.8446656878934873, + "learning_rate": 2.269376240064763e-06, + "loss": 0.9403, + "num_input_tokens_seen": 79313510, + "step": 6298 + }, + { + "epoch": 0.47339546069442356, + "grad_norm": 2.16424416561074, + "learning_rate": 2.2688938259844525e-06, + "loss": 1.0126, + "num_input_tokens_seen": 79335505, + "step": 6299 + }, + { + "epoch": 0.47347061476025853, + "grad_norm": 1.3108571079051985, + "learning_rate": 2.268411395971153e-06, + "loss": 0.8849, + "num_input_tokens_seen": 79358525, + "step": 6300 + }, + { + "epoch": 0.4735457688260935, + "grad_norm": 1.6303905429919583, + "learning_rate": 2.2679289500534504e-06, + "loss": 0.932, + "num_input_tokens_seen": 79381215, + "step": 6301 + }, + { + "epoch": 0.4736209228919285, + "grad_norm": 1.4189928628367001, + "learning_rate": 2.267446488259932e-06, + "loss": 0.7852, + "num_input_tokens_seen": 79405235, + "step": 6302 + }, + { + "epoch": 0.4736960769577634, + "grad_norm": 1.3962330838837602, + "learning_rate": 2.266964010619185e-06, + "loss": 1.0082, + "num_input_tokens_seen": 79428215, + "step": 6303 + }, + { + "epoch": 0.47377123102359836, + "grad_norm": 1.746108769772509, + "learning_rate": 2.2664815171597983e-06, + "loss": 0.9869, + "num_input_tokens_seen": 79451145, + "step": 6304 + }, + { + "epoch": 0.47384638508943333, + "grad_norm": 1.4672653323848028, + "learning_rate": 2.2659990079103604e-06, + "loss": 0.9915, + "num_input_tokens_seen": 79473975, + "step": 6305 + }, + { + "epoch": 0.4739215391552683, + "grad_norm": 1.4920378604973779, + "learning_rate": 2.2655164828994635e-06, + "loss": 1.0281, + "num_input_tokens_seen": 79496190, + "step": 6306 + }, + { + "epoch": 0.47399669322110327, + "grad_norm": 1.7043891205043606, + "learning_rate": 2.2650339421556982e-06, + "loss": 1.0176, + "num_input_tokens_seen": 79517985, + "step": 6307 + }, + { + "epoch": 0.47407184728693824, + "grad_norm": 1.6402121614831187, + "learning_rate": 2.2645513857076567e-06, + "loss": 0.9062, + "num_input_tokens_seen": 79541995, + "step": 6308 + }, + { + "epoch": 0.4741470013527732, + "grad_norm": 1.4711247178037803, + "learning_rate": 2.2640688135839326e-06, + "loss": 0.8147, + "num_input_tokens_seen": 79565385, + "step": 6309 + }, + { + "epoch": 0.4742221554186081, + "grad_norm": 1.5222237471335416, + "learning_rate": 2.26358622581312e-06, + "loss": 0.979, + "num_input_tokens_seen": 79585735, + "step": 6310 + }, + { + "epoch": 0.4742973094844431, + "grad_norm": 1.6437028624299244, + "learning_rate": 2.2631036224238144e-06, + "loss": 0.9408, + "num_input_tokens_seen": 79610690, + "step": 6311 + }, + { + "epoch": 0.47437246355027807, + "grad_norm": 1.4685941572906127, + "learning_rate": 2.262621003444611e-06, + "loss": 0.9479, + "num_input_tokens_seen": 79635820, + "step": 6312 + }, + { + "epoch": 0.47444761761611304, + "grad_norm": 1.4432545625160902, + "learning_rate": 2.2621383689041087e-06, + "loss": 0.9129, + "num_input_tokens_seen": 79659465, + "step": 6313 + }, + { + "epoch": 0.474522771681948, + "grad_norm": 1.4699025603198046, + "learning_rate": 2.2616557188309033e-06, + "loss": 0.8675, + "num_input_tokens_seen": 79682885, + "step": 6314 + }, + { + "epoch": 0.474597925747783, + "grad_norm": 1.2710141016900427, + "learning_rate": 2.261173053253595e-06, + "loss": 0.9635, + "num_input_tokens_seen": 79711365, + "step": 6315 + }, + { + "epoch": 0.4746730798136179, + "grad_norm": 1.6405188389899497, + "learning_rate": 2.260690372200783e-06, + "loss": 0.934, + "num_input_tokens_seen": 79733475, + "step": 6316 + }, + { + "epoch": 0.47474823387945286, + "grad_norm": 1.4848212926435445, + "learning_rate": 2.260207675701069e-06, + "loss": 0.923, + "num_input_tokens_seen": 79761690, + "step": 6317 + }, + { + "epoch": 0.47482338794528783, + "grad_norm": 1.7854713479047606, + "learning_rate": 2.259724963783052e-06, + "loss": 0.8684, + "num_input_tokens_seen": 79787510, + "step": 6318 + }, + { + "epoch": 0.4748985420111228, + "grad_norm": 1.2711883528474122, + "learning_rate": 2.2592422364753377e-06, + "loss": 1.0054, + "num_input_tokens_seen": 79811990, + "step": 6319 + }, + { + "epoch": 0.4749736960769578, + "grad_norm": 1.441115848885105, + "learning_rate": 2.2587594938065285e-06, + "loss": 0.9667, + "num_input_tokens_seen": 79838400, + "step": 6320 + }, + { + "epoch": 0.47504885014279274, + "grad_norm": 1.4920422548237176, + "learning_rate": 2.2582767358052272e-06, + "loss": 0.9886, + "num_input_tokens_seen": 79862850, + "step": 6321 + }, + { + "epoch": 0.47512400420862766, + "grad_norm": 1.316233049086115, + "learning_rate": 2.2577939625000414e-06, + "loss": 0.9723, + "num_input_tokens_seen": 79887795, + "step": 6322 + }, + { + "epoch": 0.47519915827446263, + "grad_norm": 0.745541752838924, + "learning_rate": 2.2573111739195756e-06, + "loss": 0.7686, + "num_input_tokens_seen": 79958695, + "step": 6323 + }, + { + "epoch": 0.4752743123402976, + "grad_norm": 2.5159978175676883, + "learning_rate": 2.2568283700924375e-06, + "loss": 1.0372, + "num_input_tokens_seen": 79981380, + "step": 6324 + }, + { + "epoch": 0.47534946640613257, + "grad_norm": 1.5824308667743172, + "learning_rate": 2.2563455510472353e-06, + "loss": 0.9432, + "num_input_tokens_seen": 80004315, + "step": 6325 + }, + { + "epoch": 0.47542462047196754, + "grad_norm": 1.4234254369632842, + "learning_rate": 2.255862716812577e-06, + "loss": 0.8477, + "num_input_tokens_seen": 80029850, + "step": 6326 + }, + { + "epoch": 0.4754997745378025, + "grad_norm": 1.5071359015674257, + "learning_rate": 2.2553798674170735e-06, + "loss": 0.9443, + "num_input_tokens_seen": 80053635, + "step": 6327 + }, + { + "epoch": 0.4755749286036375, + "grad_norm": 1.4928814936859736, + "learning_rate": 2.2548970028893348e-06, + "loss": 0.8851, + "num_input_tokens_seen": 80078170, + "step": 6328 + }, + { + "epoch": 0.4756500826694724, + "grad_norm": 1.4872838480360453, + "learning_rate": 2.254414123257973e-06, + "loss": 0.9989, + "num_input_tokens_seen": 80100115, + "step": 6329 + }, + { + "epoch": 0.47572523673530737, + "grad_norm": 1.6042326463614103, + "learning_rate": 2.2539312285516e-06, + "loss": 1.0081, + "num_input_tokens_seen": 80123525, + "step": 6330 + }, + { + "epoch": 0.47580039080114234, + "grad_norm": 1.272994637273504, + "learning_rate": 2.2534483187988288e-06, + "loss": 0.898, + "num_input_tokens_seen": 80145755, + "step": 6331 + }, + { + "epoch": 0.4758755448669773, + "grad_norm": 1.5130573034131474, + "learning_rate": 2.2529653940282743e-06, + "loss": 1.0009, + "num_input_tokens_seen": 80169135, + "step": 6332 + }, + { + "epoch": 0.4759506989328123, + "grad_norm": 2.2668229717702832, + "learning_rate": 2.2524824542685515e-06, + "loss": 1.0588, + "num_input_tokens_seen": 80184655, + "step": 6333 + }, + { + "epoch": 0.47602585299864725, + "grad_norm": 1.4057871586682973, + "learning_rate": 2.2519994995482774e-06, + "loss": 0.8344, + "num_input_tokens_seen": 80212200, + "step": 6334 + }, + { + "epoch": 0.47610100706448216, + "grad_norm": 0.9228137457632821, + "learning_rate": 2.2515165298960674e-06, + "loss": 0.8066, + "num_input_tokens_seen": 80288970, + "step": 6335 + }, + { + "epoch": 0.47617616113031713, + "grad_norm": 2.2819947372696796, + "learning_rate": 2.251033545340539e-06, + "loss": 1.0144, + "num_input_tokens_seen": 80312385, + "step": 6336 + }, + { + "epoch": 0.4762513151961521, + "grad_norm": 1.2914315850645293, + "learning_rate": 2.2505505459103133e-06, + "loss": 0.9751, + "num_input_tokens_seen": 80336755, + "step": 6337 + }, + { + "epoch": 0.4763264692619871, + "grad_norm": 1.3570184560995115, + "learning_rate": 2.250067531634007e-06, + "loss": 0.8643, + "num_input_tokens_seen": 80366850, + "step": 6338 + }, + { + "epoch": 0.47640162332782204, + "grad_norm": 1.7182039260359383, + "learning_rate": 2.249584502540242e-06, + "loss": 0.9847, + "num_input_tokens_seen": 80390785, + "step": 6339 + }, + { + "epoch": 0.476476777393657, + "grad_norm": 1.742327834152341, + "learning_rate": 2.2491014586576404e-06, + "loss": 0.8691, + "num_input_tokens_seen": 80416985, + "step": 6340 + }, + { + "epoch": 0.476551931459492, + "grad_norm": 0.9237007553333723, + "learning_rate": 2.248618400014823e-06, + "loss": 0.8305, + "num_input_tokens_seen": 80489665, + "step": 6341 + }, + { + "epoch": 0.4766270855253269, + "grad_norm": 1.5573932355166304, + "learning_rate": 2.248135326640414e-06, + "loss": 0.9081, + "num_input_tokens_seen": 80512055, + "step": 6342 + }, + { + "epoch": 0.47670223959116187, + "grad_norm": 1.5190671393428195, + "learning_rate": 2.2476522385630354e-06, + "loss": 0.9037, + "num_input_tokens_seen": 80535705, + "step": 6343 + }, + { + "epoch": 0.47677739365699684, + "grad_norm": 1.0002966083287967, + "learning_rate": 2.2471691358113146e-06, + "loss": 0.787, + "num_input_tokens_seen": 80617310, + "step": 6344 + }, + { + "epoch": 0.4768525477228318, + "grad_norm": 1.4855556410541635, + "learning_rate": 2.246686018413875e-06, + "loss": 0.9748, + "num_input_tokens_seen": 80641070, + "step": 6345 + }, + { + "epoch": 0.4769277017886668, + "grad_norm": 2.0755890021782095, + "learning_rate": 2.246202886399345e-06, + "loss": 0.8687, + "num_input_tokens_seen": 80660450, + "step": 6346 + }, + { + "epoch": 0.47700285585450175, + "grad_norm": 1.4463616764403422, + "learning_rate": 2.245719739796351e-06, + "loss": 0.8795, + "num_input_tokens_seen": 80684775, + "step": 6347 + }, + { + "epoch": 0.47707800992033667, + "grad_norm": 1.8295714695834768, + "learning_rate": 2.2452365786335214e-06, + "loss": 1.0686, + "num_input_tokens_seen": 80708615, + "step": 6348 + }, + { + "epoch": 0.47715316398617164, + "grad_norm": 3.272948890946668, + "learning_rate": 2.2447534029394856e-06, + "loss": 0.8137, + "num_input_tokens_seen": 80781260, + "step": 6349 + }, + { + "epoch": 0.4772283180520066, + "grad_norm": 1.2644334529351675, + "learning_rate": 2.244270212742873e-06, + "loss": 0.806, + "num_input_tokens_seen": 80811295, + "step": 6350 + }, + { + "epoch": 0.4773034721178416, + "grad_norm": 0.9504820027541339, + "learning_rate": 2.2437870080723153e-06, + "loss": 0.7817, + "num_input_tokens_seen": 80887380, + "step": 6351 + }, + { + "epoch": 0.47737862618367655, + "grad_norm": 1.6937989541049079, + "learning_rate": 2.243303788956444e-06, + "loss": 0.9115, + "num_input_tokens_seen": 80909475, + "step": 6352 + }, + { + "epoch": 0.4774537802495115, + "grad_norm": 1.4392055676446402, + "learning_rate": 2.2428205554238914e-06, + "loss": 0.8257, + "num_input_tokens_seen": 80936780, + "step": 6353 + }, + { + "epoch": 0.4775289343153465, + "grad_norm": 1.4243431886516071, + "learning_rate": 2.2423373075032913e-06, + "loss": 0.9691, + "num_input_tokens_seen": 80962640, + "step": 6354 + }, + { + "epoch": 0.4776040883811814, + "grad_norm": 1.6785768607379243, + "learning_rate": 2.241854045223277e-06, + "loss": 1.0309, + "num_input_tokens_seen": 80985310, + "step": 6355 + }, + { + "epoch": 0.4776792424470164, + "grad_norm": 1.5327162437274588, + "learning_rate": 2.241370768612485e-06, + "loss": 1.0322, + "num_input_tokens_seen": 81009015, + "step": 6356 + }, + { + "epoch": 0.47775439651285134, + "grad_norm": 2.441073121803694, + "learning_rate": 2.2408874776995508e-06, + "loss": 0.98, + "num_input_tokens_seen": 81033610, + "step": 6357 + }, + { + "epoch": 0.4778295505786863, + "grad_norm": 1.6998834822498006, + "learning_rate": 2.2404041725131106e-06, + "loss": 1.0177, + "num_input_tokens_seen": 81053855, + "step": 6358 + }, + { + "epoch": 0.4779047046445213, + "grad_norm": 1.356422726228219, + "learning_rate": 2.239920853081803e-06, + "loss": 0.9609, + "num_input_tokens_seen": 81079895, + "step": 6359 + }, + { + "epoch": 0.47797985871035625, + "grad_norm": 1.4805004750579205, + "learning_rate": 2.2394375194342653e-06, + "loss": 0.9656, + "num_input_tokens_seen": 81105765, + "step": 6360 + }, + { + "epoch": 0.47805501277619117, + "grad_norm": 1.3814741220359654, + "learning_rate": 2.2389541715991385e-06, + "loss": 0.9581, + "num_input_tokens_seen": 81129060, + "step": 6361 + }, + { + "epoch": 0.47813016684202614, + "grad_norm": 1.1249250810896587, + "learning_rate": 2.238470809605062e-06, + "loss": 0.9059, + "num_input_tokens_seen": 81155780, + "step": 6362 + }, + { + "epoch": 0.4782053209078611, + "grad_norm": 1.8024767922059468, + "learning_rate": 2.2379874334806764e-06, + "loss": 0.8863, + "num_input_tokens_seen": 81176800, + "step": 6363 + }, + { + "epoch": 0.4782804749736961, + "grad_norm": 1.4634165410576003, + "learning_rate": 2.237504043254625e-06, + "loss": 1.0426, + "num_input_tokens_seen": 81203115, + "step": 6364 + }, + { + "epoch": 0.47835562903953105, + "grad_norm": 1.7291044967076652, + "learning_rate": 2.2370206389555485e-06, + "loss": 0.7555, + "num_input_tokens_seen": 81231485, + "step": 6365 + }, + { + "epoch": 0.478430783105366, + "grad_norm": 1.7617125246737293, + "learning_rate": 2.2365372206120923e-06, + "loss": 0.9684, + "num_input_tokens_seen": 81254305, + "step": 6366 + }, + { + "epoch": 0.47850593717120093, + "grad_norm": 1.8894329884646768, + "learning_rate": 2.2360537882528996e-06, + "loss": 0.9989, + "num_input_tokens_seen": 81278735, + "step": 6367 + }, + { + "epoch": 0.4785810912370359, + "grad_norm": 2.5543218228466062, + "learning_rate": 2.2355703419066163e-06, + "loss": 1.0891, + "num_input_tokens_seen": 81297370, + "step": 6368 + }, + { + "epoch": 0.4786562453028709, + "grad_norm": 1.4692561312371157, + "learning_rate": 2.2350868816018886e-06, + "loss": 0.8635, + "num_input_tokens_seen": 81320825, + "step": 6369 + }, + { + "epoch": 0.47873139936870585, + "grad_norm": 1.0092833084646256, + "learning_rate": 2.234603407367362e-06, + "loss": 0.8623, + "num_input_tokens_seen": 81396755, + "step": 6370 + }, + { + "epoch": 0.4788065534345408, + "grad_norm": 1.783774678113708, + "learning_rate": 2.234119919231686e-06, + "loss": 0.9844, + "num_input_tokens_seen": 81419295, + "step": 6371 + }, + { + "epoch": 0.4788817075003758, + "grad_norm": 1.6927170112583045, + "learning_rate": 2.2336364172235074e-06, + "loss": 0.8979, + "num_input_tokens_seen": 81442125, + "step": 6372 + }, + { + "epoch": 0.47895686156621076, + "grad_norm": 1.873261026783492, + "learning_rate": 2.2331529013714775e-06, + "loss": 0.9937, + "num_input_tokens_seen": 81464400, + "step": 6373 + }, + { + "epoch": 0.47903201563204567, + "grad_norm": 1.3647881953479122, + "learning_rate": 2.2326693717042446e-06, + "loss": 0.8998, + "num_input_tokens_seen": 81491760, + "step": 6374 + }, + { + "epoch": 0.47910716969788064, + "grad_norm": 1.6585863297413723, + "learning_rate": 2.2321858282504603e-06, + "loss": 0.871, + "num_input_tokens_seen": 81517805, + "step": 6375 + }, + { + "epoch": 0.4791823237637156, + "grad_norm": 1.0494420476490836, + "learning_rate": 2.231702271038777e-06, + "loss": 0.7992, + "num_input_tokens_seen": 81591795, + "step": 6376 + }, + { + "epoch": 0.4792574778295506, + "grad_norm": 1.646089300539786, + "learning_rate": 2.2312187000978467e-06, + "loss": 0.9837, + "num_input_tokens_seen": 81613340, + "step": 6377 + }, + { + "epoch": 0.47933263189538555, + "grad_norm": 1.2213050273887056, + "learning_rate": 2.230735115456324e-06, + "loss": 0.922, + "num_input_tokens_seen": 81639905, + "step": 6378 + }, + { + "epoch": 0.4794077859612205, + "grad_norm": 1.6753707418509334, + "learning_rate": 2.2302515171428613e-06, + "loss": 0.9241, + "num_input_tokens_seen": 81665290, + "step": 6379 + }, + { + "epoch": 0.47948294002705544, + "grad_norm": 1.5001083175968712, + "learning_rate": 2.229767905186114e-06, + "loss": 0.9318, + "num_input_tokens_seen": 81689200, + "step": 6380 + }, + { + "epoch": 0.4795580940928904, + "grad_norm": 1.6815049385862624, + "learning_rate": 2.2292842796147395e-06, + "loss": 0.8497, + "num_input_tokens_seen": 81715185, + "step": 6381 + }, + { + "epoch": 0.4796332481587254, + "grad_norm": 1.5707934198275206, + "learning_rate": 2.2288006404573922e-06, + "loss": 0.9402, + "num_input_tokens_seen": 81740775, + "step": 6382 + }, + { + "epoch": 0.47970840222456035, + "grad_norm": 1.4113686123680524, + "learning_rate": 2.228316987742732e-06, + "loss": 1.0158, + "num_input_tokens_seen": 81763340, + "step": 6383 + }, + { + "epoch": 0.4797835562903953, + "grad_norm": 1.3585003637033255, + "learning_rate": 2.227833321499415e-06, + "loss": 0.9646, + "num_input_tokens_seen": 81784870, + "step": 6384 + }, + { + "epoch": 0.4798587103562303, + "grad_norm": 2.0855679416824997, + "learning_rate": 2.227349641756102e-06, + "loss": 0.7713, + "num_input_tokens_seen": 81805965, + "step": 6385 + }, + { + "epoch": 0.47993386442206526, + "grad_norm": 1.4220340346382556, + "learning_rate": 2.2268659485414526e-06, + "loss": 0.9713, + "num_input_tokens_seen": 81830490, + "step": 6386 + }, + { + "epoch": 0.4800090184879002, + "grad_norm": 1.5314985579565759, + "learning_rate": 2.226382241884126e-06, + "loss": 0.9735, + "num_input_tokens_seen": 81854335, + "step": 6387 + }, + { + "epoch": 0.48008417255373514, + "grad_norm": 1.8065807105774616, + "learning_rate": 2.225898521812785e-06, + "loss": 1.01, + "num_input_tokens_seen": 81877130, + "step": 6388 + }, + { + "epoch": 0.4801593266195701, + "grad_norm": 1.6653326497261063, + "learning_rate": 2.2254147883560916e-06, + "loss": 0.969, + "num_input_tokens_seen": 81899185, + "step": 6389 + }, + { + "epoch": 0.4802344806854051, + "grad_norm": 1.3422407501635227, + "learning_rate": 2.2249310415427087e-06, + "loss": 0.9503, + "num_input_tokens_seen": 81925175, + "step": 6390 + }, + { + "epoch": 0.48030963475124006, + "grad_norm": 1.8396789736167034, + "learning_rate": 2.224447281401301e-06, + "loss": 0.8167, + "num_input_tokens_seen": 81945720, + "step": 6391 + }, + { + "epoch": 0.480384788817075, + "grad_norm": 1.3457858609703277, + "learning_rate": 2.2239635079605316e-06, + "loss": 0.9747, + "num_input_tokens_seen": 81969575, + "step": 6392 + }, + { + "epoch": 0.48045994288290994, + "grad_norm": 1.436134685140048, + "learning_rate": 2.223479721249067e-06, + "loss": 0.8202, + "num_input_tokens_seen": 81999710, + "step": 6393 + }, + { + "epoch": 0.4805350969487449, + "grad_norm": 1.5296257123459958, + "learning_rate": 2.222995921295573e-06, + "loss": 0.8732, + "num_input_tokens_seen": 82024795, + "step": 6394 + }, + { + "epoch": 0.4806102510145799, + "grad_norm": 1.5984615171237198, + "learning_rate": 2.2225121081287174e-06, + "loss": 0.8789, + "num_input_tokens_seen": 82048345, + "step": 6395 + }, + { + "epoch": 0.48068540508041485, + "grad_norm": 1.547933890382516, + "learning_rate": 2.2220282817771668e-06, + "loss": 0.9163, + "num_input_tokens_seen": 82071535, + "step": 6396 + }, + { + "epoch": 0.4807605591462498, + "grad_norm": 1.4519379392029381, + "learning_rate": 2.2215444422695906e-06, + "loss": 0.8238, + "num_input_tokens_seen": 82095215, + "step": 6397 + }, + { + "epoch": 0.4808357132120848, + "grad_norm": 1.5144500897691728, + "learning_rate": 2.2210605896346575e-06, + "loss": 0.9636, + "num_input_tokens_seen": 82118620, + "step": 6398 + }, + { + "epoch": 0.48091086727791976, + "grad_norm": 1.5727941536927637, + "learning_rate": 2.2205767239010376e-06, + "loss": 1.0229, + "num_input_tokens_seen": 82138865, + "step": 6399 + }, + { + "epoch": 0.4809860213437547, + "grad_norm": 1.8919653436861879, + "learning_rate": 2.2200928450974028e-06, + "loss": 1.0349, + "num_input_tokens_seen": 82159055, + "step": 6400 + }, + { + "epoch": 0.48106117540958965, + "grad_norm": 1.6272173205640068, + "learning_rate": 2.2196089532524244e-06, + "loss": 0.9826, + "num_input_tokens_seen": 82181775, + "step": 6401 + }, + { + "epoch": 0.4811363294754246, + "grad_norm": 1.594797126998208, + "learning_rate": 2.2191250483947736e-06, + "loss": 1.0081, + "num_input_tokens_seen": 82204490, + "step": 6402 + }, + { + "epoch": 0.4812114835412596, + "grad_norm": 1.6883112052579525, + "learning_rate": 2.2186411305531254e-06, + "loss": 0.9889, + "num_input_tokens_seen": 82227905, + "step": 6403 + }, + { + "epoch": 0.48128663760709456, + "grad_norm": 1.1858570880793704, + "learning_rate": 2.2181571997561523e-06, + "loss": 0.9581, + "num_input_tokens_seen": 82255950, + "step": 6404 + }, + { + "epoch": 0.48136179167292953, + "grad_norm": 1.1518377712647974, + "learning_rate": 2.2176732560325302e-06, + "loss": 1.0471, + "num_input_tokens_seen": 82282015, + "step": 6405 + }, + { + "epoch": 0.48143694573876444, + "grad_norm": 1.4188861663946248, + "learning_rate": 2.217189299410934e-06, + "loss": 0.905, + "num_input_tokens_seen": 82307560, + "step": 6406 + }, + { + "epoch": 0.4815120998045994, + "grad_norm": 1.271975327267469, + "learning_rate": 2.21670532992004e-06, + "loss": 0.9432, + "num_input_tokens_seen": 82332065, + "step": 6407 + }, + { + "epoch": 0.4815872538704344, + "grad_norm": 2.130888913349346, + "learning_rate": 2.2162213475885262e-06, + "loss": 1.0552, + "num_input_tokens_seen": 82350595, + "step": 6408 + }, + { + "epoch": 0.48166240793626935, + "grad_norm": 1.4984169394760147, + "learning_rate": 2.215737352445069e-06, + "loss": 0.8246, + "num_input_tokens_seen": 82373965, + "step": 6409 + }, + { + "epoch": 0.4817375620021043, + "grad_norm": 1.4334055597752353, + "learning_rate": 2.2152533445183477e-06, + "loss": 0.9692, + "num_input_tokens_seen": 82401395, + "step": 6410 + }, + { + "epoch": 0.4818127160679393, + "grad_norm": 1.5892480582464963, + "learning_rate": 2.214769323837041e-06, + "loss": 0.9707, + "num_input_tokens_seen": 82421855, + "step": 6411 + }, + { + "epoch": 0.4818878701337742, + "grad_norm": 0.806133267727436, + "learning_rate": 2.21428529042983e-06, + "loss": 0.8114, + "num_input_tokens_seen": 82495855, + "step": 6412 + }, + { + "epoch": 0.4819630241996092, + "grad_norm": 1.6918643040454582, + "learning_rate": 2.213801244325395e-06, + "loss": 0.9599, + "num_input_tokens_seen": 82520030, + "step": 6413 + }, + { + "epoch": 0.48203817826544415, + "grad_norm": 2.469985266575906, + "learning_rate": 2.2133171855524167e-06, + "loss": 1.0049, + "num_input_tokens_seen": 82542070, + "step": 6414 + }, + { + "epoch": 0.4821133323312791, + "grad_norm": 0.7845685286972891, + "learning_rate": 2.21283311413958e-06, + "loss": 0.7963, + "num_input_tokens_seen": 82617575, + "step": 6415 + }, + { + "epoch": 0.4821884863971141, + "grad_norm": 1.3896774321136283, + "learning_rate": 2.2123490301155647e-06, + "loss": 0.828, + "num_input_tokens_seen": 82642685, + "step": 6416 + }, + { + "epoch": 0.48226364046294906, + "grad_norm": 1.5195794183701392, + "learning_rate": 2.2118649335090568e-06, + "loss": 0.9995, + "num_input_tokens_seen": 82665235, + "step": 6417 + }, + { + "epoch": 0.48233879452878403, + "grad_norm": 1.397787187988015, + "learning_rate": 2.2113808243487404e-06, + "loss": 0.9807, + "num_input_tokens_seen": 82690085, + "step": 6418 + }, + { + "epoch": 0.48241394859461895, + "grad_norm": 2.1544198271733226, + "learning_rate": 2.210896702663301e-06, + "loss": 0.977, + "num_input_tokens_seen": 82710430, + "step": 6419 + }, + { + "epoch": 0.4824891026604539, + "grad_norm": 1.4568850364650288, + "learning_rate": 2.2104125684814238e-06, + "loss": 0.8935, + "num_input_tokens_seen": 82732790, + "step": 6420 + }, + { + "epoch": 0.4825642567262889, + "grad_norm": 1.4438350239608455, + "learning_rate": 2.209928421831796e-06, + "loss": 0.9935, + "num_input_tokens_seen": 82756210, + "step": 6421 + }, + { + "epoch": 0.48263941079212386, + "grad_norm": 1.5489545003317238, + "learning_rate": 2.209444262743106e-06, + "loss": 0.9733, + "num_input_tokens_seen": 82780215, + "step": 6422 + }, + { + "epoch": 0.4827145648579588, + "grad_norm": 1.3672401636062435, + "learning_rate": 2.2089600912440413e-06, + "loss": 0.8381, + "num_input_tokens_seen": 82803975, + "step": 6423 + }, + { + "epoch": 0.4827897189237938, + "grad_norm": 1.3626701931130178, + "learning_rate": 2.2084759073632904e-06, + "loss": 0.9165, + "num_input_tokens_seen": 82828765, + "step": 6424 + }, + { + "epoch": 0.4828648729896287, + "grad_norm": 1.67087282465484, + "learning_rate": 2.2079917111295448e-06, + "loss": 0.9452, + "num_input_tokens_seen": 82852605, + "step": 6425 + }, + { + "epoch": 0.4829400270554637, + "grad_norm": 1.4960178923735363, + "learning_rate": 2.2075075025714922e-06, + "loss": 0.7973, + "num_input_tokens_seen": 82880315, + "step": 6426 + }, + { + "epoch": 0.48301518112129865, + "grad_norm": 1.6309571745095666, + "learning_rate": 2.2070232817178272e-06, + "loss": 1.0374, + "num_input_tokens_seen": 82901725, + "step": 6427 + }, + { + "epoch": 0.4830903351871336, + "grad_norm": 1.4571686135347803, + "learning_rate": 2.206539048597239e-06, + "loss": 0.9309, + "num_input_tokens_seen": 82927130, + "step": 6428 + }, + { + "epoch": 0.4831654892529686, + "grad_norm": 1.5712080342493315, + "learning_rate": 2.206054803238422e-06, + "loss": 0.9713, + "num_input_tokens_seen": 82952020, + "step": 6429 + }, + { + "epoch": 0.48324064331880356, + "grad_norm": 0.8099470745856876, + "learning_rate": 2.2055705456700686e-06, + "loss": 0.8628, + "num_input_tokens_seen": 83032620, + "step": 6430 + }, + { + "epoch": 0.48331579738463853, + "grad_norm": 0.7887931212836207, + "learning_rate": 2.2050862759208728e-06, + "loss": 0.784, + "num_input_tokens_seen": 83110285, + "step": 6431 + }, + { + "epoch": 0.48339095145047345, + "grad_norm": 1.6990180938457753, + "learning_rate": 2.2046019940195303e-06, + "loss": 1.0231, + "num_input_tokens_seen": 83130190, + "step": 6432 + }, + { + "epoch": 0.4834661055163084, + "grad_norm": 1.4830690762875083, + "learning_rate": 2.2041176999947353e-06, + "loss": 0.8755, + "num_input_tokens_seen": 83153525, + "step": 6433 + }, + { + "epoch": 0.4835412595821434, + "grad_norm": 1.3931370880026948, + "learning_rate": 2.203633393875186e-06, + "loss": 1.0127, + "num_input_tokens_seen": 83177510, + "step": 6434 + }, + { + "epoch": 0.48361641364797836, + "grad_norm": 1.5120220028018068, + "learning_rate": 2.2031490756895784e-06, + "loss": 0.9711, + "num_input_tokens_seen": 83198375, + "step": 6435 + }, + { + "epoch": 0.48369156771381333, + "grad_norm": 1.9050809683826433, + "learning_rate": 2.2026647454666097e-06, + "loss": 1.0041, + "num_input_tokens_seen": 83220470, + "step": 6436 + }, + { + "epoch": 0.4837667217796483, + "grad_norm": 2.451001262426343, + "learning_rate": 2.202180403234979e-06, + "loss": 0.9083, + "num_input_tokens_seen": 83247330, + "step": 6437 + }, + { + "epoch": 0.4838418758454832, + "grad_norm": 1.524000797301557, + "learning_rate": 2.2016960490233845e-06, + "loss": 0.9511, + "num_input_tokens_seen": 83271585, + "step": 6438 + }, + { + "epoch": 0.4839170299113182, + "grad_norm": 2.233092573577994, + "learning_rate": 2.2012116828605275e-06, + "loss": 0.9555, + "num_input_tokens_seen": 83290785, + "step": 6439 + }, + { + "epoch": 0.48399218397715316, + "grad_norm": 1.4339905898983836, + "learning_rate": 2.200727304775108e-06, + "loss": 0.9226, + "num_input_tokens_seen": 83314905, + "step": 6440 + }, + { + "epoch": 0.4840673380429881, + "grad_norm": 1.790855741696449, + "learning_rate": 2.200242914795826e-06, + "loss": 0.8855, + "num_input_tokens_seen": 83336810, + "step": 6441 + }, + { + "epoch": 0.4841424921088231, + "grad_norm": 1.3543351679896, + "learning_rate": 2.1997585129513852e-06, + "loss": 0.8457, + "num_input_tokens_seen": 83363145, + "step": 6442 + }, + { + "epoch": 0.48421764617465807, + "grad_norm": 1.7326719153073875, + "learning_rate": 2.1992740992704877e-06, + "loss": 0.9342, + "num_input_tokens_seen": 83386300, + "step": 6443 + }, + { + "epoch": 0.48429280024049304, + "grad_norm": 1.6988376935565275, + "learning_rate": 2.1987896737818365e-06, + "loss": 1.095, + "num_input_tokens_seen": 83407500, + "step": 6444 + }, + { + "epoch": 0.48436795430632795, + "grad_norm": 1.7947853210773936, + "learning_rate": 2.198305236514136e-06, + "loss": 1.081, + "num_input_tokens_seen": 83430230, + "step": 6445 + }, + { + "epoch": 0.4844431083721629, + "grad_norm": 0.755221667117216, + "learning_rate": 2.1978207874960908e-06, + "loss": 0.7652, + "num_input_tokens_seen": 83512610, + "step": 6446 + }, + { + "epoch": 0.4845182624379979, + "grad_norm": 1.3892130595292216, + "learning_rate": 2.1973363267564063e-06, + "loss": 0.9898, + "num_input_tokens_seen": 83538505, + "step": 6447 + }, + { + "epoch": 0.48459341650383286, + "grad_norm": 1.5379982768918468, + "learning_rate": 2.196851854323789e-06, + "loss": 1.0412, + "num_input_tokens_seen": 83561425, + "step": 6448 + }, + { + "epoch": 0.48466857056966783, + "grad_norm": 1.705210467573946, + "learning_rate": 2.1963673702269454e-06, + "loss": 0.9765, + "num_input_tokens_seen": 83584925, + "step": 6449 + }, + { + "epoch": 0.4847437246355028, + "grad_norm": 1.3755352105695668, + "learning_rate": 2.195882874494583e-06, + "loss": 0.9165, + "num_input_tokens_seen": 83607060, + "step": 6450 + }, + { + "epoch": 0.4848188787013377, + "grad_norm": 0.9531784355318375, + "learning_rate": 2.195398367155411e-06, + "loss": 0.849, + "num_input_tokens_seen": 83672155, + "step": 6451 + }, + { + "epoch": 0.4848940327671727, + "grad_norm": 1.1870286909715777, + "learning_rate": 2.194913848238137e-06, + "loss": 0.8969, + "num_input_tokens_seen": 83698785, + "step": 6452 + }, + { + "epoch": 0.48496918683300766, + "grad_norm": 1.4625052134103043, + "learning_rate": 2.194429317771471e-06, + "loss": 0.8916, + "num_input_tokens_seen": 83727045, + "step": 6453 + }, + { + "epoch": 0.48504434089884263, + "grad_norm": 1.5301677226716768, + "learning_rate": 2.1939447757841236e-06, + "loss": 1.0148, + "num_input_tokens_seen": 83752625, + "step": 6454 + }, + { + "epoch": 0.4851194949646776, + "grad_norm": 1.2562041335961194, + "learning_rate": 2.1934602223048054e-06, + "loss": 0.9158, + "num_input_tokens_seen": 83781450, + "step": 6455 + }, + { + "epoch": 0.48519464903051257, + "grad_norm": 1.5720738643492846, + "learning_rate": 2.1929756573622282e-06, + "loss": 0.9434, + "num_input_tokens_seen": 83809480, + "step": 6456 + }, + { + "epoch": 0.4852698030963475, + "grad_norm": 1.5380209094636483, + "learning_rate": 2.192491080985105e-06, + "loss": 0.9506, + "num_input_tokens_seen": 83833285, + "step": 6457 + }, + { + "epoch": 0.48534495716218246, + "grad_norm": 1.5234391432533205, + "learning_rate": 2.192006493202147e-06, + "loss": 0.9726, + "num_input_tokens_seen": 83856740, + "step": 6458 + }, + { + "epoch": 0.4854201112280174, + "grad_norm": 1.5887048939710495, + "learning_rate": 2.19152189404207e-06, + "loss": 0.9008, + "num_input_tokens_seen": 83882705, + "step": 6459 + }, + { + "epoch": 0.4854952652938524, + "grad_norm": 1.468104606217194, + "learning_rate": 2.191037283533587e-06, + "loss": 0.9836, + "num_input_tokens_seen": 83905835, + "step": 6460 + }, + { + "epoch": 0.48557041935968737, + "grad_norm": 1.354064037639943, + "learning_rate": 2.1905526617054136e-06, + "loss": 0.8579, + "num_input_tokens_seen": 83933140, + "step": 6461 + }, + { + "epoch": 0.48564557342552234, + "grad_norm": 1.653491476068711, + "learning_rate": 2.1900680285862655e-06, + "loss": 0.95, + "num_input_tokens_seen": 83956680, + "step": 6462 + }, + { + "epoch": 0.4857207274913573, + "grad_norm": 1.8586060031713052, + "learning_rate": 2.1895833842048583e-06, + "loss": 0.9985, + "num_input_tokens_seen": 83980170, + "step": 6463 + }, + { + "epoch": 0.4857958815571922, + "grad_norm": 1.4722722612830086, + "learning_rate": 2.18909872858991e-06, + "loss": 0.9855, + "num_input_tokens_seen": 84005015, + "step": 6464 + }, + { + "epoch": 0.4858710356230272, + "grad_norm": 2.840674277571669, + "learning_rate": 2.1886140617701374e-06, + "loss": 0.9281, + "num_input_tokens_seen": 84025760, + "step": 6465 + }, + { + "epoch": 0.48594618968886216, + "grad_norm": 3.145037838900778, + "learning_rate": 2.1881293837742604e-06, + "loss": 0.8219, + "num_input_tokens_seen": 84047690, + "step": 6466 + }, + { + "epoch": 0.48602134375469713, + "grad_norm": 1.8274675801085107, + "learning_rate": 2.1876446946309965e-06, + "loss": 1.0031, + "num_input_tokens_seen": 84073005, + "step": 6467 + }, + { + "epoch": 0.4860964978205321, + "grad_norm": 1.5179888980645466, + "learning_rate": 2.187159994369065e-06, + "loss": 0.8987, + "num_input_tokens_seen": 84096840, + "step": 6468 + }, + { + "epoch": 0.4861716518863671, + "grad_norm": 2.053955411848637, + "learning_rate": 2.1866752830171884e-06, + "loss": 0.859, + "num_input_tokens_seen": 84121555, + "step": 6469 + }, + { + "epoch": 0.486246805952202, + "grad_norm": 1.5791772035165026, + "learning_rate": 2.1861905606040857e-06, + "loss": 0.9794, + "num_input_tokens_seen": 84144200, + "step": 6470 + }, + { + "epoch": 0.48632196001803696, + "grad_norm": 1.470308815832924, + "learning_rate": 2.18570582715848e-06, + "loss": 0.9829, + "num_input_tokens_seen": 84168505, + "step": 6471 + }, + { + "epoch": 0.48639711408387193, + "grad_norm": 4.605950379687832, + "learning_rate": 2.1852210827090922e-06, + "loss": 0.9814, + "num_input_tokens_seen": 84190445, + "step": 6472 + }, + { + "epoch": 0.4864722681497069, + "grad_norm": 1.6012232723581319, + "learning_rate": 2.1847363272846465e-06, + "loss": 0.8521, + "num_input_tokens_seen": 84218255, + "step": 6473 + }, + { + "epoch": 0.48654742221554187, + "grad_norm": 1.4023121269719652, + "learning_rate": 2.184251560913866e-06, + "loss": 0.9234, + "num_input_tokens_seen": 84243530, + "step": 6474 + }, + { + "epoch": 0.48662257628137684, + "grad_norm": 1.6691394422916621, + "learning_rate": 2.1837667836254746e-06, + "loss": 0.9714, + "num_input_tokens_seen": 84266270, + "step": 6475 + }, + { + "epoch": 0.4866977303472118, + "grad_norm": 1.581167630549404, + "learning_rate": 2.183281995448198e-06, + "loss": 0.9468, + "num_input_tokens_seen": 84287670, + "step": 6476 + }, + { + "epoch": 0.4867728844130467, + "grad_norm": 1.7926452610846446, + "learning_rate": 2.1827971964107607e-06, + "loss": 0.8754, + "num_input_tokens_seen": 84305140, + "step": 6477 + }, + { + "epoch": 0.4868480384788817, + "grad_norm": 1.6103151408669278, + "learning_rate": 2.1823123865418903e-06, + "loss": 0.9605, + "num_input_tokens_seen": 84327750, + "step": 6478 + }, + { + "epoch": 0.48692319254471667, + "grad_norm": 1.5491525856159256, + "learning_rate": 2.1818275658703128e-06, + "loss": 1.0387, + "num_input_tokens_seen": 84350595, + "step": 6479 + }, + { + "epoch": 0.48699834661055164, + "grad_norm": 1.7781387060404796, + "learning_rate": 2.1813427344247557e-06, + "loss": 1.0214, + "num_input_tokens_seen": 84376965, + "step": 6480 + }, + { + "epoch": 0.4870735006763866, + "grad_norm": 0.8728160851797955, + "learning_rate": 2.1808578922339466e-06, + "loss": 0.8513, + "num_input_tokens_seen": 84447435, + "step": 6481 + }, + { + "epoch": 0.4871486547422216, + "grad_norm": 1.5887242530159134, + "learning_rate": 2.180373039326615e-06, + "loss": 0.9518, + "num_input_tokens_seen": 84478325, + "step": 6482 + }, + { + "epoch": 0.4872238088080565, + "grad_norm": 1.2174275877644183, + "learning_rate": 2.1798881757314905e-06, + "loss": 0.8783, + "num_input_tokens_seen": 84507860, + "step": 6483 + }, + { + "epoch": 0.48729896287389146, + "grad_norm": 1.3697499267639375, + "learning_rate": 2.1794033014773033e-06, + "loss": 0.9671, + "num_input_tokens_seen": 84530440, + "step": 6484 + }, + { + "epoch": 0.48737411693972643, + "grad_norm": 1.5967307546305993, + "learning_rate": 2.1789184165927824e-06, + "loss": 0.8061, + "num_input_tokens_seen": 84553800, + "step": 6485 + }, + { + "epoch": 0.4874492710055614, + "grad_norm": 1.5840752854229168, + "learning_rate": 2.178433521106661e-06, + "loss": 1.03, + "num_input_tokens_seen": 84578195, + "step": 6486 + }, + { + "epoch": 0.4875244250713964, + "grad_norm": 1.7052595428750088, + "learning_rate": 2.177948615047669e-06, + "loss": 1.0212, + "num_input_tokens_seen": 84601915, + "step": 6487 + }, + { + "epoch": 0.48759957913723134, + "grad_norm": 1.5719378965015647, + "learning_rate": 2.1774636984445417e-06, + "loss": 0.9587, + "num_input_tokens_seen": 84626725, + "step": 6488 + }, + { + "epoch": 0.4876747332030663, + "grad_norm": 1.4857290416550675, + "learning_rate": 2.1769787713260097e-06, + "loss": 1.0478, + "num_input_tokens_seen": 84649685, + "step": 6489 + }, + { + "epoch": 0.4877498872689012, + "grad_norm": 1.5480258398289732, + "learning_rate": 2.1764938337208075e-06, + "loss": 0.9982, + "num_input_tokens_seen": 84670355, + "step": 6490 + }, + { + "epoch": 0.4878250413347362, + "grad_norm": 2.8772217831641274, + "learning_rate": 2.1760088856576706e-06, + "loss": 1.0215, + "num_input_tokens_seen": 84694410, + "step": 6491 + }, + { + "epoch": 0.48790019540057117, + "grad_norm": 1.4861320135641087, + "learning_rate": 2.175523927165333e-06, + "loss": 0.8674, + "num_input_tokens_seen": 84720075, + "step": 6492 + }, + { + "epoch": 0.48797534946640614, + "grad_norm": 1.8537028371890962, + "learning_rate": 2.1750389582725307e-06, + "loss": 0.8883, + "num_input_tokens_seen": 84745265, + "step": 6493 + }, + { + "epoch": 0.4880505035322411, + "grad_norm": 0.7728991706149754, + "learning_rate": 2.1745539790079993e-06, + "loss": 0.7765, + "num_input_tokens_seen": 84825725, + "step": 6494 + }, + { + "epoch": 0.4881256575980761, + "grad_norm": 1.3413996434701674, + "learning_rate": 2.1740689894004773e-06, + "loss": 0.9823, + "num_input_tokens_seen": 84848110, + "step": 6495 + }, + { + "epoch": 0.488200811663911, + "grad_norm": 1.253046091311241, + "learning_rate": 2.1735839894787003e-06, + "loss": 0.9595, + "num_input_tokens_seen": 84873945, + "step": 6496 + }, + { + "epoch": 0.48827596572974596, + "grad_norm": 1.4182443062035741, + "learning_rate": 2.1730989792714074e-06, + "loss": 1.0088, + "num_input_tokens_seen": 84896140, + "step": 6497 + }, + { + "epoch": 0.48835111979558093, + "grad_norm": 1.4224084545103077, + "learning_rate": 2.1726139588073374e-06, + "loss": 0.893, + "num_input_tokens_seen": 84920715, + "step": 6498 + }, + { + "epoch": 0.4884262738614159, + "grad_norm": 1.7682996088827545, + "learning_rate": 2.172128928115229e-06, + "loss": 0.9135, + "num_input_tokens_seen": 84946440, + "step": 6499 + }, + { + "epoch": 0.4885014279272509, + "grad_norm": 0.7439790885791155, + "learning_rate": 2.1716438872238227e-06, + "loss": 0.749, + "num_input_tokens_seen": 85022535, + "step": 6500 + }, + { + "epoch": 0.48857658199308585, + "grad_norm": 1.9748549492602427, + "learning_rate": 2.171158836161859e-06, + "loss": 0.9375, + "num_input_tokens_seen": 85043910, + "step": 6501 + }, + { + "epoch": 0.48865173605892076, + "grad_norm": 0.8364024785364936, + "learning_rate": 2.1706737749580783e-06, + "loss": 0.8655, + "num_input_tokens_seen": 85123045, + "step": 6502 + }, + { + "epoch": 0.48872689012475573, + "grad_norm": 1.6489964843629243, + "learning_rate": 2.1701887036412236e-06, + "loss": 1.0291, + "num_input_tokens_seen": 85145115, + "step": 6503 + }, + { + "epoch": 0.4888020441905907, + "grad_norm": 1.6014710935234246, + "learning_rate": 2.1697036222400355e-06, + "loss": 0.9931, + "num_input_tokens_seen": 85168425, + "step": 6504 + }, + { + "epoch": 0.48887719825642567, + "grad_norm": 1.8271922805068568, + "learning_rate": 2.169218530783259e-06, + "loss": 0.9854, + "num_input_tokens_seen": 85188635, + "step": 6505 + }, + { + "epoch": 0.48895235232226064, + "grad_norm": 1.6844790226622908, + "learning_rate": 2.1687334292996363e-06, + "loss": 0.8786, + "num_input_tokens_seen": 85212675, + "step": 6506 + }, + { + "epoch": 0.4890275063880956, + "grad_norm": 1.318298021937555, + "learning_rate": 2.168248317817911e-06, + "loss": 0.9227, + "num_input_tokens_seen": 85236710, + "step": 6507 + }, + { + "epoch": 0.4891026604539306, + "grad_norm": 0.8630760259635403, + "learning_rate": 2.1677631963668298e-06, + "loss": 0.9259, + "num_input_tokens_seen": 85311495, + "step": 6508 + }, + { + "epoch": 0.4891778145197655, + "grad_norm": 1.8012048715395517, + "learning_rate": 2.1672780649751353e-06, + "loss": 0.9394, + "num_input_tokens_seen": 85335960, + "step": 6509 + }, + { + "epoch": 0.48925296858560047, + "grad_norm": 1.6094464036230762, + "learning_rate": 2.166792923671576e-06, + "loss": 0.9196, + "num_input_tokens_seen": 85361120, + "step": 6510 + }, + { + "epoch": 0.48932812265143544, + "grad_norm": 1.5367546011423165, + "learning_rate": 2.166307772484898e-06, + "loss": 0.9625, + "num_input_tokens_seen": 85383785, + "step": 6511 + }, + { + "epoch": 0.4894032767172704, + "grad_norm": 1.2881835252699503, + "learning_rate": 2.1658226114438457e-06, + "loss": 0.9989, + "num_input_tokens_seen": 85410585, + "step": 6512 + }, + { + "epoch": 0.4894784307831054, + "grad_norm": 1.990266359800336, + "learning_rate": 2.1653374405771696e-06, + "loss": 1.0296, + "num_input_tokens_seen": 85431795, + "step": 6513 + }, + { + "epoch": 0.48955358484894035, + "grad_norm": 1.7968116417370958, + "learning_rate": 2.1648522599136173e-06, + "loss": 0.817, + "num_input_tokens_seen": 85454735, + "step": 6514 + }, + { + "epoch": 0.48962873891477526, + "grad_norm": 1.5794927127932075, + "learning_rate": 2.1643670694819375e-06, + "loss": 0.9261, + "num_input_tokens_seen": 85478925, + "step": 6515 + }, + { + "epoch": 0.48970389298061023, + "grad_norm": 1.4594229941471173, + "learning_rate": 2.163881869310879e-06, + "loss": 0.9942, + "num_input_tokens_seen": 85499765, + "step": 6516 + }, + { + "epoch": 0.4897790470464452, + "grad_norm": 1.3947891723443373, + "learning_rate": 2.163396659429192e-06, + "loss": 0.9531, + "num_input_tokens_seen": 85523005, + "step": 6517 + }, + { + "epoch": 0.4898542011122802, + "grad_norm": 1.5890260135208982, + "learning_rate": 2.162911439865628e-06, + "loss": 0.9655, + "num_input_tokens_seen": 85541070, + "step": 6518 + }, + { + "epoch": 0.48992935517811514, + "grad_norm": 1.5365860277119834, + "learning_rate": 2.1624262106489368e-06, + "loss": 0.865, + "num_input_tokens_seen": 85564990, + "step": 6519 + }, + { + "epoch": 0.4900045092439501, + "grad_norm": 2.2886693831725, + "learning_rate": 2.161940971807871e-06, + "loss": 0.9755, + "num_input_tokens_seen": 85586460, + "step": 6520 + }, + { + "epoch": 0.4900796633097851, + "grad_norm": 2.0394650350623675, + "learning_rate": 2.1614557233711817e-06, + "loss": 0.8994, + "num_input_tokens_seen": 85610200, + "step": 6521 + }, + { + "epoch": 0.49015481737562, + "grad_norm": 1.5850905989978143, + "learning_rate": 2.1609704653676234e-06, + "loss": 0.9406, + "num_input_tokens_seen": 85634680, + "step": 6522 + }, + { + "epoch": 0.49022997144145497, + "grad_norm": 1.315488772916714, + "learning_rate": 2.1604851978259485e-06, + "loss": 0.8312, + "num_input_tokens_seen": 85656655, + "step": 6523 + }, + { + "epoch": 0.49030512550728994, + "grad_norm": 1.8235471089425452, + "learning_rate": 2.159999920774911e-06, + "loss": 0.9279, + "num_input_tokens_seen": 85679370, + "step": 6524 + }, + { + "epoch": 0.4903802795731249, + "grad_norm": 1.9567451800960445, + "learning_rate": 2.1595146342432655e-06, + "loss": 0.9732, + "num_input_tokens_seen": 85701800, + "step": 6525 + }, + { + "epoch": 0.4904554336389599, + "grad_norm": 1.4289024156922567, + "learning_rate": 2.1590293382597667e-06, + "loss": 0.9907, + "num_input_tokens_seen": 85725835, + "step": 6526 + }, + { + "epoch": 0.49053058770479485, + "grad_norm": 1.6390437907850302, + "learning_rate": 2.158544032853171e-06, + "loss": 0.9859, + "num_input_tokens_seen": 85750125, + "step": 6527 + }, + { + "epoch": 0.49060574177062977, + "grad_norm": 1.5304765889004521, + "learning_rate": 2.1580587180522345e-06, + "loss": 0.8525, + "num_input_tokens_seen": 85776380, + "step": 6528 + }, + { + "epoch": 0.49068089583646474, + "grad_norm": 1.3374650041511746, + "learning_rate": 2.1575733938857134e-06, + "loss": 1.0327, + "num_input_tokens_seen": 85801665, + "step": 6529 + }, + { + "epoch": 0.4907560499022997, + "grad_norm": 1.3887065985049767, + "learning_rate": 2.1570880603823654e-06, + "loss": 0.9585, + "num_input_tokens_seen": 85825785, + "step": 6530 + }, + { + "epoch": 0.4908312039681347, + "grad_norm": 1.5025604171157745, + "learning_rate": 2.156602717570948e-06, + "loss": 0.8384, + "num_input_tokens_seen": 85853140, + "step": 6531 + }, + { + "epoch": 0.49090635803396965, + "grad_norm": 1.6666476804923418, + "learning_rate": 2.15611736548022e-06, + "loss": 0.9145, + "num_input_tokens_seen": 85875295, + "step": 6532 + }, + { + "epoch": 0.4909815120998046, + "grad_norm": 1.6921200556536646, + "learning_rate": 2.1556320041389407e-06, + "loss": 1.0021, + "num_input_tokens_seen": 85898185, + "step": 6533 + }, + { + "epoch": 0.4910566661656396, + "grad_norm": 1.8279351029844184, + "learning_rate": 2.155146633575869e-06, + "loss": 0.9308, + "num_input_tokens_seen": 85920155, + "step": 6534 + }, + { + "epoch": 0.4911318202314745, + "grad_norm": 1.8257135642306332, + "learning_rate": 2.154661253819765e-06, + "loss": 0.7842, + "num_input_tokens_seen": 85941975, + "step": 6535 + }, + { + "epoch": 0.4912069742973095, + "grad_norm": 1.2563896424223342, + "learning_rate": 2.1541758648993895e-06, + "loss": 1.0095, + "num_input_tokens_seen": 85966480, + "step": 6536 + }, + { + "epoch": 0.49128212836314444, + "grad_norm": 1.5154115978901235, + "learning_rate": 2.1536904668435035e-06, + "loss": 0.918, + "num_input_tokens_seen": 85987725, + "step": 6537 + }, + { + "epoch": 0.4913572824289794, + "grad_norm": 1.439969470972056, + "learning_rate": 2.153205059680869e-06, + "loss": 1.0408, + "num_input_tokens_seen": 86010300, + "step": 6538 + }, + { + "epoch": 0.4914324364948144, + "grad_norm": 1.4254638318544925, + "learning_rate": 2.1527196434402483e-06, + "loss": 0.9606, + "num_input_tokens_seen": 86034865, + "step": 6539 + }, + { + "epoch": 0.49150759056064935, + "grad_norm": 1.3837870811173623, + "learning_rate": 2.152234218150404e-06, + "loss": 0.8336, + "num_input_tokens_seen": 86058340, + "step": 6540 + }, + { + "epoch": 0.49158274462648427, + "grad_norm": 1.6197492424253213, + "learning_rate": 2.1517487838400984e-06, + "loss": 0.9756, + "num_input_tokens_seen": 86080870, + "step": 6541 + }, + { + "epoch": 0.49165789869231924, + "grad_norm": 1.436633968703991, + "learning_rate": 2.1512633405380972e-06, + "loss": 0.9845, + "num_input_tokens_seen": 86103740, + "step": 6542 + }, + { + "epoch": 0.4917330527581542, + "grad_norm": 1.55936514138448, + "learning_rate": 2.1507778882731635e-06, + "loss": 0.9285, + "num_input_tokens_seen": 86122860, + "step": 6543 + }, + { + "epoch": 0.4918082068239892, + "grad_norm": 1.9537492288598939, + "learning_rate": 2.1502924270740626e-06, + "loss": 0.9081, + "num_input_tokens_seen": 86143855, + "step": 6544 + }, + { + "epoch": 0.49188336088982415, + "grad_norm": 1.763701209482305, + "learning_rate": 2.1498069569695605e-06, + "loss": 0.8919, + "num_input_tokens_seen": 86164930, + "step": 6545 + }, + { + "epoch": 0.4919585149556591, + "grad_norm": 0.819809423316556, + "learning_rate": 2.149321477988421e-06, + "loss": 0.8507, + "num_input_tokens_seen": 86251025, + "step": 6546 + }, + { + "epoch": 0.49203366902149404, + "grad_norm": 1.5491232668970762, + "learning_rate": 2.1488359901594137e-06, + "loss": 0.9418, + "num_input_tokens_seen": 86274935, + "step": 6547 + }, + { + "epoch": 0.492108823087329, + "grad_norm": 1.4342806051605104, + "learning_rate": 2.148350493511303e-06, + "loss": 0.9268, + "num_input_tokens_seen": 86297050, + "step": 6548 + }, + { + "epoch": 0.492183977153164, + "grad_norm": 1.5760556240770618, + "learning_rate": 2.1478649880728582e-06, + "loss": 1.0122, + "num_input_tokens_seen": 86319220, + "step": 6549 + }, + { + "epoch": 0.49225913121899895, + "grad_norm": 1.4040184965355305, + "learning_rate": 2.1473794738728466e-06, + "loss": 1.0013, + "num_input_tokens_seen": 86341955, + "step": 6550 + }, + { + "epoch": 0.4923342852848339, + "grad_norm": 1.302856393844645, + "learning_rate": 2.1468939509400363e-06, + "loss": 0.9908, + "num_input_tokens_seen": 86364990, + "step": 6551 + }, + { + "epoch": 0.4924094393506689, + "grad_norm": 1.3761189849331197, + "learning_rate": 2.1464084193031976e-06, + "loss": 0.8666, + "num_input_tokens_seen": 86389865, + "step": 6552 + }, + { + "epoch": 0.49248459341650386, + "grad_norm": 1.42491462184052, + "learning_rate": 2.1459228789910983e-06, + "loss": 0.9174, + "num_input_tokens_seen": 86415175, + "step": 6553 + }, + { + "epoch": 0.4925597474823388, + "grad_norm": 1.5381357723821445, + "learning_rate": 2.1454373300325103e-06, + "loss": 0.9491, + "num_input_tokens_seen": 86440550, + "step": 6554 + }, + { + "epoch": 0.49263490154817374, + "grad_norm": 1.364579377884507, + "learning_rate": 2.1449517724562037e-06, + "loss": 0.9756, + "num_input_tokens_seen": 86462020, + "step": 6555 + }, + { + "epoch": 0.4927100556140087, + "grad_norm": 1.4639311936161254, + "learning_rate": 2.1444662062909495e-06, + "loss": 0.9215, + "num_input_tokens_seen": 86486800, + "step": 6556 + }, + { + "epoch": 0.4927852096798437, + "grad_norm": 1.7629105624071668, + "learning_rate": 2.1439806315655197e-06, + "loss": 1.0243, + "num_input_tokens_seen": 86507240, + "step": 6557 + }, + { + "epoch": 0.49286036374567865, + "grad_norm": 1.6895241783062533, + "learning_rate": 2.1434950483086856e-06, + "loss": 0.9935, + "num_input_tokens_seen": 86529770, + "step": 6558 + }, + { + "epoch": 0.4929355178115136, + "grad_norm": 1.5965411858172855, + "learning_rate": 2.143009456549221e-06, + "loss": 0.9189, + "num_input_tokens_seen": 86551265, + "step": 6559 + }, + { + "epoch": 0.49301067187734854, + "grad_norm": 1.460932767319791, + "learning_rate": 2.1425238563158975e-06, + "loss": 0.9514, + "num_input_tokens_seen": 86572545, + "step": 6560 + }, + { + "epoch": 0.4930858259431835, + "grad_norm": 1.476687875097373, + "learning_rate": 2.1420382476374905e-06, + "loss": 0.9716, + "num_input_tokens_seen": 86596240, + "step": 6561 + }, + { + "epoch": 0.4931609800090185, + "grad_norm": 1.6498282834489668, + "learning_rate": 2.1415526305427735e-06, + "loss": 0.9828, + "num_input_tokens_seen": 86620695, + "step": 6562 + }, + { + "epoch": 0.49323613407485345, + "grad_norm": 1.8573098133831643, + "learning_rate": 2.1410670050605203e-06, + "loss": 0.828, + "num_input_tokens_seen": 86641680, + "step": 6563 + }, + { + "epoch": 0.4933112881406884, + "grad_norm": 0.8174021364249919, + "learning_rate": 2.140581371219508e-06, + "loss": 0.7778, + "num_input_tokens_seen": 86724415, + "step": 6564 + }, + { + "epoch": 0.4933864422065234, + "grad_norm": 1.5589739113582934, + "learning_rate": 2.14009572904851e-06, + "loss": 0.9963, + "num_input_tokens_seen": 86750745, + "step": 6565 + }, + { + "epoch": 0.49346159627235836, + "grad_norm": 2.7853670829595725, + "learning_rate": 2.139610078576305e-06, + "loss": 0.8739, + "num_input_tokens_seen": 86772300, + "step": 6566 + }, + { + "epoch": 0.4935367503381933, + "grad_norm": 6.388947432245173, + "learning_rate": 2.139124419831667e-06, + "loss": 0.941, + "num_input_tokens_seen": 86797020, + "step": 6567 + }, + { + "epoch": 0.49361190440402825, + "grad_norm": 1.7792955350399728, + "learning_rate": 2.1386387528433743e-06, + "loss": 0.9323, + "num_input_tokens_seen": 86819545, + "step": 6568 + }, + { + "epoch": 0.4936870584698632, + "grad_norm": 1.6944694708277097, + "learning_rate": 2.1381530776402058e-06, + "loss": 0.9928, + "num_input_tokens_seen": 86841300, + "step": 6569 + }, + { + "epoch": 0.4937622125356982, + "grad_norm": 1.7593125110322525, + "learning_rate": 2.1376673942509373e-06, + "loss": 0.9103, + "num_input_tokens_seen": 86862445, + "step": 6570 + }, + { + "epoch": 0.49383736660153316, + "grad_norm": 1.490602700248595, + "learning_rate": 2.1371817027043494e-06, + "loss": 1.0437, + "num_input_tokens_seen": 86885930, + "step": 6571 + }, + { + "epoch": 0.4939125206673681, + "grad_norm": 2.430175732196227, + "learning_rate": 2.1366960030292195e-06, + "loss": 0.8696, + "num_input_tokens_seen": 86906115, + "step": 6572 + }, + { + "epoch": 0.49398767473320304, + "grad_norm": 1.662694520971414, + "learning_rate": 2.1362102952543277e-06, + "loss": 0.7757, + "num_input_tokens_seen": 86929115, + "step": 6573 + }, + { + "epoch": 0.494062828799038, + "grad_norm": 1.250525030499519, + "learning_rate": 2.135724579408456e-06, + "loss": 0.964, + "num_input_tokens_seen": 86954690, + "step": 6574 + }, + { + "epoch": 0.494137982864873, + "grad_norm": 1.3857622743297298, + "learning_rate": 2.1352388555203814e-06, + "loss": 1.0025, + "num_input_tokens_seen": 86981615, + "step": 6575 + }, + { + "epoch": 0.49421313693070795, + "grad_norm": 1.3231056058432626, + "learning_rate": 2.1347531236188878e-06, + "loss": 0.9483, + "num_input_tokens_seen": 87006425, + "step": 6576 + }, + { + "epoch": 0.4942882909965429, + "grad_norm": 1.565276920815186, + "learning_rate": 2.134267383732755e-06, + "loss": 0.8887, + "num_input_tokens_seen": 87031170, + "step": 6577 + }, + { + "epoch": 0.4943634450623779, + "grad_norm": 1.429609119372108, + "learning_rate": 2.1337816358907663e-06, + "loss": 0.9085, + "num_input_tokens_seen": 87055085, + "step": 6578 + }, + { + "epoch": 0.49443859912821286, + "grad_norm": 1.469777904133355, + "learning_rate": 2.133295880121703e-06, + "loss": 0.9804, + "num_input_tokens_seen": 87081460, + "step": 6579 + }, + { + "epoch": 0.4945137531940478, + "grad_norm": 1.582304151568234, + "learning_rate": 2.132810116454348e-06, + "loss": 1.0057, + "num_input_tokens_seen": 87101735, + "step": 6580 + }, + { + "epoch": 0.49458890725988275, + "grad_norm": 1.4416951869429415, + "learning_rate": 2.132324344917486e-06, + "loss": 0.9691, + "num_input_tokens_seen": 87124155, + "step": 6581 + }, + { + "epoch": 0.4946640613257177, + "grad_norm": 2.173541191474216, + "learning_rate": 2.131838565539899e-06, + "loss": 1.001, + "num_input_tokens_seen": 87148080, + "step": 6582 + }, + { + "epoch": 0.4947392153915527, + "grad_norm": 2.7929387017614395, + "learning_rate": 2.1313527783503727e-06, + "loss": 0.8815, + "num_input_tokens_seen": 87173005, + "step": 6583 + }, + { + "epoch": 0.49481436945738766, + "grad_norm": 0.7452567633495024, + "learning_rate": 2.1308669833776907e-06, + "loss": 0.7728, + "num_input_tokens_seen": 87251850, + "step": 6584 + }, + { + "epoch": 0.49488952352322263, + "grad_norm": 1.3617897072107128, + "learning_rate": 2.130381180650639e-06, + "loss": 0.8996, + "num_input_tokens_seen": 87275660, + "step": 6585 + }, + { + "epoch": 0.49496467758905754, + "grad_norm": 1.4298364113556925, + "learning_rate": 2.1298953701980038e-06, + "loss": 1.1048, + "num_input_tokens_seen": 87298910, + "step": 6586 + }, + { + "epoch": 0.4950398316548925, + "grad_norm": 0.7959775920426382, + "learning_rate": 2.129409552048569e-06, + "loss": 0.853, + "num_input_tokens_seen": 87369850, + "step": 6587 + }, + { + "epoch": 0.4951149857207275, + "grad_norm": 1.3564519037535594, + "learning_rate": 2.1289237262311243e-06, + "loss": 0.9726, + "num_input_tokens_seen": 87394795, + "step": 6588 + }, + { + "epoch": 0.49519013978656246, + "grad_norm": 1.5774058601352765, + "learning_rate": 2.1284378927744546e-06, + "loss": 0.9511, + "num_input_tokens_seen": 87416515, + "step": 6589 + }, + { + "epoch": 0.4952652938523974, + "grad_norm": 1.9133072589245101, + "learning_rate": 2.1279520517073475e-06, + "loss": 0.9358, + "num_input_tokens_seen": 87437955, + "step": 6590 + }, + { + "epoch": 0.4953404479182324, + "grad_norm": 1.4249916712199635, + "learning_rate": 2.127466203058592e-06, + "loss": 1.0414, + "num_input_tokens_seen": 87462805, + "step": 6591 + }, + { + "epoch": 0.4954156019840673, + "grad_norm": 1.2805799267247902, + "learning_rate": 2.1269803468569756e-06, + "loss": 0.9925, + "num_input_tokens_seen": 87488135, + "step": 6592 + }, + { + "epoch": 0.4954907560499023, + "grad_norm": 0.7975055967032904, + "learning_rate": 2.1264944831312874e-06, + "loss": 0.8253, + "num_input_tokens_seen": 87568045, + "step": 6593 + }, + { + "epoch": 0.49556591011573725, + "grad_norm": 1.263927451559658, + "learning_rate": 2.1260086119103165e-06, + "loss": 0.9012, + "num_input_tokens_seen": 87591715, + "step": 6594 + }, + { + "epoch": 0.4956410641815722, + "grad_norm": 1.3719676566496162, + "learning_rate": 2.1255227332228527e-06, + "loss": 0.9976, + "num_input_tokens_seen": 87615415, + "step": 6595 + }, + { + "epoch": 0.4957162182474072, + "grad_norm": 1.3127007103905346, + "learning_rate": 2.125036847097687e-06, + "loss": 0.9587, + "num_input_tokens_seen": 87640940, + "step": 6596 + }, + { + "epoch": 0.49579137231324216, + "grad_norm": 1.521759987623022, + "learning_rate": 2.1245509535636088e-06, + "loss": 0.9989, + "num_input_tokens_seen": 87664960, + "step": 6597 + }, + { + "epoch": 0.49586652637907713, + "grad_norm": 1.700085088059014, + "learning_rate": 2.1240650526494096e-06, + "loss": 0.9246, + "num_input_tokens_seen": 87687495, + "step": 6598 + }, + { + "epoch": 0.49594168044491205, + "grad_norm": 0.8667535469328091, + "learning_rate": 2.1235791443838804e-06, + "loss": 0.8014, + "num_input_tokens_seen": 87774095, + "step": 6599 + }, + { + "epoch": 0.496016834510747, + "grad_norm": 1.6913419010881101, + "learning_rate": 2.123093228795815e-06, + "loss": 0.8789, + "num_input_tokens_seen": 87798440, + "step": 6600 + }, + { + "epoch": 0.496091988576582, + "grad_norm": 1.6784399323134198, + "learning_rate": 2.122607305914004e-06, + "loss": 0.851, + "num_input_tokens_seen": 87820305, + "step": 6601 + }, + { + "epoch": 0.49616714264241696, + "grad_norm": 1.4354504611277703, + "learning_rate": 2.1221213757672404e-06, + "loss": 0.8562, + "num_input_tokens_seen": 87845080, + "step": 6602 + }, + { + "epoch": 0.49624229670825193, + "grad_norm": 0.9592418441564929, + "learning_rate": 2.1216354383843176e-06, + "loss": 0.8796, + "num_input_tokens_seen": 87921890, + "step": 6603 + }, + { + "epoch": 0.4963174507740869, + "grad_norm": 1.8908397458022241, + "learning_rate": 2.1211494937940296e-06, + "loss": 0.8852, + "num_input_tokens_seen": 87941655, + "step": 6604 + }, + { + "epoch": 0.4963926048399218, + "grad_norm": 1.4350230396561137, + "learning_rate": 2.12066354202517e-06, + "loss": 0.8693, + "num_input_tokens_seen": 87967610, + "step": 6605 + }, + { + "epoch": 0.4964677589057568, + "grad_norm": 1.3108614730158858, + "learning_rate": 2.1201775831065336e-06, + "loss": 0.9105, + "num_input_tokens_seen": 87994760, + "step": 6606 + }, + { + "epoch": 0.49654291297159175, + "grad_norm": 1.3773343471610766, + "learning_rate": 2.119691617066915e-06, + "loss": 0.7991, + "num_input_tokens_seen": 88018250, + "step": 6607 + }, + { + "epoch": 0.4966180670374267, + "grad_norm": 1.4021017986893602, + "learning_rate": 2.11920564393511e-06, + "loss": 0.9866, + "num_input_tokens_seen": 88041310, + "step": 6608 + }, + { + "epoch": 0.4966932211032617, + "grad_norm": 1.453792131753165, + "learning_rate": 2.1187196637399138e-06, + "loss": 0.9152, + "num_input_tokens_seen": 88064320, + "step": 6609 + }, + { + "epoch": 0.49676837516909667, + "grad_norm": 1.4932366322957347, + "learning_rate": 2.118233676510123e-06, + "loss": 0.9175, + "num_input_tokens_seen": 88087110, + "step": 6610 + }, + { + "epoch": 0.49684352923493164, + "grad_norm": 1.7834328778134987, + "learning_rate": 2.1177476822745344e-06, + "loss": 1.0301, + "num_input_tokens_seen": 88107685, + "step": 6611 + }, + { + "epoch": 0.49691868330076655, + "grad_norm": 1.4450154592089681, + "learning_rate": 2.1172616810619446e-06, + "loss": 0.9212, + "num_input_tokens_seen": 88127690, + "step": 6612 + }, + { + "epoch": 0.4969938373666015, + "grad_norm": 1.4282187333012364, + "learning_rate": 2.116775672901151e-06, + "loss": 0.8909, + "num_input_tokens_seen": 88150015, + "step": 6613 + }, + { + "epoch": 0.4970689914324365, + "grad_norm": 1.3785261111078173, + "learning_rate": 2.1162896578209516e-06, + "loss": 0.8037, + "num_input_tokens_seen": 88175190, + "step": 6614 + }, + { + "epoch": 0.49714414549827146, + "grad_norm": 1.4680258406357887, + "learning_rate": 2.115803635850145e-06, + "loss": 1.0411, + "num_input_tokens_seen": 88198145, + "step": 6615 + }, + { + "epoch": 0.49721929956410643, + "grad_norm": 1.7572252437005305, + "learning_rate": 2.1153176070175297e-06, + "loss": 0.8436, + "num_input_tokens_seen": 88221780, + "step": 6616 + }, + { + "epoch": 0.4972944536299414, + "grad_norm": 1.645544034745982, + "learning_rate": 2.1148315713519036e-06, + "loss": 0.9405, + "num_input_tokens_seen": 88246995, + "step": 6617 + }, + { + "epoch": 0.4973696076957763, + "grad_norm": 1.9589238865832725, + "learning_rate": 2.114345528882068e-06, + "loss": 0.8409, + "num_input_tokens_seen": 88269760, + "step": 6618 + }, + { + "epoch": 0.4974447617616113, + "grad_norm": 2.6391380683193413, + "learning_rate": 2.1138594796368213e-06, + "loss": 0.8791, + "num_input_tokens_seen": 88292995, + "step": 6619 + }, + { + "epoch": 0.49751991582744626, + "grad_norm": 1.9701856192651317, + "learning_rate": 2.1133734236449654e-06, + "loss": 0.9243, + "num_input_tokens_seen": 88317985, + "step": 6620 + }, + { + "epoch": 0.4975950698932812, + "grad_norm": 1.759602089246796, + "learning_rate": 2.112887360935299e-06, + "loss": 0.9308, + "num_input_tokens_seen": 88339550, + "step": 6621 + }, + { + "epoch": 0.4976702239591162, + "grad_norm": 1.6582543822365328, + "learning_rate": 2.112401291536625e-06, + "loss": 0.9492, + "num_input_tokens_seen": 88365225, + "step": 6622 + }, + { + "epoch": 0.49774537802495117, + "grad_norm": 0.7796189639255535, + "learning_rate": 2.1119152154777442e-06, + "loss": 0.8901, + "num_input_tokens_seen": 88436110, + "step": 6623 + }, + { + "epoch": 0.49782053209078614, + "grad_norm": 1.6954478548221472, + "learning_rate": 2.1114291327874578e-06, + "loss": 1.0549, + "num_input_tokens_seen": 88457725, + "step": 6624 + }, + { + "epoch": 0.49789568615662105, + "grad_norm": 2.1866588610222912, + "learning_rate": 2.1109430434945685e-06, + "loss": 0.976, + "num_input_tokens_seen": 88481065, + "step": 6625 + }, + { + "epoch": 0.497970840222456, + "grad_norm": 1.830336252286037, + "learning_rate": 2.1104569476278794e-06, + "loss": 1.0158, + "num_input_tokens_seen": 88503760, + "step": 6626 + }, + { + "epoch": 0.498045994288291, + "grad_norm": 1.3761569704217231, + "learning_rate": 2.109970845216193e-06, + "loss": 0.9038, + "num_input_tokens_seen": 88529495, + "step": 6627 + }, + { + "epoch": 0.49812114835412596, + "grad_norm": 1.5271070057836966, + "learning_rate": 2.109484736288313e-06, + "loss": 1.0026, + "num_input_tokens_seen": 88551635, + "step": 6628 + }, + { + "epoch": 0.49819630241996093, + "grad_norm": 1.7635917096173936, + "learning_rate": 2.108998620873043e-06, + "loss": 1.0113, + "num_input_tokens_seen": 88573015, + "step": 6629 + }, + { + "epoch": 0.4982714564857959, + "grad_norm": 0.7110159694418048, + "learning_rate": 2.1085124989991876e-06, + "loss": 0.8027, + "num_input_tokens_seen": 88648300, + "step": 6630 + }, + { + "epoch": 0.4983466105516308, + "grad_norm": 1.621866138705998, + "learning_rate": 2.108026370695551e-06, + "loss": 0.9918, + "num_input_tokens_seen": 88669455, + "step": 6631 + }, + { + "epoch": 0.4984217646174658, + "grad_norm": 3.531523972407378, + "learning_rate": 2.107540235990938e-06, + "loss": 1.0318, + "num_input_tokens_seen": 88692215, + "step": 6632 + }, + { + "epoch": 0.49849691868330076, + "grad_norm": 1.8798380422834475, + "learning_rate": 2.107054094914155e-06, + "loss": 0.8767, + "num_input_tokens_seen": 88716960, + "step": 6633 + }, + { + "epoch": 0.49857207274913573, + "grad_norm": 1.4609099196824264, + "learning_rate": 2.106567947494006e-06, + "loss": 1.0709, + "num_input_tokens_seen": 88740820, + "step": 6634 + }, + { + "epoch": 0.4986472268149707, + "grad_norm": 1.3007251166068343, + "learning_rate": 2.106081793759298e-06, + "loss": 0.9745, + "num_input_tokens_seen": 88765475, + "step": 6635 + }, + { + "epoch": 0.49872238088080567, + "grad_norm": 1.2384087048710102, + "learning_rate": 2.1055956337388376e-06, + "loss": 0.9815, + "num_input_tokens_seen": 88793810, + "step": 6636 + }, + { + "epoch": 0.4987975349466406, + "grad_norm": 1.7169138724221293, + "learning_rate": 2.1051094674614327e-06, + "loss": 1.0326, + "num_input_tokens_seen": 88815825, + "step": 6637 + }, + { + "epoch": 0.49887268901247556, + "grad_norm": 1.4487368869928607, + "learning_rate": 2.1046232949558887e-06, + "loss": 0.9372, + "num_input_tokens_seen": 88839510, + "step": 6638 + }, + { + "epoch": 0.4989478430783105, + "grad_norm": 1.613552486349745, + "learning_rate": 2.104137116251013e-06, + "loss": 0.9036, + "num_input_tokens_seen": 88862400, + "step": 6639 + }, + { + "epoch": 0.4990229971441455, + "grad_norm": 1.3911666243968162, + "learning_rate": 2.103650931375615e-06, + "loss": 1.0104, + "num_input_tokens_seen": 88887865, + "step": 6640 + }, + { + "epoch": 0.49909815120998047, + "grad_norm": 0.7825557478037685, + "learning_rate": 2.103164740358502e-06, + "loss": 0.8339, + "num_input_tokens_seen": 88963800, + "step": 6641 + }, + { + "epoch": 0.49917330527581544, + "grad_norm": 2.0837542935094326, + "learning_rate": 2.1026785432284837e-06, + "loss": 0.9097, + "num_input_tokens_seen": 88986140, + "step": 6642 + }, + { + "epoch": 0.4992484593416504, + "grad_norm": 1.4823865722955973, + "learning_rate": 2.1021923400143683e-06, + "loss": 1.0256, + "num_input_tokens_seen": 89005535, + "step": 6643 + }, + { + "epoch": 0.4993236134074853, + "grad_norm": 1.6727550008844716, + "learning_rate": 2.101706130744966e-06, + "loss": 1.0318, + "num_input_tokens_seen": 89027780, + "step": 6644 + }, + { + "epoch": 0.4993987674733203, + "grad_norm": 1.4653298347149064, + "learning_rate": 2.1012199154490852e-06, + "loss": 0.891, + "num_input_tokens_seen": 89049625, + "step": 6645 + }, + { + "epoch": 0.49947392153915526, + "grad_norm": 2.2095239656507815, + "learning_rate": 2.1007336941555374e-06, + "loss": 0.9692, + "num_input_tokens_seen": 89071470, + "step": 6646 + }, + { + "epoch": 0.49954907560499023, + "grad_norm": 1.76776600887671, + "learning_rate": 2.100247466893132e-06, + "loss": 0.9148, + "num_input_tokens_seen": 89095730, + "step": 6647 + }, + { + "epoch": 0.4996242296708252, + "grad_norm": 2.09952329947109, + "learning_rate": 2.0997612336906805e-06, + "loss": 0.9355, + "num_input_tokens_seen": 89119145, + "step": 6648 + }, + { + "epoch": 0.4996993837366602, + "grad_norm": 1.35203120753708, + "learning_rate": 2.099274994576994e-06, + "loss": 0.8966, + "num_input_tokens_seen": 89143305, + "step": 6649 + }, + { + "epoch": 0.4997745378024951, + "grad_norm": 1.6542297133466812, + "learning_rate": 2.098788749580884e-06, + "loss": 0.841, + "num_input_tokens_seen": 89169720, + "step": 6650 + }, + { + "epoch": 0.49984969186833006, + "grad_norm": 1.456873908253199, + "learning_rate": 2.098302498731162e-06, + "loss": 0.8902, + "num_input_tokens_seen": 89192565, + "step": 6651 + }, + { + "epoch": 0.49992484593416503, + "grad_norm": 1.5001303298279542, + "learning_rate": 2.0978162420566406e-06, + "loss": 0.997, + "num_input_tokens_seen": 89217070, + "step": 6652 + }, + { + "epoch": 0.5, + "grad_norm": 0.7139561880758659, + "learning_rate": 2.0973299795861322e-06, + "loss": 0.8205, + "num_input_tokens_seen": 89299005, + "step": 6653 + }, + { + "epoch": 0.5000751540658349, + "grad_norm": 1.3390464523582373, + "learning_rate": 2.09684371134845e-06, + "loss": 0.9083, + "num_input_tokens_seen": 89324585, + "step": 6654 + }, + { + "epoch": 0.5001503081316699, + "grad_norm": 1.499743996072901, + "learning_rate": 2.0963574373724074e-06, + "loss": 1.008, + "num_input_tokens_seen": 89346970, + "step": 6655 + }, + { + "epoch": 0.5002254621975049, + "grad_norm": 1.665432791149477, + "learning_rate": 2.095871157686817e-06, + "loss": 0.9836, + "num_input_tokens_seen": 89366855, + "step": 6656 + }, + { + "epoch": 0.5003006162633399, + "grad_norm": 1.582705657757209, + "learning_rate": 2.095384872320494e-06, + "loss": 0.8838, + "num_input_tokens_seen": 89391880, + "step": 6657 + }, + { + "epoch": 0.5003757703291748, + "grad_norm": 1.7707453387968033, + "learning_rate": 2.094898581302251e-06, + "loss": 0.9176, + "num_input_tokens_seen": 89413755, + "step": 6658 + }, + { + "epoch": 0.5004509243950098, + "grad_norm": 1.4662790819488931, + "learning_rate": 2.094412284660905e-06, + "loss": 0.9277, + "num_input_tokens_seen": 89436050, + "step": 6659 + }, + { + "epoch": 0.5005260784608447, + "grad_norm": 1.4587983979317038, + "learning_rate": 2.093925982425269e-06, + "loss": 0.9601, + "num_input_tokens_seen": 89460140, + "step": 6660 + }, + { + "epoch": 0.5006012325266797, + "grad_norm": 1.2405465281385308, + "learning_rate": 2.093439674624158e-06, + "loss": 0.8856, + "num_input_tokens_seen": 89487150, + "step": 6661 + }, + { + "epoch": 0.5006763865925147, + "grad_norm": 1.631134046620833, + "learning_rate": 2.09295336128639e-06, + "loss": 0.9407, + "num_input_tokens_seen": 89513195, + "step": 6662 + }, + { + "epoch": 0.5007515406583496, + "grad_norm": 1.3158169114866012, + "learning_rate": 2.0924670424407785e-06, + "loss": 0.8848, + "num_input_tokens_seen": 89542360, + "step": 6663 + }, + { + "epoch": 0.5008266947241846, + "grad_norm": 1.4232971289688365, + "learning_rate": 2.091980718116141e-06, + "loss": 0.9298, + "num_input_tokens_seen": 89568755, + "step": 6664 + }, + { + "epoch": 0.5009018487900195, + "grad_norm": 1.610273536286248, + "learning_rate": 2.0914943883412935e-06, + "loss": 0.8819, + "num_input_tokens_seen": 89594255, + "step": 6665 + }, + { + "epoch": 0.5009770028558544, + "grad_norm": 2.2545607226517594, + "learning_rate": 2.0910080531450534e-06, + "loss": 0.9935, + "num_input_tokens_seen": 89619240, + "step": 6666 + }, + { + "epoch": 0.5010521569216895, + "grad_norm": 1.5162305113946501, + "learning_rate": 2.0905217125562378e-06, + "loss": 0.9641, + "num_input_tokens_seen": 89640745, + "step": 6667 + }, + { + "epoch": 0.5011273109875244, + "grad_norm": 0.9579632638884003, + "learning_rate": 2.0900353666036635e-06, + "loss": 0.8955, + "num_input_tokens_seen": 89701685, + "step": 6668 + }, + { + "epoch": 0.5012024650533594, + "grad_norm": 1.322676353955236, + "learning_rate": 2.0895490153161496e-06, + "loss": 1.0482, + "num_input_tokens_seen": 89723155, + "step": 6669 + }, + { + "epoch": 0.5012776191191943, + "grad_norm": 1.6037479283474798, + "learning_rate": 2.089062658722513e-06, + "loss": 0.9975, + "num_input_tokens_seen": 89744585, + "step": 6670 + }, + { + "epoch": 0.5013527731850294, + "grad_norm": 1.590201147899422, + "learning_rate": 2.0885762968515737e-06, + "loss": 1.146, + "num_input_tokens_seen": 89766450, + "step": 6671 + }, + { + "epoch": 0.5014279272508643, + "grad_norm": 1.8278802561157095, + "learning_rate": 2.088089929732149e-06, + "loss": 0.9141, + "num_input_tokens_seen": 89788470, + "step": 6672 + }, + { + "epoch": 0.5015030813166992, + "grad_norm": 1.6554963178441038, + "learning_rate": 2.0876035573930587e-06, + "loss": 0.995, + "num_input_tokens_seen": 89810695, + "step": 6673 + }, + { + "epoch": 0.5015782353825342, + "grad_norm": 1.7640418998387188, + "learning_rate": 2.0871171798631224e-06, + "loss": 0.9347, + "num_input_tokens_seen": 89832930, + "step": 6674 + }, + { + "epoch": 0.5016533894483691, + "grad_norm": 1.5226907318109748, + "learning_rate": 2.0866307971711594e-06, + "loss": 0.9853, + "num_input_tokens_seen": 89854900, + "step": 6675 + }, + { + "epoch": 0.5017285435142042, + "grad_norm": 1.6089989676558156, + "learning_rate": 2.08614440934599e-06, + "loss": 0.8858, + "num_input_tokens_seen": 89878280, + "step": 6676 + }, + { + "epoch": 0.5018036975800391, + "grad_norm": 1.6872499068691575, + "learning_rate": 2.0856580164164344e-06, + "loss": 0.9592, + "num_input_tokens_seen": 89905730, + "step": 6677 + }, + { + "epoch": 0.5018788516458741, + "grad_norm": 1.7449598620370153, + "learning_rate": 2.085171618411313e-06, + "loss": 0.9329, + "num_input_tokens_seen": 89928875, + "step": 6678 + }, + { + "epoch": 0.501954005711709, + "grad_norm": 1.5544137378217093, + "learning_rate": 2.0846852153594477e-06, + "loss": 1.0027, + "num_input_tokens_seen": 89954070, + "step": 6679 + }, + { + "epoch": 0.5020291597775439, + "grad_norm": 2.052712190836656, + "learning_rate": 2.0841988072896585e-06, + "loss": 0.952, + "num_input_tokens_seen": 89976910, + "step": 6680 + }, + { + "epoch": 0.502104313843379, + "grad_norm": 1.7446287695913745, + "learning_rate": 2.0837123942307677e-06, + "loss": 0.9503, + "num_input_tokens_seen": 90000690, + "step": 6681 + }, + { + "epoch": 0.5021794679092139, + "grad_norm": 1.4593600972821714, + "learning_rate": 2.0832259762115977e-06, + "loss": 0.9277, + "num_input_tokens_seen": 90021880, + "step": 6682 + }, + { + "epoch": 0.5022546219750489, + "grad_norm": 6.3952127791402305, + "learning_rate": 2.0827395532609685e-06, + "loss": 1.0017, + "num_input_tokens_seen": 90045925, + "step": 6683 + }, + { + "epoch": 0.5023297760408838, + "grad_norm": 1.8044225985489244, + "learning_rate": 2.082253125407705e-06, + "loss": 0.9929, + "num_input_tokens_seen": 90068015, + "step": 6684 + }, + { + "epoch": 0.5024049301067188, + "grad_norm": 1.3663451733469756, + "learning_rate": 2.0817666926806287e-06, + "loss": 0.9867, + "num_input_tokens_seen": 90091645, + "step": 6685 + }, + { + "epoch": 0.5024800841725537, + "grad_norm": 1.2019576813044581, + "learning_rate": 2.0812802551085633e-06, + "loss": 0.9709, + "num_input_tokens_seen": 90120135, + "step": 6686 + }, + { + "epoch": 0.5025552382383887, + "grad_norm": 1.824836446670876, + "learning_rate": 2.0807938127203304e-06, + "loss": 0.9677, + "num_input_tokens_seen": 90143385, + "step": 6687 + }, + { + "epoch": 0.5026303923042237, + "grad_norm": 1.3570170066318796, + "learning_rate": 2.080307365544755e-06, + "loss": 0.9487, + "num_input_tokens_seen": 90167390, + "step": 6688 + }, + { + "epoch": 0.5027055463700586, + "grad_norm": 1.5239145167899124, + "learning_rate": 2.0798209136106615e-06, + "loss": 1.0643, + "num_input_tokens_seen": 90189115, + "step": 6689 + }, + { + "epoch": 0.5027807004358936, + "grad_norm": 2.1568802727872765, + "learning_rate": 2.0793344569468725e-06, + "loss": 0.9455, + "num_input_tokens_seen": 90211920, + "step": 6690 + }, + { + "epoch": 0.5028558545017285, + "grad_norm": 1.359994067712075, + "learning_rate": 2.0788479955822136e-06, + "loss": 0.9375, + "num_input_tokens_seen": 90237415, + "step": 6691 + }, + { + "epoch": 0.5029310085675635, + "grad_norm": 1.674004811383697, + "learning_rate": 2.0783615295455082e-06, + "loss": 0.9559, + "num_input_tokens_seen": 90258400, + "step": 6692 + }, + { + "epoch": 0.5030061626333985, + "grad_norm": 1.4073215640506718, + "learning_rate": 2.077875058865583e-06, + "loss": 0.9277, + "num_input_tokens_seen": 90282785, + "step": 6693 + }, + { + "epoch": 0.5030813166992334, + "grad_norm": 3.017139588128214, + "learning_rate": 2.077388583571262e-06, + "loss": 0.9276, + "num_input_tokens_seen": 90306300, + "step": 6694 + }, + { + "epoch": 0.5031564707650684, + "grad_norm": 1.6164036569625242, + "learning_rate": 2.076902103691371e-06, + "loss": 0.9199, + "num_input_tokens_seen": 90330735, + "step": 6695 + }, + { + "epoch": 0.5032316248309033, + "grad_norm": 1.766498113484891, + "learning_rate": 2.076415619254736e-06, + "loss": 1.0719, + "num_input_tokens_seen": 90356135, + "step": 6696 + }, + { + "epoch": 0.5033067788967384, + "grad_norm": 1.952226844270219, + "learning_rate": 2.075929130290183e-06, + "loss": 0.9552, + "num_input_tokens_seen": 90380035, + "step": 6697 + }, + { + "epoch": 0.5033819329625733, + "grad_norm": 0.6624650253204866, + "learning_rate": 2.075442636826538e-06, + "loss": 0.8054, + "num_input_tokens_seen": 90461820, + "step": 6698 + }, + { + "epoch": 0.5034570870284082, + "grad_norm": 1.3309305515184058, + "learning_rate": 2.0749561388926283e-06, + "loss": 0.9668, + "num_input_tokens_seen": 90484050, + "step": 6699 + }, + { + "epoch": 0.5035322410942432, + "grad_norm": 1.5536122653398448, + "learning_rate": 2.07446963651728e-06, + "loss": 0.9083, + "num_input_tokens_seen": 90507075, + "step": 6700 + }, + { + "epoch": 0.5036073951600781, + "grad_norm": 1.783299187603677, + "learning_rate": 2.073983129729321e-06, + "loss": 0.8453, + "num_input_tokens_seen": 90529265, + "step": 6701 + }, + { + "epoch": 0.5036825492259132, + "grad_norm": 1.6814899089055577, + "learning_rate": 2.073496618557577e-06, + "loss": 0.9084, + "num_input_tokens_seen": 90552635, + "step": 6702 + }, + { + "epoch": 0.5037577032917481, + "grad_norm": 1.7352858892152527, + "learning_rate": 2.073010103030878e-06, + "loss": 0.9449, + "num_input_tokens_seen": 90576620, + "step": 6703 + }, + { + "epoch": 0.5038328573575831, + "grad_norm": 1.9492934829971664, + "learning_rate": 2.0725235831780516e-06, + "loss": 0.7922, + "num_input_tokens_seen": 90599125, + "step": 6704 + }, + { + "epoch": 0.503908011423418, + "grad_norm": 1.5378723965107863, + "learning_rate": 2.0720370590279234e-06, + "loss": 0.9715, + "num_input_tokens_seen": 90622180, + "step": 6705 + }, + { + "epoch": 0.5039831654892529, + "grad_norm": 1.4854721832957967, + "learning_rate": 2.0715505306093247e-06, + "loss": 1.0038, + "num_input_tokens_seen": 90647090, + "step": 6706 + }, + { + "epoch": 0.504058319555088, + "grad_norm": 0.7205019207385456, + "learning_rate": 2.071063997951082e-06, + "loss": 0.8207, + "num_input_tokens_seen": 90728045, + "step": 6707 + }, + { + "epoch": 0.5041334736209229, + "grad_norm": 1.728891933018168, + "learning_rate": 2.0705774610820267e-06, + "loss": 0.8808, + "num_input_tokens_seen": 90748905, + "step": 6708 + }, + { + "epoch": 0.5042086276867579, + "grad_norm": 1.5381904106383848, + "learning_rate": 2.070090920030986e-06, + "loss": 0.9042, + "num_input_tokens_seen": 90771860, + "step": 6709 + }, + { + "epoch": 0.5042837817525928, + "grad_norm": 1.5289526073228865, + "learning_rate": 2.0696043748267897e-06, + "loss": 1.0486, + "num_input_tokens_seen": 90797335, + "step": 6710 + }, + { + "epoch": 0.5043589358184277, + "grad_norm": 1.7368689125066916, + "learning_rate": 2.0691178254982684e-06, + "loss": 0.9024, + "num_input_tokens_seen": 90820570, + "step": 6711 + }, + { + "epoch": 0.5044340898842627, + "grad_norm": 0.8501503110263477, + "learning_rate": 2.0686312720742504e-06, + "loss": 0.7898, + "num_input_tokens_seen": 90896230, + "step": 6712 + }, + { + "epoch": 0.5045092439500977, + "grad_norm": 1.7100567227152332, + "learning_rate": 2.068144714583567e-06, + "loss": 1.0259, + "num_input_tokens_seen": 90916725, + "step": 6713 + }, + { + "epoch": 0.5045843980159327, + "grad_norm": 1.5228535633433613, + "learning_rate": 2.0676581530550485e-06, + "loss": 0.9731, + "num_input_tokens_seen": 90941020, + "step": 6714 + }, + { + "epoch": 0.5046595520817676, + "grad_norm": 1.5235603478613637, + "learning_rate": 2.067171587517525e-06, + "loss": 1.0066, + "num_input_tokens_seen": 90962220, + "step": 6715 + }, + { + "epoch": 0.5047347061476026, + "grad_norm": 1.7596246491421204, + "learning_rate": 2.066685017999828e-06, + "loss": 0.9515, + "num_input_tokens_seen": 90981640, + "step": 6716 + }, + { + "epoch": 0.5048098602134375, + "grad_norm": 1.7078328989107239, + "learning_rate": 2.0661984445307886e-06, + "loss": 0.9216, + "num_input_tokens_seen": 91002240, + "step": 6717 + }, + { + "epoch": 0.5048850142792725, + "grad_norm": 1.646342170701396, + "learning_rate": 2.0657118671392373e-06, + "loss": 0.9363, + "num_input_tokens_seen": 91026655, + "step": 6718 + }, + { + "epoch": 0.5049601683451075, + "grad_norm": 1.814832370572946, + "learning_rate": 2.0652252858540064e-06, + "loss": 0.9354, + "num_input_tokens_seen": 91046805, + "step": 6719 + }, + { + "epoch": 0.5050353224109424, + "grad_norm": 1.4874927905252597, + "learning_rate": 2.0647387007039277e-06, + "loss": 0.9495, + "num_input_tokens_seen": 91073265, + "step": 6720 + }, + { + "epoch": 0.5051104764767774, + "grad_norm": 2.5275629765091367, + "learning_rate": 2.0642521117178332e-06, + "loss": 1.0222, + "num_input_tokens_seen": 91098375, + "step": 6721 + }, + { + "epoch": 0.5051856305426123, + "grad_norm": 1.52375970025296, + "learning_rate": 2.0637655189245548e-06, + "loss": 0.9361, + "num_input_tokens_seen": 91121490, + "step": 6722 + }, + { + "epoch": 0.5052607846084474, + "grad_norm": 1.4754309784307218, + "learning_rate": 2.0632789223529254e-06, + "loss": 0.9969, + "num_input_tokens_seen": 91148575, + "step": 6723 + }, + { + "epoch": 0.5053359386742823, + "grad_norm": 1.4482621894427496, + "learning_rate": 2.0627923220317767e-06, + "loss": 1.0389, + "num_input_tokens_seen": 91174795, + "step": 6724 + }, + { + "epoch": 0.5054110927401172, + "grad_norm": 1.4273672819155236, + "learning_rate": 2.062305717989943e-06, + "loss": 0.9554, + "num_input_tokens_seen": 91198645, + "step": 6725 + }, + { + "epoch": 0.5054862468059522, + "grad_norm": 1.7859284381519744, + "learning_rate": 2.0618191102562575e-06, + "loss": 0.9202, + "num_input_tokens_seen": 91221455, + "step": 6726 + }, + { + "epoch": 0.5055614008717871, + "grad_norm": 2.0773590511979374, + "learning_rate": 2.061332498859553e-06, + "loss": 0.8971, + "num_input_tokens_seen": 91239580, + "step": 6727 + }, + { + "epoch": 0.5056365549376222, + "grad_norm": 1.333421848259794, + "learning_rate": 2.060845883828663e-06, + "loss": 0.8898, + "num_input_tokens_seen": 91263180, + "step": 6728 + }, + { + "epoch": 0.5057117090034571, + "grad_norm": 0.7250304117072542, + "learning_rate": 2.0603592651924206e-06, + "loss": 0.7473, + "num_input_tokens_seen": 91336745, + "step": 6729 + }, + { + "epoch": 0.5057868630692921, + "grad_norm": 1.5399662985148486, + "learning_rate": 2.059872642979661e-06, + "loss": 0.8273, + "num_input_tokens_seen": 91363375, + "step": 6730 + }, + { + "epoch": 0.505862017135127, + "grad_norm": 1.8017594004058044, + "learning_rate": 2.0593860172192178e-06, + "loss": 0.8498, + "num_input_tokens_seen": 91386110, + "step": 6731 + }, + { + "epoch": 0.5059371712009619, + "grad_norm": 1.3848797706723603, + "learning_rate": 2.0588993879399265e-06, + "loss": 0.9732, + "num_input_tokens_seen": 91411090, + "step": 6732 + }, + { + "epoch": 0.506012325266797, + "grad_norm": 1.3400517516320039, + "learning_rate": 2.0584127551706202e-06, + "loss": 0.9355, + "num_input_tokens_seen": 91435115, + "step": 6733 + }, + { + "epoch": 0.5060874793326319, + "grad_norm": 1.6595681977816936, + "learning_rate": 2.0579261189401345e-06, + "loss": 0.9427, + "num_input_tokens_seen": 91458420, + "step": 6734 + }, + { + "epoch": 0.5061626333984669, + "grad_norm": 1.6920397411305639, + "learning_rate": 2.0574394792773048e-06, + "loss": 1.016, + "num_input_tokens_seen": 91481335, + "step": 6735 + }, + { + "epoch": 0.5062377874643018, + "grad_norm": 1.6381294525796224, + "learning_rate": 2.0569528362109662e-06, + "loss": 1.0102, + "num_input_tokens_seen": 91504305, + "step": 6736 + }, + { + "epoch": 0.5063129415301367, + "grad_norm": 1.5216874463325811, + "learning_rate": 2.056466189769953e-06, + "loss": 0.8522, + "num_input_tokens_seen": 91532420, + "step": 6737 + }, + { + "epoch": 0.5063880955959718, + "grad_norm": 1.7970245962481632, + "learning_rate": 2.055979539983103e-06, + "loss": 0.8671, + "num_input_tokens_seen": 91556090, + "step": 6738 + }, + { + "epoch": 0.5064632496618067, + "grad_norm": 0.8808273343435638, + "learning_rate": 2.05549288687925e-06, + "loss": 0.7897, + "num_input_tokens_seen": 91632235, + "step": 6739 + }, + { + "epoch": 0.5065384037276417, + "grad_norm": 1.3793457454730138, + "learning_rate": 2.0550062304872317e-06, + "loss": 0.9669, + "num_input_tokens_seen": 91656265, + "step": 6740 + }, + { + "epoch": 0.5066135577934766, + "grad_norm": 1.4278884992371295, + "learning_rate": 2.054519570835883e-06, + "loss": 0.8265, + "num_input_tokens_seen": 91680765, + "step": 6741 + }, + { + "epoch": 0.5066887118593116, + "grad_norm": 1.5219251119208086, + "learning_rate": 2.0540329079540414e-06, + "loss": 1.0075, + "num_input_tokens_seen": 91706510, + "step": 6742 + }, + { + "epoch": 0.5067638659251466, + "grad_norm": 1.58561101850902, + "learning_rate": 2.053546241870543e-06, + "loss": 0.9437, + "num_input_tokens_seen": 91729630, + "step": 6743 + }, + { + "epoch": 0.5068390199909815, + "grad_norm": 1.5369510011624852, + "learning_rate": 2.053059572614224e-06, + "loss": 1.0339, + "num_input_tokens_seen": 91753270, + "step": 6744 + }, + { + "epoch": 0.5069141740568165, + "grad_norm": 1.9713476694084813, + "learning_rate": 2.0525729002139233e-06, + "loss": 0.9913, + "num_input_tokens_seen": 91774595, + "step": 6745 + }, + { + "epoch": 0.5069893281226514, + "grad_norm": 1.8617394424165765, + "learning_rate": 2.052086224698476e-06, + "loss": 0.8763, + "num_input_tokens_seen": 91796320, + "step": 6746 + }, + { + "epoch": 0.5070644821884864, + "grad_norm": 1.8505808485459383, + "learning_rate": 2.0515995460967204e-06, + "loss": 0.9551, + "num_input_tokens_seen": 91821065, + "step": 6747 + }, + { + "epoch": 0.5071396362543213, + "grad_norm": 1.387118042962018, + "learning_rate": 2.0511128644374953e-06, + "loss": 0.9716, + "num_input_tokens_seen": 91846180, + "step": 6748 + }, + { + "epoch": 0.5072147903201564, + "grad_norm": 1.464095837713255, + "learning_rate": 2.0506261797496357e-06, + "loss": 0.8523, + "num_input_tokens_seen": 91871485, + "step": 6749 + }, + { + "epoch": 0.5072899443859913, + "grad_norm": 1.592936289383129, + "learning_rate": 2.0501394920619822e-06, + "loss": 0.9759, + "num_input_tokens_seen": 91894735, + "step": 6750 + }, + { + "epoch": 0.5073650984518262, + "grad_norm": 1.5304959834637135, + "learning_rate": 2.0496528014033717e-06, + "loss": 1.0003, + "num_input_tokens_seen": 91917790, + "step": 6751 + }, + { + "epoch": 0.5074402525176612, + "grad_norm": 1.5393408509097462, + "learning_rate": 2.0491661078026423e-06, + "loss": 0.8986, + "num_input_tokens_seen": 91940680, + "step": 6752 + }, + { + "epoch": 0.5075154065834961, + "grad_norm": 0.8165833130617279, + "learning_rate": 2.0486794112886328e-06, + "loss": 0.8302, + "num_input_tokens_seen": 92014495, + "step": 6753 + }, + { + "epoch": 0.5075905606493312, + "grad_norm": 1.482481300660643, + "learning_rate": 2.0481927118901817e-06, + "loss": 0.9386, + "num_input_tokens_seen": 92038500, + "step": 6754 + }, + { + "epoch": 0.5076657147151661, + "grad_norm": 1.3691514214757983, + "learning_rate": 2.047706009636128e-06, + "loss": 0.9707, + "num_input_tokens_seen": 92061310, + "step": 6755 + }, + { + "epoch": 0.507740868781001, + "grad_norm": 1.4309537522675357, + "learning_rate": 2.0472193045553104e-06, + "loss": 1.0028, + "num_input_tokens_seen": 92086185, + "step": 6756 + }, + { + "epoch": 0.507816022846836, + "grad_norm": 1.3768303566170015, + "learning_rate": 2.0467325966765683e-06, + "loss": 0.9602, + "num_input_tokens_seen": 92110400, + "step": 6757 + }, + { + "epoch": 0.5078911769126709, + "grad_norm": 1.6536765348533062, + "learning_rate": 2.046245886028741e-06, + "loss": 1.0225, + "num_input_tokens_seen": 92132830, + "step": 6758 + }, + { + "epoch": 0.507966330978506, + "grad_norm": 0.7650865490890857, + "learning_rate": 2.045759172640668e-06, + "loss": 0.7925, + "num_input_tokens_seen": 92210955, + "step": 6759 + }, + { + "epoch": 0.5080414850443409, + "grad_norm": 1.542994400004363, + "learning_rate": 2.0452724565411886e-06, + "loss": 0.9202, + "num_input_tokens_seen": 92234505, + "step": 6760 + }, + { + "epoch": 0.5081166391101759, + "grad_norm": 1.417392750337646, + "learning_rate": 2.044785737759143e-06, + "loss": 0.9655, + "num_input_tokens_seen": 92260350, + "step": 6761 + }, + { + "epoch": 0.5081917931760108, + "grad_norm": 1.5970432439215045, + "learning_rate": 2.0442990163233704e-06, + "loss": 0.9078, + "num_input_tokens_seen": 92283725, + "step": 6762 + }, + { + "epoch": 0.5082669472418457, + "grad_norm": 1.6536481321624237, + "learning_rate": 2.0438122922627114e-06, + "loss": 0.9896, + "num_input_tokens_seen": 92307425, + "step": 6763 + }, + { + "epoch": 0.5083421013076808, + "grad_norm": 1.9840722641589714, + "learning_rate": 2.0433255656060066e-06, + "loss": 0.9456, + "num_input_tokens_seen": 92330075, + "step": 6764 + }, + { + "epoch": 0.5084172553735157, + "grad_norm": 2.403882434864616, + "learning_rate": 2.0428388363820966e-06, + "loss": 0.8376, + "num_input_tokens_seen": 92353395, + "step": 6765 + }, + { + "epoch": 0.5084924094393507, + "grad_norm": 1.3385463472079169, + "learning_rate": 2.0423521046198206e-06, + "loss": 0.9876, + "num_input_tokens_seen": 92380420, + "step": 6766 + }, + { + "epoch": 0.5085675635051856, + "grad_norm": 0.8913586255255842, + "learning_rate": 2.041865370348021e-06, + "loss": 0.815, + "num_input_tokens_seen": 92454055, + "step": 6767 + }, + { + "epoch": 0.5086427175710206, + "grad_norm": 1.4452218517022974, + "learning_rate": 2.0413786335955374e-06, + "loss": 0.9307, + "num_input_tokens_seen": 92479240, + "step": 6768 + }, + { + "epoch": 0.5087178716368556, + "grad_norm": 1.5634546796612347, + "learning_rate": 2.0408918943912113e-06, + "loss": 0.8726, + "num_input_tokens_seen": 92502755, + "step": 6769 + }, + { + "epoch": 0.5087930257026905, + "grad_norm": 1.5775104496967962, + "learning_rate": 2.0404051527638844e-06, + "loss": 0.9563, + "num_input_tokens_seen": 92525620, + "step": 6770 + }, + { + "epoch": 0.5088681797685255, + "grad_norm": 1.3526977005342922, + "learning_rate": 2.039918408742397e-06, + "loss": 0.9019, + "num_input_tokens_seen": 92553535, + "step": 6771 + }, + { + "epoch": 0.5089433338343604, + "grad_norm": 1.4813084297084018, + "learning_rate": 2.039431662355591e-06, + "loss": 0.9573, + "num_input_tokens_seen": 92576360, + "step": 6772 + }, + { + "epoch": 0.5090184879001954, + "grad_norm": 1.5691034017738836, + "learning_rate": 2.0389449136323082e-06, + "loss": 0.9281, + "num_input_tokens_seen": 92597300, + "step": 6773 + }, + { + "epoch": 0.5090936419660304, + "grad_norm": 1.5211137312804688, + "learning_rate": 2.0384581626013905e-06, + "loss": 0.9266, + "num_input_tokens_seen": 92618765, + "step": 6774 + }, + { + "epoch": 0.5091687960318654, + "grad_norm": 1.7426457508051005, + "learning_rate": 2.037971409291679e-06, + "loss": 1.0105, + "num_input_tokens_seen": 92641290, + "step": 6775 + }, + { + "epoch": 0.5092439500977003, + "grad_norm": 1.6048092566545535, + "learning_rate": 2.037484653732016e-06, + "loss": 0.9695, + "num_input_tokens_seen": 92665195, + "step": 6776 + }, + { + "epoch": 0.5093191041635352, + "grad_norm": 1.6067367706727613, + "learning_rate": 2.036997895951244e-06, + "loss": 0.9859, + "num_input_tokens_seen": 92689680, + "step": 6777 + }, + { + "epoch": 0.5093942582293702, + "grad_norm": 1.8801720176282255, + "learning_rate": 2.0365111359782046e-06, + "loss": 0.9266, + "num_input_tokens_seen": 92710590, + "step": 6778 + }, + { + "epoch": 0.5094694122952051, + "grad_norm": 1.8180560095566054, + "learning_rate": 2.0360243738417414e-06, + "loss": 0.8823, + "num_input_tokens_seen": 92732345, + "step": 6779 + }, + { + "epoch": 0.5095445663610402, + "grad_norm": 1.600173198743171, + "learning_rate": 2.035537609570695e-06, + "loss": 0.9816, + "num_input_tokens_seen": 92755800, + "step": 6780 + }, + { + "epoch": 0.5096197204268751, + "grad_norm": 0.8094005889971498, + "learning_rate": 2.03505084319391e-06, + "loss": 0.7898, + "num_input_tokens_seen": 92829595, + "step": 6781 + }, + { + "epoch": 0.50969487449271, + "grad_norm": 1.410834321128528, + "learning_rate": 2.0345640747402283e-06, + "loss": 0.9496, + "num_input_tokens_seen": 92852590, + "step": 6782 + }, + { + "epoch": 0.509770028558545, + "grad_norm": 1.362194207199677, + "learning_rate": 2.034077304238492e-06, + "loss": 0.9615, + "num_input_tokens_seen": 92877210, + "step": 6783 + }, + { + "epoch": 0.50984518262438, + "grad_norm": 1.5651583373924214, + "learning_rate": 2.0335905317175457e-06, + "loss": 0.9939, + "num_input_tokens_seen": 92903730, + "step": 6784 + }, + { + "epoch": 0.509920336690215, + "grad_norm": 1.9388744032475655, + "learning_rate": 2.0331037572062314e-06, + "loss": 0.9951, + "num_input_tokens_seen": 92924100, + "step": 6785 + }, + { + "epoch": 0.5099954907560499, + "grad_norm": 1.6381437157497254, + "learning_rate": 2.032616980733393e-06, + "loss": 0.8866, + "num_input_tokens_seen": 92947630, + "step": 6786 + }, + { + "epoch": 0.5100706448218849, + "grad_norm": 1.8898338364920264, + "learning_rate": 2.0321302023278734e-06, + "loss": 0.9678, + "num_input_tokens_seen": 92969775, + "step": 6787 + }, + { + "epoch": 0.5101457988877198, + "grad_norm": 1.652479450253876, + "learning_rate": 2.031643422018516e-06, + "loss": 0.9094, + "num_input_tokens_seen": 92992870, + "step": 6788 + }, + { + "epoch": 0.5102209529535547, + "grad_norm": 2.1483895591242073, + "learning_rate": 2.0311566398341653e-06, + "loss": 0.878, + "num_input_tokens_seen": 93018095, + "step": 6789 + }, + { + "epoch": 0.5102961070193898, + "grad_norm": 1.4783026719524435, + "learning_rate": 2.0306698558036635e-06, + "loss": 1.0046, + "num_input_tokens_seen": 93040730, + "step": 6790 + }, + { + "epoch": 0.5103712610852247, + "grad_norm": 1.4112762483870671, + "learning_rate": 2.0301830699558563e-06, + "loss": 0.9314, + "num_input_tokens_seen": 93064925, + "step": 6791 + }, + { + "epoch": 0.5104464151510597, + "grad_norm": 1.527329700922438, + "learning_rate": 2.029696282319586e-06, + "loss": 1.0293, + "num_input_tokens_seen": 93088110, + "step": 6792 + }, + { + "epoch": 0.5105215692168946, + "grad_norm": 1.5787445967121945, + "learning_rate": 2.0292094929236976e-06, + "loss": 0.9779, + "num_input_tokens_seen": 93108195, + "step": 6793 + }, + { + "epoch": 0.5105967232827296, + "grad_norm": 1.278010949453571, + "learning_rate": 2.028722701797035e-06, + "loss": 0.8729, + "num_input_tokens_seen": 93134470, + "step": 6794 + }, + { + "epoch": 0.5106718773485646, + "grad_norm": 0.7858303080393007, + "learning_rate": 2.0282359089684417e-06, + "loss": 0.7782, + "num_input_tokens_seen": 93209750, + "step": 6795 + }, + { + "epoch": 0.5107470314143995, + "grad_norm": 2.045001737971605, + "learning_rate": 2.027749114466763e-06, + "loss": 0.9221, + "num_input_tokens_seen": 93230160, + "step": 6796 + }, + { + "epoch": 0.5108221854802345, + "grad_norm": 1.57904675446977, + "learning_rate": 2.0272623183208433e-06, + "loss": 0.9279, + "num_input_tokens_seen": 93253310, + "step": 6797 + }, + { + "epoch": 0.5108973395460694, + "grad_norm": 0.89074478682353, + "learning_rate": 2.0267755205595266e-06, + "loss": 0.8538, + "num_input_tokens_seen": 93330295, + "step": 6798 + }, + { + "epoch": 0.5109724936119044, + "grad_norm": 1.657737783461492, + "learning_rate": 2.026288721211658e-06, + "loss": 0.911, + "num_input_tokens_seen": 93351060, + "step": 6799 + }, + { + "epoch": 0.5110476476777394, + "grad_norm": 1.3931076091846342, + "learning_rate": 2.0258019203060816e-06, + "loss": 0.9514, + "num_input_tokens_seen": 93373910, + "step": 6800 + }, + { + "epoch": 0.5111228017435743, + "grad_norm": 2.4611699070256767, + "learning_rate": 2.025315117871643e-06, + "loss": 0.9314, + "num_input_tokens_seen": 93395115, + "step": 6801 + }, + { + "epoch": 0.5111979558094093, + "grad_norm": 1.7706793624660242, + "learning_rate": 2.0248283139371862e-06, + "loss": 0.9142, + "num_input_tokens_seen": 93418200, + "step": 6802 + }, + { + "epoch": 0.5112731098752442, + "grad_norm": 1.5442044102845283, + "learning_rate": 2.0243415085315573e-06, + "loss": 1.0464, + "num_input_tokens_seen": 93441785, + "step": 6803 + }, + { + "epoch": 0.5113482639410792, + "grad_norm": 1.6688272379741613, + "learning_rate": 2.023854701683601e-06, + "loss": 0.8373, + "num_input_tokens_seen": 93464695, + "step": 6804 + }, + { + "epoch": 0.5114234180069142, + "grad_norm": 1.7766984040441713, + "learning_rate": 2.0233678934221615e-06, + "loss": 0.9295, + "num_input_tokens_seen": 93486400, + "step": 6805 + }, + { + "epoch": 0.5114985720727492, + "grad_norm": 1.7668076327002942, + "learning_rate": 2.0228810837760853e-06, + "loss": 0.8415, + "num_input_tokens_seen": 93511840, + "step": 6806 + }, + { + "epoch": 0.5115737261385841, + "grad_norm": 1.7627730841136189, + "learning_rate": 2.0223942727742168e-06, + "loss": 0.9297, + "num_input_tokens_seen": 93532890, + "step": 6807 + }, + { + "epoch": 0.511648880204419, + "grad_norm": 1.6839147558224878, + "learning_rate": 2.0219074604454026e-06, + "loss": 0.943, + "num_input_tokens_seen": 93556790, + "step": 6808 + }, + { + "epoch": 0.511724034270254, + "grad_norm": 1.8913326988116064, + "learning_rate": 2.021420646818487e-06, + "loss": 0.9803, + "num_input_tokens_seen": 93578720, + "step": 6809 + }, + { + "epoch": 0.511799188336089, + "grad_norm": 1.5475203583938903, + "learning_rate": 2.0209338319223155e-06, + "loss": 1.0237, + "num_input_tokens_seen": 93603000, + "step": 6810 + }, + { + "epoch": 0.511874342401924, + "grad_norm": 1.7968623948691704, + "learning_rate": 2.0204470157857354e-06, + "loss": 0.8667, + "num_input_tokens_seen": 93628660, + "step": 6811 + }, + { + "epoch": 0.5119494964677589, + "grad_norm": 0.7553665170857637, + "learning_rate": 2.0199601984375907e-06, + "loss": 0.8253, + "num_input_tokens_seen": 93702065, + "step": 6812 + }, + { + "epoch": 0.5120246505335939, + "grad_norm": 1.6017304123106058, + "learning_rate": 2.0194733799067284e-06, + "loss": 0.9187, + "num_input_tokens_seen": 93727005, + "step": 6813 + }, + { + "epoch": 0.5120998045994288, + "grad_norm": 1.4318440333147797, + "learning_rate": 2.0189865602219934e-06, + "loss": 0.9499, + "num_input_tokens_seen": 93749985, + "step": 6814 + }, + { + "epoch": 0.5121749586652637, + "grad_norm": 1.5553361315979348, + "learning_rate": 2.0184997394122317e-06, + "loss": 1.0406, + "num_input_tokens_seen": 93771840, + "step": 6815 + }, + { + "epoch": 0.5122501127310988, + "grad_norm": 1.6734418567988993, + "learning_rate": 2.01801291750629e-06, + "loss": 1.0334, + "num_input_tokens_seen": 93796705, + "step": 6816 + }, + { + "epoch": 0.5123252667969337, + "grad_norm": 1.5947515295332018, + "learning_rate": 2.0175260945330134e-06, + "loss": 0.8966, + "num_input_tokens_seen": 93823960, + "step": 6817 + }, + { + "epoch": 0.5124004208627687, + "grad_norm": 1.394372156295031, + "learning_rate": 2.0170392705212495e-06, + "loss": 0.9442, + "num_input_tokens_seen": 93846230, + "step": 6818 + }, + { + "epoch": 0.5124755749286036, + "grad_norm": 1.8221784859089072, + "learning_rate": 2.016552445499843e-06, + "loss": 0.8998, + "num_input_tokens_seen": 93870690, + "step": 6819 + }, + { + "epoch": 0.5125507289944387, + "grad_norm": 1.4740052861619672, + "learning_rate": 2.0160656194976407e-06, + "loss": 0.9888, + "num_input_tokens_seen": 93894320, + "step": 6820 + }, + { + "epoch": 0.5126258830602736, + "grad_norm": 1.4909820480903386, + "learning_rate": 2.0155787925434893e-06, + "loss": 1.0709, + "num_input_tokens_seen": 93918410, + "step": 6821 + }, + { + "epoch": 0.5127010371261085, + "grad_norm": 1.9134960971390769, + "learning_rate": 2.0150919646662342e-06, + "loss": 0.9088, + "num_input_tokens_seen": 93941895, + "step": 6822 + }, + { + "epoch": 0.5127761911919435, + "grad_norm": 1.5295095868241997, + "learning_rate": 2.014605135894723e-06, + "loss": 0.9594, + "num_input_tokens_seen": 93963875, + "step": 6823 + }, + { + "epoch": 0.5128513452577784, + "grad_norm": 1.3020195551193334, + "learning_rate": 2.0141183062578013e-06, + "loss": 0.8856, + "num_input_tokens_seen": 93989875, + "step": 6824 + }, + { + "epoch": 0.5129264993236134, + "grad_norm": 1.6608290083325326, + "learning_rate": 2.013631475784316e-06, + "loss": 0.8367, + "num_input_tokens_seen": 94014135, + "step": 6825 + }, + { + "epoch": 0.5130016533894484, + "grad_norm": 1.5995801613251155, + "learning_rate": 2.0131446445031134e-06, + "loss": 0.9213, + "num_input_tokens_seen": 94037435, + "step": 6826 + }, + { + "epoch": 0.5130768074552833, + "grad_norm": 0.6899417950088474, + "learning_rate": 2.0126578124430402e-06, + "loss": 0.7309, + "num_input_tokens_seen": 94120790, + "step": 6827 + }, + { + "epoch": 0.5131519615211183, + "grad_norm": 1.4523827082275689, + "learning_rate": 2.012170979632944e-06, + "loss": 1.042, + "num_input_tokens_seen": 94144340, + "step": 6828 + }, + { + "epoch": 0.5132271155869532, + "grad_norm": 1.464965733853231, + "learning_rate": 2.0116841461016685e-06, + "loss": 0.9805, + "num_input_tokens_seen": 94170995, + "step": 6829 + }, + { + "epoch": 0.5133022696527882, + "grad_norm": 1.6380244398052093, + "learning_rate": 2.0111973118780653e-06, + "loss": 0.9608, + "num_input_tokens_seen": 94194040, + "step": 6830 + }, + { + "epoch": 0.5133774237186232, + "grad_norm": 1.551829409042923, + "learning_rate": 2.0107104769909773e-06, + "loss": 0.9806, + "num_input_tokens_seen": 94214145, + "step": 6831 + }, + { + "epoch": 0.5134525777844582, + "grad_norm": 1.3679085710099845, + "learning_rate": 2.0102236414692515e-06, + "loss": 0.8254, + "num_input_tokens_seen": 94239935, + "step": 6832 + }, + { + "epoch": 0.5135277318502931, + "grad_norm": 1.4169931970404743, + "learning_rate": 2.009736805341737e-06, + "loss": 0.9282, + "num_input_tokens_seen": 94260850, + "step": 6833 + }, + { + "epoch": 0.513602885916128, + "grad_norm": 3.3450822404408087, + "learning_rate": 2.0092499686372794e-06, + "loss": 0.8648, + "num_input_tokens_seen": 94285385, + "step": 6834 + }, + { + "epoch": 0.513678039981963, + "grad_norm": 1.6893469981117843, + "learning_rate": 2.0087631313847252e-06, + "loss": 0.8775, + "num_input_tokens_seen": 94310290, + "step": 6835 + }, + { + "epoch": 0.513753194047798, + "grad_norm": 1.2950055079545528, + "learning_rate": 2.0082762936129226e-06, + "loss": 0.9319, + "num_input_tokens_seen": 94337255, + "step": 6836 + }, + { + "epoch": 0.513828348113633, + "grad_norm": 1.4102558301599906, + "learning_rate": 2.0077894553507174e-06, + "loss": 0.8598, + "num_input_tokens_seen": 94360820, + "step": 6837 + }, + { + "epoch": 0.5139035021794679, + "grad_norm": 1.5262586246281162, + "learning_rate": 2.0073026166269577e-06, + "loss": 0.8909, + "num_input_tokens_seen": 94386370, + "step": 6838 + }, + { + "epoch": 0.5139786562453029, + "grad_norm": 1.9932320642529164, + "learning_rate": 2.006815777470489e-06, + "loss": 1.0446, + "num_input_tokens_seen": 94410865, + "step": 6839 + }, + { + "epoch": 0.5140538103111378, + "grad_norm": 1.5087870082409154, + "learning_rate": 2.0063289379101606e-06, + "loss": 0.9232, + "num_input_tokens_seen": 94434330, + "step": 6840 + }, + { + "epoch": 0.5141289643769728, + "grad_norm": 1.441285001622545, + "learning_rate": 2.0058420979748172e-06, + "loss": 1.0524, + "num_input_tokens_seen": 94458320, + "step": 6841 + }, + { + "epoch": 0.5142041184428078, + "grad_norm": 1.4494285740066155, + "learning_rate": 2.005355257693308e-06, + "loss": 0.9118, + "num_input_tokens_seen": 94480310, + "step": 6842 + }, + { + "epoch": 0.5142792725086427, + "grad_norm": 1.5703474202235113, + "learning_rate": 2.0048684170944795e-06, + "loss": 1.0763, + "num_input_tokens_seen": 94501355, + "step": 6843 + }, + { + "epoch": 0.5143544265744777, + "grad_norm": 2.3030663947383276, + "learning_rate": 2.004381576207178e-06, + "loss": 1.0418, + "num_input_tokens_seen": 94523685, + "step": 6844 + }, + { + "epoch": 0.5144295806403126, + "grad_norm": 1.6042703206976492, + "learning_rate": 2.0038947350602516e-06, + "loss": 0.9297, + "num_input_tokens_seen": 94547295, + "step": 6845 + }, + { + "epoch": 0.5145047347061475, + "grad_norm": 1.6893774820493699, + "learning_rate": 2.0034078936825467e-06, + "loss": 0.9395, + "num_input_tokens_seen": 94571640, + "step": 6846 + }, + { + "epoch": 0.5145798887719826, + "grad_norm": 1.6538310830789318, + "learning_rate": 2.002921052102912e-06, + "loss": 0.9212, + "num_input_tokens_seen": 94594520, + "step": 6847 + }, + { + "epoch": 0.5146550428378175, + "grad_norm": 1.8240990936441002, + "learning_rate": 2.0024342103501934e-06, + "loss": 0.9396, + "num_input_tokens_seen": 94620630, + "step": 6848 + }, + { + "epoch": 0.5147301969036525, + "grad_norm": 1.5707267101066331, + "learning_rate": 2.001947368453238e-06, + "loss": 0.9905, + "num_input_tokens_seen": 94641895, + "step": 6849 + }, + { + "epoch": 0.5148053509694874, + "grad_norm": 0.7848183963059265, + "learning_rate": 2.001460526440894e-06, + "loss": 0.8905, + "num_input_tokens_seen": 94715460, + "step": 6850 + }, + { + "epoch": 0.5148805050353225, + "grad_norm": 1.9132976015699863, + "learning_rate": 2.0009736843420076e-06, + "loss": 1.0294, + "num_input_tokens_seen": 94736585, + "step": 6851 + }, + { + "epoch": 0.5149556591011574, + "grad_norm": 1.4418870077637702, + "learning_rate": 2.0004868421854274e-06, + "loss": 0.9122, + "num_input_tokens_seen": 94760205, + "step": 6852 + }, + { + "epoch": 0.5150308131669923, + "grad_norm": 3.18085460191678, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9711, + "num_input_tokens_seen": 94784640, + "step": 6853 + }, + { + "epoch": 0.5151059672328273, + "grad_norm": 1.3930982391539068, + "learning_rate": 1.999513157814572e-06, + "loss": 0.9711, + "num_input_tokens_seen": 94811665, + "step": 6854 + }, + { + "epoch": 0.5151811212986622, + "grad_norm": 1.5229226831960052, + "learning_rate": 1.9990263156579922e-06, + "loss": 1.0283, + "num_input_tokens_seen": 94834640, + "step": 6855 + }, + { + "epoch": 0.5152562753644973, + "grad_norm": 2.0860458517329445, + "learning_rate": 1.998539473559106e-06, + "loss": 0.8432, + "num_input_tokens_seen": 94855775, + "step": 6856 + }, + { + "epoch": 0.5153314294303322, + "grad_norm": 2.2714567699378594, + "learning_rate": 1.998052631546762e-06, + "loss": 0.7827, + "num_input_tokens_seen": 94875110, + "step": 6857 + }, + { + "epoch": 0.5154065834961672, + "grad_norm": 1.4899037245554538, + "learning_rate": 1.9975657896498073e-06, + "loss": 0.9848, + "num_input_tokens_seen": 94899040, + "step": 6858 + }, + { + "epoch": 0.5154817375620021, + "grad_norm": 1.563502714759219, + "learning_rate": 1.9970789478970882e-06, + "loss": 1.0281, + "num_input_tokens_seen": 94921845, + "step": 6859 + }, + { + "epoch": 0.515556891627837, + "grad_norm": 1.4175485035504827, + "learning_rate": 1.996592106317453e-06, + "loss": 0.9474, + "num_input_tokens_seen": 94945620, + "step": 6860 + }, + { + "epoch": 0.515632045693672, + "grad_norm": 1.5693200614425253, + "learning_rate": 1.9961052649397486e-06, + "loss": 0.8534, + "num_input_tokens_seen": 94969350, + "step": 6861 + }, + { + "epoch": 0.515707199759507, + "grad_norm": 1.6270477890189534, + "learning_rate": 1.995618423792822e-06, + "loss": 0.9122, + "num_input_tokens_seen": 94994730, + "step": 6862 + }, + { + "epoch": 0.515782353825342, + "grad_norm": 1.6673243258244517, + "learning_rate": 1.9951315829055208e-06, + "loss": 0.9796, + "num_input_tokens_seen": 95017830, + "step": 6863 + }, + { + "epoch": 0.5158575078911769, + "grad_norm": 1.409192418961567, + "learning_rate": 1.994644742306692e-06, + "loss": 0.9446, + "num_input_tokens_seen": 95043350, + "step": 6864 + }, + { + "epoch": 0.5159326619570119, + "grad_norm": 1.5644718697671507, + "learning_rate": 1.9941579020251826e-06, + "loss": 0.9074, + "num_input_tokens_seen": 95070655, + "step": 6865 + }, + { + "epoch": 0.5160078160228468, + "grad_norm": 1.3860804418501842, + "learning_rate": 1.9936710620898396e-06, + "loss": 0.9101, + "num_input_tokens_seen": 95095010, + "step": 6866 + }, + { + "epoch": 0.5160829700886818, + "grad_norm": 1.6523066667584454, + "learning_rate": 1.9931842225295108e-06, + "loss": 0.948, + "num_input_tokens_seen": 95120110, + "step": 6867 + }, + { + "epoch": 0.5161581241545168, + "grad_norm": 1.4063118391215566, + "learning_rate": 1.9926973833730426e-06, + "loss": 1.0494, + "num_input_tokens_seen": 95144635, + "step": 6868 + }, + { + "epoch": 0.5162332782203517, + "grad_norm": 1.8265531906791346, + "learning_rate": 1.9922105446492824e-06, + "loss": 0.9534, + "num_input_tokens_seen": 95167795, + "step": 6869 + }, + { + "epoch": 0.5163084322861867, + "grad_norm": 1.5219265218234388, + "learning_rate": 1.9917237063870777e-06, + "loss": 1.0357, + "num_input_tokens_seen": 95191415, + "step": 6870 + }, + { + "epoch": 0.5163835863520216, + "grad_norm": 1.3979210779127096, + "learning_rate": 1.9912368686152746e-06, + "loss": 0.9693, + "num_input_tokens_seen": 95211175, + "step": 6871 + }, + { + "epoch": 0.5164587404178566, + "grad_norm": 1.726962220194538, + "learning_rate": 1.990750031362721e-06, + "loss": 1.0, + "num_input_tokens_seen": 95232970, + "step": 6872 + }, + { + "epoch": 0.5165338944836916, + "grad_norm": 1.5031346310883953, + "learning_rate": 1.9902631946582627e-06, + "loss": 0.9423, + "num_input_tokens_seen": 95257225, + "step": 6873 + }, + { + "epoch": 0.5166090485495265, + "grad_norm": 1.727983123308349, + "learning_rate": 1.9897763585307483e-06, + "loss": 0.9471, + "num_input_tokens_seen": 95281285, + "step": 6874 + }, + { + "epoch": 0.5166842026153615, + "grad_norm": 1.6438243370507453, + "learning_rate": 1.989289523009024e-06, + "loss": 1.0124, + "num_input_tokens_seen": 95304700, + "step": 6875 + }, + { + "epoch": 0.5167593566811964, + "grad_norm": 1.6731472700458747, + "learning_rate": 1.988802688121935e-06, + "loss": 0.9183, + "num_input_tokens_seen": 95325315, + "step": 6876 + }, + { + "epoch": 0.5168345107470315, + "grad_norm": 1.4929570315872769, + "learning_rate": 1.988315853898331e-06, + "loss": 0.9111, + "num_input_tokens_seen": 95351825, + "step": 6877 + }, + { + "epoch": 0.5169096648128664, + "grad_norm": 1.4385400409767566, + "learning_rate": 1.9878290203670563e-06, + "loss": 1.0226, + "num_input_tokens_seen": 95376815, + "step": 6878 + }, + { + "epoch": 0.5169848188787013, + "grad_norm": 1.6031739850210174, + "learning_rate": 1.98734218755696e-06, + "loss": 0.9806, + "num_input_tokens_seen": 95397855, + "step": 6879 + }, + { + "epoch": 0.5170599729445363, + "grad_norm": 1.4332121879185038, + "learning_rate": 1.986855355496887e-06, + "loss": 1.0497, + "num_input_tokens_seen": 95419370, + "step": 6880 + }, + { + "epoch": 0.5171351270103712, + "grad_norm": 1.7435924527221305, + "learning_rate": 1.986368524215684e-06, + "loss": 0.9717, + "num_input_tokens_seen": 95442275, + "step": 6881 + }, + { + "epoch": 0.5172102810762063, + "grad_norm": 1.4436896208320602, + "learning_rate": 1.985881693742199e-06, + "loss": 0.948, + "num_input_tokens_seen": 95467670, + "step": 6882 + }, + { + "epoch": 0.5172854351420412, + "grad_norm": 1.4888152518653923, + "learning_rate": 1.985394864105277e-06, + "loss": 0.9061, + "num_input_tokens_seen": 95493235, + "step": 6883 + }, + { + "epoch": 0.5173605892078762, + "grad_norm": 1.0628127030616663, + "learning_rate": 1.9849080353337656e-06, + "loss": 0.8227, + "num_input_tokens_seen": 95571350, + "step": 6884 + }, + { + "epoch": 0.5174357432737111, + "grad_norm": 1.8075425324771384, + "learning_rate": 1.984421207456511e-06, + "loss": 0.7953, + "num_input_tokens_seen": 95597470, + "step": 6885 + }, + { + "epoch": 0.517510897339546, + "grad_norm": 1.59337634025907, + "learning_rate": 1.983934380502359e-06, + "loss": 0.9256, + "num_input_tokens_seen": 95619380, + "step": 6886 + }, + { + "epoch": 0.517586051405381, + "grad_norm": 1.3065762126196638, + "learning_rate": 1.983447554500157e-06, + "loss": 0.8418, + "num_input_tokens_seen": 95642910, + "step": 6887 + }, + { + "epoch": 0.517661205471216, + "grad_norm": 1.5300780502058318, + "learning_rate": 1.9829607294787503e-06, + "loss": 0.9171, + "num_input_tokens_seen": 95667050, + "step": 6888 + }, + { + "epoch": 0.517736359537051, + "grad_norm": 1.76545383670233, + "learning_rate": 1.9824739054669864e-06, + "loss": 0.8375, + "num_input_tokens_seen": 95690110, + "step": 6889 + }, + { + "epoch": 0.5178115136028859, + "grad_norm": 1.7373741955117505, + "learning_rate": 1.98198708249371e-06, + "loss": 0.9598, + "num_input_tokens_seen": 95712650, + "step": 6890 + }, + { + "epoch": 0.5178866676687208, + "grad_norm": 1.3656466275036738, + "learning_rate": 1.9815002605877685e-06, + "loss": 0.8938, + "num_input_tokens_seen": 95740625, + "step": 6891 + }, + { + "epoch": 0.5179618217345558, + "grad_norm": 1.5685066733542097, + "learning_rate": 1.9810134397780073e-06, + "loss": 0.7256, + "num_input_tokens_seen": 95764515, + "step": 6892 + }, + { + "epoch": 0.5180369758003908, + "grad_norm": 1.6524836343482923, + "learning_rate": 1.980526620093272e-06, + "loss": 1.0267, + "num_input_tokens_seen": 95787085, + "step": 6893 + }, + { + "epoch": 0.5181121298662258, + "grad_norm": 1.5590973998762956, + "learning_rate": 1.9800398015624095e-06, + "loss": 0.9501, + "num_input_tokens_seen": 95810145, + "step": 6894 + }, + { + "epoch": 0.5181872839320607, + "grad_norm": 1.3427593216587586, + "learning_rate": 1.9795529842142644e-06, + "loss": 1.0205, + "num_input_tokens_seen": 95834280, + "step": 6895 + }, + { + "epoch": 0.5182624379978957, + "grad_norm": 1.4826520859158037, + "learning_rate": 1.979066168077684e-06, + "loss": 0.8963, + "num_input_tokens_seen": 95856270, + "step": 6896 + }, + { + "epoch": 0.5183375920637306, + "grad_norm": 1.5760855007021928, + "learning_rate": 1.978579353181513e-06, + "loss": 0.8902, + "num_input_tokens_seen": 95882765, + "step": 6897 + }, + { + "epoch": 0.5184127461295656, + "grad_norm": 1.7376514451144256, + "learning_rate": 1.9780925395545977e-06, + "loss": 0.9515, + "num_input_tokens_seen": 95904860, + "step": 6898 + }, + { + "epoch": 0.5184879001954006, + "grad_norm": 1.7343519054627934, + "learning_rate": 1.977605727225783e-06, + "loss": 0.9962, + "num_input_tokens_seen": 95927890, + "step": 6899 + }, + { + "epoch": 0.5185630542612355, + "grad_norm": 1.7954790082961642, + "learning_rate": 1.977118916223915e-06, + "loss": 0.9862, + "num_input_tokens_seen": 95946760, + "step": 6900 + }, + { + "epoch": 0.5186382083270705, + "grad_norm": 1.3873917356987597, + "learning_rate": 1.9766321065778387e-06, + "loss": 0.9833, + "num_input_tokens_seen": 95968760, + "step": 6901 + }, + { + "epoch": 0.5187133623929054, + "grad_norm": 1.396789218735025, + "learning_rate": 1.9761452983163996e-06, + "loss": 0.9749, + "num_input_tokens_seen": 95991080, + "step": 6902 + }, + { + "epoch": 0.5187885164587405, + "grad_norm": 1.5787202071446667, + "learning_rate": 1.9756584914684425e-06, + "loss": 0.9955, + "num_input_tokens_seen": 96014255, + "step": 6903 + }, + { + "epoch": 0.5188636705245754, + "grad_norm": 1.6894564413578361, + "learning_rate": 1.975171686062814e-06, + "loss": 0.9887, + "num_input_tokens_seen": 96037015, + "step": 6904 + }, + { + "epoch": 0.5189388245904103, + "grad_norm": 1.7845821333881233, + "learning_rate": 1.974684882128357e-06, + "loss": 0.9708, + "num_input_tokens_seen": 96059605, + "step": 6905 + }, + { + "epoch": 0.5190139786562453, + "grad_norm": 1.4039102799448497, + "learning_rate": 1.974198079693918e-06, + "loss": 1.0292, + "num_input_tokens_seen": 96081795, + "step": 6906 + }, + { + "epoch": 0.5190891327220802, + "grad_norm": 1.545055668145869, + "learning_rate": 1.973711278788342e-06, + "loss": 0.9098, + "num_input_tokens_seen": 96105510, + "step": 6907 + }, + { + "epoch": 0.5191642867879153, + "grad_norm": 1.7684509482835495, + "learning_rate": 1.973224479440473e-06, + "loss": 0.8564, + "num_input_tokens_seen": 96130010, + "step": 6908 + }, + { + "epoch": 0.5192394408537502, + "grad_norm": 1.4510318899087733, + "learning_rate": 1.972737681679157e-06, + "loss": 0.8763, + "num_input_tokens_seen": 96153900, + "step": 6909 + }, + { + "epoch": 0.5193145949195852, + "grad_norm": 1.7608602593176548, + "learning_rate": 1.9722508855332367e-06, + "loss": 0.9366, + "num_input_tokens_seen": 96176760, + "step": 6910 + }, + { + "epoch": 0.5193897489854201, + "grad_norm": 1.5818479078513232, + "learning_rate": 1.971764091031558e-06, + "loss": 0.9045, + "num_input_tokens_seen": 96200335, + "step": 6911 + }, + { + "epoch": 0.519464903051255, + "grad_norm": 1.2360740267921186, + "learning_rate": 1.971277298202965e-06, + "loss": 0.9403, + "num_input_tokens_seen": 96224415, + "step": 6912 + }, + { + "epoch": 0.5195400571170901, + "grad_norm": 1.5618268661134451, + "learning_rate": 1.9707905070763027e-06, + "loss": 0.8739, + "num_input_tokens_seen": 96248145, + "step": 6913 + }, + { + "epoch": 0.519615211182925, + "grad_norm": 1.5215480727449757, + "learning_rate": 1.970303717680414e-06, + "loss": 0.9599, + "num_input_tokens_seen": 96272780, + "step": 6914 + }, + { + "epoch": 0.51969036524876, + "grad_norm": 1.5718769232736276, + "learning_rate": 1.9698169300441435e-06, + "loss": 0.9888, + "num_input_tokens_seen": 96293140, + "step": 6915 + }, + { + "epoch": 0.5197655193145949, + "grad_norm": 1.8030944001825997, + "learning_rate": 1.9693301441963363e-06, + "loss": 0.9647, + "num_input_tokens_seen": 96314315, + "step": 6916 + }, + { + "epoch": 0.5198406733804298, + "grad_norm": 0.7520539055206943, + "learning_rate": 1.9688433601658345e-06, + "loss": 0.8223, + "num_input_tokens_seen": 96383360, + "step": 6917 + }, + { + "epoch": 0.5199158274462649, + "grad_norm": 1.2274888057936388, + "learning_rate": 1.9683565779814838e-06, + "loss": 0.9813, + "num_input_tokens_seen": 96410225, + "step": 6918 + }, + { + "epoch": 0.5199909815120998, + "grad_norm": 1.4977624416761013, + "learning_rate": 1.967869797672127e-06, + "loss": 0.9938, + "num_input_tokens_seen": 96430685, + "step": 6919 + }, + { + "epoch": 0.5200661355779348, + "grad_norm": 1.9691098202273027, + "learning_rate": 1.967383019266607e-06, + "loss": 0.9723, + "num_input_tokens_seen": 96453880, + "step": 6920 + }, + { + "epoch": 0.5201412896437697, + "grad_norm": 1.447023111429035, + "learning_rate": 1.966896242793769e-06, + "loss": 0.9486, + "num_input_tokens_seen": 96476500, + "step": 6921 + }, + { + "epoch": 0.5202164437096047, + "grad_norm": 1.4955141224553037, + "learning_rate": 1.9664094682824545e-06, + "loss": 0.9771, + "num_input_tokens_seen": 96498745, + "step": 6922 + }, + { + "epoch": 0.5202915977754397, + "grad_norm": 2.2221327233835586, + "learning_rate": 1.965922695761508e-06, + "loss": 0.9408, + "num_input_tokens_seen": 96520820, + "step": 6923 + }, + { + "epoch": 0.5203667518412746, + "grad_norm": 1.6584127448667372, + "learning_rate": 1.9654359252597723e-06, + "loss": 1.0136, + "num_input_tokens_seen": 96543955, + "step": 6924 + }, + { + "epoch": 0.5204419059071096, + "grad_norm": 1.4913224513958996, + "learning_rate": 1.96494915680609e-06, + "loss": 0.8523, + "num_input_tokens_seen": 96569210, + "step": 6925 + }, + { + "epoch": 0.5205170599729445, + "grad_norm": 1.5580537477215555, + "learning_rate": 1.964462390429305e-06, + "loss": 0.8615, + "num_input_tokens_seen": 96591300, + "step": 6926 + }, + { + "epoch": 0.5205922140387795, + "grad_norm": 1.7816023227193463, + "learning_rate": 1.963975626158259e-06, + "loss": 0.9841, + "num_input_tokens_seen": 96616945, + "step": 6927 + }, + { + "epoch": 0.5206673681046144, + "grad_norm": 3.352800474078637, + "learning_rate": 1.963488864021795e-06, + "loss": 1.0549, + "num_input_tokens_seen": 96640930, + "step": 6928 + }, + { + "epoch": 0.5207425221704495, + "grad_norm": 1.9406233193210567, + "learning_rate": 1.9630021040487557e-06, + "loss": 0.9195, + "num_input_tokens_seen": 96663210, + "step": 6929 + }, + { + "epoch": 0.5208176762362844, + "grad_norm": 1.5220521545697667, + "learning_rate": 1.962515346267984e-06, + "loss": 1.0012, + "num_input_tokens_seen": 96687855, + "step": 6930 + }, + { + "epoch": 0.5208928303021193, + "grad_norm": 1.3480683069949002, + "learning_rate": 1.9620285907083213e-06, + "loss": 0.9949, + "num_input_tokens_seen": 96713655, + "step": 6931 + }, + { + "epoch": 0.5209679843679543, + "grad_norm": 1.5520011658221713, + "learning_rate": 1.9615418373986097e-06, + "loss": 0.9743, + "num_input_tokens_seen": 96736805, + "step": 6932 + }, + { + "epoch": 0.5210431384337892, + "grad_norm": 1.3942351040153, + "learning_rate": 1.961055086367692e-06, + "loss": 0.9928, + "num_input_tokens_seen": 96760565, + "step": 6933 + }, + { + "epoch": 0.5211182924996243, + "grad_norm": 1.5328661894655118, + "learning_rate": 1.960568337644409e-06, + "loss": 0.8945, + "num_input_tokens_seen": 96784340, + "step": 6934 + }, + { + "epoch": 0.5211934465654592, + "grad_norm": 1.5285629518602375, + "learning_rate": 1.9600815912576034e-06, + "loss": 0.8846, + "num_input_tokens_seen": 96808725, + "step": 6935 + }, + { + "epoch": 0.5212686006312941, + "grad_norm": 1.6257007995087274, + "learning_rate": 1.9595948472361163e-06, + "loss": 0.9372, + "num_input_tokens_seen": 96829715, + "step": 6936 + }, + { + "epoch": 0.5213437546971291, + "grad_norm": 1.6762958438984934, + "learning_rate": 1.959108105608788e-06, + "loss": 0.9592, + "num_input_tokens_seen": 96853095, + "step": 6937 + }, + { + "epoch": 0.521418908762964, + "grad_norm": 0.7625903107363489, + "learning_rate": 1.958621366404463e-06, + "loss": 0.8039, + "num_input_tokens_seen": 96925565, + "step": 6938 + }, + { + "epoch": 0.5214940628287991, + "grad_norm": 1.2796069519254791, + "learning_rate": 1.958134629651979e-06, + "loss": 1.0369, + "num_input_tokens_seen": 96949905, + "step": 6939 + }, + { + "epoch": 0.521569216894634, + "grad_norm": 1.5837739364874561, + "learning_rate": 1.957647895380179e-06, + "loss": 0.9344, + "num_input_tokens_seen": 96971455, + "step": 6940 + }, + { + "epoch": 0.521644370960469, + "grad_norm": 1.3296875233095482, + "learning_rate": 1.9571611636179037e-06, + "loss": 0.9609, + "num_input_tokens_seen": 96999010, + "step": 6941 + }, + { + "epoch": 0.5217195250263039, + "grad_norm": 1.5531319355186177, + "learning_rate": 1.956674434393993e-06, + "loss": 1.0133, + "num_input_tokens_seen": 97022505, + "step": 6942 + }, + { + "epoch": 0.5217946790921388, + "grad_norm": 1.4779972593826673, + "learning_rate": 1.9561877077372884e-06, + "loss": 0.937, + "num_input_tokens_seen": 97045380, + "step": 6943 + }, + { + "epoch": 0.5218698331579739, + "grad_norm": 1.4264138686433903, + "learning_rate": 1.9557009836766294e-06, + "loss": 0.8647, + "num_input_tokens_seen": 97070645, + "step": 6944 + }, + { + "epoch": 0.5219449872238088, + "grad_norm": 1.547205263581736, + "learning_rate": 1.9552142622408574e-06, + "loss": 1.0313, + "num_input_tokens_seen": 97092115, + "step": 6945 + }, + { + "epoch": 0.5220201412896438, + "grad_norm": 1.9088897562981024, + "learning_rate": 1.954727543458812e-06, + "loss": 0.8849, + "num_input_tokens_seen": 97116695, + "step": 6946 + }, + { + "epoch": 0.5220952953554787, + "grad_norm": 1.3879696643933264, + "learning_rate": 1.9542408273593324e-06, + "loss": 0.9592, + "num_input_tokens_seen": 97142240, + "step": 6947 + }, + { + "epoch": 0.5221704494213137, + "grad_norm": 1.8364768027510847, + "learning_rate": 1.9537541139712594e-06, + "loss": 1.0706, + "num_input_tokens_seen": 97164750, + "step": 6948 + }, + { + "epoch": 0.5222456034871487, + "grad_norm": 1.5383232393825632, + "learning_rate": 1.9532674033234315e-06, + "loss": 1.0024, + "num_input_tokens_seen": 97186725, + "step": 6949 + }, + { + "epoch": 0.5223207575529836, + "grad_norm": 1.5669189527301814, + "learning_rate": 1.95278069544469e-06, + "loss": 0.9521, + "num_input_tokens_seen": 97208490, + "step": 6950 + }, + { + "epoch": 0.5223959116188186, + "grad_norm": 2.086922745068508, + "learning_rate": 1.952293990363872e-06, + "loss": 1.0447, + "num_input_tokens_seen": 97229490, + "step": 6951 + }, + { + "epoch": 0.5224710656846535, + "grad_norm": 1.5483759524280183, + "learning_rate": 1.9518072881098185e-06, + "loss": 0.9458, + "num_input_tokens_seen": 97251800, + "step": 6952 + }, + { + "epoch": 0.5225462197504885, + "grad_norm": 0.9551116687836293, + "learning_rate": 1.9513205887113675e-06, + "loss": 0.9084, + "num_input_tokens_seen": 97332865, + "step": 6953 + }, + { + "epoch": 0.5226213738163235, + "grad_norm": 1.2805175548513463, + "learning_rate": 1.9508338921973576e-06, + "loss": 0.863, + "num_input_tokens_seen": 97356965, + "step": 6954 + }, + { + "epoch": 0.5226965278821585, + "grad_norm": 1.7097092491723036, + "learning_rate": 1.9503471985966285e-06, + "loss": 0.9966, + "num_input_tokens_seen": 97379870, + "step": 6955 + }, + { + "epoch": 0.5227716819479934, + "grad_norm": 1.4343046250043816, + "learning_rate": 1.9498605079380176e-06, + "loss": 0.9064, + "num_input_tokens_seen": 97405260, + "step": 6956 + }, + { + "epoch": 0.5228468360138283, + "grad_norm": 1.4650517430460421, + "learning_rate": 1.949373820250364e-06, + "loss": 0.9286, + "num_input_tokens_seen": 97428405, + "step": 6957 + }, + { + "epoch": 0.5229219900796633, + "grad_norm": 1.5535521842754327, + "learning_rate": 1.9488871355625054e-06, + "loss": 0.9641, + "num_input_tokens_seen": 97450120, + "step": 6958 + }, + { + "epoch": 0.5229971441454982, + "grad_norm": 1.4943006641720629, + "learning_rate": 1.9484004539032786e-06, + "loss": 0.9719, + "num_input_tokens_seen": 97472460, + "step": 6959 + }, + { + "epoch": 0.5230722982113333, + "grad_norm": 1.486916785148113, + "learning_rate": 1.947913775301524e-06, + "loss": 0.9371, + "num_input_tokens_seen": 97495355, + "step": 6960 + }, + { + "epoch": 0.5231474522771682, + "grad_norm": 1.6558232747377015, + "learning_rate": 1.9474270997860766e-06, + "loss": 0.9394, + "num_input_tokens_seen": 97515775, + "step": 6961 + }, + { + "epoch": 0.5232226063430031, + "grad_norm": 1.9225948195271494, + "learning_rate": 1.946940427385776e-06, + "loss": 1.0035, + "num_input_tokens_seen": 97533705, + "step": 6962 + }, + { + "epoch": 0.5232977604088381, + "grad_norm": 1.392118444988543, + "learning_rate": 1.9464537581294576e-06, + "loss": 0.9076, + "num_input_tokens_seen": 97556880, + "step": 6963 + }, + { + "epoch": 0.523372914474673, + "grad_norm": 0.714622046271868, + "learning_rate": 1.945967092045959e-06, + "loss": 0.7654, + "num_input_tokens_seen": 97636605, + "step": 6964 + }, + { + "epoch": 0.5234480685405081, + "grad_norm": 1.4976224335308672, + "learning_rate": 1.945480429164117e-06, + "loss": 0.8978, + "num_input_tokens_seen": 97658675, + "step": 6965 + }, + { + "epoch": 0.523523222606343, + "grad_norm": 1.506964331328179, + "learning_rate": 1.944993769512768e-06, + "loss": 0.9273, + "num_input_tokens_seen": 97684580, + "step": 6966 + }, + { + "epoch": 0.523598376672178, + "grad_norm": 2.0475689825377636, + "learning_rate": 1.9445071131207497e-06, + "loss": 0.9597, + "num_input_tokens_seen": 97704460, + "step": 6967 + }, + { + "epoch": 0.5236735307380129, + "grad_norm": 1.435051948191911, + "learning_rate": 1.9440204600168975e-06, + "loss": 0.9751, + "num_input_tokens_seen": 97728090, + "step": 6968 + }, + { + "epoch": 0.5237486848038478, + "grad_norm": 1.6324764800312515, + "learning_rate": 1.9435338102300467e-06, + "loss": 0.9725, + "num_input_tokens_seen": 97750670, + "step": 6969 + }, + { + "epoch": 0.5238238388696829, + "grad_norm": 1.662844431483974, + "learning_rate": 1.943047163789034e-06, + "loss": 0.9642, + "num_input_tokens_seen": 97774040, + "step": 6970 + }, + { + "epoch": 0.5238989929355178, + "grad_norm": 1.720735062509546, + "learning_rate": 1.942560520722695e-06, + "loss": 0.8668, + "num_input_tokens_seen": 97794510, + "step": 6971 + }, + { + "epoch": 0.5239741470013528, + "grad_norm": 1.4831016298840816, + "learning_rate": 1.9420738810598653e-06, + "loss": 0.9348, + "num_input_tokens_seen": 97820345, + "step": 6972 + }, + { + "epoch": 0.5240493010671877, + "grad_norm": 1.5546551418291676, + "learning_rate": 1.9415872448293796e-06, + "loss": 1.0321, + "num_input_tokens_seen": 97841900, + "step": 6973 + }, + { + "epoch": 0.5241244551330227, + "grad_norm": 1.3828174138385945, + "learning_rate": 1.9411006120600737e-06, + "loss": 0.9243, + "num_input_tokens_seen": 97867030, + "step": 6974 + }, + { + "epoch": 0.5241996091988577, + "grad_norm": 1.6726312798494138, + "learning_rate": 1.940613982780782e-06, + "loss": 0.9169, + "num_input_tokens_seen": 97889490, + "step": 6975 + }, + { + "epoch": 0.5242747632646926, + "grad_norm": 1.5417563781754806, + "learning_rate": 1.940127357020339e-06, + "loss": 0.8958, + "num_input_tokens_seen": 97911860, + "step": 6976 + }, + { + "epoch": 0.5243499173305276, + "grad_norm": 1.506821776327859, + "learning_rate": 1.9396407348075796e-06, + "loss": 1.05, + "num_input_tokens_seen": 97932410, + "step": 6977 + }, + { + "epoch": 0.5244250713963625, + "grad_norm": 1.6900115925769366, + "learning_rate": 1.939154116171337e-06, + "loss": 0.9936, + "num_input_tokens_seen": 97949430, + "step": 6978 + }, + { + "epoch": 0.5245002254621975, + "grad_norm": 1.6844183723371746, + "learning_rate": 1.9386675011404473e-06, + "loss": 0.9391, + "num_input_tokens_seen": 97972070, + "step": 6979 + }, + { + "epoch": 0.5245753795280325, + "grad_norm": 1.533758716387491, + "learning_rate": 1.9381808897437427e-06, + "loss": 0.9829, + "num_input_tokens_seen": 97994240, + "step": 6980 + }, + { + "epoch": 0.5246505335938674, + "grad_norm": 1.7657575979489244, + "learning_rate": 1.9376942820100563e-06, + "loss": 0.8823, + "num_input_tokens_seen": 98018945, + "step": 6981 + }, + { + "epoch": 0.5247256876597024, + "grad_norm": 1.9107122255729294, + "learning_rate": 1.937207677968223e-06, + "loss": 0.9326, + "num_input_tokens_seen": 98044350, + "step": 6982 + }, + { + "epoch": 0.5248008417255373, + "grad_norm": 1.9892647639636762, + "learning_rate": 1.9367210776470744e-06, + "loss": 0.9901, + "num_input_tokens_seen": 98066015, + "step": 6983 + }, + { + "epoch": 0.5248759957913723, + "grad_norm": 1.6151561418289542, + "learning_rate": 1.9362344810754455e-06, + "loss": 0.934, + "num_input_tokens_seen": 98092160, + "step": 6984 + }, + { + "epoch": 0.5249511498572073, + "grad_norm": 1.428941292226181, + "learning_rate": 1.935747888282167e-06, + "loss": 0.9489, + "num_input_tokens_seen": 98116500, + "step": 6985 + }, + { + "epoch": 0.5250263039230423, + "grad_norm": 1.1470833924267292, + "learning_rate": 1.935261299296072e-06, + "loss": 0.9561, + "num_input_tokens_seen": 98142450, + "step": 6986 + }, + { + "epoch": 0.5251014579888772, + "grad_norm": 1.4129288486565652, + "learning_rate": 1.9347747141459934e-06, + "loss": 0.9589, + "num_input_tokens_seen": 98168050, + "step": 6987 + }, + { + "epoch": 0.5251766120547121, + "grad_norm": 1.6850770580658538, + "learning_rate": 1.9342881328607625e-06, + "loss": 0.9202, + "num_input_tokens_seen": 98192080, + "step": 6988 + }, + { + "epoch": 0.5252517661205471, + "grad_norm": 1.4297506182969917, + "learning_rate": 1.9338015554692116e-06, + "loss": 0.9029, + "num_input_tokens_seen": 98217285, + "step": 6989 + }, + { + "epoch": 0.525326920186382, + "grad_norm": 1.6771168044010076, + "learning_rate": 1.933314982000172e-06, + "loss": 0.9611, + "num_input_tokens_seen": 98239750, + "step": 6990 + }, + { + "epoch": 0.5254020742522171, + "grad_norm": 1.1262783628696482, + "learning_rate": 1.932828412482475e-06, + "loss": 0.9804, + "num_input_tokens_seen": 98266410, + "step": 6991 + }, + { + "epoch": 0.525477228318052, + "grad_norm": 1.7071647679119333, + "learning_rate": 1.9323418469449517e-06, + "loss": 0.9586, + "num_input_tokens_seen": 98287460, + "step": 6992 + }, + { + "epoch": 0.525552382383887, + "grad_norm": 1.4025763528613242, + "learning_rate": 1.931855285416433e-06, + "loss": 0.9999, + "num_input_tokens_seen": 98310360, + "step": 6993 + }, + { + "epoch": 0.5256275364497219, + "grad_norm": 1.8509573933557824, + "learning_rate": 1.93136872792575e-06, + "loss": 0.9143, + "num_input_tokens_seen": 98332405, + "step": 6994 + }, + { + "epoch": 0.5257026905155568, + "grad_norm": 1.4402091164542883, + "learning_rate": 1.930882174501732e-06, + "loss": 0.8769, + "num_input_tokens_seen": 98357220, + "step": 6995 + }, + { + "epoch": 0.5257778445813919, + "grad_norm": 1.57160516929388, + "learning_rate": 1.93039562517321e-06, + "loss": 0.9281, + "num_input_tokens_seen": 98380310, + "step": 6996 + }, + { + "epoch": 0.5258529986472268, + "grad_norm": 1.4616369521202817, + "learning_rate": 1.929909079969014e-06, + "loss": 0.9258, + "num_input_tokens_seen": 98403700, + "step": 6997 + }, + { + "epoch": 0.5259281527130618, + "grad_norm": 0.7327364233419265, + "learning_rate": 1.929422538917973e-06, + "loss": 0.7882, + "num_input_tokens_seen": 98484205, + "step": 6998 + }, + { + "epoch": 0.5260033067788967, + "grad_norm": 1.707075664021039, + "learning_rate": 1.9289360020489177e-06, + "loss": 0.901, + "num_input_tokens_seen": 98508965, + "step": 6999 + }, + { + "epoch": 0.5260784608447318, + "grad_norm": 1.3824760329391994, + "learning_rate": 1.928449469390675e-06, + "loss": 1.0192, + "num_input_tokens_seen": 98534385, + "step": 7000 + }, + { + "epoch": 0.5261536149105667, + "grad_norm": 1.3689717013079319, + "learning_rate": 1.927962940972077e-06, + "loss": 0.929, + "num_input_tokens_seen": 98562180, + "step": 7001 + }, + { + "epoch": 0.5262287689764016, + "grad_norm": 1.5791928295024884, + "learning_rate": 1.9274764168219495e-06, + "loss": 0.9078, + "num_input_tokens_seen": 98584410, + "step": 7002 + }, + { + "epoch": 0.5263039230422366, + "grad_norm": 1.2086172702032, + "learning_rate": 1.9269898969691214e-06, + "loss": 0.9189, + "num_input_tokens_seen": 98611175, + "step": 7003 + }, + { + "epoch": 0.5263790771080715, + "grad_norm": 1.4541312394658799, + "learning_rate": 1.9265033814424227e-06, + "loss": 0.9446, + "num_input_tokens_seen": 98633840, + "step": 7004 + }, + { + "epoch": 0.5264542311739066, + "grad_norm": 1.6602175712480023, + "learning_rate": 1.9260168702706794e-06, + "loss": 0.9122, + "num_input_tokens_seen": 98657830, + "step": 7005 + }, + { + "epoch": 0.5265293852397415, + "grad_norm": 1.4164767231481281, + "learning_rate": 1.9255303634827204e-06, + "loss": 0.9957, + "num_input_tokens_seen": 98682560, + "step": 7006 + }, + { + "epoch": 0.5266045393055764, + "grad_norm": 4.787297033883003, + "learning_rate": 1.9250438611073724e-06, + "loss": 0.8181, + "num_input_tokens_seen": 98703690, + "step": 7007 + }, + { + "epoch": 0.5266796933714114, + "grad_norm": 1.534521617216331, + "learning_rate": 1.924557363173462e-06, + "loss": 0.9405, + "num_input_tokens_seen": 98726490, + "step": 7008 + }, + { + "epoch": 0.5267548474372463, + "grad_norm": 1.8084743781686015, + "learning_rate": 1.9240708697098174e-06, + "loss": 1.0156, + "num_input_tokens_seen": 98747715, + "step": 7009 + }, + { + "epoch": 0.5268300015030813, + "grad_norm": 1.4717533166561918, + "learning_rate": 1.9235843807452642e-06, + "loss": 0.991, + "num_input_tokens_seen": 98770060, + "step": 7010 + }, + { + "epoch": 0.5269051555689163, + "grad_norm": 1.4632522278701452, + "learning_rate": 1.923097896308629e-06, + "loss": 0.8952, + "num_input_tokens_seen": 98795255, + "step": 7011 + }, + { + "epoch": 0.5269803096347513, + "grad_norm": 1.4000535716297529, + "learning_rate": 1.9226114164287384e-06, + "loss": 0.9761, + "num_input_tokens_seen": 98819370, + "step": 7012 + }, + { + "epoch": 0.5270554637005862, + "grad_norm": 0.7580267219966499, + "learning_rate": 1.9221249411344173e-06, + "loss": 0.7575, + "num_input_tokens_seen": 98894855, + "step": 7013 + }, + { + "epoch": 0.5271306177664211, + "grad_norm": 1.7403317134191814, + "learning_rate": 1.921638470454492e-06, + "loss": 0.8202, + "num_input_tokens_seen": 98919840, + "step": 7014 + }, + { + "epoch": 0.5272057718322561, + "grad_norm": 1.7524771868735083, + "learning_rate": 1.9211520044177866e-06, + "loss": 1.0505, + "num_input_tokens_seen": 98941405, + "step": 7015 + }, + { + "epoch": 0.5272809258980911, + "grad_norm": 1.8029613083397318, + "learning_rate": 1.9206655430531277e-06, + "loss": 0.8698, + "num_input_tokens_seen": 98962680, + "step": 7016 + }, + { + "epoch": 0.5273560799639261, + "grad_norm": 1.6529001151505414, + "learning_rate": 1.9201790863893387e-06, + "loss": 0.9497, + "num_input_tokens_seen": 98985465, + "step": 7017 + }, + { + "epoch": 0.527431234029761, + "grad_norm": 1.4011754346706073, + "learning_rate": 1.919692634455245e-06, + "loss": 0.8363, + "num_input_tokens_seen": 99012990, + "step": 7018 + }, + { + "epoch": 0.527506388095596, + "grad_norm": 1.456758448039289, + "learning_rate": 1.91920618727967e-06, + "loss": 0.937, + "num_input_tokens_seen": 99035730, + "step": 7019 + }, + { + "epoch": 0.5275815421614309, + "grad_norm": 1.51775328657523, + "learning_rate": 1.9187197448914374e-06, + "loss": 0.9058, + "num_input_tokens_seen": 99057915, + "step": 7020 + }, + { + "epoch": 0.5276566962272659, + "grad_norm": 1.495173477518497, + "learning_rate": 1.918233307319371e-06, + "loss": 0.9387, + "num_input_tokens_seen": 99081870, + "step": 7021 + }, + { + "epoch": 0.5277318502931009, + "grad_norm": 3.3367035676311114, + "learning_rate": 1.9177468745922944e-06, + "loss": 0.986, + "num_input_tokens_seen": 99107580, + "step": 7022 + }, + { + "epoch": 0.5278070043589358, + "grad_norm": 1.5117854613732342, + "learning_rate": 1.917260446739031e-06, + "loss": 0.9457, + "num_input_tokens_seen": 99130660, + "step": 7023 + }, + { + "epoch": 0.5278821584247708, + "grad_norm": 1.8024310913882065, + "learning_rate": 1.916774023788403e-06, + "loss": 0.9849, + "num_input_tokens_seen": 99152355, + "step": 7024 + }, + { + "epoch": 0.5279573124906057, + "grad_norm": 1.443964148818706, + "learning_rate": 1.9162876057692317e-06, + "loss": 0.9258, + "num_input_tokens_seen": 99177875, + "step": 7025 + }, + { + "epoch": 0.5280324665564406, + "grad_norm": 1.5135676784953702, + "learning_rate": 1.9158011927103413e-06, + "loss": 0.8747, + "num_input_tokens_seen": 99203430, + "step": 7026 + }, + { + "epoch": 0.5281076206222757, + "grad_norm": 1.5886117721680455, + "learning_rate": 1.9153147846405525e-06, + "loss": 0.9513, + "num_input_tokens_seen": 99226915, + "step": 7027 + }, + { + "epoch": 0.5281827746881106, + "grad_norm": 1.8255250489045218, + "learning_rate": 1.914828381588687e-06, + "loss": 0.8693, + "num_input_tokens_seen": 99250655, + "step": 7028 + }, + { + "epoch": 0.5282579287539456, + "grad_norm": 1.818978864814399, + "learning_rate": 1.9143419835835663e-06, + "loss": 0.9851, + "num_input_tokens_seen": 99271080, + "step": 7029 + }, + { + "epoch": 0.5283330828197805, + "grad_norm": 1.4825169227496238, + "learning_rate": 1.9138555906540103e-06, + "loss": 0.9978, + "num_input_tokens_seen": 99297900, + "step": 7030 + }, + { + "epoch": 0.5284082368856156, + "grad_norm": 1.6330028952241777, + "learning_rate": 1.9133692028288413e-06, + "loss": 0.9283, + "num_input_tokens_seen": 99319325, + "step": 7031 + }, + { + "epoch": 0.5284833909514505, + "grad_norm": 1.8933260200699524, + "learning_rate": 1.912882820136878e-06, + "loss": 0.8464, + "num_input_tokens_seen": 99342060, + "step": 7032 + }, + { + "epoch": 0.5285585450172854, + "grad_norm": 1.3848989231533757, + "learning_rate": 1.9123964426069416e-06, + "loss": 0.8666, + "num_input_tokens_seen": 99366840, + "step": 7033 + }, + { + "epoch": 0.5286336990831204, + "grad_norm": 1.7543895709686697, + "learning_rate": 1.9119100702678515e-06, + "loss": 0.9201, + "num_input_tokens_seen": 99392535, + "step": 7034 + }, + { + "epoch": 0.5287088531489553, + "grad_norm": 1.8039639165953207, + "learning_rate": 1.9114237031484266e-06, + "loss": 0.9573, + "num_input_tokens_seen": 99413980, + "step": 7035 + }, + { + "epoch": 0.5287840072147904, + "grad_norm": 1.2688075423698366, + "learning_rate": 1.9109373412774867e-06, + "loss": 1.0084, + "num_input_tokens_seen": 99437475, + "step": 7036 + }, + { + "epoch": 0.5288591612806253, + "grad_norm": 1.83735511104751, + "learning_rate": 1.91045098468385e-06, + "loss": 1.0473, + "num_input_tokens_seen": 99460555, + "step": 7037 + }, + { + "epoch": 0.5289343153464603, + "grad_norm": 1.3456774793597892, + "learning_rate": 1.9099646333963363e-06, + "loss": 0.9221, + "num_input_tokens_seen": 99487605, + "step": 7038 + }, + { + "epoch": 0.5290094694122952, + "grad_norm": 0.9148376634038592, + "learning_rate": 1.9094782874437625e-06, + "loss": 0.8023, + "num_input_tokens_seen": 99563715, + "step": 7039 + }, + { + "epoch": 0.5290846234781301, + "grad_norm": 1.527158291769428, + "learning_rate": 1.9089919468549464e-06, + "loss": 0.9246, + "num_input_tokens_seen": 99589290, + "step": 7040 + }, + { + "epoch": 0.5291597775439651, + "grad_norm": 0.8157822263748706, + "learning_rate": 1.9085056116587068e-06, + "loss": 0.8311, + "num_input_tokens_seen": 99666095, + "step": 7041 + }, + { + "epoch": 0.5292349316098001, + "grad_norm": 0.8264932210405385, + "learning_rate": 1.908019281883859e-06, + "loss": 0.8298, + "num_input_tokens_seen": 99737015, + "step": 7042 + }, + { + "epoch": 0.5293100856756351, + "grad_norm": 2.2926890317967232, + "learning_rate": 1.9075329575592217e-06, + "loss": 0.9084, + "num_input_tokens_seen": 99762305, + "step": 7043 + }, + { + "epoch": 0.52938523974147, + "grad_norm": 1.595887639223121, + "learning_rate": 1.9070466387136095e-06, + "loss": 0.9593, + "num_input_tokens_seen": 99786005, + "step": 7044 + }, + { + "epoch": 0.529460393807305, + "grad_norm": 1.3586344839507882, + "learning_rate": 1.906560325375841e-06, + "loss": 0.924, + "num_input_tokens_seen": 99809985, + "step": 7045 + }, + { + "epoch": 0.52953554787314, + "grad_norm": 1.5576528515215677, + "learning_rate": 1.9060740175747317e-06, + "loss": 0.9861, + "num_input_tokens_seen": 99831585, + "step": 7046 + }, + { + "epoch": 0.5296107019389749, + "grad_norm": 1.429616874242799, + "learning_rate": 1.9055877153390948e-06, + "loss": 1.0272, + "num_input_tokens_seen": 99854975, + "step": 7047 + }, + { + "epoch": 0.5296858560048099, + "grad_norm": 1.7836207624638125, + "learning_rate": 1.9051014186977485e-06, + "loss": 0.9457, + "num_input_tokens_seen": 99876095, + "step": 7048 + }, + { + "epoch": 0.5297610100706448, + "grad_norm": 1.629373826354628, + "learning_rate": 1.9046151276795062e-06, + "loss": 0.9858, + "num_input_tokens_seen": 99897585, + "step": 7049 + }, + { + "epoch": 0.5298361641364798, + "grad_norm": 1.4683574395396595, + "learning_rate": 1.904128842313183e-06, + "loss": 0.829, + "num_input_tokens_seen": 99927530, + "step": 7050 + }, + { + "epoch": 0.5299113182023147, + "grad_norm": 2.0831237560397833, + "learning_rate": 1.9036425626275929e-06, + "loss": 1.0004, + "num_input_tokens_seen": 99950400, + "step": 7051 + }, + { + "epoch": 0.5299864722681497, + "grad_norm": 1.598896095812504, + "learning_rate": 1.9031562886515497e-06, + "loss": 0.9252, + "num_input_tokens_seen": 99973075, + "step": 7052 + }, + { + "epoch": 0.5300616263339847, + "grad_norm": 2.021019156914943, + "learning_rate": 1.9026700204138676e-06, + "loss": 0.9586, + "num_input_tokens_seen": 99993975, + "step": 7053 + }, + { + "epoch": 0.5301367803998196, + "grad_norm": 1.2279603124567222, + "learning_rate": 1.9021837579433593e-06, + "loss": 0.9946, + "num_input_tokens_seen": 100019435, + "step": 7054 + }, + { + "epoch": 0.5302119344656546, + "grad_norm": 1.3915929961359144, + "learning_rate": 1.9016975012688382e-06, + "loss": 0.9818, + "num_input_tokens_seen": 100043520, + "step": 7055 + }, + { + "epoch": 0.5302870885314895, + "grad_norm": 0.8500312546986367, + "learning_rate": 1.901211250419116e-06, + "loss": 0.7975, + "num_input_tokens_seen": 100111000, + "step": 7056 + }, + { + "epoch": 0.5303622425973246, + "grad_norm": 1.647074275835865, + "learning_rate": 1.900725005423006e-06, + "loss": 1.0118, + "num_input_tokens_seen": 100131035, + "step": 7057 + }, + { + "epoch": 0.5304373966631595, + "grad_norm": 1.7977845916624158, + "learning_rate": 1.9002387663093195e-06, + "loss": 0.9509, + "num_input_tokens_seen": 100154920, + "step": 7058 + }, + { + "epoch": 0.5305125507289944, + "grad_norm": 1.4390095786968122, + "learning_rate": 1.899752533106868e-06, + "loss": 1.0802, + "num_input_tokens_seen": 100179555, + "step": 7059 + }, + { + "epoch": 0.5305877047948294, + "grad_norm": 1.290406607191526, + "learning_rate": 1.8992663058444629e-06, + "loss": 0.8671, + "num_input_tokens_seen": 100202955, + "step": 7060 + }, + { + "epoch": 0.5306628588606643, + "grad_norm": 0.797345564786149, + "learning_rate": 1.8987800845509146e-06, + "loss": 0.8108, + "num_input_tokens_seen": 100279920, + "step": 7061 + }, + { + "epoch": 0.5307380129264994, + "grad_norm": 1.5427921242360347, + "learning_rate": 1.8982938692550344e-06, + "loss": 0.8404, + "num_input_tokens_seen": 100304345, + "step": 7062 + }, + { + "epoch": 0.5308131669923343, + "grad_norm": 1.3366369457618983, + "learning_rate": 1.8978076599856317e-06, + "loss": 1.0053, + "num_input_tokens_seen": 100328000, + "step": 7063 + }, + { + "epoch": 0.5308883210581693, + "grad_norm": 1.6344930952596703, + "learning_rate": 1.897321456771516e-06, + "loss": 0.9516, + "num_input_tokens_seen": 100349525, + "step": 7064 + }, + { + "epoch": 0.5309634751240042, + "grad_norm": 1.406430126945733, + "learning_rate": 1.8968352596414977e-06, + "loss": 0.9301, + "num_input_tokens_seen": 100374580, + "step": 7065 + }, + { + "epoch": 0.5310386291898391, + "grad_norm": 1.997841266035338, + "learning_rate": 1.8963490686243847e-06, + "loss": 0.943, + "num_input_tokens_seen": 100395515, + "step": 7066 + }, + { + "epoch": 0.5311137832556742, + "grad_norm": 1.5804423323878092, + "learning_rate": 1.895862883748987e-06, + "loss": 1.0205, + "num_input_tokens_seen": 100418360, + "step": 7067 + }, + { + "epoch": 0.5311889373215091, + "grad_norm": 2.504766783527373, + "learning_rate": 1.895376705044112e-06, + "loss": 0.9831, + "num_input_tokens_seen": 100441025, + "step": 7068 + }, + { + "epoch": 0.5312640913873441, + "grad_norm": 1.666274342455699, + "learning_rate": 1.8948905325385675e-06, + "loss": 0.9573, + "num_input_tokens_seen": 100463530, + "step": 7069 + }, + { + "epoch": 0.531339245453179, + "grad_norm": 1.3526325289935608, + "learning_rate": 1.894404366261162e-06, + "loss": 0.8727, + "num_input_tokens_seen": 100486035, + "step": 7070 + }, + { + "epoch": 0.5314143995190139, + "grad_norm": 1.8264965401153834, + "learning_rate": 1.8939182062407017e-06, + "loss": 0.9363, + "num_input_tokens_seen": 100510900, + "step": 7071 + }, + { + "epoch": 0.531489553584849, + "grad_norm": 0.7239729561918384, + "learning_rate": 1.8934320525059944e-06, + "loss": 0.7502, + "num_input_tokens_seen": 100584785, + "step": 7072 + }, + { + "epoch": 0.5315647076506839, + "grad_norm": 1.257621655883529, + "learning_rate": 1.8929459050858458e-06, + "loss": 0.9323, + "num_input_tokens_seen": 100610335, + "step": 7073 + }, + { + "epoch": 0.5316398617165189, + "grad_norm": 1.535926838990134, + "learning_rate": 1.892459764009062e-06, + "loss": 0.9134, + "num_input_tokens_seen": 100633885, + "step": 7074 + }, + { + "epoch": 0.5317150157823538, + "grad_norm": 2.5443519311251275, + "learning_rate": 1.8919736293044495e-06, + "loss": 0.8835, + "num_input_tokens_seen": 100659905, + "step": 7075 + }, + { + "epoch": 0.5317901698481888, + "grad_norm": 1.2190379756210077, + "learning_rate": 1.8914875010008124e-06, + "loss": 0.9712, + "num_input_tokens_seen": 100684560, + "step": 7076 + }, + { + "epoch": 0.5318653239140237, + "grad_norm": 1.628126071922257, + "learning_rate": 1.891001379126957e-06, + "loss": 0.9625, + "num_input_tokens_seen": 100710035, + "step": 7077 + }, + { + "epoch": 0.5319404779798587, + "grad_norm": 1.329804424519722, + "learning_rate": 1.8905152637116868e-06, + "loss": 0.9016, + "num_input_tokens_seen": 100734395, + "step": 7078 + }, + { + "epoch": 0.5320156320456937, + "grad_norm": 1.8610823670187204, + "learning_rate": 1.890029154783807e-06, + "loss": 0.82, + "num_input_tokens_seen": 100757280, + "step": 7079 + }, + { + "epoch": 0.5320907861115286, + "grad_norm": 1.5420453749139347, + "learning_rate": 1.889543052372121e-06, + "loss": 1.0271, + "num_input_tokens_seen": 100778415, + "step": 7080 + }, + { + "epoch": 0.5321659401773636, + "grad_norm": 1.6559249990912501, + "learning_rate": 1.8890569565054313e-06, + "loss": 0.9997, + "num_input_tokens_seen": 100802170, + "step": 7081 + }, + { + "epoch": 0.5322410942431985, + "grad_norm": 2.115322065428979, + "learning_rate": 1.8885708672125425e-06, + "loss": 0.882, + "num_input_tokens_seen": 100824485, + "step": 7082 + }, + { + "epoch": 0.5323162483090336, + "grad_norm": 1.7470514116283002, + "learning_rate": 1.888084784522256e-06, + "loss": 1.0179, + "num_input_tokens_seen": 100848055, + "step": 7083 + }, + { + "epoch": 0.5323914023748685, + "grad_norm": 1.4532154014113565, + "learning_rate": 1.8875987084633748e-06, + "loss": 0.959, + "num_input_tokens_seen": 100869395, + "step": 7084 + }, + { + "epoch": 0.5324665564407034, + "grad_norm": 1.6031660286633462, + "learning_rate": 1.887112639064701e-06, + "loss": 0.8843, + "num_input_tokens_seen": 100896265, + "step": 7085 + }, + { + "epoch": 0.5325417105065384, + "grad_norm": 2.2383580485911154, + "learning_rate": 1.8866265763550344e-06, + "loss": 0.8532, + "num_input_tokens_seen": 100926875, + "step": 7086 + }, + { + "epoch": 0.5326168645723733, + "grad_norm": 1.4861948201657644, + "learning_rate": 1.8861405203631786e-06, + "loss": 1.0819, + "num_input_tokens_seen": 100949130, + "step": 7087 + }, + { + "epoch": 0.5326920186382084, + "grad_norm": 1.6266254218662402, + "learning_rate": 1.8856544711179317e-06, + "loss": 0.9237, + "num_input_tokens_seen": 100972075, + "step": 7088 + }, + { + "epoch": 0.5327671727040433, + "grad_norm": 2.1071334797094026, + "learning_rate": 1.8851684286480962e-06, + "loss": 0.8724, + "num_input_tokens_seen": 100994505, + "step": 7089 + }, + { + "epoch": 0.5328423267698783, + "grad_norm": 1.6068091818150458, + "learning_rate": 1.884682392982471e-06, + "loss": 0.8656, + "num_input_tokens_seen": 101016650, + "step": 7090 + }, + { + "epoch": 0.5329174808357132, + "grad_norm": 1.661245262650164, + "learning_rate": 1.884196364149855e-06, + "loss": 0.9134, + "num_input_tokens_seen": 101040340, + "step": 7091 + }, + { + "epoch": 0.5329926349015481, + "grad_norm": 1.5550030718708667, + "learning_rate": 1.8837103421790483e-06, + "loss": 0.9368, + "num_input_tokens_seen": 101063755, + "step": 7092 + }, + { + "epoch": 0.5330677889673832, + "grad_norm": 1.875361344169068, + "learning_rate": 1.8832243270988488e-06, + "loss": 0.8774, + "num_input_tokens_seen": 101086775, + "step": 7093 + }, + { + "epoch": 0.5331429430332181, + "grad_norm": 1.4842307572803273, + "learning_rate": 1.8827383189380556e-06, + "loss": 0.9502, + "num_input_tokens_seen": 101111240, + "step": 7094 + }, + { + "epoch": 0.5332180970990531, + "grad_norm": 1.349152814869796, + "learning_rate": 1.8822523177254658e-06, + "loss": 0.8969, + "num_input_tokens_seen": 101134425, + "step": 7095 + }, + { + "epoch": 0.533293251164888, + "grad_norm": 4.411815402169753, + "learning_rate": 1.881766323489877e-06, + "loss": 0.9693, + "num_input_tokens_seen": 101158990, + "step": 7096 + }, + { + "epoch": 0.5333684052307229, + "grad_norm": 1.536880030117669, + "learning_rate": 1.8812803362600865e-06, + "loss": 0.9156, + "num_input_tokens_seen": 101184270, + "step": 7097 + }, + { + "epoch": 0.533443559296558, + "grad_norm": 1.93607099194451, + "learning_rate": 1.8807943560648903e-06, + "loss": 0.9035, + "num_input_tokens_seen": 101203890, + "step": 7098 + }, + { + "epoch": 0.5335187133623929, + "grad_norm": 1.4793613287838352, + "learning_rate": 1.8803083829330853e-06, + "loss": 0.8859, + "num_input_tokens_seen": 101228535, + "step": 7099 + }, + { + "epoch": 0.5335938674282279, + "grad_norm": 1.541855113222474, + "learning_rate": 1.8798224168934664e-06, + "loss": 0.9749, + "num_input_tokens_seen": 101254330, + "step": 7100 + }, + { + "epoch": 0.5336690214940628, + "grad_norm": 1.4570836934732907, + "learning_rate": 1.87933645797483e-06, + "loss": 0.862, + "num_input_tokens_seen": 101275390, + "step": 7101 + }, + { + "epoch": 0.5337441755598978, + "grad_norm": 1.4315595200529896, + "learning_rate": 1.8788505062059708e-06, + "loss": 0.9657, + "num_input_tokens_seen": 101298825, + "step": 7102 + }, + { + "epoch": 0.5338193296257328, + "grad_norm": 1.5488308188570126, + "learning_rate": 1.8783645616156822e-06, + "loss": 0.9373, + "num_input_tokens_seen": 101321365, + "step": 7103 + }, + { + "epoch": 0.5338944836915677, + "grad_norm": 1.6559131207734918, + "learning_rate": 1.8778786242327598e-06, + "loss": 0.9915, + "num_input_tokens_seen": 101344595, + "step": 7104 + }, + { + "epoch": 0.5339696377574027, + "grad_norm": 1.5089523671483576, + "learning_rate": 1.877392694085996e-06, + "loss": 0.9643, + "num_input_tokens_seen": 101369505, + "step": 7105 + }, + { + "epoch": 0.5340447918232376, + "grad_norm": 1.5273647453250454, + "learning_rate": 1.876906771204185e-06, + "loss": 1.0444, + "num_input_tokens_seen": 101393955, + "step": 7106 + }, + { + "epoch": 0.5341199458890726, + "grad_norm": 1.5022561272407164, + "learning_rate": 1.8764208556161192e-06, + "loss": 0.9406, + "num_input_tokens_seen": 101415010, + "step": 7107 + }, + { + "epoch": 0.5341950999549075, + "grad_norm": 1.7377798664641992, + "learning_rate": 1.87593494735059e-06, + "loss": 0.8849, + "num_input_tokens_seen": 101441255, + "step": 7108 + }, + { + "epoch": 0.5342702540207426, + "grad_norm": 2.554398453252212, + "learning_rate": 1.8754490464363917e-06, + "loss": 0.8589, + "num_input_tokens_seen": 101462005, + "step": 7109 + }, + { + "epoch": 0.5343454080865775, + "grad_norm": 1.554257203943064, + "learning_rate": 1.8749631529023129e-06, + "loss": 0.9491, + "num_input_tokens_seen": 101485310, + "step": 7110 + }, + { + "epoch": 0.5344205621524124, + "grad_norm": 2.122913233561137, + "learning_rate": 1.874477266777147e-06, + "loss": 0.9355, + "num_input_tokens_seen": 101503515, + "step": 7111 + }, + { + "epoch": 0.5344957162182474, + "grad_norm": 1.5768397167853303, + "learning_rate": 1.8739913880896835e-06, + "loss": 0.9505, + "num_input_tokens_seen": 101527445, + "step": 7112 + }, + { + "epoch": 0.5345708702840823, + "grad_norm": 1.4722895886744942, + "learning_rate": 1.8735055168687126e-06, + "loss": 0.9776, + "num_input_tokens_seen": 101552100, + "step": 7113 + }, + { + "epoch": 0.5346460243499174, + "grad_norm": 1.264371604518885, + "learning_rate": 1.8730196531430246e-06, + "loss": 0.9635, + "num_input_tokens_seen": 101578280, + "step": 7114 + }, + { + "epoch": 0.5347211784157523, + "grad_norm": 1.5761282346372945, + "learning_rate": 1.872533796941408e-06, + "loss": 0.9088, + "num_input_tokens_seen": 101602655, + "step": 7115 + }, + { + "epoch": 0.5347963324815872, + "grad_norm": 1.7349299324904293, + "learning_rate": 1.8720479482926523e-06, + "loss": 0.9945, + "num_input_tokens_seen": 101625070, + "step": 7116 + }, + { + "epoch": 0.5348714865474222, + "grad_norm": 1.5018049348068088, + "learning_rate": 1.8715621072255457e-06, + "loss": 0.9508, + "num_input_tokens_seen": 101649300, + "step": 7117 + }, + { + "epoch": 0.5349466406132571, + "grad_norm": 1.5560257102783885, + "learning_rate": 1.8710762737688757e-06, + "loss": 0.9612, + "num_input_tokens_seen": 101670920, + "step": 7118 + }, + { + "epoch": 0.5350217946790922, + "grad_norm": 1.270592630239544, + "learning_rate": 1.8705904479514305e-06, + "loss": 0.9507, + "num_input_tokens_seen": 101695940, + "step": 7119 + }, + { + "epoch": 0.5350969487449271, + "grad_norm": 1.5959683854980196, + "learning_rate": 1.8701046298019965e-06, + "loss": 1.0105, + "num_input_tokens_seen": 101717915, + "step": 7120 + }, + { + "epoch": 0.5351721028107621, + "grad_norm": 1.6924549402186193, + "learning_rate": 1.869618819349361e-06, + "loss": 0.9442, + "num_input_tokens_seen": 101738830, + "step": 7121 + }, + { + "epoch": 0.535247256876597, + "grad_norm": 1.5878801146062056, + "learning_rate": 1.8691330166223091e-06, + "loss": 0.9144, + "num_input_tokens_seen": 101762025, + "step": 7122 + }, + { + "epoch": 0.5353224109424319, + "grad_norm": 2.0094547902545976, + "learning_rate": 1.8686472216496275e-06, + "loss": 0.904, + "num_input_tokens_seen": 101785470, + "step": 7123 + }, + { + "epoch": 0.535397565008267, + "grad_norm": 1.4293951350546565, + "learning_rate": 1.8681614344601013e-06, + "loss": 0.9894, + "num_input_tokens_seen": 101807285, + "step": 7124 + }, + { + "epoch": 0.5354727190741019, + "grad_norm": 1.5811427505972107, + "learning_rate": 1.8676756550825144e-06, + "loss": 0.9324, + "num_input_tokens_seen": 101830420, + "step": 7125 + }, + { + "epoch": 0.5355478731399369, + "grad_norm": 1.4868710061557233, + "learning_rate": 1.8671898835456518e-06, + "loss": 0.9389, + "num_input_tokens_seen": 101854500, + "step": 7126 + }, + { + "epoch": 0.5356230272057718, + "grad_norm": 1.44125464657831, + "learning_rate": 1.8667041198782972e-06, + "loss": 0.9553, + "num_input_tokens_seen": 101878310, + "step": 7127 + }, + { + "epoch": 0.5356981812716068, + "grad_norm": 1.53596347231943, + "learning_rate": 1.866218364109234e-06, + "loss": 0.9939, + "num_input_tokens_seen": 101900930, + "step": 7128 + }, + { + "epoch": 0.5357733353374418, + "grad_norm": 1.28109088933211, + "learning_rate": 1.8657326162672452e-06, + "loss": 0.9764, + "num_input_tokens_seen": 101924185, + "step": 7129 + }, + { + "epoch": 0.5358484894032767, + "grad_norm": 1.3929732138090984, + "learning_rate": 1.865246876381112e-06, + "loss": 0.9384, + "num_input_tokens_seen": 101946975, + "step": 7130 + }, + { + "epoch": 0.5359236434691117, + "grad_norm": 1.4910408127825678, + "learning_rate": 1.8647611444796182e-06, + "loss": 0.7907, + "num_input_tokens_seen": 101973980, + "step": 7131 + }, + { + "epoch": 0.5359987975349466, + "grad_norm": 1.566704700224742, + "learning_rate": 1.8642754205915444e-06, + "loss": 1.0264, + "num_input_tokens_seen": 101996095, + "step": 7132 + }, + { + "epoch": 0.5360739516007816, + "grad_norm": 1.7532821938390037, + "learning_rate": 1.8637897047456717e-06, + "loss": 0.9515, + "num_input_tokens_seen": 102018645, + "step": 7133 + }, + { + "epoch": 0.5361491056666166, + "grad_norm": 1.4297852196155048, + "learning_rate": 1.8633039969707808e-06, + "loss": 0.9482, + "num_input_tokens_seen": 102041345, + "step": 7134 + }, + { + "epoch": 0.5362242597324516, + "grad_norm": 1.6232468611514883, + "learning_rate": 1.8628182972956509e-06, + "loss": 0.9352, + "num_input_tokens_seen": 102065550, + "step": 7135 + }, + { + "epoch": 0.5362994137982865, + "grad_norm": 1.448704877781011, + "learning_rate": 1.8623326057490627e-06, + "loss": 0.9342, + "num_input_tokens_seen": 102089085, + "step": 7136 + }, + { + "epoch": 0.5363745678641214, + "grad_norm": 1.34010499236873, + "learning_rate": 1.8618469223597943e-06, + "loss": 1.0753, + "num_input_tokens_seen": 102112865, + "step": 7137 + }, + { + "epoch": 0.5364497219299564, + "grad_norm": 1.7541146951015396, + "learning_rate": 1.8613612471566253e-06, + "loss": 1.0123, + "num_input_tokens_seen": 102136145, + "step": 7138 + }, + { + "epoch": 0.5365248759957914, + "grad_norm": 1.4721978483441651, + "learning_rate": 1.8608755801683334e-06, + "loss": 0.9408, + "num_input_tokens_seen": 102162515, + "step": 7139 + }, + { + "epoch": 0.5366000300616264, + "grad_norm": 1.6299164564569857, + "learning_rate": 1.8603899214236956e-06, + "loss": 0.9645, + "num_input_tokens_seen": 102185915, + "step": 7140 + }, + { + "epoch": 0.5366751841274613, + "grad_norm": 1.615637953686019, + "learning_rate": 1.85990427095149e-06, + "loss": 0.9473, + "num_input_tokens_seen": 102208825, + "step": 7141 + }, + { + "epoch": 0.5367503381932962, + "grad_norm": 1.49619016012998, + "learning_rate": 1.8594186287804923e-06, + "loss": 0.9651, + "num_input_tokens_seen": 102231230, + "step": 7142 + }, + { + "epoch": 0.5368254922591312, + "grad_norm": 1.6637945541503645, + "learning_rate": 1.8589329949394793e-06, + "loss": 0.9475, + "num_input_tokens_seen": 102253195, + "step": 7143 + }, + { + "epoch": 0.5369006463249661, + "grad_norm": 0.80596577861351, + "learning_rate": 1.8584473694572268e-06, + "loss": 0.7979, + "num_input_tokens_seen": 102320590, + "step": 7144 + }, + { + "epoch": 0.5369758003908012, + "grad_norm": 1.7324470595601087, + "learning_rate": 1.8579617523625096e-06, + "loss": 0.9333, + "num_input_tokens_seen": 102340410, + "step": 7145 + }, + { + "epoch": 0.5370509544566361, + "grad_norm": 1.7289879105180022, + "learning_rate": 1.8574761436841027e-06, + "loss": 0.9249, + "num_input_tokens_seen": 102362315, + "step": 7146 + }, + { + "epoch": 0.5371261085224711, + "grad_norm": 1.4542696146698864, + "learning_rate": 1.8569905434507796e-06, + "loss": 1.0335, + "num_input_tokens_seen": 102385890, + "step": 7147 + }, + { + "epoch": 0.537201262588306, + "grad_norm": 1.8406266087782457, + "learning_rate": 1.8565049516913146e-06, + "loss": 0.8374, + "num_input_tokens_seen": 102410795, + "step": 7148 + }, + { + "epoch": 0.5372764166541409, + "grad_norm": 1.5668386114739188, + "learning_rate": 1.85601936843448e-06, + "loss": 0.7974, + "num_input_tokens_seen": 102436405, + "step": 7149 + }, + { + "epoch": 0.537351570719976, + "grad_norm": 1.7615815170531124, + "learning_rate": 1.8555337937090506e-06, + "loss": 1.004, + "num_input_tokens_seen": 102458645, + "step": 7150 + }, + { + "epoch": 0.5374267247858109, + "grad_norm": 1.4999645546857834, + "learning_rate": 1.8550482275437964e-06, + "loss": 0.9293, + "num_input_tokens_seen": 102483885, + "step": 7151 + }, + { + "epoch": 0.5375018788516459, + "grad_norm": 1.661786309121187, + "learning_rate": 1.854562669967489e-06, + "loss": 1.0126, + "num_input_tokens_seen": 102506220, + "step": 7152 + }, + { + "epoch": 0.5375770329174808, + "grad_norm": 1.3498818858010164, + "learning_rate": 1.8540771210089016e-06, + "loss": 0.8059, + "num_input_tokens_seen": 102532135, + "step": 7153 + }, + { + "epoch": 0.5376521869833158, + "grad_norm": 1.4602149359684908, + "learning_rate": 1.8535915806968026e-06, + "loss": 0.9907, + "num_input_tokens_seen": 102556020, + "step": 7154 + }, + { + "epoch": 0.5377273410491508, + "grad_norm": 1.905495165852541, + "learning_rate": 1.8531060490599637e-06, + "loss": 1.1111, + "num_input_tokens_seen": 102574370, + "step": 7155 + }, + { + "epoch": 0.5378024951149857, + "grad_norm": 1.5368946124140583, + "learning_rate": 1.8526205261271538e-06, + "loss": 0.9264, + "num_input_tokens_seen": 102595370, + "step": 7156 + }, + { + "epoch": 0.5378776491808207, + "grad_norm": 1.5588066701368972, + "learning_rate": 1.8521350119271418e-06, + "loss": 0.8903, + "num_input_tokens_seen": 102619515, + "step": 7157 + }, + { + "epoch": 0.5379528032466556, + "grad_norm": 1.7521791514331635, + "learning_rate": 1.8516495064886967e-06, + "loss": 0.9011, + "num_input_tokens_seen": 102643645, + "step": 7158 + }, + { + "epoch": 0.5380279573124906, + "grad_norm": 1.5441189498789016, + "learning_rate": 1.8511640098405863e-06, + "loss": 0.9234, + "num_input_tokens_seen": 102671225, + "step": 7159 + }, + { + "epoch": 0.5381031113783256, + "grad_norm": 2.036665519718195, + "learning_rate": 1.8506785220115787e-06, + "loss": 0.9025, + "num_input_tokens_seen": 102697090, + "step": 7160 + }, + { + "epoch": 0.5381782654441605, + "grad_norm": 1.8231556181744484, + "learning_rate": 1.8501930430304402e-06, + "loss": 1.0024, + "num_input_tokens_seen": 102722480, + "step": 7161 + }, + { + "epoch": 0.5382534195099955, + "grad_norm": 1.4494699430109403, + "learning_rate": 1.8497075729259372e-06, + "loss": 1.0471, + "num_input_tokens_seen": 102745630, + "step": 7162 + }, + { + "epoch": 0.5383285735758304, + "grad_norm": 0.7925005126975534, + "learning_rate": 1.8492221117268367e-06, + "loss": 0.8242, + "num_input_tokens_seen": 102820720, + "step": 7163 + }, + { + "epoch": 0.5384037276416654, + "grad_norm": 1.6026164982862319, + "learning_rate": 1.8487366594619028e-06, + "loss": 0.9525, + "num_input_tokens_seen": 102848145, + "step": 7164 + }, + { + "epoch": 0.5384788817075004, + "grad_norm": 1.3870324867915975, + "learning_rate": 1.8482512161599016e-06, + "loss": 0.8967, + "num_input_tokens_seen": 102873030, + "step": 7165 + }, + { + "epoch": 0.5385540357733354, + "grad_norm": 1.485253164629927, + "learning_rate": 1.8477657818495963e-06, + "loss": 0.9565, + "num_input_tokens_seen": 102897150, + "step": 7166 + }, + { + "epoch": 0.5386291898391703, + "grad_norm": 1.608700583172468, + "learning_rate": 1.847280356559752e-06, + "loss": 0.9144, + "num_input_tokens_seen": 102921315, + "step": 7167 + }, + { + "epoch": 0.5387043439050052, + "grad_norm": 1.6829079206893771, + "learning_rate": 1.8467949403191312e-06, + "loss": 0.9549, + "num_input_tokens_seen": 102944210, + "step": 7168 + }, + { + "epoch": 0.5387794979708402, + "grad_norm": 1.4855257091473852, + "learning_rate": 1.8463095331564965e-06, + "loss": 1.0198, + "num_input_tokens_seen": 102968810, + "step": 7169 + }, + { + "epoch": 0.5388546520366752, + "grad_norm": 1.5412546414864627, + "learning_rate": 1.8458241351006107e-06, + "loss": 0.8577, + "num_input_tokens_seen": 102993915, + "step": 7170 + }, + { + "epoch": 0.5389298061025102, + "grad_norm": 1.7881520049193647, + "learning_rate": 1.8453387461802347e-06, + "loss": 0.9846, + "num_input_tokens_seen": 103016525, + "step": 7171 + }, + { + "epoch": 0.5390049601683451, + "grad_norm": 1.4600720621640688, + "learning_rate": 1.8448533664241316e-06, + "loss": 0.8839, + "num_input_tokens_seen": 103041975, + "step": 7172 + }, + { + "epoch": 0.5390801142341801, + "grad_norm": 1.4359001711039465, + "learning_rate": 1.84436799586106e-06, + "loss": 0.9357, + "num_input_tokens_seen": 103065915, + "step": 7173 + }, + { + "epoch": 0.539155268300015, + "grad_norm": 1.828254955921944, + "learning_rate": 1.8438826345197796e-06, + "loss": 1.0271, + "num_input_tokens_seen": 103087980, + "step": 7174 + }, + { + "epoch": 0.53923042236585, + "grad_norm": 1.5567504384591713, + "learning_rate": 1.843397282429052e-06, + "loss": 0.9925, + "num_input_tokens_seen": 103109585, + "step": 7175 + }, + { + "epoch": 0.539305576431685, + "grad_norm": 1.260977181681192, + "learning_rate": 1.8429119396176348e-06, + "loss": 1.0065, + "num_input_tokens_seen": 103133905, + "step": 7176 + }, + { + "epoch": 0.5393807304975199, + "grad_norm": 1.6440943049138992, + "learning_rate": 1.8424266061142869e-06, + "loss": 0.9604, + "num_input_tokens_seen": 103155750, + "step": 7177 + }, + { + "epoch": 0.5394558845633549, + "grad_norm": 1.9991910013506944, + "learning_rate": 1.841941281947766e-06, + "loss": 0.971, + "num_input_tokens_seen": 103174920, + "step": 7178 + }, + { + "epoch": 0.5395310386291898, + "grad_norm": 2.1609912658754977, + "learning_rate": 1.8414559671468288e-06, + "loss": 0.9466, + "num_input_tokens_seen": 103196615, + "step": 7179 + }, + { + "epoch": 0.5396061926950249, + "grad_norm": 1.3776166899700177, + "learning_rate": 1.8409706617402333e-06, + "loss": 0.9817, + "num_input_tokens_seen": 103219535, + "step": 7180 + }, + { + "epoch": 0.5396813467608598, + "grad_norm": 1.435787009571067, + "learning_rate": 1.8404853657567347e-06, + "loss": 0.9608, + "num_input_tokens_seen": 103248390, + "step": 7181 + }, + { + "epoch": 0.5397565008266947, + "grad_norm": 1.2287134153831962, + "learning_rate": 1.8400000792250894e-06, + "loss": 0.9855, + "num_input_tokens_seen": 103271480, + "step": 7182 + }, + { + "epoch": 0.5398316548925297, + "grad_norm": 1.7026510016518075, + "learning_rate": 1.8395148021740518e-06, + "loss": 0.9695, + "num_input_tokens_seen": 103293860, + "step": 7183 + }, + { + "epoch": 0.5399068089583646, + "grad_norm": 1.8043423937820182, + "learning_rate": 1.8390295346323765e-06, + "loss": 0.8706, + "num_input_tokens_seen": 103312345, + "step": 7184 + }, + { + "epoch": 0.5399819630241997, + "grad_norm": 1.522602732908401, + "learning_rate": 1.8385442766288181e-06, + "loss": 0.9527, + "num_input_tokens_seen": 103335015, + "step": 7185 + }, + { + "epoch": 0.5400571170900346, + "grad_norm": 1.5552371792972313, + "learning_rate": 1.8380590281921294e-06, + "loss": 0.9922, + "num_input_tokens_seen": 103360920, + "step": 7186 + }, + { + "epoch": 0.5401322711558695, + "grad_norm": 1.4712045329027732, + "learning_rate": 1.8375737893510635e-06, + "loss": 1.0248, + "num_input_tokens_seen": 103384065, + "step": 7187 + }, + { + "epoch": 0.5402074252217045, + "grad_norm": 1.5795117319123588, + "learning_rate": 1.837088560134372e-06, + "loss": 0.9485, + "num_input_tokens_seen": 103405775, + "step": 7188 + }, + { + "epoch": 0.5402825792875394, + "grad_norm": 1.5231994442910695, + "learning_rate": 1.8366033405708076e-06, + "loss": 0.8574, + "num_input_tokens_seen": 103427895, + "step": 7189 + }, + { + "epoch": 0.5403577333533744, + "grad_norm": 0.8267209368114283, + "learning_rate": 1.8361181306891214e-06, + "loss": 0.8038, + "num_input_tokens_seen": 103509455, + "step": 7190 + }, + { + "epoch": 0.5404328874192094, + "grad_norm": 1.5495406793090318, + "learning_rate": 1.8356329305180626e-06, + "loss": 0.9789, + "num_input_tokens_seen": 103532975, + "step": 7191 + }, + { + "epoch": 0.5405080414850444, + "grad_norm": 1.5769501647089892, + "learning_rate": 1.835147740086383e-06, + "loss": 0.965, + "num_input_tokens_seen": 103555825, + "step": 7192 + }, + { + "epoch": 0.5405831955508793, + "grad_norm": 2.1310645685604026, + "learning_rate": 1.8346625594228295e-06, + "loss": 0.9551, + "num_input_tokens_seen": 103579110, + "step": 7193 + }, + { + "epoch": 0.5406583496167142, + "grad_norm": 0.73612074625759, + "learning_rate": 1.8341773885561539e-06, + "loss": 0.7655, + "num_input_tokens_seen": 103657030, + "step": 7194 + }, + { + "epoch": 0.5407335036825492, + "grad_norm": 1.469947570203141, + "learning_rate": 1.8336922275151032e-06, + "loss": 0.9792, + "num_input_tokens_seen": 103682925, + "step": 7195 + }, + { + "epoch": 0.5408086577483842, + "grad_norm": 1.6002595392964014, + "learning_rate": 1.8332070763284236e-06, + "loss": 1.0187, + "num_input_tokens_seen": 103706345, + "step": 7196 + }, + { + "epoch": 0.5408838118142192, + "grad_norm": 1.3155948390810752, + "learning_rate": 1.8327219350248643e-06, + "loss": 1.027, + "num_input_tokens_seen": 103731665, + "step": 7197 + }, + { + "epoch": 0.5409589658800541, + "grad_norm": 1.7419406733127942, + "learning_rate": 1.8322368036331705e-06, + "loss": 0.9433, + "num_input_tokens_seen": 103754365, + "step": 7198 + }, + { + "epoch": 0.5410341199458891, + "grad_norm": 1.4009894842288213, + "learning_rate": 1.8317516821820888e-06, + "loss": 0.9821, + "num_input_tokens_seen": 103777020, + "step": 7199 + }, + { + "epoch": 0.541109274011724, + "grad_norm": 1.5202090754521755, + "learning_rate": 1.8312665707003643e-06, + "loss": 0.8838, + "num_input_tokens_seen": 103801810, + "step": 7200 + }, + { + "epoch": 0.541184428077559, + "grad_norm": 1.6950002637654777, + "learning_rate": 1.8307814692167412e-06, + "loss": 0.7731, + "num_input_tokens_seen": 103827805, + "step": 7201 + }, + { + "epoch": 0.541259582143394, + "grad_norm": 1.3479192946976872, + "learning_rate": 1.8302963777599645e-06, + "loss": 1.014, + "num_input_tokens_seen": 103851395, + "step": 7202 + }, + { + "epoch": 0.5413347362092289, + "grad_norm": 0.7033455396861568, + "learning_rate": 1.8298112963587766e-06, + "loss": 0.8006, + "num_input_tokens_seen": 103933820, + "step": 7203 + }, + { + "epoch": 0.5414098902750639, + "grad_norm": 1.3896072177845555, + "learning_rate": 1.8293262250419217e-06, + "loss": 0.9147, + "num_input_tokens_seen": 103957005, + "step": 7204 + }, + { + "epoch": 0.5414850443408988, + "grad_norm": 1.503807480776546, + "learning_rate": 1.8288411638381415e-06, + "loss": 0.9454, + "num_input_tokens_seen": 103981035, + "step": 7205 + }, + { + "epoch": 0.5415601984067338, + "grad_norm": 1.7522455522130007, + "learning_rate": 1.8283561127761773e-06, + "loss": 0.9809, + "num_input_tokens_seen": 104003765, + "step": 7206 + }, + { + "epoch": 0.5416353524725688, + "grad_norm": 1.2628610831102425, + "learning_rate": 1.8278710718847711e-06, + "loss": 0.9092, + "num_input_tokens_seen": 104029345, + "step": 7207 + }, + { + "epoch": 0.5417105065384037, + "grad_norm": 1.6470985217479366, + "learning_rate": 1.8273860411926627e-06, + "loss": 0.9358, + "num_input_tokens_seen": 104052215, + "step": 7208 + }, + { + "epoch": 0.5417856606042387, + "grad_norm": 1.5440137969818748, + "learning_rate": 1.8269010207285927e-06, + "loss": 0.9151, + "num_input_tokens_seen": 104077360, + "step": 7209 + }, + { + "epoch": 0.5418608146700736, + "grad_norm": 1.4274689183634357, + "learning_rate": 1.8264160105212995e-06, + "loss": 0.9177, + "num_input_tokens_seen": 104101660, + "step": 7210 + }, + { + "epoch": 0.5419359687359087, + "grad_norm": 1.4455098610234522, + "learning_rate": 1.825931010599523e-06, + "loss": 1.0341, + "num_input_tokens_seen": 104126140, + "step": 7211 + }, + { + "epoch": 0.5420111228017436, + "grad_norm": 1.6997593513026183, + "learning_rate": 1.8254460209920007e-06, + "loss": 0.9692, + "num_input_tokens_seen": 104148895, + "step": 7212 + }, + { + "epoch": 0.5420862768675785, + "grad_norm": 1.4969847891728534, + "learning_rate": 1.8249610417274695e-06, + "loss": 0.894, + "num_input_tokens_seen": 104170565, + "step": 7213 + }, + { + "epoch": 0.5421614309334135, + "grad_norm": 1.6210261687014396, + "learning_rate": 1.8244760728346674e-06, + "loss": 0.991, + "num_input_tokens_seen": 104192470, + "step": 7214 + }, + { + "epoch": 0.5422365849992484, + "grad_norm": 1.5654294590464977, + "learning_rate": 1.823991114342329e-06, + "loss": 0.9397, + "num_input_tokens_seen": 104215515, + "step": 7215 + }, + { + "epoch": 0.5423117390650835, + "grad_norm": 1.7162269106184878, + "learning_rate": 1.823506166279192e-06, + "loss": 0.787, + "num_input_tokens_seen": 104239640, + "step": 7216 + }, + { + "epoch": 0.5423868931309184, + "grad_norm": 1.2674637612122597, + "learning_rate": 1.823021228673991e-06, + "loss": 0.9102, + "num_input_tokens_seen": 104264295, + "step": 7217 + }, + { + "epoch": 0.5424620471967534, + "grad_norm": 1.6977486400506117, + "learning_rate": 1.8225363015554586e-06, + "loss": 0.8689, + "num_input_tokens_seen": 104286800, + "step": 7218 + }, + { + "epoch": 0.5425372012625883, + "grad_norm": 2.0945227036947403, + "learning_rate": 1.822051384952331e-06, + "loss": 0.9502, + "num_input_tokens_seen": 104307645, + "step": 7219 + }, + { + "epoch": 0.5426123553284232, + "grad_norm": 1.9027580421086339, + "learning_rate": 1.8215664788933394e-06, + "loss": 0.9875, + "num_input_tokens_seen": 104330600, + "step": 7220 + }, + { + "epoch": 0.5426875093942582, + "grad_norm": 0.8306721435555625, + "learning_rate": 1.8210815834072177e-06, + "loss": 0.8138, + "num_input_tokens_seen": 104401755, + "step": 7221 + }, + { + "epoch": 0.5427626634600932, + "grad_norm": 1.9109080576714457, + "learning_rate": 1.8205966985226975e-06, + "loss": 0.9543, + "num_input_tokens_seen": 104421635, + "step": 7222 + }, + { + "epoch": 0.5428378175259282, + "grad_norm": 1.4937563333416992, + "learning_rate": 1.8201118242685093e-06, + "loss": 0.9784, + "num_input_tokens_seen": 104444565, + "step": 7223 + }, + { + "epoch": 0.5429129715917631, + "grad_norm": 1.4285285296129766, + "learning_rate": 1.819626960673385e-06, + "loss": 0.8903, + "num_input_tokens_seen": 104467135, + "step": 7224 + }, + { + "epoch": 0.5429881256575981, + "grad_norm": 0.7088016242950506, + "learning_rate": 1.8191421077660535e-06, + "loss": 0.7985, + "num_input_tokens_seen": 104547345, + "step": 7225 + }, + { + "epoch": 0.543063279723433, + "grad_norm": 1.698719451158041, + "learning_rate": 1.8186572655752448e-06, + "loss": 0.9572, + "num_input_tokens_seen": 104570615, + "step": 7226 + }, + { + "epoch": 0.543138433789268, + "grad_norm": 1.3543416814926252, + "learning_rate": 1.8181724341296877e-06, + "loss": 0.8759, + "num_input_tokens_seen": 104593850, + "step": 7227 + }, + { + "epoch": 0.543213587855103, + "grad_norm": 1.8071179558402353, + "learning_rate": 1.8176876134581098e-06, + "loss": 1.0162, + "num_input_tokens_seen": 104611825, + "step": 7228 + }, + { + "epoch": 0.5432887419209379, + "grad_norm": 2.1076313865677525, + "learning_rate": 1.8172028035892394e-06, + "loss": 0.9241, + "num_input_tokens_seen": 104633320, + "step": 7229 + }, + { + "epoch": 0.5433638959867729, + "grad_norm": 1.340734113685543, + "learning_rate": 1.816718004551802e-06, + "loss": 0.9415, + "num_input_tokens_seen": 104656185, + "step": 7230 + }, + { + "epoch": 0.5434390500526078, + "grad_norm": 1.5607809910481394, + "learning_rate": 1.8162332163745254e-06, + "loss": 0.8481, + "num_input_tokens_seen": 104684640, + "step": 7231 + }, + { + "epoch": 0.5435142041184428, + "grad_norm": 0.8429076263639386, + "learning_rate": 1.8157484390861342e-06, + "loss": 0.8516, + "num_input_tokens_seen": 104752050, + "step": 7232 + }, + { + "epoch": 0.5435893581842778, + "grad_norm": 1.91573456891145, + "learning_rate": 1.8152636727153536e-06, + "loss": 0.9535, + "num_input_tokens_seen": 104773340, + "step": 7233 + }, + { + "epoch": 0.5436645122501127, + "grad_norm": 1.6148095085304435, + "learning_rate": 1.814778917290908e-06, + "loss": 0.9477, + "num_input_tokens_seen": 104797200, + "step": 7234 + }, + { + "epoch": 0.5437396663159477, + "grad_norm": 0.8680861344779467, + "learning_rate": 1.8142941728415204e-06, + "loss": 0.8577, + "num_input_tokens_seen": 104876595, + "step": 7235 + }, + { + "epoch": 0.5438148203817826, + "grad_norm": 1.635792545283614, + "learning_rate": 1.8138094393959144e-06, + "loss": 0.9093, + "num_input_tokens_seen": 104901420, + "step": 7236 + }, + { + "epoch": 0.5438899744476177, + "grad_norm": 1.6096817761221125, + "learning_rate": 1.8133247169828114e-06, + "loss": 0.9507, + "num_input_tokens_seen": 104925220, + "step": 7237 + }, + { + "epoch": 0.5439651285134526, + "grad_norm": 1.5954715929649546, + "learning_rate": 1.8128400056309345e-06, + "loss": 0.897, + "num_input_tokens_seen": 104952515, + "step": 7238 + }, + { + "epoch": 0.5440402825792875, + "grad_norm": 1.2669411867878553, + "learning_rate": 1.8123553053690046e-06, + "loss": 0.9764, + "num_input_tokens_seen": 104976130, + "step": 7239 + }, + { + "epoch": 0.5441154366451225, + "grad_norm": 1.8420349161440184, + "learning_rate": 1.81187061622574e-06, + "loss": 0.8578, + "num_input_tokens_seen": 105004285, + "step": 7240 + }, + { + "epoch": 0.5441905907109574, + "grad_norm": 1.3839873156407292, + "learning_rate": 1.8113859382298627e-06, + "loss": 0.9456, + "num_input_tokens_seen": 105028465, + "step": 7241 + }, + { + "epoch": 0.5442657447767925, + "grad_norm": 1.5564078009698878, + "learning_rate": 1.81090127141009e-06, + "loss": 0.8732, + "num_input_tokens_seen": 105050905, + "step": 7242 + }, + { + "epoch": 0.5443408988426274, + "grad_norm": 1.7500753386493069, + "learning_rate": 1.8104166157951419e-06, + "loss": 0.9699, + "num_input_tokens_seen": 105072270, + "step": 7243 + }, + { + "epoch": 0.5444160529084624, + "grad_norm": 1.4213961958364494, + "learning_rate": 1.809931971413735e-06, + "loss": 1.0463, + "num_input_tokens_seen": 105095715, + "step": 7244 + }, + { + "epoch": 0.5444912069742973, + "grad_norm": 1.7106647709553284, + "learning_rate": 1.8094473382945866e-06, + "loss": 0.9891, + "num_input_tokens_seen": 105118875, + "step": 7245 + }, + { + "epoch": 0.5445663610401322, + "grad_norm": 1.4710303926524353, + "learning_rate": 1.8089627164664132e-06, + "loss": 0.9644, + "num_input_tokens_seen": 105141140, + "step": 7246 + }, + { + "epoch": 0.5446415151059673, + "grad_norm": 1.705001147652964, + "learning_rate": 1.80847810595793e-06, + "loss": 0.8413, + "num_input_tokens_seen": 105163120, + "step": 7247 + }, + { + "epoch": 0.5447166691718022, + "grad_norm": 1.5400599621580442, + "learning_rate": 1.8079935067978528e-06, + "loss": 0.9021, + "num_input_tokens_seen": 105185035, + "step": 7248 + }, + { + "epoch": 0.5447918232376372, + "grad_norm": 1.5293722511706418, + "learning_rate": 1.8075089190148956e-06, + "loss": 0.8587, + "num_input_tokens_seen": 105209700, + "step": 7249 + }, + { + "epoch": 0.5448669773034721, + "grad_norm": 1.6432745474359214, + "learning_rate": 1.8070243426377716e-06, + "loss": 0.914, + "num_input_tokens_seen": 105232555, + "step": 7250 + }, + { + "epoch": 0.544942131369307, + "grad_norm": 2.3566483664262976, + "learning_rate": 1.8065397776951946e-06, + "loss": 0.9017, + "num_input_tokens_seen": 105262560, + "step": 7251 + }, + { + "epoch": 0.545017285435142, + "grad_norm": 1.6026800954851705, + "learning_rate": 1.8060552242158765e-06, + "loss": 0.9603, + "num_input_tokens_seen": 105285880, + "step": 7252 + }, + { + "epoch": 0.545092439500977, + "grad_norm": 1.591574436864275, + "learning_rate": 1.8055706822285291e-06, + "loss": 0.9906, + "num_input_tokens_seen": 105307765, + "step": 7253 + }, + { + "epoch": 0.545167593566812, + "grad_norm": 1.6128541690534004, + "learning_rate": 1.8050861517618629e-06, + "loss": 0.9392, + "num_input_tokens_seen": 105328945, + "step": 7254 + }, + { + "epoch": 0.5452427476326469, + "grad_norm": 0.8555596708292335, + "learning_rate": 1.8046016328445893e-06, + "loss": 0.8306, + "num_input_tokens_seen": 105407165, + "step": 7255 + }, + { + "epoch": 0.5453179016984819, + "grad_norm": 1.6810435654068505, + "learning_rate": 1.804117125505417e-06, + "loss": 0.8921, + "num_input_tokens_seen": 105431335, + "step": 7256 + }, + { + "epoch": 0.5453930557643168, + "grad_norm": 1.4571414527432167, + "learning_rate": 1.803632629773054e-06, + "loss": 0.9214, + "num_input_tokens_seen": 105456315, + "step": 7257 + }, + { + "epoch": 0.5454682098301518, + "grad_norm": 1.4443342867848918, + "learning_rate": 1.8031481456762112e-06, + "loss": 1.0131, + "num_input_tokens_seen": 105483850, + "step": 7258 + }, + { + "epoch": 0.5455433638959868, + "grad_norm": 2.260507948009455, + "learning_rate": 1.802663673243593e-06, + "loss": 0.9541, + "num_input_tokens_seen": 105506805, + "step": 7259 + }, + { + "epoch": 0.5456185179618217, + "grad_norm": 1.703341339090026, + "learning_rate": 1.802179212503909e-06, + "loss": 0.9515, + "num_input_tokens_seen": 105531015, + "step": 7260 + }, + { + "epoch": 0.5456936720276567, + "grad_norm": 1.6699963290922284, + "learning_rate": 1.801694763485864e-06, + "loss": 0.889, + "num_input_tokens_seen": 105554865, + "step": 7261 + }, + { + "epoch": 0.5457688260934916, + "grad_norm": 1.8833304678070864, + "learning_rate": 1.8012103262181635e-06, + "loss": 1.0101, + "num_input_tokens_seen": 105578640, + "step": 7262 + }, + { + "epoch": 0.5458439801593267, + "grad_norm": 1.3359687756898457, + "learning_rate": 1.8007259007295125e-06, + "loss": 0.9382, + "num_input_tokens_seen": 105603920, + "step": 7263 + }, + { + "epoch": 0.5459191342251616, + "grad_norm": 1.9209222680588396, + "learning_rate": 1.8002414870486144e-06, + "loss": 0.9153, + "num_input_tokens_seen": 105621045, + "step": 7264 + }, + { + "epoch": 0.5459942882909965, + "grad_norm": 1.5833832415026488, + "learning_rate": 1.7997570852041739e-06, + "loss": 0.9396, + "num_input_tokens_seen": 105644505, + "step": 7265 + }, + { + "epoch": 0.5460694423568315, + "grad_norm": 1.673429817873065, + "learning_rate": 1.7992726952248926e-06, + "loss": 0.9083, + "num_input_tokens_seen": 105668130, + "step": 7266 + }, + { + "epoch": 0.5461445964226664, + "grad_norm": 1.7115795724855531, + "learning_rate": 1.7987883171394724e-06, + "loss": 0.9704, + "num_input_tokens_seen": 105692375, + "step": 7267 + }, + { + "epoch": 0.5462197504885015, + "grad_norm": 1.7406761559138006, + "learning_rate": 1.7983039509766156e-06, + "loss": 0.9225, + "num_input_tokens_seen": 105718715, + "step": 7268 + }, + { + "epoch": 0.5462949045543364, + "grad_norm": 1.997289788234097, + "learning_rate": 1.7978195967650214e-06, + "loss": 0.9843, + "num_input_tokens_seen": 105741085, + "step": 7269 + }, + { + "epoch": 0.5463700586201714, + "grad_norm": 1.7235875801873044, + "learning_rate": 1.7973352545333905e-06, + "loss": 0.8311, + "num_input_tokens_seen": 105762720, + "step": 7270 + }, + { + "epoch": 0.5464452126860063, + "grad_norm": 2.838837625082575, + "learning_rate": 1.796850924310422e-06, + "loss": 0.9579, + "num_input_tokens_seen": 105785640, + "step": 7271 + }, + { + "epoch": 0.5465203667518412, + "grad_norm": 1.6599231511447448, + "learning_rate": 1.796366606124814e-06, + "loss": 1.0554, + "num_input_tokens_seen": 105806810, + "step": 7272 + }, + { + "epoch": 0.5465955208176763, + "grad_norm": 1.4696546163090167, + "learning_rate": 1.7958823000052643e-06, + "loss": 0.9721, + "num_input_tokens_seen": 105832760, + "step": 7273 + }, + { + "epoch": 0.5466706748835112, + "grad_norm": 1.545525241321108, + "learning_rate": 1.79539800598047e-06, + "loss": 0.9718, + "num_input_tokens_seen": 105856520, + "step": 7274 + }, + { + "epoch": 0.5467458289493462, + "grad_norm": 1.5232420967181262, + "learning_rate": 1.7949137240791275e-06, + "loss": 1.0325, + "num_input_tokens_seen": 105880005, + "step": 7275 + }, + { + "epoch": 0.5468209830151811, + "grad_norm": 1.484332114152274, + "learning_rate": 1.7944294543299317e-06, + "loss": 0.9446, + "num_input_tokens_seen": 105904250, + "step": 7276 + }, + { + "epoch": 0.546896137081016, + "grad_norm": 1.5333337424457392, + "learning_rate": 1.7939451967615783e-06, + "loss": 0.8827, + "num_input_tokens_seen": 105927400, + "step": 7277 + }, + { + "epoch": 0.5469712911468511, + "grad_norm": 1.54306895256617, + "learning_rate": 1.793460951402761e-06, + "loss": 0.8801, + "num_input_tokens_seen": 105953410, + "step": 7278 + }, + { + "epoch": 0.547046445212686, + "grad_norm": 1.5979441637494145, + "learning_rate": 1.7929767182821724e-06, + "loss": 0.8804, + "num_input_tokens_seen": 105978325, + "step": 7279 + }, + { + "epoch": 0.547121599278521, + "grad_norm": 1.503195061904649, + "learning_rate": 1.7924924974285074e-06, + "loss": 0.9755, + "num_input_tokens_seen": 105999575, + "step": 7280 + }, + { + "epoch": 0.5471967533443559, + "grad_norm": 2.0171686220356846, + "learning_rate": 1.7920082888704553e-06, + "loss": 0.8946, + "num_input_tokens_seen": 106023095, + "step": 7281 + }, + { + "epoch": 0.5472719074101909, + "grad_norm": 1.7445977478063535, + "learning_rate": 1.7915240926367092e-06, + "loss": 0.9632, + "num_input_tokens_seen": 106048855, + "step": 7282 + }, + { + "epoch": 0.5473470614760259, + "grad_norm": 1.4233016517710517, + "learning_rate": 1.791039908755959e-06, + "loss": 0.9355, + "num_input_tokens_seen": 106070580, + "step": 7283 + }, + { + "epoch": 0.5474222155418608, + "grad_norm": 1.4470996426886957, + "learning_rate": 1.790555737256894e-06, + "loss": 0.8602, + "num_input_tokens_seen": 106093680, + "step": 7284 + }, + { + "epoch": 0.5474973696076958, + "grad_norm": 0.7403115104002033, + "learning_rate": 1.7900715781682039e-06, + "loss": 0.7731, + "num_input_tokens_seen": 106173125, + "step": 7285 + }, + { + "epoch": 0.5475725236735307, + "grad_norm": 1.854976227291808, + "learning_rate": 1.7895874315185763e-06, + "loss": 0.9671, + "num_input_tokens_seen": 106193275, + "step": 7286 + }, + { + "epoch": 0.5476476777393657, + "grad_norm": 1.7048836124106699, + "learning_rate": 1.7891032973366996e-06, + "loss": 0.9508, + "num_input_tokens_seen": 106218115, + "step": 7287 + }, + { + "epoch": 0.5477228318052006, + "grad_norm": 1.8157407792645213, + "learning_rate": 1.7886191756512598e-06, + "loss": 0.9089, + "num_input_tokens_seen": 106239765, + "step": 7288 + }, + { + "epoch": 0.5477979858710357, + "grad_norm": 1.8122053893486256, + "learning_rate": 1.788135066490943e-06, + "loss": 0.903, + "num_input_tokens_seen": 106264870, + "step": 7289 + }, + { + "epoch": 0.5478731399368706, + "grad_norm": 1.4858407262823947, + "learning_rate": 1.7876509698844356e-06, + "loss": 0.8753, + "num_input_tokens_seen": 106288095, + "step": 7290 + }, + { + "epoch": 0.5479482940027055, + "grad_norm": 2.141599335954155, + "learning_rate": 1.7871668858604206e-06, + "loss": 0.9273, + "num_input_tokens_seen": 106310365, + "step": 7291 + }, + { + "epoch": 0.5480234480685405, + "grad_norm": 1.5077366785148167, + "learning_rate": 1.786682814447583e-06, + "loss": 0.9628, + "num_input_tokens_seen": 106333460, + "step": 7292 + }, + { + "epoch": 0.5480986021343754, + "grad_norm": 1.6518187353891225, + "learning_rate": 1.7861987556746056e-06, + "loss": 0.9653, + "num_input_tokens_seen": 106357200, + "step": 7293 + }, + { + "epoch": 0.5481737562002105, + "grad_norm": 1.7467337508580747, + "learning_rate": 1.78571470957017e-06, + "loss": 0.906, + "num_input_tokens_seen": 106382525, + "step": 7294 + }, + { + "epoch": 0.5482489102660454, + "grad_norm": 1.4454556780854855, + "learning_rate": 1.7852306761629592e-06, + "loss": 0.9456, + "num_input_tokens_seen": 106406080, + "step": 7295 + }, + { + "epoch": 0.5483240643318803, + "grad_norm": 1.6894632857353415, + "learning_rate": 1.7847466554816526e-06, + "loss": 0.9472, + "num_input_tokens_seen": 106430325, + "step": 7296 + }, + { + "epoch": 0.5483992183977153, + "grad_norm": 1.6535350932690365, + "learning_rate": 1.7842626475549314e-06, + "loss": 1.029, + "num_input_tokens_seen": 106450770, + "step": 7297 + }, + { + "epoch": 0.5484743724635502, + "grad_norm": 1.8944478754004432, + "learning_rate": 1.783778652411474e-06, + "loss": 0.9011, + "num_input_tokens_seen": 106476695, + "step": 7298 + }, + { + "epoch": 0.5485495265293853, + "grad_norm": 1.6761873904603364, + "learning_rate": 1.7832946700799596e-06, + "loss": 1.0476, + "num_input_tokens_seen": 106498810, + "step": 7299 + }, + { + "epoch": 0.5486246805952202, + "grad_norm": 1.5259021091985847, + "learning_rate": 1.7828107005890663e-06, + "loss": 0.8571, + "num_input_tokens_seen": 106523275, + "step": 7300 + }, + { + "epoch": 0.5486998346610552, + "grad_norm": 1.8062693611348128, + "learning_rate": 1.7823267439674694e-06, + "loss": 1.0465, + "num_input_tokens_seen": 106547050, + "step": 7301 + }, + { + "epoch": 0.5487749887268901, + "grad_norm": 2.052427956807084, + "learning_rate": 1.7818428002438475e-06, + "loss": 1.0208, + "num_input_tokens_seen": 106568905, + "step": 7302 + }, + { + "epoch": 0.548850142792725, + "grad_norm": 0.7841801435533388, + "learning_rate": 1.7813588694468745e-06, + "loss": 0.8363, + "num_input_tokens_seen": 106642805, + "step": 7303 + }, + { + "epoch": 0.5489252968585601, + "grad_norm": 1.288156318043218, + "learning_rate": 1.780874951605226e-06, + "loss": 0.9125, + "num_input_tokens_seen": 106667655, + "step": 7304 + }, + { + "epoch": 0.549000450924395, + "grad_norm": 5.318063851732463, + "learning_rate": 1.7803910467475763e-06, + "loss": 0.9416, + "num_input_tokens_seen": 106692625, + "step": 7305 + }, + { + "epoch": 0.54907560499023, + "grad_norm": 1.935772648797697, + "learning_rate": 1.779907154902597e-06, + "loss": 0.8476, + "num_input_tokens_seen": 106717715, + "step": 7306 + }, + { + "epoch": 0.5491507590560649, + "grad_norm": 1.6974983011947917, + "learning_rate": 1.7794232760989623e-06, + "loss": 0.9627, + "num_input_tokens_seen": 106740215, + "step": 7307 + }, + { + "epoch": 0.5492259131219, + "grad_norm": 1.4341891764696133, + "learning_rate": 1.7789394103653425e-06, + "loss": 1.0344, + "num_input_tokens_seen": 106762380, + "step": 7308 + }, + { + "epoch": 0.5493010671877349, + "grad_norm": 0.9048110455897023, + "learning_rate": 1.7784555577304099e-06, + "loss": 0.7782, + "num_input_tokens_seen": 106843175, + "step": 7309 + }, + { + "epoch": 0.5493762212535698, + "grad_norm": 0.8213941436566239, + "learning_rate": 1.7779717182228335e-06, + "loss": 0.8349, + "num_input_tokens_seen": 106905595, + "step": 7310 + }, + { + "epoch": 0.5494513753194048, + "grad_norm": 2.640779479434564, + "learning_rate": 1.7774878918712828e-06, + "loss": 0.9641, + "num_input_tokens_seen": 106929565, + "step": 7311 + }, + { + "epoch": 0.5495265293852397, + "grad_norm": 1.3926498587217833, + "learning_rate": 1.777004078704427e-06, + "loss": 0.9241, + "num_input_tokens_seen": 106952455, + "step": 7312 + }, + { + "epoch": 0.5496016834510747, + "grad_norm": 2.1644034547447224, + "learning_rate": 1.7765202787509327e-06, + "loss": 0.8362, + "num_input_tokens_seen": 106973580, + "step": 7313 + }, + { + "epoch": 0.5496768375169097, + "grad_norm": 1.4961186897450085, + "learning_rate": 1.7760364920394684e-06, + "loss": 0.8292, + "num_input_tokens_seen": 106997510, + "step": 7314 + }, + { + "epoch": 0.5497519915827447, + "grad_norm": 1.4548869781905645, + "learning_rate": 1.7755527185986996e-06, + "loss": 0.9559, + "num_input_tokens_seen": 107021960, + "step": 7315 + }, + { + "epoch": 0.5498271456485796, + "grad_norm": 1.403939022473502, + "learning_rate": 1.775068958457291e-06, + "loss": 1.0307, + "num_input_tokens_seen": 107046080, + "step": 7316 + }, + { + "epoch": 0.5499022997144145, + "grad_norm": 1.486427815085495, + "learning_rate": 1.7745852116439087e-06, + "loss": 1.0683, + "num_input_tokens_seen": 107070415, + "step": 7317 + }, + { + "epoch": 0.5499774537802495, + "grad_norm": 1.314352272551885, + "learning_rate": 1.774101478187215e-06, + "loss": 0.9883, + "num_input_tokens_seen": 107095855, + "step": 7318 + }, + { + "epoch": 0.5500526078460845, + "grad_norm": 0.6907069231008075, + "learning_rate": 1.7736177581158742e-06, + "loss": 0.7574, + "num_input_tokens_seen": 107182805, + "step": 7319 + }, + { + "epoch": 0.5501277619119195, + "grad_norm": 1.6817971399289056, + "learning_rate": 1.7731340514585474e-06, + "loss": 0.8913, + "num_input_tokens_seen": 107206505, + "step": 7320 + }, + { + "epoch": 0.5502029159777544, + "grad_norm": 1.6157309934574486, + "learning_rate": 1.7726503582438982e-06, + "loss": 0.853, + "num_input_tokens_seen": 107231505, + "step": 7321 + }, + { + "epoch": 0.5502780700435893, + "grad_norm": 1.3271399042277634, + "learning_rate": 1.772166678500585e-06, + "loss": 0.9406, + "num_input_tokens_seen": 107260155, + "step": 7322 + }, + { + "epoch": 0.5503532241094243, + "grad_norm": 1.7895872187840562, + "learning_rate": 1.771683012257268e-06, + "loss": 1.0128, + "num_input_tokens_seen": 107280310, + "step": 7323 + }, + { + "epoch": 0.5504283781752592, + "grad_norm": 1.9536682594074568, + "learning_rate": 1.7711993595426076e-06, + "loss": 0.8515, + "num_input_tokens_seen": 107300595, + "step": 7324 + }, + { + "epoch": 0.5505035322410943, + "grad_norm": 1.313861277229418, + "learning_rate": 1.7707157203852608e-06, + "loss": 0.9659, + "num_input_tokens_seen": 107326640, + "step": 7325 + }, + { + "epoch": 0.5505786863069292, + "grad_norm": 1.5497323051140277, + "learning_rate": 1.770232094813886e-06, + "loss": 0.9414, + "num_input_tokens_seen": 107348970, + "step": 7326 + }, + { + "epoch": 0.5506538403727642, + "grad_norm": 1.8479472086495898, + "learning_rate": 1.7697484828571394e-06, + "loss": 0.9327, + "num_input_tokens_seen": 107374735, + "step": 7327 + }, + { + "epoch": 0.5507289944385991, + "grad_norm": 2.037055302868776, + "learning_rate": 1.7692648845436764e-06, + "loss": 0.9368, + "num_input_tokens_seen": 107398185, + "step": 7328 + }, + { + "epoch": 0.550804148504434, + "grad_norm": 1.3223398658274395, + "learning_rate": 1.7687812999021531e-06, + "loss": 0.8433, + "num_input_tokens_seen": 107425575, + "step": 7329 + }, + { + "epoch": 0.5508793025702691, + "grad_norm": 2.0727957174265006, + "learning_rate": 1.7682977289612226e-06, + "loss": 0.9411, + "num_input_tokens_seen": 107448885, + "step": 7330 + }, + { + "epoch": 0.550954456636104, + "grad_norm": 1.6825971375265454, + "learning_rate": 1.7678141717495395e-06, + "loss": 0.9407, + "num_input_tokens_seen": 107470625, + "step": 7331 + }, + { + "epoch": 0.551029610701939, + "grad_norm": 1.5627979757377377, + "learning_rate": 1.7673306282957559e-06, + "loss": 0.9165, + "num_input_tokens_seen": 107493345, + "step": 7332 + }, + { + "epoch": 0.5511047647677739, + "grad_norm": 1.6547466148644987, + "learning_rate": 1.766847098628523e-06, + "loss": 1.0644, + "num_input_tokens_seen": 107517535, + "step": 7333 + }, + { + "epoch": 0.551179918833609, + "grad_norm": 1.4991862155868887, + "learning_rate": 1.7663635827764924e-06, + "loss": 0.8646, + "num_input_tokens_seen": 107542375, + "step": 7334 + }, + { + "epoch": 0.5512550728994439, + "grad_norm": 1.2597581966683848, + "learning_rate": 1.7658800807683142e-06, + "loss": 0.9661, + "num_input_tokens_seen": 107570865, + "step": 7335 + }, + { + "epoch": 0.5513302269652788, + "grad_norm": 1.7011988535864846, + "learning_rate": 1.7653965926326379e-06, + "loss": 0.8836, + "num_input_tokens_seen": 107593110, + "step": 7336 + }, + { + "epoch": 0.5514053810311138, + "grad_norm": 1.541621293439827, + "learning_rate": 1.764913118398112e-06, + "loss": 1.0152, + "num_input_tokens_seen": 107614400, + "step": 7337 + }, + { + "epoch": 0.5514805350969487, + "grad_norm": 1.2776921348388814, + "learning_rate": 1.7644296580933835e-06, + "loss": 0.892, + "num_input_tokens_seen": 107639245, + "step": 7338 + }, + { + "epoch": 0.5515556891627837, + "grad_norm": 1.3944228528205718, + "learning_rate": 1.7639462117471004e-06, + "loss": 0.8976, + "num_input_tokens_seen": 107663375, + "step": 7339 + }, + { + "epoch": 0.5516308432286187, + "grad_norm": 1.8088872354217667, + "learning_rate": 1.7634627793879075e-06, + "loss": 0.9528, + "num_input_tokens_seen": 107688070, + "step": 7340 + }, + { + "epoch": 0.5517059972944536, + "grad_norm": 1.463074941190238, + "learning_rate": 1.7629793610444513e-06, + "loss": 0.8563, + "num_input_tokens_seen": 107715320, + "step": 7341 + }, + { + "epoch": 0.5517811513602886, + "grad_norm": 1.5661706913796922, + "learning_rate": 1.7624959567453746e-06, + "loss": 0.9688, + "num_input_tokens_seen": 107737800, + "step": 7342 + }, + { + "epoch": 0.5518563054261235, + "grad_norm": 1.840815067030822, + "learning_rate": 1.7620125665193232e-06, + "loss": 0.8911, + "num_input_tokens_seen": 107758285, + "step": 7343 + }, + { + "epoch": 0.5519314594919585, + "grad_norm": 1.7286001066263244, + "learning_rate": 1.7615291903949382e-06, + "loss": 0.8919, + "num_input_tokens_seen": 107778525, + "step": 7344 + }, + { + "epoch": 0.5520066135577935, + "grad_norm": 0.7093115858058162, + "learning_rate": 1.761045828400861e-06, + "loss": 0.8342, + "num_input_tokens_seen": 107862750, + "step": 7345 + }, + { + "epoch": 0.5520817676236285, + "grad_norm": 0.8444495656530232, + "learning_rate": 1.7605624805657343e-06, + "loss": 0.8669, + "num_input_tokens_seen": 107934790, + "step": 7346 + }, + { + "epoch": 0.5521569216894634, + "grad_norm": 1.5950273369147536, + "learning_rate": 1.760079146918197e-06, + "loss": 0.9646, + "num_input_tokens_seen": 107959230, + "step": 7347 + }, + { + "epoch": 0.5522320757552983, + "grad_norm": 1.7623820272151762, + "learning_rate": 1.7595958274868896e-06, + "loss": 0.9279, + "num_input_tokens_seen": 107982570, + "step": 7348 + }, + { + "epoch": 0.5523072298211333, + "grad_norm": 1.5448869973741646, + "learning_rate": 1.75911252230045e-06, + "loss": 1.0322, + "num_input_tokens_seen": 108003270, + "step": 7349 + }, + { + "epoch": 0.5523823838869683, + "grad_norm": 2.541183384042365, + "learning_rate": 1.758629231387515e-06, + "loss": 0.8371, + "num_input_tokens_seen": 108025650, + "step": 7350 + }, + { + "epoch": 0.5524575379528033, + "grad_norm": 2.015925540252912, + "learning_rate": 1.7581459547767233e-06, + "loss": 0.9758, + "num_input_tokens_seen": 108049590, + "step": 7351 + }, + { + "epoch": 0.5525326920186382, + "grad_norm": 1.4848976419795312, + "learning_rate": 1.7576626924967091e-06, + "loss": 0.9789, + "num_input_tokens_seen": 108071775, + "step": 7352 + }, + { + "epoch": 0.5526078460844732, + "grad_norm": 1.2957754874529661, + "learning_rate": 1.7571794445761089e-06, + "loss": 0.8953, + "num_input_tokens_seen": 108095020, + "step": 7353 + }, + { + "epoch": 0.5526830001503081, + "grad_norm": 1.5544321435134574, + "learning_rate": 1.7566962110435563e-06, + "loss": 0.9055, + "num_input_tokens_seen": 108121720, + "step": 7354 + }, + { + "epoch": 0.552758154216143, + "grad_norm": 1.6589555774234523, + "learning_rate": 1.7562129919276845e-06, + "loss": 0.9223, + "num_input_tokens_seen": 108143705, + "step": 7355 + }, + { + "epoch": 0.5528333082819781, + "grad_norm": 1.6550727114882628, + "learning_rate": 1.7557297872571272e-06, + "loss": 0.9551, + "num_input_tokens_seen": 108165335, + "step": 7356 + }, + { + "epoch": 0.552908462347813, + "grad_norm": 1.8182936839889563, + "learning_rate": 1.7552465970605145e-06, + "loss": 1.0317, + "num_input_tokens_seen": 108189805, + "step": 7357 + }, + { + "epoch": 0.552983616413648, + "grad_norm": 2.1056019134025807, + "learning_rate": 1.7547634213664786e-06, + "loss": 0.8945, + "num_input_tokens_seen": 108211870, + "step": 7358 + }, + { + "epoch": 0.5530587704794829, + "grad_norm": 1.4747196383651318, + "learning_rate": 1.7542802602036492e-06, + "loss": 0.9141, + "num_input_tokens_seen": 108234000, + "step": 7359 + }, + { + "epoch": 0.553133924545318, + "grad_norm": 2.0612876103048405, + "learning_rate": 1.753797113600655e-06, + "loss": 0.9905, + "num_input_tokens_seen": 108255390, + "step": 7360 + }, + { + "epoch": 0.5532090786111529, + "grad_norm": 1.2938766288522265, + "learning_rate": 1.7533139815861248e-06, + "loss": 1.0022, + "num_input_tokens_seen": 108279645, + "step": 7361 + }, + { + "epoch": 0.5532842326769878, + "grad_norm": 1.6284607934539446, + "learning_rate": 1.7528308641886856e-06, + "loss": 0.9917, + "num_input_tokens_seen": 108301385, + "step": 7362 + }, + { + "epoch": 0.5533593867428228, + "grad_norm": 1.4264992774081748, + "learning_rate": 1.7523477614369645e-06, + "loss": 0.8608, + "num_input_tokens_seen": 108326025, + "step": 7363 + }, + { + "epoch": 0.5534345408086577, + "grad_norm": 1.5627182617333202, + "learning_rate": 1.751864673359586e-06, + "loss": 0.9443, + "num_input_tokens_seen": 108350955, + "step": 7364 + }, + { + "epoch": 0.5535096948744928, + "grad_norm": 1.8910819518165594, + "learning_rate": 1.7513815999851767e-06, + "loss": 0.8241, + "num_input_tokens_seen": 108372505, + "step": 7365 + }, + { + "epoch": 0.5535848489403277, + "grad_norm": 0.7885283371182139, + "learning_rate": 1.7508985413423599e-06, + "loss": 0.8284, + "num_input_tokens_seen": 108444665, + "step": 7366 + }, + { + "epoch": 0.5536600030061626, + "grad_norm": 1.5963938605977839, + "learning_rate": 1.7504154974597572e-06, + "loss": 0.873, + "num_input_tokens_seen": 108466740, + "step": 7367 + }, + { + "epoch": 0.5537351570719976, + "grad_norm": 0.7784375110872774, + "learning_rate": 1.7499324683659928e-06, + "loss": 0.817, + "num_input_tokens_seen": 108544845, + "step": 7368 + }, + { + "epoch": 0.5538103111378325, + "grad_norm": 1.6191889223043991, + "learning_rate": 1.749449454089687e-06, + "loss": 0.8992, + "num_input_tokens_seen": 108568710, + "step": 7369 + }, + { + "epoch": 0.5538854652036675, + "grad_norm": 1.3652011501651662, + "learning_rate": 1.7489664546594606e-06, + "loss": 0.9304, + "num_input_tokens_seen": 108591235, + "step": 7370 + }, + { + "epoch": 0.5539606192695025, + "grad_norm": 1.4925560620895366, + "learning_rate": 1.7484834701039333e-06, + "loss": 0.869, + "num_input_tokens_seen": 108617825, + "step": 7371 + }, + { + "epoch": 0.5540357733353375, + "grad_norm": 1.4227857921430396, + "learning_rate": 1.7480005004517228e-06, + "loss": 0.974, + "num_input_tokens_seen": 108642555, + "step": 7372 + }, + { + "epoch": 0.5541109274011724, + "grad_norm": 1.6257297637867312, + "learning_rate": 1.7475175457314481e-06, + "loss": 0.8834, + "num_input_tokens_seen": 108664790, + "step": 7373 + }, + { + "epoch": 0.5541860814670073, + "grad_norm": 1.5413541047391703, + "learning_rate": 1.7470346059717253e-06, + "loss": 0.937, + "num_input_tokens_seen": 108687515, + "step": 7374 + }, + { + "epoch": 0.5542612355328423, + "grad_norm": 1.6340889937586984, + "learning_rate": 1.7465516812011713e-06, + "loss": 0.926, + "num_input_tokens_seen": 108714880, + "step": 7375 + }, + { + "epoch": 0.5543363895986773, + "grad_norm": 1.534093046989621, + "learning_rate": 1.7460687714484008e-06, + "loss": 0.926, + "num_input_tokens_seen": 108738780, + "step": 7376 + }, + { + "epoch": 0.5544115436645123, + "grad_norm": 1.8973420675741943, + "learning_rate": 1.7455858767420272e-06, + "loss": 0.9127, + "num_input_tokens_seen": 108763090, + "step": 7377 + }, + { + "epoch": 0.5544866977303472, + "grad_norm": 1.6479483506942159, + "learning_rate": 1.7451029971106653e-06, + "loss": 1.0704, + "num_input_tokens_seen": 108784095, + "step": 7378 + }, + { + "epoch": 0.5545618517961822, + "grad_norm": 1.7428960343003839, + "learning_rate": 1.7446201325829261e-06, + "loss": 0.9618, + "num_input_tokens_seen": 108807450, + "step": 7379 + }, + { + "epoch": 0.5546370058620171, + "grad_norm": 3.02965983731811, + "learning_rate": 1.7441372831874228e-06, + "loss": 0.9454, + "num_input_tokens_seen": 108830920, + "step": 7380 + }, + { + "epoch": 0.554712159927852, + "grad_norm": 0.8256484734247067, + "learning_rate": 1.7436544489527652e-06, + "loss": 0.8159, + "num_input_tokens_seen": 108899740, + "step": 7381 + }, + { + "epoch": 0.5547873139936871, + "grad_norm": 0.6968713281303799, + "learning_rate": 1.7431716299075625e-06, + "loss": 0.8678, + "num_input_tokens_seen": 108986665, + "step": 7382 + }, + { + "epoch": 0.554862468059522, + "grad_norm": 1.6705843498173734, + "learning_rate": 1.7426888260804247e-06, + "loss": 0.9612, + "num_input_tokens_seen": 109008065, + "step": 7383 + }, + { + "epoch": 0.554937622125357, + "grad_norm": 0.8535742488604717, + "learning_rate": 1.7422060374999587e-06, + "loss": 0.8203, + "num_input_tokens_seen": 109090400, + "step": 7384 + }, + { + "epoch": 0.5550127761911919, + "grad_norm": 1.6740248218459888, + "learning_rate": 1.7417232641947728e-06, + "loss": 0.825, + "num_input_tokens_seen": 109113860, + "step": 7385 + }, + { + "epoch": 0.5550879302570269, + "grad_norm": 1.8516458098263846, + "learning_rate": 1.7412405061934714e-06, + "loss": 0.9857, + "num_input_tokens_seen": 109133070, + "step": 7386 + }, + { + "epoch": 0.5551630843228619, + "grad_norm": 1.7459468225025743, + "learning_rate": 1.740757763524662e-06, + "loss": 0.931, + "num_input_tokens_seen": 109158235, + "step": 7387 + }, + { + "epoch": 0.5552382383886968, + "grad_norm": 1.3589361951444796, + "learning_rate": 1.740275036216948e-06, + "loss": 0.9985, + "num_input_tokens_seen": 109181390, + "step": 7388 + }, + { + "epoch": 0.5553133924545318, + "grad_norm": 1.4553354330223922, + "learning_rate": 1.7397923242989314e-06, + "loss": 0.9193, + "num_input_tokens_seen": 109204785, + "step": 7389 + }, + { + "epoch": 0.5553885465203667, + "grad_norm": 1.561913113046987, + "learning_rate": 1.739309627799217e-06, + "loss": 0.9983, + "num_input_tokens_seen": 109227365, + "step": 7390 + }, + { + "epoch": 0.5554637005862018, + "grad_norm": 1.6803315103515066, + "learning_rate": 1.7388269467464047e-06, + "loss": 1.0613, + "num_input_tokens_seen": 109249995, + "step": 7391 + }, + { + "epoch": 0.5555388546520367, + "grad_norm": 1.4527588249625158, + "learning_rate": 1.7383442811690967e-06, + "loss": 0.8778, + "num_input_tokens_seen": 109275095, + "step": 7392 + }, + { + "epoch": 0.5556140087178716, + "grad_norm": 1.487611394606029, + "learning_rate": 1.7378616310958917e-06, + "loss": 0.9515, + "num_input_tokens_seen": 109298350, + "step": 7393 + }, + { + "epoch": 0.5556891627837066, + "grad_norm": 2.111856653789164, + "learning_rate": 1.7373789965553886e-06, + "loss": 0.9292, + "num_input_tokens_seen": 109320955, + "step": 7394 + }, + { + "epoch": 0.5557643168495415, + "grad_norm": 1.5807384341961737, + "learning_rate": 1.736896377576186e-06, + "loss": 0.8327, + "num_input_tokens_seen": 109346515, + "step": 7395 + }, + { + "epoch": 0.5558394709153766, + "grad_norm": 1.3222699975261207, + "learning_rate": 1.73641377418688e-06, + "loss": 0.9705, + "num_input_tokens_seen": 109370595, + "step": 7396 + }, + { + "epoch": 0.5559146249812115, + "grad_norm": 1.5098070778035415, + "learning_rate": 1.7359311864160677e-06, + "loss": 0.9491, + "num_input_tokens_seen": 109394270, + "step": 7397 + }, + { + "epoch": 0.5559897790470465, + "grad_norm": 1.283084997490984, + "learning_rate": 1.7354486142923438e-06, + "loss": 0.9569, + "num_input_tokens_seen": 109417990, + "step": 7398 + }, + { + "epoch": 0.5560649331128814, + "grad_norm": 1.4950119370204025, + "learning_rate": 1.7349660578443022e-06, + "loss": 0.9046, + "num_input_tokens_seen": 109442400, + "step": 7399 + }, + { + "epoch": 0.5561400871787163, + "grad_norm": 1.4109719577438635, + "learning_rate": 1.7344835171005368e-06, + "loss": 0.9679, + "num_input_tokens_seen": 109465980, + "step": 7400 + }, + { + "epoch": 0.5562152412445514, + "grad_norm": 1.7331771150519766, + "learning_rate": 1.7340009920896392e-06, + "loss": 0.9135, + "num_input_tokens_seen": 109488160, + "step": 7401 + }, + { + "epoch": 0.5562903953103863, + "grad_norm": 1.4188655822995198, + "learning_rate": 1.7335184828402022e-06, + "loss": 0.9146, + "num_input_tokens_seen": 109511050, + "step": 7402 + }, + { + "epoch": 0.5563655493762213, + "grad_norm": 1.7779120909431794, + "learning_rate": 1.7330359893808154e-06, + "loss": 1.004, + "num_input_tokens_seen": 109533870, + "step": 7403 + }, + { + "epoch": 0.5564407034420562, + "grad_norm": 1.4191950599904832, + "learning_rate": 1.732553511740068e-06, + "loss": 0.9607, + "num_input_tokens_seen": 109557470, + "step": 7404 + }, + { + "epoch": 0.5565158575078912, + "grad_norm": 1.6508781841539384, + "learning_rate": 1.7320710499465494e-06, + "loss": 0.9872, + "num_input_tokens_seen": 109581230, + "step": 7405 + }, + { + "epoch": 0.5565910115737261, + "grad_norm": 1.8265694414958122, + "learning_rate": 1.7315886040288468e-06, + "loss": 0.9297, + "num_input_tokens_seen": 109603955, + "step": 7406 + }, + { + "epoch": 0.5566661656395611, + "grad_norm": 1.4494168128216394, + "learning_rate": 1.7311061740155477e-06, + "loss": 0.8736, + "num_input_tokens_seen": 109629160, + "step": 7407 + }, + { + "epoch": 0.5567413197053961, + "grad_norm": 1.4783751649038026, + "learning_rate": 1.7306237599352365e-06, + "loss": 0.8335, + "num_input_tokens_seen": 109652995, + "step": 7408 + }, + { + "epoch": 0.556816473771231, + "grad_norm": 1.5620960476367118, + "learning_rate": 1.7301413618165e-06, + "loss": 0.942, + "num_input_tokens_seen": 109676705, + "step": 7409 + }, + { + "epoch": 0.556891627837066, + "grad_norm": 0.7025418088221381, + "learning_rate": 1.7296589796879215e-06, + "loss": 0.8206, + "num_input_tokens_seen": 109756765, + "step": 7410 + }, + { + "epoch": 0.5569667819029009, + "grad_norm": 2.0743482876144155, + "learning_rate": 1.7291766135780825e-06, + "loss": 0.9221, + "num_input_tokens_seen": 109778915, + "step": 7411 + }, + { + "epoch": 0.5570419359687359, + "grad_norm": 1.2453346451497844, + "learning_rate": 1.728694263515567e-06, + "loss": 0.9637, + "num_input_tokens_seen": 109803920, + "step": 7412 + }, + { + "epoch": 0.5571170900345709, + "grad_norm": 1.6677003595406568, + "learning_rate": 1.728211929528955e-06, + "loss": 0.9292, + "num_input_tokens_seen": 109827960, + "step": 7413 + }, + { + "epoch": 0.5571922441004058, + "grad_norm": 1.5607965720770112, + "learning_rate": 1.727729611646827e-06, + "loss": 0.8882, + "num_input_tokens_seen": 109852145, + "step": 7414 + }, + { + "epoch": 0.5572673981662408, + "grad_norm": 1.5310973266472585, + "learning_rate": 1.7272473098977623e-06, + "loss": 0.9375, + "num_input_tokens_seen": 109875485, + "step": 7415 + }, + { + "epoch": 0.5573425522320757, + "grad_norm": 2.0837915933779385, + "learning_rate": 1.7267650243103384e-06, + "loss": 0.9611, + "num_input_tokens_seen": 109897260, + "step": 7416 + }, + { + "epoch": 0.5574177062979108, + "grad_norm": 0.7828318031961508, + "learning_rate": 1.7262827549131337e-06, + "loss": 0.8103, + "num_input_tokens_seen": 109970460, + "step": 7417 + }, + { + "epoch": 0.5574928603637457, + "grad_norm": 1.4516405296089556, + "learning_rate": 1.7258005017347234e-06, + "loss": 0.9824, + "num_input_tokens_seen": 109993490, + "step": 7418 + }, + { + "epoch": 0.5575680144295806, + "grad_norm": 1.45515947562546, + "learning_rate": 1.725318264803684e-06, + "loss": 0.8569, + "num_input_tokens_seen": 110017300, + "step": 7419 + }, + { + "epoch": 0.5576431684954156, + "grad_norm": 1.3238282654763038, + "learning_rate": 1.724836044148589e-06, + "loss": 0.917, + "num_input_tokens_seen": 110041475, + "step": 7420 + }, + { + "epoch": 0.5577183225612505, + "grad_norm": 1.6356994076728624, + "learning_rate": 1.7243538397980115e-06, + "loss": 0.9902, + "num_input_tokens_seen": 110063245, + "step": 7421 + }, + { + "epoch": 0.5577934766270856, + "grad_norm": 1.9491734929459275, + "learning_rate": 1.7238716517805249e-06, + "loss": 0.9626, + "num_input_tokens_seen": 110092565, + "step": 7422 + }, + { + "epoch": 0.5578686306929205, + "grad_norm": 1.518038293302732, + "learning_rate": 1.7233894801247002e-06, + "loss": 0.9369, + "num_input_tokens_seen": 110116375, + "step": 7423 + }, + { + "epoch": 0.5579437847587555, + "grad_norm": 1.3281763852220592, + "learning_rate": 1.7229073248591084e-06, + "loss": 0.9847, + "num_input_tokens_seen": 110139910, + "step": 7424 + }, + { + "epoch": 0.5580189388245904, + "grad_norm": 1.5094536265237202, + "learning_rate": 1.7224251860123185e-06, + "loss": 0.9552, + "num_input_tokens_seen": 110163685, + "step": 7425 + }, + { + "epoch": 0.5580940928904253, + "grad_norm": 1.615578334584502, + "learning_rate": 1.7219430636128989e-06, + "loss": 0.9512, + "num_input_tokens_seen": 110185515, + "step": 7426 + }, + { + "epoch": 0.5581692469562604, + "grad_norm": 1.529283311571797, + "learning_rate": 1.721460957689418e-06, + "loss": 0.9411, + "num_input_tokens_seen": 110210150, + "step": 7427 + }, + { + "epoch": 0.5582444010220953, + "grad_norm": 1.719762954456833, + "learning_rate": 1.720978868270441e-06, + "loss": 1.0505, + "num_input_tokens_seen": 110234155, + "step": 7428 + }, + { + "epoch": 0.5583195550879303, + "grad_norm": 1.7087809588760874, + "learning_rate": 1.7204967953845358e-06, + "loss": 0.9739, + "num_input_tokens_seen": 110257315, + "step": 7429 + }, + { + "epoch": 0.5583947091537652, + "grad_norm": 1.6683043381301936, + "learning_rate": 1.7200147390602643e-06, + "loss": 0.9616, + "num_input_tokens_seen": 110283110, + "step": 7430 + }, + { + "epoch": 0.5584698632196001, + "grad_norm": 2.0719334428694633, + "learning_rate": 1.7195326993261927e-06, + "loss": 0.9073, + "num_input_tokens_seen": 110304690, + "step": 7431 + }, + { + "epoch": 0.5585450172854352, + "grad_norm": 1.748790595542561, + "learning_rate": 1.7190506762108828e-06, + "loss": 0.8796, + "num_input_tokens_seen": 110328050, + "step": 7432 + }, + { + "epoch": 0.5586201713512701, + "grad_norm": 1.5048542156138762, + "learning_rate": 1.7185686697428954e-06, + "loss": 0.9582, + "num_input_tokens_seen": 110354390, + "step": 7433 + }, + { + "epoch": 0.5586953254171051, + "grad_norm": 1.4549996374070806, + "learning_rate": 1.7180866799507925e-06, + "loss": 0.8979, + "num_input_tokens_seen": 110376955, + "step": 7434 + }, + { + "epoch": 0.55877047948294, + "grad_norm": 1.3673922576163982, + "learning_rate": 1.717604706863133e-06, + "loss": 0.9133, + "num_input_tokens_seen": 110402890, + "step": 7435 + }, + { + "epoch": 0.558845633548775, + "grad_norm": 1.4117463675469626, + "learning_rate": 1.7171227505084764e-06, + "loss": 0.9387, + "num_input_tokens_seen": 110425965, + "step": 7436 + }, + { + "epoch": 0.55892078761461, + "grad_norm": 2.106301675496524, + "learning_rate": 1.71664081091538e-06, + "loss": 0.9906, + "num_input_tokens_seen": 110447765, + "step": 7437 + }, + { + "epoch": 0.5589959416804449, + "grad_norm": 1.500410500622959, + "learning_rate": 1.7161588881124003e-06, + "loss": 1.0262, + "num_input_tokens_seen": 110471765, + "step": 7438 + }, + { + "epoch": 0.5590710957462799, + "grad_norm": 2.222044312455013, + "learning_rate": 1.7156769821280937e-06, + "loss": 0.9235, + "num_input_tokens_seen": 110496805, + "step": 7439 + }, + { + "epoch": 0.5591462498121148, + "grad_norm": 1.3772418646039841, + "learning_rate": 1.7151950929910145e-06, + "loss": 0.8729, + "num_input_tokens_seen": 110522780, + "step": 7440 + }, + { + "epoch": 0.5592214038779498, + "grad_norm": 1.5840937979987657, + "learning_rate": 1.7147132207297165e-06, + "loss": 0.9247, + "num_input_tokens_seen": 110545395, + "step": 7441 + }, + { + "epoch": 0.5592965579437847, + "grad_norm": 0.7788395125132206, + "learning_rate": 1.7142313653727531e-06, + "loss": 0.7751, + "num_input_tokens_seen": 110615800, + "step": 7442 + }, + { + "epoch": 0.5593717120096198, + "grad_norm": 1.8558944695618522, + "learning_rate": 1.7137495269486749e-06, + "loss": 1.0672, + "num_input_tokens_seen": 110636440, + "step": 7443 + }, + { + "epoch": 0.5594468660754547, + "grad_norm": 1.2696696044652203, + "learning_rate": 1.7132677054860339e-06, + "loss": 0.8904, + "num_input_tokens_seen": 110663255, + "step": 7444 + }, + { + "epoch": 0.5595220201412896, + "grad_norm": 3.1229380098970374, + "learning_rate": 1.7127859010133788e-06, + "loss": 0.8834, + "num_input_tokens_seen": 110686805, + "step": 7445 + }, + { + "epoch": 0.5595971742071246, + "grad_norm": 1.5995961096729419, + "learning_rate": 1.7123041135592593e-06, + "loss": 0.7986, + "num_input_tokens_seen": 110709405, + "step": 7446 + }, + { + "epoch": 0.5596723282729595, + "grad_norm": 1.5794938448900433, + "learning_rate": 1.7118223431522227e-06, + "loss": 0.8742, + "num_input_tokens_seen": 110732330, + "step": 7447 + }, + { + "epoch": 0.5597474823387946, + "grad_norm": 1.5971201997690643, + "learning_rate": 1.7113405898208156e-06, + "loss": 0.965, + "num_input_tokens_seen": 110755815, + "step": 7448 + }, + { + "epoch": 0.5598226364046295, + "grad_norm": 1.2476413885962059, + "learning_rate": 1.710858853593584e-06, + "loss": 0.9659, + "num_input_tokens_seen": 110782620, + "step": 7449 + }, + { + "epoch": 0.5598977904704645, + "grad_norm": 2.318657692771043, + "learning_rate": 1.710377134499072e-06, + "loss": 0.9085, + "num_input_tokens_seen": 110807285, + "step": 7450 + }, + { + "epoch": 0.5599729445362994, + "grad_norm": 1.8321721850358106, + "learning_rate": 1.7098954325658249e-06, + "loss": 0.9686, + "num_input_tokens_seen": 110830420, + "step": 7451 + }, + { + "epoch": 0.5600480986021343, + "grad_norm": 1.6735464991984084, + "learning_rate": 1.7094137478223831e-06, + "loss": 0.9603, + "num_input_tokens_seen": 110850815, + "step": 7452 + }, + { + "epoch": 0.5601232526679694, + "grad_norm": 1.8558463585745781, + "learning_rate": 1.7089320802972901e-06, + "loss": 0.763, + "num_input_tokens_seen": 110874560, + "step": 7453 + }, + { + "epoch": 0.5601984067338043, + "grad_norm": 1.6553137670234497, + "learning_rate": 1.7084504300190862e-06, + "loss": 0.9325, + "num_input_tokens_seen": 110899975, + "step": 7454 + }, + { + "epoch": 0.5602735607996393, + "grad_norm": 1.7691334662584242, + "learning_rate": 1.7079687970163105e-06, + "loss": 1.0311, + "num_input_tokens_seen": 110921920, + "step": 7455 + }, + { + "epoch": 0.5603487148654742, + "grad_norm": 1.6547561242243398, + "learning_rate": 1.7074871813175018e-06, + "loss": 0.9532, + "num_input_tokens_seen": 110945270, + "step": 7456 + }, + { + "epoch": 0.5604238689313091, + "grad_norm": 1.4331621149100773, + "learning_rate": 1.7070055829511973e-06, + "loss": 0.8892, + "num_input_tokens_seen": 110965850, + "step": 7457 + }, + { + "epoch": 0.5604990229971442, + "grad_norm": 2.1520064240186025, + "learning_rate": 1.7065240019459347e-06, + "loss": 0.9494, + "num_input_tokens_seen": 110987125, + "step": 7458 + }, + { + "epoch": 0.5605741770629791, + "grad_norm": 1.3820653933830027, + "learning_rate": 1.7060424383302485e-06, + "loss": 0.9547, + "num_input_tokens_seen": 111011635, + "step": 7459 + }, + { + "epoch": 0.5606493311288141, + "grad_norm": 1.50560080439539, + "learning_rate": 1.7055608921326731e-06, + "loss": 0.9285, + "num_input_tokens_seen": 111038180, + "step": 7460 + }, + { + "epoch": 0.560724485194649, + "grad_norm": 1.566346050827528, + "learning_rate": 1.7050793633817431e-06, + "loss": 0.8208, + "num_input_tokens_seen": 111062965, + "step": 7461 + }, + { + "epoch": 0.560799639260484, + "grad_norm": 1.5422242505178998, + "learning_rate": 1.7045978521059894e-06, + "loss": 0.9194, + "num_input_tokens_seen": 111087550, + "step": 7462 + }, + { + "epoch": 0.560874793326319, + "grad_norm": 0.8111181245425136, + "learning_rate": 1.7041163583339446e-06, + "loss": 0.7818, + "num_input_tokens_seen": 111153330, + "step": 7463 + }, + { + "epoch": 0.5609499473921539, + "grad_norm": 0.778934212721235, + "learning_rate": 1.7036348820941386e-06, + "loss": 0.904, + "num_input_tokens_seen": 111238425, + "step": 7464 + }, + { + "epoch": 0.5610251014579889, + "grad_norm": 1.6199354330782845, + "learning_rate": 1.7031534234151001e-06, + "loss": 0.9324, + "num_input_tokens_seen": 111259810, + "step": 7465 + }, + { + "epoch": 0.5611002555238238, + "grad_norm": 1.6009570210041295, + "learning_rate": 1.7026719823253585e-06, + "loss": 0.9248, + "num_input_tokens_seen": 111282040, + "step": 7466 + }, + { + "epoch": 0.5611754095896588, + "grad_norm": 1.7558363593675452, + "learning_rate": 1.7021905588534402e-06, + "loss": 1.0279, + "num_input_tokens_seen": 111303160, + "step": 7467 + }, + { + "epoch": 0.5612505636554938, + "grad_norm": 1.707525674897798, + "learning_rate": 1.701709153027872e-06, + "loss": 0.8864, + "num_input_tokens_seen": 111327190, + "step": 7468 + }, + { + "epoch": 0.5613257177213288, + "grad_norm": 1.7837037702957395, + "learning_rate": 1.7012277648771787e-06, + "loss": 0.8593, + "num_input_tokens_seen": 111349920, + "step": 7469 + }, + { + "epoch": 0.5614008717871637, + "grad_norm": 2.695898373824983, + "learning_rate": 1.700746394429884e-06, + "loss": 0.9672, + "num_input_tokens_seen": 111372455, + "step": 7470 + }, + { + "epoch": 0.5614760258529986, + "grad_norm": 1.63369814567902, + "learning_rate": 1.7002650417145119e-06, + "loss": 0.8466, + "num_input_tokens_seen": 111396050, + "step": 7471 + }, + { + "epoch": 0.5615511799188336, + "grad_norm": 1.7035226620128818, + "learning_rate": 1.699783706759583e-06, + "loss": 0.9542, + "num_input_tokens_seen": 111418850, + "step": 7472 + }, + { + "epoch": 0.5616263339846685, + "grad_norm": 1.3264970563330343, + "learning_rate": 1.6993023895936196e-06, + "loss": 1.0053, + "num_input_tokens_seen": 111444175, + "step": 7473 + }, + { + "epoch": 0.5617014880505036, + "grad_norm": 1.5469787254621412, + "learning_rate": 1.6988210902451407e-06, + "loss": 1.0161, + "num_input_tokens_seen": 111466380, + "step": 7474 + }, + { + "epoch": 0.5617766421163385, + "grad_norm": 1.8619790315072249, + "learning_rate": 1.698339808742666e-06, + "loss": 0.9086, + "num_input_tokens_seen": 111490445, + "step": 7475 + }, + { + "epoch": 0.5618517961821734, + "grad_norm": 1.3545159500776993, + "learning_rate": 1.697858545114713e-06, + "loss": 1.0378, + "num_input_tokens_seen": 111514305, + "step": 7476 + }, + { + "epoch": 0.5619269502480084, + "grad_norm": 1.5507744085127348, + "learning_rate": 1.6973772993897978e-06, + "loss": 0.9635, + "num_input_tokens_seen": 111535460, + "step": 7477 + }, + { + "epoch": 0.5620021043138433, + "grad_norm": 1.6676607584635472, + "learning_rate": 1.6968960715964364e-06, + "loss": 0.9861, + "num_input_tokens_seen": 111558630, + "step": 7478 + }, + { + "epoch": 0.5620772583796784, + "grad_norm": 2.0575177676058285, + "learning_rate": 1.6964148617631432e-06, + "loss": 0.9109, + "num_input_tokens_seen": 111585180, + "step": 7479 + }, + { + "epoch": 0.5621524124455133, + "grad_norm": 1.3874229684702213, + "learning_rate": 1.6959336699184323e-06, + "loss": 0.9337, + "num_input_tokens_seen": 111611205, + "step": 7480 + }, + { + "epoch": 0.5622275665113483, + "grad_norm": 1.9178160938802993, + "learning_rate": 1.695452496090816e-06, + "loss": 0.9687, + "num_input_tokens_seen": 111635930, + "step": 7481 + }, + { + "epoch": 0.5623027205771832, + "grad_norm": 1.6505256191323512, + "learning_rate": 1.694971340308805e-06, + "loss": 0.9073, + "num_input_tokens_seen": 111659075, + "step": 7482 + }, + { + "epoch": 0.5623778746430181, + "grad_norm": 1.4515372184793813, + "learning_rate": 1.6944902026009107e-06, + "loss": 0.9348, + "num_input_tokens_seen": 111680805, + "step": 7483 + }, + { + "epoch": 0.5624530287088532, + "grad_norm": 2.2276132346334303, + "learning_rate": 1.694009082995641e-06, + "loss": 0.9562, + "num_input_tokens_seen": 111701575, + "step": 7484 + }, + { + "epoch": 0.5625281827746881, + "grad_norm": 2.4194184673309937, + "learning_rate": 1.6935279815215056e-06, + "loss": 0.9149, + "num_input_tokens_seen": 111725295, + "step": 7485 + }, + { + "epoch": 0.5626033368405231, + "grad_norm": 1.3442417841712213, + "learning_rate": 1.6930468982070106e-06, + "loss": 0.869, + "num_input_tokens_seen": 111749365, + "step": 7486 + }, + { + "epoch": 0.562678490906358, + "grad_norm": 1.7853241889473472, + "learning_rate": 1.6925658330806618e-06, + "loss": 0.9555, + "num_input_tokens_seen": 111772900, + "step": 7487 + }, + { + "epoch": 0.562753644972193, + "grad_norm": 1.6741205269414765, + "learning_rate": 1.6920847861709653e-06, + "loss": 0.9825, + "num_input_tokens_seen": 111793155, + "step": 7488 + }, + { + "epoch": 0.562828799038028, + "grad_norm": 1.7835269228713464, + "learning_rate": 1.6916037575064238e-06, + "loss": 0.8961, + "num_input_tokens_seen": 111814980, + "step": 7489 + }, + { + "epoch": 0.5629039531038629, + "grad_norm": 1.3651113386473963, + "learning_rate": 1.6911227471155408e-06, + "loss": 0.9323, + "num_input_tokens_seen": 111839840, + "step": 7490 + }, + { + "epoch": 0.5629791071696979, + "grad_norm": 1.8572426758653982, + "learning_rate": 1.6906417550268182e-06, + "loss": 0.9218, + "num_input_tokens_seen": 111860615, + "step": 7491 + }, + { + "epoch": 0.5630542612355328, + "grad_norm": 2.0213118180050027, + "learning_rate": 1.6901607812687558e-06, + "loss": 0.9527, + "num_input_tokens_seen": 111882850, + "step": 7492 + }, + { + "epoch": 0.5631294153013678, + "grad_norm": 2.0392629006078176, + "learning_rate": 1.6896798258698538e-06, + "loss": 0.8867, + "num_input_tokens_seen": 111908135, + "step": 7493 + }, + { + "epoch": 0.5632045693672028, + "grad_norm": 2.0504702377633697, + "learning_rate": 1.6891988888586094e-06, + "loss": 0.9594, + "num_input_tokens_seen": 111929255, + "step": 7494 + }, + { + "epoch": 0.5632797234330378, + "grad_norm": 1.3638694827625726, + "learning_rate": 1.6887179702635219e-06, + "loss": 0.9599, + "num_input_tokens_seen": 111953715, + "step": 7495 + }, + { + "epoch": 0.5633548774988727, + "grad_norm": 1.5691454901796698, + "learning_rate": 1.6882370701130863e-06, + "loss": 0.9111, + "num_input_tokens_seen": 111980735, + "step": 7496 + }, + { + "epoch": 0.5634300315647076, + "grad_norm": 1.4288314174038699, + "learning_rate": 1.6877561884357987e-06, + "loss": 0.8886, + "num_input_tokens_seen": 112005415, + "step": 7497 + }, + { + "epoch": 0.5635051856305426, + "grad_norm": 1.4240845500351458, + "learning_rate": 1.6872753252601525e-06, + "loss": 0.884, + "num_input_tokens_seen": 112030040, + "step": 7498 + }, + { + "epoch": 0.5635803396963776, + "grad_norm": 1.3855006065849151, + "learning_rate": 1.6867944806146403e-06, + "loss": 0.8845, + "num_input_tokens_seen": 112054490, + "step": 7499 + }, + { + "epoch": 0.5636554937622126, + "grad_norm": 1.8048604246579119, + "learning_rate": 1.6863136545277547e-06, + "loss": 0.8471, + "num_input_tokens_seen": 112077800, + "step": 7500 + }, + { + "epoch": 0.5637306478280475, + "grad_norm": 1.4468740739263286, + "learning_rate": 1.685832847027986e-06, + "loss": 0.8788, + "num_input_tokens_seen": 112102295, + "step": 7501 + }, + { + "epoch": 0.5638058018938824, + "grad_norm": 1.645955753199854, + "learning_rate": 1.6853520581438246e-06, + "loss": 0.9421, + "num_input_tokens_seen": 112128510, + "step": 7502 + }, + { + "epoch": 0.5638809559597174, + "grad_norm": 1.668952312421321, + "learning_rate": 1.6848712879037588e-06, + "loss": 0.9311, + "num_input_tokens_seen": 112148430, + "step": 7503 + }, + { + "epoch": 0.5639561100255523, + "grad_norm": 1.4944047681905983, + "learning_rate": 1.6843905363362754e-06, + "loss": 0.846, + "num_input_tokens_seen": 112172940, + "step": 7504 + }, + { + "epoch": 0.5640312640913874, + "grad_norm": 1.640689666926808, + "learning_rate": 1.6839098034698616e-06, + "loss": 0.8935, + "num_input_tokens_seen": 112193920, + "step": 7505 + }, + { + "epoch": 0.5641064181572223, + "grad_norm": 1.9909462446488657, + "learning_rate": 1.6834290893330017e-06, + "loss": 0.9334, + "num_input_tokens_seen": 112215315, + "step": 7506 + }, + { + "epoch": 0.5641815722230573, + "grad_norm": 0.7843475732149716, + "learning_rate": 1.682948393954181e-06, + "loss": 0.8666, + "num_input_tokens_seen": 112295315, + "step": 7507 + }, + { + "epoch": 0.5642567262888922, + "grad_norm": 1.6776468345176787, + "learning_rate": 1.6824677173618822e-06, + "loss": 1.0276, + "num_input_tokens_seen": 112320145, + "step": 7508 + }, + { + "epoch": 0.5643318803547271, + "grad_norm": 1.428414700630149, + "learning_rate": 1.6819870595845867e-06, + "loss": 0.9632, + "num_input_tokens_seen": 112344080, + "step": 7509 + }, + { + "epoch": 0.5644070344205622, + "grad_norm": 1.6232281341513546, + "learning_rate": 1.681506420650776e-06, + "loss": 1.0401, + "num_input_tokens_seen": 112367885, + "step": 7510 + }, + { + "epoch": 0.5644821884863971, + "grad_norm": 1.4294794483662197, + "learning_rate": 1.6810258005889287e-06, + "loss": 0.8482, + "num_input_tokens_seen": 112391350, + "step": 7511 + }, + { + "epoch": 0.5645573425522321, + "grad_norm": 1.88697285648912, + "learning_rate": 1.680545199427525e-06, + "loss": 0.9977, + "num_input_tokens_seen": 112416015, + "step": 7512 + }, + { + "epoch": 0.564632496618067, + "grad_norm": 1.708111314380255, + "learning_rate": 1.6800646171950415e-06, + "loss": 0.9919, + "num_input_tokens_seen": 112436680, + "step": 7513 + }, + { + "epoch": 0.564707650683902, + "grad_norm": 1.7019685791378636, + "learning_rate": 1.6795840539199538e-06, + "loss": 0.9426, + "num_input_tokens_seen": 112458080, + "step": 7514 + }, + { + "epoch": 0.564782804749737, + "grad_norm": 1.5713836660409413, + "learning_rate": 1.6791035096307387e-06, + "loss": 0.8343, + "num_input_tokens_seen": 112481565, + "step": 7515 + }, + { + "epoch": 0.5648579588155719, + "grad_norm": 1.5280167055664786, + "learning_rate": 1.6786229843558685e-06, + "loss": 1.0362, + "num_input_tokens_seen": 112505230, + "step": 7516 + }, + { + "epoch": 0.5649331128814069, + "grad_norm": 1.5626415188597267, + "learning_rate": 1.6781424781238178e-06, + "loss": 0.9715, + "num_input_tokens_seen": 112528765, + "step": 7517 + }, + { + "epoch": 0.5650082669472418, + "grad_norm": 0.7224461842990254, + "learning_rate": 1.6776619909630574e-06, + "loss": 0.7721, + "num_input_tokens_seen": 112598025, + "step": 7518 + }, + { + "epoch": 0.5650834210130768, + "grad_norm": 1.4146392657646125, + "learning_rate": 1.6771815229020586e-06, + "loss": 0.8852, + "num_input_tokens_seen": 112627370, + "step": 7519 + }, + { + "epoch": 0.5651585750789118, + "grad_norm": 1.6323667218890598, + "learning_rate": 1.676701073969291e-06, + "loss": 0.9813, + "num_input_tokens_seen": 112652430, + "step": 7520 + }, + { + "epoch": 0.5652337291447467, + "grad_norm": 0.8594002113112096, + "learning_rate": 1.676220644193222e-06, + "loss": 0.8462, + "num_input_tokens_seen": 112722195, + "step": 7521 + }, + { + "epoch": 0.5653088832105817, + "grad_norm": 1.6366430733968467, + "learning_rate": 1.6757402336023205e-06, + "loss": 0.9985, + "num_input_tokens_seen": 112744475, + "step": 7522 + }, + { + "epoch": 0.5653840372764166, + "grad_norm": 1.6068033207886359, + "learning_rate": 1.6752598422250512e-06, + "loss": 0.9655, + "num_input_tokens_seen": 112768440, + "step": 7523 + }, + { + "epoch": 0.5654591913422516, + "grad_norm": 1.4635329421610321, + "learning_rate": 1.6747794700898803e-06, + "loss": 1.0804, + "num_input_tokens_seen": 112791540, + "step": 7524 + }, + { + "epoch": 0.5655343454080866, + "grad_norm": 1.5976450576949803, + "learning_rate": 1.674299117225271e-06, + "loss": 0.9306, + "num_input_tokens_seen": 112816400, + "step": 7525 + }, + { + "epoch": 0.5656094994739216, + "grad_norm": 1.8584680348773197, + "learning_rate": 1.6738187836596858e-06, + "loss": 0.8235, + "num_input_tokens_seen": 112838265, + "step": 7526 + }, + { + "epoch": 0.5656846535397565, + "grad_norm": 1.635646714797747, + "learning_rate": 1.6733384694215872e-06, + "loss": 0.9926, + "num_input_tokens_seen": 112861630, + "step": 7527 + }, + { + "epoch": 0.5657598076055914, + "grad_norm": 0.7710105116729992, + "learning_rate": 1.6728581745394346e-06, + "loss": 0.7639, + "num_input_tokens_seen": 112940335, + "step": 7528 + }, + { + "epoch": 0.5658349616714264, + "grad_norm": 1.696405014792354, + "learning_rate": 1.6723778990416883e-06, + "loss": 1.0996, + "num_input_tokens_seen": 112962375, + "step": 7529 + }, + { + "epoch": 0.5659101157372614, + "grad_norm": 1.2130017805534559, + "learning_rate": 1.671897642956806e-06, + "loss": 0.9506, + "num_input_tokens_seen": 112989175, + "step": 7530 + }, + { + "epoch": 0.5659852698030964, + "grad_norm": 0.884478726723675, + "learning_rate": 1.6714174063132447e-06, + "loss": 0.8313, + "num_input_tokens_seen": 113062150, + "step": 7531 + }, + { + "epoch": 0.5660604238689313, + "grad_norm": 1.5399003435067247, + "learning_rate": 1.6709371891394605e-06, + "loss": 0.9974, + "num_input_tokens_seen": 113085605, + "step": 7532 + }, + { + "epoch": 0.5661355779347663, + "grad_norm": 1.4026084373835004, + "learning_rate": 1.6704569914639073e-06, + "loss": 0.9015, + "num_input_tokens_seen": 113109505, + "step": 7533 + }, + { + "epoch": 0.5662107320006012, + "grad_norm": 1.3129792473802504, + "learning_rate": 1.6699768133150399e-06, + "loss": 0.9672, + "num_input_tokens_seen": 113131585, + "step": 7534 + }, + { + "epoch": 0.5662858860664362, + "grad_norm": 1.3435863572658242, + "learning_rate": 1.6694966547213098e-06, + "loss": 0.9016, + "num_input_tokens_seen": 113155970, + "step": 7535 + }, + { + "epoch": 0.5663610401322712, + "grad_norm": 1.8804583889557338, + "learning_rate": 1.669016515711168e-06, + "loss": 0.918, + "num_input_tokens_seen": 113179115, + "step": 7536 + }, + { + "epoch": 0.5664361941981061, + "grad_norm": 1.44791265070882, + "learning_rate": 1.668536396313066e-06, + "loss": 1.0264, + "num_input_tokens_seen": 113201295, + "step": 7537 + }, + { + "epoch": 0.5665113482639411, + "grad_norm": 1.9051417271515674, + "learning_rate": 1.6680562965554508e-06, + "loss": 0.9324, + "num_input_tokens_seen": 113224975, + "step": 7538 + }, + { + "epoch": 0.566586502329776, + "grad_norm": 1.6144926312308856, + "learning_rate": 1.6675762164667717e-06, + "loss": 0.9815, + "num_input_tokens_seen": 113246680, + "step": 7539 + }, + { + "epoch": 0.5666616563956111, + "grad_norm": 1.493469326831987, + "learning_rate": 1.6670961560754744e-06, + "loss": 0.9173, + "num_input_tokens_seen": 113271100, + "step": 7540 + }, + { + "epoch": 0.566736810461446, + "grad_norm": 1.3018052020250457, + "learning_rate": 1.6666161154100052e-06, + "loss": 0.9643, + "num_input_tokens_seen": 113297000, + "step": 7541 + }, + { + "epoch": 0.5668119645272809, + "grad_norm": 0.7993453759855033, + "learning_rate": 1.6661360944988076e-06, + "loss": 0.7938, + "num_input_tokens_seen": 113374510, + "step": 7542 + }, + { + "epoch": 0.5668871185931159, + "grad_norm": 1.8549041210186572, + "learning_rate": 1.6656560933703244e-06, + "loss": 1.0683, + "num_input_tokens_seen": 113396830, + "step": 7543 + }, + { + "epoch": 0.5669622726589508, + "grad_norm": 1.4235649545935372, + "learning_rate": 1.6651761120529983e-06, + "loss": 0.8692, + "num_input_tokens_seen": 113421980, + "step": 7544 + }, + { + "epoch": 0.5670374267247859, + "grad_norm": 3.4201187434412685, + "learning_rate": 1.6646961505752696e-06, + "loss": 0.9509, + "num_input_tokens_seen": 113440935, + "step": 7545 + }, + { + "epoch": 0.5671125807906208, + "grad_norm": 1.4110699596729102, + "learning_rate": 1.6642162089655782e-06, + "loss": 1.0222, + "num_input_tokens_seen": 113463870, + "step": 7546 + }, + { + "epoch": 0.5671877348564557, + "grad_norm": 1.805922972961815, + "learning_rate": 1.6637362872523621e-06, + "loss": 0.8418, + "num_input_tokens_seen": 113491050, + "step": 7547 + }, + { + "epoch": 0.5672628889222907, + "grad_norm": 1.784617202809088, + "learning_rate": 1.6632563854640583e-06, + "loss": 1.0143, + "num_input_tokens_seen": 113513050, + "step": 7548 + }, + { + "epoch": 0.5673380429881256, + "grad_norm": 1.4115107152399022, + "learning_rate": 1.6627765036291034e-06, + "loss": 0.8877, + "num_input_tokens_seen": 113539515, + "step": 7549 + }, + { + "epoch": 0.5674131970539606, + "grad_norm": 1.4039180494013226, + "learning_rate": 1.6622966417759319e-06, + "loss": 0.9788, + "num_input_tokens_seen": 113563695, + "step": 7550 + }, + { + "epoch": 0.5674883511197956, + "grad_norm": 1.5133526471595828, + "learning_rate": 1.6618167999329778e-06, + "loss": 0.9076, + "num_input_tokens_seen": 113584715, + "step": 7551 + }, + { + "epoch": 0.5675635051856306, + "grad_norm": 1.7126965124609137, + "learning_rate": 1.6613369781286732e-06, + "loss": 0.9238, + "num_input_tokens_seen": 113608320, + "step": 7552 + }, + { + "epoch": 0.5676386592514655, + "grad_norm": 1.4710060000268075, + "learning_rate": 1.660857176391449e-06, + "loss": 0.9306, + "num_input_tokens_seen": 113632810, + "step": 7553 + }, + { + "epoch": 0.5677138133173004, + "grad_norm": 1.5933476856548265, + "learning_rate": 1.6603773947497364e-06, + "loss": 0.9139, + "num_input_tokens_seen": 113656440, + "step": 7554 + }, + { + "epoch": 0.5677889673831354, + "grad_norm": 1.6353487458268021, + "learning_rate": 1.6598976332319631e-06, + "loss": 0.9133, + "num_input_tokens_seen": 113680245, + "step": 7555 + }, + { + "epoch": 0.5678641214489704, + "grad_norm": 1.4921545853904212, + "learning_rate": 1.6594178918665578e-06, + "loss": 0.9804, + "num_input_tokens_seen": 113705280, + "step": 7556 + }, + { + "epoch": 0.5679392755148054, + "grad_norm": 1.3575399443508531, + "learning_rate": 1.6589381706819467e-06, + "loss": 0.9846, + "num_input_tokens_seen": 113734260, + "step": 7557 + }, + { + "epoch": 0.5680144295806403, + "grad_norm": 1.5490029081264358, + "learning_rate": 1.658458469706554e-06, + "loss": 0.9674, + "num_input_tokens_seen": 113758235, + "step": 7558 + }, + { + "epoch": 0.5680895836464753, + "grad_norm": 2.6139582967196984, + "learning_rate": 1.6579787889688062e-06, + "loss": 1.0505, + "num_input_tokens_seen": 113781320, + "step": 7559 + }, + { + "epoch": 0.5681647377123102, + "grad_norm": 1.50818427844006, + "learning_rate": 1.6574991284971235e-06, + "loss": 0.9113, + "num_input_tokens_seen": 113804305, + "step": 7560 + }, + { + "epoch": 0.5682398917781452, + "grad_norm": 1.5892332062229246, + "learning_rate": 1.6570194883199298e-06, + "loss": 0.906, + "num_input_tokens_seen": 113830240, + "step": 7561 + }, + { + "epoch": 0.5683150458439802, + "grad_norm": 1.2126036963412627, + "learning_rate": 1.6565398684656442e-06, + "loss": 0.9063, + "num_input_tokens_seen": 113855795, + "step": 7562 + }, + { + "epoch": 0.5683901999098151, + "grad_norm": 1.374502829053735, + "learning_rate": 1.6560602689626872e-06, + "loss": 0.8738, + "num_input_tokens_seen": 113878570, + "step": 7563 + }, + { + "epoch": 0.5684653539756501, + "grad_norm": 1.4767194392043546, + "learning_rate": 1.6555806898394764e-06, + "loss": 0.9391, + "num_input_tokens_seen": 113901355, + "step": 7564 + }, + { + "epoch": 0.568540508041485, + "grad_norm": 1.7934019940762214, + "learning_rate": 1.655101131124428e-06, + "loss": 0.9087, + "num_input_tokens_seen": 113925400, + "step": 7565 + }, + { + "epoch": 0.56861566210732, + "grad_norm": 1.5394133347056533, + "learning_rate": 1.6546215928459589e-06, + "loss": 0.8578, + "num_input_tokens_seen": 113948915, + "step": 7566 + }, + { + "epoch": 0.568690816173155, + "grad_norm": 0.8103905952267525, + "learning_rate": 1.6541420750324825e-06, + "loss": 0.7947, + "num_input_tokens_seen": 114031670, + "step": 7567 + }, + { + "epoch": 0.5687659702389899, + "grad_norm": 1.6727929135634485, + "learning_rate": 1.6536625777124128e-06, + "loss": 0.7977, + "num_input_tokens_seen": 114053380, + "step": 7568 + }, + { + "epoch": 0.5688411243048249, + "grad_norm": 1.1395137024039284, + "learning_rate": 1.6531831009141616e-06, + "loss": 0.8382, + "num_input_tokens_seen": 114119180, + "step": 7569 + }, + { + "epoch": 0.5689162783706598, + "grad_norm": 1.659084988698693, + "learning_rate": 1.6527036446661393e-06, + "loss": 0.8753, + "num_input_tokens_seen": 114143575, + "step": 7570 + }, + { + "epoch": 0.5689914324364949, + "grad_norm": 1.7181829730952474, + "learning_rate": 1.6522242089967564e-06, + "loss": 0.8463, + "num_input_tokens_seen": 114170595, + "step": 7571 + }, + { + "epoch": 0.5690665865023298, + "grad_norm": 1.4704363455590121, + "learning_rate": 1.6517447939344205e-06, + "loss": 0.8291, + "num_input_tokens_seen": 114193485, + "step": 7572 + }, + { + "epoch": 0.5691417405681647, + "grad_norm": 1.2795965644469534, + "learning_rate": 1.6512653995075393e-06, + "loss": 0.9048, + "num_input_tokens_seen": 114220235, + "step": 7573 + }, + { + "epoch": 0.5692168946339997, + "grad_norm": 1.5765806138479121, + "learning_rate": 1.650786025744519e-06, + "loss": 0.9619, + "num_input_tokens_seen": 114243625, + "step": 7574 + }, + { + "epoch": 0.5692920486998346, + "grad_norm": 1.5869734063423193, + "learning_rate": 1.6503066726737632e-06, + "loss": 0.9859, + "num_input_tokens_seen": 114269420, + "step": 7575 + }, + { + "epoch": 0.5693672027656697, + "grad_norm": 1.6577897022632067, + "learning_rate": 1.6498273403236764e-06, + "loss": 0.856, + "num_input_tokens_seen": 114293850, + "step": 7576 + }, + { + "epoch": 0.5694423568315046, + "grad_norm": 1.4396480599116823, + "learning_rate": 1.6493480287226605e-06, + "loss": 0.8289, + "num_input_tokens_seen": 114321305, + "step": 7577 + }, + { + "epoch": 0.5695175108973396, + "grad_norm": 1.4864158654601183, + "learning_rate": 1.6488687378991168e-06, + "loss": 0.9735, + "num_input_tokens_seen": 114346360, + "step": 7578 + }, + { + "epoch": 0.5695926649631745, + "grad_norm": 1.8446131480033268, + "learning_rate": 1.648389467881444e-06, + "loss": 0.9288, + "num_input_tokens_seen": 114366685, + "step": 7579 + }, + { + "epoch": 0.5696678190290094, + "grad_norm": 1.5089018054306937, + "learning_rate": 1.6479102186980428e-06, + "loss": 0.8869, + "num_input_tokens_seen": 114393300, + "step": 7580 + }, + { + "epoch": 0.5697429730948445, + "grad_norm": 0.7157449929864679, + "learning_rate": 1.6474309903773098e-06, + "loss": 0.7451, + "num_input_tokens_seen": 114473135, + "step": 7581 + }, + { + "epoch": 0.5698181271606794, + "grad_norm": 1.3625537056793746, + "learning_rate": 1.6469517829476391e-06, + "loss": 0.9868, + "num_input_tokens_seen": 114498770, + "step": 7582 + }, + { + "epoch": 0.5698932812265144, + "grad_norm": 1.4121644575210617, + "learning_rate": 1.6464725964374285e-06, + "loss": 0.9669, + "num_input_tokens_seen": 114524510, + "step": 7583 + }, + { + "epoch": 0.5699684352923493, + "grad_norm": 1.5281458941008306, + "learning_rate": 1.6459934308750694e-06, + "loss": 1.1166, + "num_input_tokens_seen": 114546260, + "step": 7584 + }, + { + "epoch": 0.5700435893581843, + "grad_norm": 1.21721983488371, + "learning_rate": 1.6455142862889557e-06, + "loss": 1.0018, + "num_input_tokens_seen": 114571240, + "step": 7585 + }, + { + "epoch": 0.5701187434240192, + "grad_norm": 0.7987376490609638, + "learning_rate": 1.6450351627074781e-06, + "loss": 0.7943, + "num_input_tokens_seen": 114653240, + "step": 7586 + }, + { + "epoch": 0.5701938974898542, + "grad_norm": 1.8363167878243016, + "learning_rate": 1.6445560601590257e-06, + "loss": 0.9405, + "num_input_tokens_seen": 114675570, + "step": 7587 + }, + { + "epoch": 0.5702690515556892, + "grad_norm": 1.6051541876484914, + "learning_rate": 1.6440769786719883e-06, + "loss": 0.8261, + "num_input_tokens_seen": 114701375, + "step": 7588 + }, + { + "epoch": 0.5703442056215241, + "grad_norm": 1.5211271324471918, + "learning_rate": 1.6435979182747526e-06, + "loss": 1.0002, + "num_input_tokens_seen": 114723695, + "step": 7589 + }, + { + "epoch": 0.5704193596873591, + "grad_norm": 1.5786582121515274, + "learning_rate": 1.6431188789957053e-06, + "loss": 0.898, + "num_input_tokens_seen": 114745540, + "step": 7590 + }, + { + "epoch": 0.570494513753194, + "grad_norm": 1.9503531967535501, + "learning_rate": 1.642639860863231e-06, + "loss": 0.8913, + "num_input_tokens_seen": 114767900, + "step": 7591 + }, + { + "epoch": 0.570569667819029, + "grad_norm": 1.196043501582707, + "learning_rate": 1.642160863905713e-06, + "loss": 0.9584, + "num_input_tokens_seen": 114795125, + "step": 7592 + }, + { + "epoch": 0.570644821884864, + "grad_norm": 1.9374394561168007, + "learning_rate": 1.6416818881515344e-06, + "loss": 0.8597, + "num_input_tokens_seen": 114816435, + "step": 7593 + }, + { + "epoch": 0.5707199759506989, + "grad_norm": 1.4969350177622789, + "learning_rate": 1.6412029336290755e-06, + "loss": 0.8882, + "num_input_tokens_seen": 114838160, + "step": 7594 + }, + { + "epoch": 0.5707951300165339, + "grad_norm": 1.4461284911497454, + "learning_rate": 1.6407240003667172e-06, + "loss": 0.9605, + "num_input_tokens_seen": 114862305, + "step": 7595 + }, + { + "epoch": 0.5708702840823688, + "grad_norm": 1.2353533959657836, + "learning_rate": 1.640245088392838e-06, + "loss": 0.9337, + "num_input_tokens_seen": 114888345, + "step": 7596 + }, + { + "epoch": 0.5709454381482039, + "grad_norm": 1.240667841097156, + "learning_rate": 1.6397661977358142e-06, + "loss": 0.9201, + "num_input_tokens_seen": 114914300, + "step": 7597 + }, + { + "epoch": 0.5710205922140388, + "grad_norm": 1.4943013821556481, + "learning_rate": 1.639287328424023e-06, + "loss": 1.0166, + "num_input_tokens_seen": 114938860, + "step": 7598 + }, + { + "epoch": 0.5710957462798737, + "grad_norm": 1.8031004165216498, + "learning_rate": 1.638808480485838e-06, + "loss": 0.9725, + "num_input_tokens_seen": 114962860, + "step": 7599 + }, + { + "epoch": 0.5711709003457087, + "grad_norm": 2.106499414363755, + "learning_rate": 1.638329653949635e-06, + "loss": 0.9015, + "num_input_tokens_seen": 114983960, + "step": 7600 + }, + { + "epoch": 0.5712460544115436, + "grad_norm": 0.8036030619313725, + "learning_rate": 1.6378508488437835e-06, + "loss": 0.8041, + "num_input_tokens_seen": 115056330, + "step": 7601 + }, + { + "epoch": 0.5713212084773787, + "grad_norm": 1.8368977085097364, + "learning_rate": 1.6373720651966569e-06, + "loss": 0.9704, + "num_input_tokens_seen": 115078055, + "step": 7602 + }, + { + "epoch": 0.5713963625432136, + "grad_norm": 0.7958075910336093, + "learning_rate": 1.6368933030366241e-06, + "loss": 0.7942, + "num_input_tokens_seen": 115152220, + "step": 7603 + }, + { + "epoch": 0.5714715166090486, + "grad_norm": 1.6353915348268218, + "learning_rate": 1.6364145623920528e-06, + "loss": 0.9572, + "num_input_tokens_seen": 115175215, + "step": 7604 + }, + { + "epoch": 0.5715466706748835, + "grad_norm": 1.511445723134216, + "learning_rate": 1.6359358432913118e-06, + "loss": 0.9733, + "num_input_tokens_seen": 115199295, + "step": 7605 + }, + { + "epoch": 0.5716218247407184, + "grad_norm": 0.7256929259662063, + "learning_rate": 1.6354571457627656e-06, + "loss": 0.8129, + "num_input_tokens_seen": 115277640, + "step": 7606 + }, + { + "epoch": 0.5716969788065535, + "grad_norm": 1.5694740296372574, + "learning_rate": 1.6349784698347797e-06, + "loss": 0.8305, + "num_input_tokens_seen": 115304805, + "step": 7607 + }, + { + "epoch": 0.5717721328723884, + "grad_norm": 0.7443619446526429, + "learning_rate": 1.6344998155357175e-06, + "loss": 0.7173, + "num_input_tokens_seen": 115384150, + "step": 7608 + }, + { + "epoch": 0.5718472869382234, + "grad_norm": 1.5048481159307285, + "learning_rate": 1.63402118289394e-06, + "loss": 0.9003, + "num_input_tokens_seen": 115408175, + "step": 7609 + }, + { + "epoch": 0.5719224410040583, + "grad_norm": 1.5536911421880866, + "learning_rate": 1.6335425719378097e-06, + "loss": 0.9384, + "num_input_tokens_seen": 115432615, + "step": 7610 + }, + { + "epoch": 0.5719975950698932, + "grad_norm": 0.7228618020190773, + "learning_rate": 1.6330639826956848e-06, + "loss": 0.8184, + "num_input_tokens_seen": 115511860, + "step": 7611 + }, + { + "epoch": 0.5720727491357283, + "grad_norm": 1.4163338139270503, + "learning_rate": 1.632585415195924e-06, + "loss": 0.9108, + "num_input_tokens_seen": 115537205, + "step": 7612 + }, + { + "epoch": 0.5721479032015632, + "grad_norm": 2.1418639447183208, + "learning_rate": 1.6321068694668846e-06, + "loss": 0.8888, + "num_input_tokens_seen": 115557855, + "step": 7613 + }, + { + "epoch": 0.5722230572673982, + "grad_norm": 2.1950315248141603, + "learning_rate": 1.6316283455369215e-06, + "loss": 0.9609, + "num_input_tokens_seen": 115581715, + "step": 7614 + }, + { + "epoch": 0.5722982113332331, + "grad_norm": 1.6210449946781607, + "learning_rate": 1.63114984343439e-06, + "loss": 0.9229, + "num_input_tokens_seen": 115604900, + "step": 7615 + }, + { + "epoch": 0.5723733653990681, + "grad_norm": 1.8695699902165956, + "learning_rate": 1.630671363187642e-06, + "loss": 0.9006, + "num_input_tokens_seen": 115627070, + "step": 7616 + }, + { + "epoch": 0.572448519464903, + "grad_norm": 1.8259403183951393, + "learning_rate": 1.6301929048250306e-06, + "loss": 0.8859, + "num_input_tokens_seen": 115652855, + "step": 7617 + }, + { + "epoch": 0.572523673530738, + "grad_norm": 1.8010868632387707, + "learning_rate": 1.6297144683749057e-06, + "loss": 1.0733, + "num_input_tokens_seen": 115673440, + "step": 7618 + }, + { + "epoch": 0.572598827596573, + "grad_norm": 1.5764612171013463, + "learning_rate": 1.6292360538656162e-06, + "loss": 0.9674, + "num_input_tokens_seen": 115697075, + "step": 7619 + }, + { + "epoch": 0.5726739816624079, + "grad_norm": 2.5358062992249075, + "learning_rate": 1.6287576613255105e-06, + "loss": 0.8441, + "num_input_tokens_seen": 115717660, + "step": 7620 + }, + { + "epoch": 0.5727491357282429, + "grad_norm": 1.3370215496047029, + "learning_rate": 1.6282792907829341e-06, + "loss": 0.9632, + "num_input_tokens_seen": 115741960, + "step": 7621 + }, + { + "epoch": 0.5728242897940778, + "grad_norm": 1.3120710035132217, + "learning_rate": 1.6278009422662345e-06, + "loss": 1.0304, + "num_input_tokens_seen": 115765365, + "step": 7622 + }, + { + "epoch": 0.5728994438599129, + "grad_norm": 1.4110008098808144, + "learning_rate": 1.6273226158037528e-06, + "loss": 0.8607, + "num_input_tokens_seen": 115794390, + "step": 7623 + }, + { + "epoch": 0.5729745979257478, + "grad_norm": 1.4071220660938568, + "learning_rate": 1.6268443114238345e-06, + "loss": 0.8893, + "num_input_tokens_seen": 115819045, + "step": 7624 + }, + { + "epoch": 0.5730497519915827, + "grad_norm": 1.3375252551965418, + "learning_rate": 1.6263660291548191e-06, + "loss": 0.9546, + "num_input_tokens_seen": 115841005, + "step": 7625 + }, + { + "epoch": 0.5731249060574177, + "grad_norm": 1.9471412856216894, + "learning_rate": 1.6258877690250472e-06, + "loss": 1.0009, + "num_input_tokens_seen": 115862695, + "step": 7626 + }, + { + "epoch": 0.5732000601232526, + "grad_norm": 1.529621815659361, + "learning_rate": 1.6254095310628578e-06, + "loss": 0.9987, + "num_input_tokens_seen": 115887820, + "step": 7627 + }, + { + "epoch": 0.5732752141890877, + "grad_norm": 1.4569531130868696, + "learning_rate": 1.6249313152965876e-06, + "loss": 1.0156, + "num_input_tokens_seen": 115911140, + "step": 7628 + }, + { + "epoch": 0.5733503682549226, + "grad_norm": 2.2913806621216644, + "learning_rate": 1.6244531217545738e-06, + "loss": 0.9811, + "num_input_tokens_seen": 115934155, + "step": 7629 + }, + { + "epoch": 0.5734255223207576, + "grad_norm": 0.7609625559604043, + "learning_rate": 1.6239749504651505e-06, + "loss": 0.7913, + "num_input_tokens_seen": 116016005, + "step": 7630 + }, + { + "epoch": 0.5735006763865925, + "grad_norm": 1.2266798509963857, + "learning_rate": 1.6234968014566509e-06, + "loss": 0.9666, + "num_input_tokens_seen": 116045905, + "step": 7631 + }, + { + "epoch": 0.5735758304524274, + "grad_norm": 1.5291472808828683, + "learning_rate": 1.6230186747574077e-06, + "loss": 0.9137, + "num_input_tokens_seen": 116069790, + "step": 7632 + }, + { + "epoch": 0.5736509845182625, + "grad_norm": 1.3933448767221905, + "learning_rate": 1.6225405703957515e-06, + "loss": 0.901, + "num_input_tokens_seen": 116094030, + "step": 7633 + }, + { + "epoch": 0.5737261385840974, + "grad_norm": 1.5307986314329354, + "learning_rate": 1.6220624884000123e-06, + "loss": 0.9694, + "num_input_tokens_seen": 116111870, + "step": 7634 + }, + { + "epoch": 0.5738012926499324, + "grad_norm": 1.6495328039472927, + "learning_rate": 1.6215844287985178e-06, + "loss": 0.8884, + "num_input_tokens_seen": 116137135, + "step": 7635 + }, + { + "epoch": 0.5738764467157673, + "grad_norm": 1.6429451658231358, + "learning_rate": 1.6211063916195945e-06, + "loss": 0.928, + "num_input_tokens_seen": 116159740, + "step": 7636 + }, + { + "epoch": 0.5739516007816022, + "grad_norm": 1.9554686256375122, + "learning_rate": 1.6206283768915687e-06, + "loss": 0.8957, + "num_input_tokens_seen": 116181905, + "step": 7637 + }, + { + "epoch": 0.5740267548474373, + "grad_norm": 1.879554684228325, + "learning_rate": 1.620150384642764e-06, + "loss": 0.9719, + "num_input_tokens_seen": 116204435, + "step": 7638 + }, + { + "epoch": 0.5741019089132722, + "grad_norm": 1.6433663852506766, + "learning_rate": 1.619672414901504e-06, + "loss": 0.9636, + "num_input_tokens_seen": 116225720, + "step": 7639 + }, + { + "epoch": 0.5741770629791072, + "grad_norm": 1.959259652814296, + "learning_rate": 1.6191944676961097e-06, + "loss": 0.9165, + "num_input_tokens_seen": 116248460, + "step": 7640 + }, + { + "epoch": 0.5742522170449421, + "grad_norm": 1.5948219434313036, + "learning_rate": 1.6187165430549011e-06, + "loss": 1.021, + "num_input_tokens_seen": 116272395, + "step": 7641 + }, + { + "epoch": 0.5743273711107771, + "grad_norm": 1.7010798642911082, + "learning_rate": 1.6182386410061976e-06, + "loss": 0.8811, + "num_input_tokens_seen": 116293850, + "step": 7642 + }, + { + "epoch": 0.574402525176612, + "grad_norm": 1.5165881213435688, + "learning_rate": 1.6177607615783158e-06, + "loss": 1.0015, + "num_input_tokens_seen": 116319495, + "step": 7643 + }, + { + "epoch": 0.574477679242447, + "grad_norm": 1.5268686633733926, + "learning_rate": 1.6172829047995733e-06, + "loss": 0.9467, + "num_input_tokens_seen": 116342635, + "step": 7644 + }, + { + "epoch": 0.574552833308282, + "grad_norm": 1.5568499071448973, + "learning_rate": 1.616805070698284e-06, + "loss": 0.8899, + "num_input_tokens_seen": 116365110, + "step": 7645 + }, + { + "epoch": 0.5746279873741169, + "grad_norm": 1.8676597604500782, + "learning_rate": 1.6163272593027615e-06, + "loss": 0.8854, + "num_input_tokens_seen": 116390955, + "step": 7646 + }, + { + "epoch": 0.5747031414399519, + "grad_norm": 1.6349977900405555, + "learning_rate": 1.6158494706413187e-06, + "loss": 0.9361, + "num_input_tokens_seen": 116412010, + "step": 7647 + }, + { + "epoch": 0.5747782955057869, + "grad_norm": 1.7480935202400847, + "learning_rate": 1.6153717047422648e-06, + "loss": 0.9736, + "num_input_tokens_seen": 116434655, + "step": 7648 + }, + { + "epoch": 0.5748534495716219, + "grad_norm": 1.4798469948936945, + "learning_rate": 1.614893961633911e-06, + "loss": 1.046, + "num_input_tokens_seen": 116456330, + "step": 7649 + }, + { + "epoch": 0.5749286036374568, + "grad_norm": 1.654028356168368, + "learning_rate": 1.6144162413445642e-06, + "loss": 1.0064, + "num_input_tokens_seen": 116476965, + "step": 7650 + }, + { + "epoch": 0.5750037577032917, + "grad_norm": 1.8420593786619914, + "learning_rate": 1.6139385439025319e-06, + "loss": 1.013, + "num_input_tokens_seen": 116500845, + "step": 7651 + }, + { + "epoch": 0.5750789117691267, + "grad_norm": 1.5583957938394044, + "learning_rate": 1.6134608693361193e-06, + "loss": 0.9096, + "num_input_tokens_seen": 116530080, + "step": 7652 + }, + { + "epoch": 0.5751540658349616, + "grad_norm": 1.25322079565849, + "learning_rate": 1.61298321767363e-06, + "loss": 0.9437, + "num_input_tokens_seen": 116553380, + "step": 7653 + }, + { + "epoch": 0.5752292199007967, + "grad_norm": 0.8056849251560173, + "learning_rate": 1.6125055889433674e-06, + "loss": 0.8187, + "num_input_tokens_seen": 116635550, + "step": 7654 + }, + { + "epoch": 0.5753043739666316, + "grad_norm": 1.5157031796142102, + "learning_rate": 1.612027983173632e-06, + "loss": 0.9921, + "num_input_tokens_seen": 116659545, + "step": 7655 + }, + { + "epoch": 0.5753795280324665, + "grad_norm": 1.3088788077902251, + "learning_rate": 1.6115504003927245e-06, + "loss": 0.9359, + "num_input_tokens_seen": 116683920, + "step": 7656 + }, + { + "epoch": 0.5754546820983015, + "grad_norm": 1.5697829079420558, + "learning_rate": 1.6110728406289436e-06, + "loss": 0.7941, + "num_input_tokens_seen": 116708285, + "step": 7657 + }, + { + "epoch": 0.5755298361641364, + "grad_norm": 1.43183012953228, + "learning_rate": 1.6105953039105855e-06, + "loss": 1.0021, + "num_input_tokens_seen": 116730975, + "step": 7658 + }, + { + "epoch": 0.5756049902299715, + "grad_norm": 1.717629639731647, + "learning_rate": 1.6101177902659474e-06, + "loss": 0.8809, + "num_input_tokens_seen": 116754965, + "step": 7659 + }, + { + "epoch": 0.5756801442958064, + "grad_norm": 1.625626516609115, + "learning_rate": 1.6096402997233225e-06, + "loss": 0.9302, + "num_input_tokens_seen": 116777810, + "step": 7660 + }, + { + "epoch": 0.5757552983616414, + "grad_norm": 1.6177120371635108, + "learning_rate": 1.6091628323110053e-06, + "loss": 0.8526, + "num_input_tokens_seen": 116801810, + "step": 7661 + }, + { + "epoch": 0.5758304524274763, + "grad_norm": 1.3340049284081876, + "learning_rate": 1.6086853880572868e-06, + "loss": 0.9068, + "num_input_tokens_seen": 116829195, + "step": 7662 + }, + { + "epoch": 0.5759056064933112, + "grad_norm": 1.6207859712298607, + "learning_rate": 1.6082079669904572e-06, + "loss": 0.8678, + "num_input_tokens_seen": 116851345, + "step": 7663 + }, + { + "epoch": 0.5759807605591463, + "grad_norm": 1.691243646219277, + "learning_rate": 1.607730569138806e-06, + "loss": 0.9432, + "num_input_tokens_seen": 116872550, + "step": 7664 + }, + { + "epoch": 0.5760559146249812, + "grad_norm": 1.2568476986872008, + "learning_rate": 1.60725319453062e-06, + "loss": 0.9374, + "num_input_tokens_seen": 116897630, + "step": 7665 + }, + { + "epoch": 0.5761310686908162, + "grad_norm": 1.5841220932251077, + "learning_rate": 1.606775843194187e-06, + "loss": 1.0082, + "num_input_tokens_seen": 116918510, + "step": 7666 + }, + { + "epoch": 0.5762062227566511, + "grad_norm": 1.7597224053464153, + "learning_rate": 1.6062985151577904e-06, + "loss": 1.039, + "num_input_tokens_seen": 116937315, + "step": 7667 + }, + { + "epoch": 0.5762813768224861, + "grad_norm": 1.5011211020394994, + "learning_rate": 1.605821210449715e-06, + "loss": 0.954, + "num_input_tokens_seen": 116959675, + "step": 7668 + }, + { + "epoch": 0.5763565308883211, + "grad_norm": 1.499488743438649, + "learning_rate": 1.6053439290982422e-06, + "loss": 1.0173, + "num_input_tokens_seen": 116981820, + "step": 7669 + }, + { + "epoch": 0.576431684954156, + "grad_norm": 1.421984322505754, + "learning_rate": 1.6048666711316526e-06, + "loss": 1.0046, + "num_input_tokens_seen": 117004565, + "step": 7670 + }, + { + "epoch": 0.576506839019991, + "grad_norm": 1.5945989554206224, + "learning_rate": 1.6043894365782262e-06, + "loss": 1.0226, + "num_input_tokens_seen": 117025395, + "step": 7671 + }, + { + "epoch": 0.5765819930858259, + "grad_norm": 1.5529583847355533, + "learning_rate": 1.60391222546624e-06, + "loss": 0.9228, + "num_input_tokens_seen": 117047325, + "step": 7672 + }, + { + "epoch": 0.5766571471516609, + "grad_norm": 1.8340472940817831, + "learning_rate": 1.6034350378239715e-06, + "loss": 0.9806, + "num_input_tokens_seen": 117069095, + "step": 7673 + }, + { + "epoch": 0.5767323012174959, + "grad_norm": 1.423428368146562, + "learning_rate": 1.6029578736796958e-06, + "loss": 0.9018, + "num_input_tokens_seen": 117093800, + "step": 7674 + }, + { + "epoch": 0.5768074552833309, + "grad_norm": 2.1899354728354683, + "learning_rate": 1.6024807330616858e-06, + "loss": 0.9272, + "num_input_tokens_seen": 117120440, + "step": 7675 + }, + { + "epoch": 0.5768826093491658, + "grad_norm": 1.4391512301879328, + "learning_rate": 1.6020036159982154e-06, + "loss": 0.9214, + "num_input_tokens_seen": 117149245, + "step": 7676 + }, + { + "epoch": 0.5769577634150007, + "grad_norm": 1.4544809229580922, + "learning_rate": 1.601526522517554e-06, + "loss": 1.002, + "num_input_tokens_seen": 117174200, + "step": 7677 + }, + { + "epoch": 0.5770329174808357, + "grad_norm": 1.4622401993009426, + "learning_rate": 1.6010494526479726e-06, + "loss": 0.9783, + "num_input_tokens_seen": 117197225, + "step": 7678 + }, + { + "epoch": 0.5771080715466707, + "grad_norm": 1.289209484621383, + "learning_rate": 1.6005724064177387e-06, + "loss": 0.9432, + "num_input_tokens_seen": 117219490, + "step": 7679 + }, + { + "epoch": 0.5771832256125057, + "grad_norm": 1.4623039505453432, + "learning_rate": 1.6000953838551187e-06, + "loss": 0.8334, + "num_input_tokens_seen": 117243385, + "step": 7680 + }, + { + "epoch": 0.5772583796783406, + "grad_norm": 2.1481147662427618, + "learning_rate": 1.5996183849883793e-06, + "loss": 0.9313, + "num_input_tokens_seen": 117262785, + "step": 7681 + }, + { + "epoch": 0.5773335337441755, + "grad_norm": 1.8206604719127606, + "learning_rate": 1.599141409845783e-06, + "loss": 0.9573, + "num_input_tokens_seen": 117284555, + "step": 7682 + }, + { + "epoch": 0.5774086878100105, + "grad_norm": 1.5811055053310128, + "learning_rate": 1.5986644584555937e-06, + "loss": 0.949, + "num_input_tokens_seen": 117305705, + "step": 7683 + }, + { + "epoch": 0.5774838418758454, + "grad_norm": 1.4856850715531664, + "learning_rate": 1.598187530846072e-06, + "loss": 0.8615, + "num_input_tokens_seen": 117331235, + "step": 7684 + }, + { + "epoch": 0.5775589959416805, + "grad_norm": 2.0452472533564396, + "learning_rate": 1.5977106270454775e-06, + "loss": 0.8615, + "num_input_tokens_seen": 117357155, + "step": 7685 + }, + { + "epoch": 0.5776341500075154, + "grad_norm": 1.4033412095586577, + "learning_rate": 1.597233747082069e-06, + "loss": 0.9742, + "num_input_tokens_seen": 117383440, + "step": 7686 + }, + { + "epoch": 0.5777093040733504, + "grad_norm": 1.603844110666679, + "learning_rate": 1.5967568909841026e-06, + "loss": 0.9331, + "num_input_tokens_seen": 117407990, + "step": 7687 + }, + { + "epoch": 0.5777844581391853, + "grad_norm": 1.3517533536897275, + "learning_rate": 1.5962800587798352e-06, + "loss": 0.9499, + "num_input_tokens_seen": 117433120, + "step": 7688 + }, + { + "epoch": 0.5778596122050202, + "grad_norm": 1.44561203610635, + "learning_rate": 1.59580325049752e-06, + "loss": 0.9647, + "num_input_tokens_seen": 117456925, + "step": 7689 + }, + { + "epoch": 0.5779347662708553, + "grad_norm": 1.7509075944788424, + "learning_rate": 1.5953264661654104e-06, + "loss": 0.976, + "num_input_tokens_seen": 117476275, + "step": 7690 + }, + { + "epoch": 0.5780099203366902, + "grad_norm": 1.6236065245122042, + "learning_rate": 1.5948497058117574e-06, + "loss": 0.8897, + "num_input_tokens_seen": 117499445, + "step": 7691 + }, + { + "epoch": 0.5780850744025252, + "grad_norm": 1.6250604471554835, + "learning_rate": 1.59437296946481e-06, + "loss": 0.8699, + "num_input_tokens_seen": 117522980, + "step": 7692 + }, + { + "epoch": 0.5781602284683601, + "grad_norm": 1.6790874740188804, + "learning_rate": 1.593896257152818e-06, + "loss": 0.8502, + "num_input_tokens_seen": 117547320, + "step": 7693 + }, + { + "epoch": 0.5782353825341952, + "grad_norm": 1.5813584392290456, + "learning_rate": 1.5934195689040276e-06, + "loss": 0.9262, + "num_input_tokens_seen": 117576815, + "step": 7694 + }, + { + "epoch": 0.5783105366000301, + "grad_norm": 1.7626792167287972, + "learning_rate": 1.592942904746685e-06, + "loss": 0.9169, + "num_input_tokens_seen": 117603480, + "step": 7695 + }, + { + "epoch": 0.578385690665865, + "grad_norm": 1.8861873028284646, + "learning_rate": 1.592466264709034e-06, + "loss": 0.9846, + "num_input_tokens_seen": 117626620, + "step": 7696 + }, + { + "epoch": 0.5784608447317, + "grad_norm": 1.541762254518117, + "learning_rate": 1.5919896488193166e-06, + "loss": 1.0167, + "num_input_tokens_seen": 117649905, + "step": 7697 + }, + { + "epoch": 0.5785359987975349, + "grad_norm": 0.8780829409372491, + "learning_rate": 1.5915130571057755e-06, + "loss": 0.8754, + "num_input_tokens_seen": 117717030, + "step": 7698 + }, + { + "epoch": 0.57861115286337, + "grad_norm": 0.8103868441418599, + "learning_rate": 1.5910364895966498e-06, + "loss": 0.7949, + "num_input_tokens_seen": 117803645, + "step": 7699 + }, + { + "epoch": 0.5786863069292049, + "grad_norm": 1.6995478870348746, + "learning_rate": 1.5905599463201785e-06, + "loss": 0.9751, + "num_input_tokens_seen": 117825945, + "step": 7700 + }, + { + "epoch": 0.5787614609950398, + "grad_norm": 1.620802372863102, + "learning_rate": 1.590083427304598e-06, + "loss": 0.7342, + "num_input_tokens_seen": 117849315, + "step": 7701 + }, + { + "epoch": 0.5788366150608748, + "grad_norm": 1.8482671457633375, + "learning_rate": 1.5896069325781435e-06, + "loss": 0.8683, + "num_input_tokens_seen": 117872750, + "step": 7702 + }, + { + "epoch": 0.5789117691267097, + "grad_norm": 1.718305079353967, + "learning_rate": 1.5891304621690508e-06, + "loss": 0.9432, + "num_input_tokens_seen": 117895375, + "step": 7703 + }, + { + "epoch": 0.5789869231925447, + "grad_norm": 1.523564103563495, + "learning_rate": 1.5886540161055507e-06, + "loss": 0.9419, + "num_input_tokens_seen": 117916805, + "step": 7704 + }, + { + "epoch": 0.5790620772583797, + "grad_norm": 1.8486993599353285, + "learning_rate": 1.5881775944158755e-06, + "loss": 0.8353, + "num_input_tokens_seen": 117941335, + "step": 7705 + }, + { + "epoch": 0.5791372313242147, + "grad_norm": 1.4544468271889892, + "learning_rate": 1.5877011971282553e-06, + "loss": 0.9929, + "num_input_tokens_seen": 117963755, + "step": 7706 + }, + { + "epoch": 0.5792123853900496, + "grad_norm": 1.3376341191009395, + "learning_rate": 1.5872248242709168e-06, + "loss": 0.9057, + "num_input_tokens_seen": 117988260, + "step": 7707 + }, + { + "epoch": 0.5792875394558845, + "grad_norm": 1.8313388451385781, + "learning_rate": 1.5867484758720894e-06, + "loss": 0.9349, + "num_input_tokens_seen": 118012430, + "step": 7708 + }, + { + "epoch": 0.5793626935217195, + "grad_norm": 1.6469172115815625, + "learning_rate": 1.5862721519599963e-06, + "loss": 0.9793, + "num_input_tokens_seen": 118036420, + "step": 7709 + }, + { + "epoch": 0.5794378475875545, + "grad_norm": 1.6901909600666691, + "learning_rate": 1.585795852562863e-06, + "loss": 0.9153, + "num_input_tokens_seen": 118061315, + "step": 7710 + }, + { + "epoch": 0.5795130016533895, + "grad_norm": 1.6339834294873847, + "learning_rate": 1.585319577708911e-06, + "loss": 0.8622, + "num_input_tokens_seen": 118082505, + "step": 7711 + }, + { + "epoch": 0.5795881557192244, + "grad_norm": 0.7960662477663681, + "learning_rate": 1.5848433274263627e-06, + "loss": 0.8056, + "num_input_tokens_seen": 118166360, + "step": 7712 + }, + { + "epoch": 0.5796633097850594, + "grad_norm": 2.1053199503165034, + "learning_rate": 1.5843671017434366e-06, + "loss": 0.9064, + "num_input_tokens_seen": 118191020, + "step": 7713 + }, + { + "epoch": 0.5797384638508943, + "grad_norm": 2.4088713994140067, + "learning_rate": 1.583890900688351e-06, + "loss": 0.9136, + "num_input_tokens_seen": 118215215, + "step": 7714 + }, + { + "epoch": 0.5798136179167293, + "grad_norm": 2.1758858948340207, + "learning_rate": 1.5834147242893234e-06, + "loss": 0.9248, + "num_input_tokens_seen": 118238105, + "step": 7715 + }, + { + "epoch": 0.5798887719825643, + "grad_norm": 1.5103194983786103, + "learning_rate": 1.5829385725745684e-06, + "loss": 0.8957, + "num_input_tokens_seen": 118262895, + "step": 7716 + }, + { + "epoch": 0.5799639260483992, + "grad_norm": 1.8685166484365665, + "learning_rate": 1.5824624455723e-06, + "loss": 0.9656, + "num_input_tokens_seen": 118287975, + "step": 7717 + }, + { + "epoch": 0.5800390801142342, + "grad_norm": 1.5053874066582882, + "learning_rate": 1.581986343310731e-06, + "loss": 0.98, + "num_input_tokens_seen": 118311825, + "step": 7718 + }, + { + "epoch": 0.5801142341800691, + "grad_norm": 1.389149558189472, + "learning_rate": 1.581510265818071e-06, + "loss": 0.8099, + "num_input_tokens_seen": 118337835, + "step": 7719 + }, + { + "epoch": 0.5801893882459042, + "grad_norm": 1.7664781382871018, + "learning_rate": 1.5810342131225308e-06, + "loss": 0.9144, + "num_input_tokens_seen": 118361595, + "step": 7720 + }, + { + "epoch": 0.5802645423117391, + "grad_norm": 2.055873977925289, + "learning_rate": 1.5805581852523176e-06, + "loss": 0.8587, + "num_input_tokens_seen": 118382295, + "step": 7721 + }, + { + "epoch": 0.580339696377574, + "grad_norm": 1.6143566177508992, + "learning_rate": 1.5800821822356383e-06, + "loss": 0.8433, + "num_input_tokens_seen": 118410870, + "step": 7722 + }, + { + "epoch": 0.580414850443409, + "grad_norm": 1.9818348884673729, + "learning_rate": 1.5796062041006978e-06, + "loss": 1.0243, + "num_input_tokens_seen": 118429645, + "step": 7723 + }, + { + "epoch": 0.5804900045092439, + "grad_norm": 0.8780991981489654, + "learning_rate": 1.579130250875699e-06, + "loss": 0.867, + "num_input_tokens_seen": 118498800, + "step": 7724 + }, + { + "epoch": 0.580565158575079, + "grad_norm": 1.776400540098002, + "learning_rate": 1.578654322588845e-06, + "loss": 0.9156, + "num_input_tokens_seen": 118522545, + "step": 7725 + }, + { + "epoch": 0.5806403126409139, + "grad_norm": 1.6087151072644723, + "learning_rate": 1.5781784192683351e-06, + "loss": 0.9633, + "num_input_tokens_seen": 118543990, + "step": 7726 + }, + { + "epoch": 0.5807154667067488, + "grad_norm": 1.4340194361179568, + "learning_rate": 1.57770254094237e-06, + "loss": 0.9501, + "num_input_tokens_seen": 118568325, + "step": 7727 + }, + { + "epoch": 0.5807906207725838, + "grad_norm": 1.6225876141184865, + "learning_rate": 1.577226687639146e-06, + "loss": 0.853, + "num_input_tokens_seen": 118591550, + "step": 7728 + }, + { + "epoch": 0.5808657748384187, + "grad_norm": 1.3722634393385225, + "learning_rate": 1.5767508593868588e-06, + "loss": 0.9642, + "num_input_tokens_seen": 118616015, + "step": 7729 + }, + { + "epoch": 0.5809409289042538, + "grad_norm": 1.5846058934976606, + "learning_rate": 1.5762750562137056e-06, + "loss": 1.0175, + "num_input_tokens_seen": 118637995, + "step": 7730 + }, + { + "epoch": 0.5810160829700887, + "grad_norm": 1.9024377443250844, + "learning_rate": 1.5757992781478762e-06, + "loss": 0.9326, + "num_input_tokens_seen": 118659060, + "step": 7731 + }, + { + "epoch": 0.5810912370359237, + "grad_norm": 1.5338617744537781, + "learning_rate": 1.575323525217565e-06, + "loss": 0.8743, + "num_input_tokens_seen": 118683885, + "step": 7732 + }, + { + "epoch": 0.5811663911017586, + "grad_norm": 2.5923000097433047, + "learning_rate": 1.5748477974509606e-06, + "loss": 0.9686, + "num_input_tokens_seen": 118707150, + "step": 7733 + }, + { + "epoch": 0.5812415451675935, + "grad_norm": 1.9165893069149205, + "learning_rate": 1.5743720948762527e-06, + "loss": 0.9341, + "num_input_tokens_seen": 118728710, + "step": 7734 + }, + { + "epoch": 0.5813166992334285, + "grad_norm": 1.3615443579248923, + "learning_rate": 1.573896417521628e-06, + "loss": 1.0087, + "num_input_tokens_seen": 118756660, + "step": 7735 + }, + { + "epoch": 0.5813918532992635, + "grad_norm": 1.5833483829117683, + "learning_rate": 1.5734207654152718e-06, + "loss": 0.9238, + "num_input_tokens_seen": 118781960, + "step": 7736 + }, + { + "epoch": 0.5814670073650985, + "grad_norm": 2.188014923526005, + "learning_rate": 1.572945138585369e-06, + "loss": 0.9833, + "num_input_tokens_seen": 118805720, + "step": 7737 + }, + { + "epoch": 0.5815421614309334, + "grad_norm": 1.4720180757801802, + "learning_rate": 1.572469537060102e-06, + "loss": 0.9515, + "num_input_tokens_seen": 118830560, + "step": 7738 + }, + { + "epoch": 0.5816173154967684, + "grad_norm": 0.8054316514922448, + "learning_rate": 1.5719939608676523e-06, + "loss": 0.7668, + "num_input_tokens_seen": 118907615, + "step": 7739 + }, + { + "epoch": 0.5816924695626033, + "grad_norm": 1.510077795841462, + "learning_rate": 1.5715184100361992e-06, + "loss": 0.9398, + "num_input_tokens_seen": 118931520, + "step": 7740 + }, + { + "epoch": 0.5817676236284383, + "grad_norm": 1.7006549246838885, + "learning_rate": 1.5710428845939207e-06, + "loss": 0.9458, + "num_input_tokens_seen": 118955170, + "step": 7741 + }, + { + "epoch": 0.5818427776942733, + "grad_norm": 1.6938850264551963, + "learning_rate": 1.5705673845689945e-06, + "loss": 1.0252, + "num_input_tokens_seen": 118977035, + "step": 7742 + }, + { + "epoch": 0.5819179317601082, + "grad_norm": 1.4587318784083376, + "learning_rate": 1.5700919099895943e-06, + "loss": 0.9336, + "num_input_tokens_seen": 118999685, + "step": 7743 + }, + { + "epoch": 0.5819930858259432, + "grad_norm": 1.525585284249436, + "learning_rate": 1.5696164608838956e-06, + "loss": 1.0374, + "num_input_tokens_seen": 119022940, + "step": 7744 + }, + { + "epoch": 0.5820682398917781, + "grad_norm": 1.5064230728436656, + "learning_rate": 1.5691410372800696e-06, + "loss": 0.9504, + "num_input_tokens_seen": 119047955, + "step": 7745 + }, + { + "epoch": 0.582143393957613, + "grad_norm": 1.590246426061828, + "learning_rate": 1.5686656392062863e-06, + "loss": 1.0106, + "num_input_tokens_seen": 119071440, + "step": 7746 + }, + { + "epoch": 0.5822185480234481, + "grad_norm": 1.7567917636409014, + "learning_rate": 1.5681902666907161e-06, + "loss": 0.844, + "num_input_tokens_seen": 119095630, + "step": 7747 + }, + { + "epoch": 0.582293702089283, + "grad_norm": 1.8630547535785262, + "learning_rate": 1.5677149197615257e-06, + "loss": 0.9803, + "num_input_tokens_seen": 119120225, + "step": 7748 + }, + { + "epoch": 0.582368856155118, + "grad_norm": 1.9918247266588835, + "learning_rate": 1.567239598446882e-06, + "loss": 0.8795, + "num_input_tokens_seen": 119144460, + "step": 7749 + }, + { + "epoch": 0.5824440102209529, + "grad_norm": 2.005580745314574, + "learning_rate": 1.5667643027749492e-06, + "loss": 0.8131, + "num_input_tokens_seen": 119166335, + "step": 7750 + }, + { + "epoch": 0.582519164286788, + "grad_norm": 1.5187546184453875, + "learning_rate": 1.5662890327738897e-06, + "loss": 0.8596, + "num_input_tokens_seen": 119189465, + "step": 7751 + }, + { + "epoch": 0.5825943183526229, + "grad_norm": 1.5790147444922655, + "learning_rate": 1.5658137884718672e-06, + "loss": 0.9359, + "num_input_tokens_seen": 119209400, + "step": 7752 + }, + { + "epoch": 0.5826694724184578, + "grad_norm": 1.7622758952844086, + "learning_rate": 1.565338569897039e-06, + "loss": 0.8363, + "num_input_tokens_seen": 119232245, + "step": 7753 + }, + { + "epoch": 0.5827446264842928, + "grad_norm": 1.515160164880627, + "learning_rate": 1.5648633770775656e-06, + "loss": 0.9306, + "num_input_tokens_seen": 119256845, + "step": 7754 + }, + { + "epoch": 0.5828197805501277, + "grad_norm": 1.3963550427709286, + "learning_rate": 1.564388210041603e-06, + "loss": 0.989, + "num_input_tokens_seen": 119279020, + "step": 7755 + }, + { + "epoch": 0.5828949346159628, + "grad_norm": 1.876239748220881, + "learning_rate": 1.5639130688173077e-06, + "loss": 1.0668, + "num_input_tokens_seen": 119298375, + "step": 7756 + }, + { + "epoch": 0.5829700886817977, + "grad_norm": 1.5634655829954895, + "learning_rate": 1.5634379534328326e-06, + "loss": 0.967, + "num_input_tokens_seen": 119321705, + "step": 7757 + }, + { + "epoch": 0.5830452427476327, + "grad_norm": 1.6757198580364792, + "learning_rate": 1.5629628639163304e-06, + "loss": 0.9147, + "num_input_tokens_seen": 119344595, + "step": 7758 + }, + { + "epoch": 0.5831203968134676, + "grad_norm": 1.60192703192333, + "learning_rate": 1.5624878002959521e-06, + "loss": 0.842, + "num_input_tokens_seen": 119368810, + "step": 7759 + }, + { + "epoch": 0.5831955508793025, + "grad_norm": 1.8334124793684001, + "learning_rate": 1.5620127625998469e-06, + "loss": 0.9148, + "num_input_tokens_seen": 119392205, + "step": 7760 + }, + { + "epoch": 0.5832707049451376, + "grad_norm": 1.6844605517910358, + "learning_rate": 1.5615377508561628e-06, + "loss": 0.9337, + "num_input_tokens_seen": 119414050, + "step": 7761 + }, + { + "epoch": 0.5833458590109725, + "grad_norm": 0.8035390491622638, + "learning_rate": 1.561062765093046e-06, + "loss": 0.8373, + "num_input_tokens_seen": 119486290, + "step": 7762 + }, + { + "epoch": 0.5834210130768075, + "grad_norm": 1.4752388324376273, + "learning_rate": 1.560587805338641e-06, + "loss": 0.959, + "num_input_tokens_seen": 119510325, + "step": 7763 + }, + { + "epoch": 0.5834961671426424, + "grad_norm": 1.4411049297866534, + "learning_rate": 1.5601128716210915e-06, + "loss": 0.9614, + "num_input_tokens_seen": 119535915, + "step": 7764 + }, + { + "epoch": 0.5835713212084774, + "grad_norm": 1.4904563408529106, + "learning_rate": 1.5596379639685382e-06, + "loss": 0.9643, + "num_input_tokens_seen": 119558700, + "step": 7765 + }, + { + "epoch": 0.5836464752743123, + "grad_norm": 1.3621391168426273, + "learning_rate": 1.5591630824091224e-06, + "loss": 0.9219, + "num_input_tokens_seen": 119583905, + "step": 7766 + }, + { + "epoch": 0.5837216293401473, + "grad_norm": 1.359495267261434, + "learning_rate": 1.5586882269709819e-06, + "loss": 0.9272, + "num_input_tokens_seen": 119608775, + "step": 7767 + }, + { + "epoch": 0.5837967834059823, + "grad_norm": 1.5132045492054953, + "learning_rate": 1.5582133976822534e-06, + "loss": 0.9761, + "num_input_tokens_seen": 119630355, + "step": 7768 + }, + { + "epoch": 0.5838719374718172, + "grad_norm": 1.8208499491593249, + "learning_rate": 1.5577385945710732e-06, + "loss": 0.96, + "num_input_tokens_seen": 119652450, + "step": 7769 + }, + { + "epoch": 0.5839470915376522, + "grad_norm": 0.9053609038301988, + "learning_rate": 1.5572638176655742e-06, + "loss": 0.8969, + "num_input_tokens_seen": 119717530, + "step": 7770 + }, + { + "epoch": 0.5840222456034871, + "grad_norm": 1.390870490533014, + "learning_rate": 1.5567890669938905e-06, + "loss": 0.9657, + "num_input_tokens_seen": 119742735, + "step": 7771 + }, + { + "epoch": 0.5840973996693221, + "grad_norm": 1.7983940669431935, + "learning_rate": 1.5563143425841512e-06, + "loss": 0.8534, + "num_input_tokens_seen": 119767340, + "step": 7772 + }, + { + "epoch": 0.5841725537351571, + "grad_norm": 1.569998714543685, + "learning_rate": 1.5558396444644854e-06, + "loss": 0.9846, + "num_input_tokens_seen": 119790660, + "step": 7773 + }, + { + "epoch": 0.584247707800992, + "grad_norm": 1.4400366088399261, + "learning_rate": 1.5553649726630222e-06, + "loss": 0.9446, + "num_input_tokens_seen": 119812870, + "step": 7774 + }, + { + "epoch": 0.584322861866827, + "grad_norm": 1.4865658303063718, + "learning_rate": 1.5548903272078865e-06, + "loss": 0.9135, + "num_input_tokens_seen": 119834910, + "step": 7775 + }, + { + "epoch": 0.5843980159326619, + "grad_norm": 1.477607155672353, + "learning_rate": 1.554415708127204e-06, + "loss": 0.8487, + "num_input_tokens_seen": 119858140, + "step": 7776 + }, + { + "epoch": 0.584473169998497, + "grad_norm": 1.426644260031486, + "learning_rate": 1.5539411154490967e-06, + "loss": 0.9571, + "num_input_tokens_seen": 119881290, + "step": 7777 + }, + { + "epoch": 0.5845483240643319, + "grad_norm": 1.6084855223417107, + "learning_rate": 1.5534665492016865e-06, + "loss": 0.9089, + "num_input_tokens_seen": 119904195, + "step": 7778 + }, + { + "epoch": 0.5846234781301668, + "grad_norm": 1.8093011185447099, + "learning_rate": 1.5529920094130932e-06, + "loss": 1.0188, + "num_input_tokens_seen": 119924130, + "step": 7779 + }, + { + "epoch": 0.5846986321960018, + "grad_norm": 1.5500206607549107, + "learning_rate": 1.552517496111435e-06, + "loss": 0.7899, + "num_input_tokens_seen": 119948520, + "step": 7780 + }, + { + "epoch": 0.5847737862618367, + "grad_norm": 2.4834338630792683, + "learning_rate": 1.5520430093248286e-06, + "loss": 0.8709, + "num_input_tokens_seen": 119971460, + "step": 7781 + }, + { + "epoch": 0.5848489403276718, + "grad_norm": 1.5522742783350705, + "learning_rate": 1.5515685490813891e-06, + "loss": 0.8899, + "num_input_tokens_seen": 119993770, + "step": 7782 + }, + { + "epoch": 0.5849240943935067, + "grad_norm": 1.8759515255275523, + "learning_rate": 1.5510941154092304e-06, + "loss": 0.9472, + "num_input_tokens_seen": 120015050, + "step": 7783 + }, + { + "epoch": 0.5849992484593417, + "grad_norm": 1.666328928423805, + "learning_rate": 1.5506197083364647e-06, + "loss": 0.975, + "num_input_tokens_seen": 120038910, + "step": 7784 + }, + { + "epoch": 0.5850744025251766, + "grad_norm": 1.6004009966226729, + "learning_rate": 1.5501453278912013e-06, + "loss": 0.899, + "num_input_tokens_seen": 120063155, + "step": 7785 + }, + { + "epoch": 0.5851495565910115, + "grad_norm": 1.6056033632595796, + "learning_rate": 1.5496709741015505e-06, + "loss": 0.9239, + "num_input_tokens_seen": 120087460, + "step": 7786 + }, + { + "epoch": 0.5852247106568466, + "grad_norm": 1.4314681673977494, + "learning_rate": 1.5491966469956187e-06, + "loss": 0.9134, + "num_input_tokens_seen": 120116100, + "step": 7787 + }, + { + "epoch": 0.5852998647226815, + "grad_norm": 0.7514590138586601, + "learning_rate": 1.5487223466015118e-06, + "loss": 0.8319, + "num_input_tokens_seen": 120192465, + "step": 7788 + }, + { + "epoch": 0.5853750187885165, + "grad_norm": 2.216587893680595, + "learning_rate": 1.5482480729473339e-06, + "loss": 1.0557, + "num_input_tokens_seen": 120213285, + "step": 7789 + }, + { + "epoch": 0.5854501728543514, + "grad_norm": 1.3790897755159, + "learning_rate": 1.5477738260611875e-06, + "loss": 0.9475, + "num_input_tokens_seen": 120237740, + "step": 7790 + }, + { + "epoch": 0.5855253269201863, + "grad_norm": 1.5064828969628405, + "learning_rate": 1.5472996059711738e-06, + "loss": 0.9464, + "num_input_tokens_seen": 120259835, + "step": 7791 + }, + { + "epoch": 0.5856004809860214, + "grad_norm": 1.5959255852524359, + "learning_rate": 1.546825412705391e-06, + "loss": 0.9661, + "num_input_tokens_seen": 120285250, + "step": 7792 + }, + { + "epoch": 0.5856756350518563, + "grad_norm": 1.6353559624487344, + "learning_rate": 1.5463512462919393e-06, + "loss": 0.9342, + "num_input_tokens_seen": 120308300, + "step": 7793 + }, + { + "epoch": 0.5857507891176913, + "grad_norm": 1.759475328420433, + "learning_rate": 1.5458771067589128e-06, + "loss": 0.9097, + "num_input_tokens_seen": 120331785, + "step": 7794 + }, + { + "epoch": 0.5858259431835262, + "grad_norm": 1.6128691731211469, + "learning_rate": 1.545402994134406e-06, + "loss": 1.0074, + "num_input_tokens_seen": 120352350, + "step": 7795 + }, + { + "epoch": 0.5859010972493612, + "grad_norm": 1.3428433930640975, + "learning_rate": 1.544928908446513e-06, + "loss": 0.9573, + "num_input_tokens_seen": 120376430, + "step": 7796 + }, + { + "epoch": 0.5859762513151962, + "grad_norm": 1.3749890760508001, + "learning_rate": 1.544454849723325e-06, + "loss": 0.8333, + "num_input_tokens_seen": 120400995, + "step": 7797 + }, + { + "epoch": 0.5860514053810311, + "grad_norm": 1.565206624854418, + "learning_rate": 1.5439808179929316e-06, + "loss": 0.8766, + "num_input_tokens_seen": 120429120, + "step": 7798 + }, + { + "epoch": 0.5861265594468661, + "grad_norm": 1.5125291017617306, + "learning_rate": 1.5435068132834204e-06, + "loss": 0.9035, + "num_input_tokens_seen": 120451725, + "step": 7799 + }, + { + "epoch": 0.586201713512701, + "grad_norm": 1.50017125423665, + "learning_rate": 1.543032835622879e-06, + "loss": 0.8773, + "num_input_tokens_seen": 120475255, + "step": 7800 + }, + { + "epoch": 0.586276867578536, + "grad_norm": 1.3610508115001771, + "learning_rate": 1.5425588850393922e-06, + "loss": 0.9496, + "num_input_tokens_seen": 120498305, + "step": 7801 + }, + { + "epoch": 0.586352021644371, + "grad_norm": 1.424300420264035, + "learning_rate": 1.5420849615610424e-06, + "loss": 0.9414, + "num_input_tokens_seen": 120521885, + "step": 7802 + }, + { + "epoch": 0.586427175710206, + "grad_norm": 1.6646481768312376, + "learning_rate": 1.541611065215913e-06, + "loss": 0.8716, + "num_input_tokens_seen": 120544935, + "step": 7803 + }, + { + "epoch": 0.5865023297760409, + "grad_norm": 1.5563652915072734, + "learning_rate": 1.5411371960320822e-06, + "loss": 1.022, + "num_input_tokens_seen": 120565255, + "step": 7804 + }, + { + "epoch": 0.5865774838418758, + "grad_norm": 1.5261426334591872, + "learning_rate": 1.5406633540376307e-06, + "loss": 0.9044, + "num_input_tokens_seen": 120588595, + "step": 7805 + }, + { + "epoch": 0.5866526379077108, + "grad_norm": 1.7769724028273923, + "learning_rate": 1.5401895392606339e-06, + "loss": 0.9563, + "num_input_tokens_seen": 120609675, + "step": 7806 + }, + { + "epoch": 0.5867277919735457, + "grad_norm": 1.6416016351176739, + "learning_rate": 1.5397157517291674e-06, + "loss": 0.9788, + "num_input_tokens_seen": 120632135, + "step": 7807 + }, + { + "epoch": 0.5868029460393808, + "grad_norm": 1.4109401056512745, + "learning_rate": 1.5392419914713054e-06, + "loss": 1.0382, + "num_input_tokens_seen": 120655420, + "step": 7808 + }, + { + "epoch": 0.5868781001052157, + "grad_norm": 1.5571190303514426, + "learning_rate": 1.5387682585151195e-06, + "loss": 0.9365, + "num_input_tokens_seen": 120679740, + "step": 7809 + }, + { + "epoch": 0.5869532541710507, + "grad_norm": 1.4608696090508686, + "learning_rate": 1.538294552888681e-06, + "loss": 0.9789, + "num_input_tokens_seen": 120703460, + "step": 7810 + }, + { + "epoch": 0.5870284082368856, + "grad_norm": 1.684544129099051, + "learning_rate": 1.537820874620058e-06, + "loss": 0.828, + "num_input_tokens_seen": 120727910, + "step": 7811 + }, + { + "epoch": 0.5871035623027205, + "grad_norm": 1.5461725508861057, + "learning_rate": 1.537347223737318e-06, + "loss": 0.9853, + "num_input_tokens_seen": 120751795, + "step": 7812 + }, + { + "epoch": 0.5871787163685556, + "grad_norm": 0.8481890853924362, + "learning_rate": 1.5368736002685266e-06, + "loss": 0.7929, + "num_input_tokens_seen": 120822435, + "step": 7813 + }, + { + "epoch": 0.5872538704343905, + "grad_norm": 1.6107040398905919, + "learning_rate": 1.5364000042417468e-06, + "loss": 0.8445, + "num_input_tokens_seen": 120846605, + "step": 7814 + }, + { + "epoch": 0.5873290245002255, + "grad_norm": 1.966705956578439, + "learning_rate": 1.5359264356850435e-06, + "loss": 0.9453, + "num_input_tokens_seen": 120869640, + "step": 7815 + }, + { + "epoch": 0.5874041785660604, + "grad_norm": 1.588004207888969, + "learning_rate": 1.5354528946264757e-06, + "loss": 0.8573, + "num_input_tokens_seen": 120891420, + "step": 7816 + }, + { + "epoch": 0.5874793326318953, + "grad_norm": 1.3945349257818764, + "learning_rate": 1.534979381094102e-06, + "loss": 0.8731, + "num_input_tokens_seen": 120915245, + "step": 7817 + }, + { + "epoch": 0.5875544866977304, + "grad_norm": 1.9064068338883982, + "learning_rate": 1.534505895115981e-06, + "loss": 0.9452, + "num_input_tokens_seen": 120936110, + "step": 7818 + }, + { + "epoch": 0.5876296407635653, + "grad_norm": 1.744050197670709, + "learning_rate": 1.5340324367201681e-06, + "loss": 0.9115, + "num_input_tokens_seen": 120958080, + "step": 7819 + }, + { + "epoch": 0.5877047948294003, + "grad_norm": 1.8149902074460347, + "learning_rate": 1.533559005934718e-06, + "loss": 0.9145, + "num_input_tokens_seen": 120978875, + "step": 7820 + }, + { + "epoch": 0.5877799488952352, + "grad_norm": 1.5877870195097736, + "learning_rate": 1.5330856027876827e-06, + "loss": 1.0142, + "num_input_tokens_seen": 121002095, + "step": 7821 + }, + { + "epoch": 0.5878551029610702, + "grad_norm": 1.863100566940545, + "learning_rate": 1.5326122273071133e-06, + "loss": 0.993, + "num_input_tokens_seen": 121027140, + "step": 7822 + }, + { + "epoch": 0.5879302570269052, + "grad_norm": 5.534402240016249, + "learning_rate": 1.5321388795210597e-06, + "loss": 1.0252, + "num_input_tokens_seen": 121047175, + "step": 7823 + }, + { + "epoch": 0.5880054110927401, + "grad_norm": 1.7109875258008298, + "learning_rate": 1.5316655594575685e-06, + "loss": 0.9308, + "num_input_tokens_seen": 121071145, + "step": 7824 + }, + { + "epoch": 0.5880805651585751, + "grad_norm": 1.5256352930530472, + "learning_rate": 1.5311922671446864e-06, + "loss": 0.8404, + "num_input_tokens_seen": 121095525, + "step": 7825 + }, + { + "epoch": 0.58815571922441, + "grad_norm": 1.5752119618036726, + "learning_rate": 1.5307190026104574e-06, + "loss": 0.9856, + "num_input_tokens_seen": 121116560, + "step": 7826 + }, + { + "epoch": 0.588230873290245, + "grad_norm": 1.4452056845294767, + "learning_rate": 1.530245765882925e-06, + "loss": 0.9788, + "num_input_tokens_seen": 121140935, + "step": 7827 + }, + { + "epoch": 0.58830602735608, + "grad_norm": 1.4605155784353328, + "learning_rate": 1.5297725569901298e-06, + "loss": 0.984, + "num_input_tokens_seen": 121164635, + "step": 7828 + }, + { + "epoch": 0.588381181421915, + "grad_norm": 1.5618494586909326, + "learning_rate": 1.5292993759601107e-06, + "loss": 0.9471, + "num_input_tokens_seen": 121187830, + "step": 7829 + }, + { + "epoch": 0.5884563354877499, + "grad_norm": 1.5535073713429146, + "learning_rate": 1.5288262228209066e-06, + "loss": 0.8909, + "num_input_tokens_seen": 121213035, + "step": 7830 + }, + { + "epoch": 0.5885314895535848, + "grad_norm": 1.2642750074565938, + "learning_rate": 1.5283530976005524e-06, + "loss": 0.9357, + "num_input_tokens_seen": 121239410, + "step": 7831 + }, + { + "epoch": 0.5886066436194198, + "grad_norm": 1.5628791349103535, + "learning_rate": 1.5278800003270838e-06, + "loss": 0.965, + "num_input_tokens_seen": 121260510, + "step": 7832 + }, + { + "epoch": 0.5886817976852547, + "grad_norm": 1.624933314789132, + "learning_rate": 1.527406931028533e-06, + "loss": 0.8933, + "num_input_tokens_seen": 121282675, + "step": 7833 + }, + { + "epoch": 0.5887569517510898, + "grad_norm": 1.494449837795832, + "learning_rate": 1.5269338897329308e-06, + "loss": 0.9171, + "num_input_tokens_seen": 121308700, + "step": 7834 + }, + { + "epoch": 0.5888321058169247, + "grad_norm": 1.339967594039562, + "learning_rate": 1.5264608764683074e-06, + "loss": 0.9817, + "num_input_tokens_seen": 121332365, + "step": 7835 + }, + { + "epoch": 0.5889072598827596, + "grad_norm": 1.7419512122337446, + "learning_rate": 1.5259878912626896e-06, + "loss": 1.0359, + "num_input_tokens_seen": 121353245, + "step": 7836 + }, + { + "epoch": 0.5889824139485946, + "grad_norm": 1.59917185928485, + "learning_rate": 1.5255149341441053e-06, + "loss": 0.9035, + "num_input_tokens_seen": 121378645, + "step": 7837 + }, + { + "epoch": 0.5890575680144295, + "grad_norm": 1.4898578773152789, + "learning_rate": 1.5250420051405783e-06, + "loss": 0.9175, + "num_input_tokens_seen": 121402525, + "step": 7838 + }, + { + "epoch": 0.5891327220802646, + "grad_norm": 1.4186994703259994, + "learning_rate": 1.5245691042801302e-06, + "loss": 1.0767, + "num_input_tokens_seen": 121425275, + "step": 7839 + }, + { + "epoch": 0.5892078761460995, + "grad_norm": 1.5370892107754592, + "learning_rate": 1.524096231590784e-06, + "loss": 0.9451, + "num_input_tokens_seen": 121448095, + "step": 7840 + }, + { + "epoch": 0.5892830302119345, + "grad_norm": 1.7419264388248368, + "learning_rate": 1.523623387100558e-06, + "loss": 0.9905, + "num_input_tokens_seen": 121471625, + "step": 7841 + }, + { + "epoch": 0.5893581842777694, + "grad_norm": 1.3514299272555201, + "learning_rate": 1.5231505708374707e-06, + "loss": 1.0074, + "num_input_tokens_seen": 121496740, + "step": 7842 + }, + { + "epoch": 0.5894333383436043, + "grad_norm": 1.9771999492226204, + "learning_rate": 1.5226777828295378e-06, + "loss": 1.0352, + "num_input_tokens_seen": 121516300, + "step": 7843 + }, + { + "epoch": 0.5895084924094394, + "grad_norm": 1.8836823033228518, + "learning_rate": 1.5222050231047747e-06, + "loss": 0.9642, + "num_input_tokens_seen": 121539330, + "step": 7844 + }, + { + "epoch": 0.5895836464752743, + "grad_norm": 1.6162876445165615, + "learning_rate": 1.5217322916911934e-06, + "loss": 0.9412, + "num_input_tokens_seen": 121564040, + "step": 7845 + }, + { + "epoch": 0.5896588005411093, + "grad_norm": 1.6644139165049427, + "learning_rate": 1.5212595886168046e-06, + "loss": 0.9878, + "num_input_tokens_seen": 121587435, + "step": 7846 + }, + { + "epoch": 0.5897339546069442, + "grad_norm": 1.6891937233116538, + "learning_rate": 1.5207869139096191e-06, + "loss": 1.0174, + "num_input_tokens_seen": 121609200, + "step": 7847 + }, + { + "epoch": 0.5898091086727792, + "grad_norm": 1.3590642639696047, + "learning_rate": 1.5203142675976434e-06, + "loss": 0.9034, + "num_input_tokens_seen": 121634260, + "step": 7848 + }, + { + "epoch": 0.5898842627386142, + "grad_norm": 2.1126221706117962, + "learning_rate": 1.5198416497088849e-06, + "loss": 0.944, + "num_input_tokens_seen": 121653195, + "step": 7849 + }, + { + "epoch": 0.5899594168044491, + "grad_norm": 1.5198979658463467, + "learning_rate": 1.519369060271347e-06, + "loss": 1.0356, + "num_input_tokens_seen": 121678685, + "step": 7850 + }, + { + "epoch": 0.5900345708702841, + "grad_norm": 1.8031933697146192, + "learning_rate": 1.5188964993130321e-06, + "loss": 0.8396, + "num_input_tokens_seen": 121705090, + "step": 7851 + }, + { + "epoch": 0.590109724936119, + "grad_norm": 1.4248975549648386, + "learning_rate": 1.5184239668619427e-06, + "loss": 0.9926, + "num_input_tokens_seen": 121726590, + "step": 7852 + }, + { + "epoch": 0.590184879001954, + "grad_norm": 1.440458156498327, + "learning_rate": 1.517951462946077e-06, + "loss": 0.8466, + "num_input_tokens_seen": 121749335, + "step": 7853 + }, + { + "epoch": 0.590260033067789, + "grad_norm": 1.5079620761504282, + "learning_rate": 1.5174789875934332e-06, + "loss": 1.0327, + "num_input_tokens_seen": 121774865, + "step": 7854 + }, + { + "epoch": 0.590335187133624, + "grad_norm": 1.5075366306440996, + "learning_rate": 1.517006540832007e-06, + "loss": 0.8977, + "num_input_tokens_seen": 121799490, + "step": 7855 + }, + { + "epoch": 0.5904103411994589, + "grad_norm": 1.4548584637836577, + "learning_rate": 1.5165341226897926e-06, + "loss": 0.9664, + "num_input_tokens_seen": 121821530, + "step": 7856 + }, + { + "epoch": 0.5904854952652938, + "grad_norm": 1.7573642413475266, + "learning_rate": 1.5160617331947828e-06, + "loss": 0.9362, + "num_input_tokens_seen": 121842820, + "step": 7857 + }, + { + "epoch": 0.5905606493311288, + "grad_norm": 1.52822834749763, + "learning_rate": 1.515589372374968e-06, + "loss": 1.066, + "num_input_tokens_seen": 121864875, + "step": 7858 + }, + { + "epoch": 0.5906358033969638, + "grad_norm": 1.3964276495663237, + "learning_rate": 1.5151170402583384e-06, + "loss": 0.8974, + "num_input_tokens_seen": 121887435, + "step": 7859 + }, + { + "epoch": 0.5907109574627988, + "grad_norm": 1.3654612951558625, + "learning_rate": 1.5146447368728814e-06, + "loss": 1.0105, + "num_input_tokens_seen": 121911775, + "step": 7860 + }, + { + "epoch": 0.5907861115286337, + "grad_norm": 1.4425265292896585, + "learning_rate": 1.514172462246581e-06, + "loss": 0.9927, + "num_input_tokens_seen": 121935105, + "step": 7861 + }, + { + "epoch": 0.5908612655944686, + "grad_norm": 9.724436185129314, + "learning_rate": 1.5137002164074234e-06, + "loss": 0.9075, + "num_input_tokens_seen": 121962880, + "step": 7862 + }, + { + "epoch": 0.5909364196603036, + "grad_norm": 1.656157077125953, + "learning_rate": 1.5132279993833898e-06, + "loss": 0.9204, + "num_input_tokens_seen": 121984470, + "step": 7863 + }, + { + "epoch": 0.5910115737261386, + "grad_norm": 1.4627452416493356, + "learning_rate": 1.5127558112024617e-06, + "loss": 1.0058, + "num_input_tokens_seen": 122008675, + "step": 7864 + }, + { + "epoch": 0.5910867277919736, + "grad_norm": 1.5826890119637207, + "learning_rate": 1.512283651892617e-06, + "loss": 0.8764, + "num_input_tokens_seen": 122031245, + "step": 7865 + }, + { + "epoch": 0.5911618818578085, + "grad_norm": 1.5757376865763453, + "learning_rate": 1.5118115214818339e-06, + "loss": 0.9774, + "num_input_tokens_seen": 122053670, + "step": 7866 + }, + { + "epoch": 0.5912370359236435, + "grad_norm": 1.2937068618732792, + "learning_rate": 1.5113394199980877e-06, + "loss": 1.0117, + "num_input_tokens_seen": 122078875, + "step": 7867 + }, + { + "epoch": 0.5913121899894784, + "grad_norm": 1.6879807246676026, + "learning_rate": 1.5108673474693516e-06, + "loss": 1.0011, + "num_input_tokens_seen": 122100110, + "step": 7868 + }, + { + "epoch": 0.5913873440553133, + "grad_norm": 1.471650607423583, + "learning_rate": 1.5103953039235986e-06, + "loss": 0.8998, + "num_input_tokens_seen": 122124940, + "step": 7869 + }, + { + "epoch": 0.5914624981211484, + "grad_norm": 1.8849568798688323, + "learning_rate": 1.5099232893887983e-06, + "loss": 0.9509, + "num_input_tokens_seen": 122147125, + "step": 7870 + }, + { + "epoch": 0.5915376521869833, + "grad_norm": 1.6157774744449893, + "learning_rate": 1.5094513038929199e-06, + "loss": 1.0409, + "num_input_tokens_seen": 122169295, + "step": 7871 + }, + { + "epoch": 0.5916128062528183, + "grad_norm": 1.4781908210148376, + "learning_rate": 1.5089793474639305e-06, + "loss": 0.8688, + "num_input_tokens_seen": 122192815, + "step": 7872 + }, + { + "epoch": 0.5916879603186532, + "grad_norm": 1.5919569063775896, + "learning_rate": 1.5085074201297943e-06, + "loss": 1.0317, + "num_input_tokens_seen": 122211915, + "step": 7873 + }, + { + "epoch": 0.5917631143844883, + "grad_norm": 1.6025028352518933, + "learning_rate": 1.5080355219184762e-06, + "loss": 0.9331, + "num_input_tokens_seen": 122235710, + "step": 7874 + }, + { + "epoch": 0.5918382684503232, + "grad_norm": 1.6536612522477718, + "learning_rate": 1.5075636528579366e-06, + "loss": 0.998, + "num_input_tokens_seen": 122258755, + "step": 7875 + }, + { + "epoch": 0.5919134225161581, + "grad_norm": 1.5735153254758742, + "learning_rate": 1.507091812976137e-06, + "loss": 1.021, + "num_input_tokens_seen": 122281730, + "step": 7876 + }, + { + "epoch": 0.5919885765819931, + "grad_norm": 1.4807689835379563, + "learning_rate": 1.5066200023010347e-06, + "loss": 0.8875, + "num_input_tokens_seen": 122304075, + "step": 7877 + }, + { + "epoch": 0.592063730647828, + "grad_norm": 1.60832632027054, + "learning_rate": 1.5061482208605856e-06, + "loss": 0.9331, + "num_input_tokens_seen": 122328730, + "step": 7878 + }, + { + "epoch": 0.592138884713663, + "grad_norm": 1.5839419282806826, + "learning_rate": 1.505676468682747e-06, + "loss": 0.9364, + "num_input_tokens_seen": 122353010, + "step": 7879 + }, + { + "epoch": 0.592214038779498, + "grad_norm": 1.2903130678529937, + "learning_rate": 1.5052047457954691e-06, + "loss": 0.9946, + "num_input_tokens_seen": 122378670, + "step": 7880 + }, + { + "epoch": 0.5922891928453329, + "grad_norm": 1.588517369152251, + "learning_rate": 1.5047330522267056e-06, + "loss": 0.9552, + "num_input_tokens_seen": 122402865, + "step": 7881 + }, + { + "epoch": 0.5923643469111679, + "grad_norm": 1.3905619917750378, + "learning_rate": 1.5042613880044053e-06, + "loss": 0.9555, + "num_input_tokens_seen": 122424400, + "step": 7882 + }, + { + "epoch": 0.5924395009770028, + "grad_norm": 1.7360506640613533, + "learning_rate": 1.5037897531565155e-06, + "loss": 0.9632, + "num_input_tokens_seen": 122447905, + "step": 7883 + }, + { + "epoch": 0.5925146550428378, + "grad_norm": 1.503296408922277, + "learning_rate": 1.5033181477109835e-06, + "loss": 0.9597, + "num_input_tokens_seen": 122469485, + "step": 7884 + }, + { + "epoch": 0.5925898091086728, + "grad_norm": 1.5045489634656548, + "learning_rate": 1.5028465716957527e-06, + "loss": 0.9023, + "num_input_tokens_seen": 122492535, + "step": 7885 + }, + { + "epoch": 0.5926649631745078, + "grad_norm": 1.3666079791579295, + "learning_rate": 1.5023750251387668e-06, + "loss": 0.9228, + "num_input_tokens_seen": 122521020, + "step": 7886 + }, + { + "epoch": 0.5927401172403427, + "grad_norm": 1.4535111765292381, + "learning_rate": 1.501903508067966e-06, + "loss": 0.901, + "num_input_tokens_seen": 122545705, + "step": 7887 + }, + { + "epoch": 0.5928152713061776, + "grad_norm": 1.5118695955152295, + "learning_rate": 1.5014320205112897e-06, + "loss": 0.7777, + "num_input_tokens_seen": 122569755, + "step": 7888 + }, + { + "epoch": 0.5928904253720126, + "grad_norm": 4.389083863339117, + "learning_rate": 1.5009605624966753e-06, + "loss": 0.899, + "num_input_tokens_seen": 122593380, + "step": 7889 + }, + { + "epoch": 0.5929655794378476, + "grad_norm": 1.6260602133371542, + "learning_rate": 1.5004891340520583e-06, + "loss": 0.945, + "num_input_tokens_seen": 122614725, + "step": 7890 + }, + { + "epoch": 0.5930407335036826, + "grad_norm": 1.5854391433118415, + "learning_rate": 1.5000177352053732e-06, + "loss": 0.9094, + "num_input_tokens_seen": 122638050, + "step": 7891 + }, + { + "epoch": 0.5931158875695175, + "grad_norm": 1.4031557160479688, + "learning_rate": 1.4995463659845512e-06, + "loss": 0.9568, + "num_input_tokens_seen": 122661380, + "step": 7892 + }, + { + "epoch": 0.5931910416353525, + "grad_norm": 1.400595720454063, + "learning_rate": 1.499075026417524e-06, + "loss": 0.9295, + "num_input_tokens_seen": 122686280, + "step": 7893 + }, + { + "epoch": 0.5932661957011874, + "grad_norm": 1.6513424989376393, + "learning_rate": 1.4986037165322199e-06, + "loss": 1.0005, + "num_input_tokens_seen": 122709980, + "step": 7894 + }, + { + "epoch": 0.5933413497670224, + "grad_norm": 0.8080141874545509, + "learning_rate": 1.498132436356565e-06, + "loss": 0.8506, + "num_input_tokens_seen": 122782905, + "step": 7895 + }, + { + "epoch": 0.5934165038328574, + "grad_norm": 1.4082466362144892, + "learning_rate": 1.4976611859184852e-06, + "loss": 0.9844, + "num_input_tokens_seen": 122805995, + "step": 7896 + }, + { + "epoch": 0.5934916578986923, + "grad_norm": 1.4962197990205854, + "learning_rate": 1.4971899652459034e-06, + "loss": 0.8549, + "num_input_tokens_seen": 122828765, + "step": 7897 + }, + { + "epoch": 0.5935668119645273, + "grad_norm": 1.4696551841055974, + "learning_rate": 1.4967187743667423e-06, + "loss": 0.9354, + "num_input_tokens_seen": 122855030, + "step": 7898 + }, + { + "epoch": 0.5936419660303622, + "grad_norm": 1.605143938829653, + "learning_rate": 1.4962476133089207e-06, + "loss": 0.9825, + "num_input_tokens_seen": 122878565, + "step": 7899 + }, + { + "epoch": 0.5937171200961973, + "grad_norm": 1.6011571603821537, + "learning_rate": 1.4957764821003562e-06, + "loss": 0.8995, + "num_input_tokens_seen": 122900365, + "step": 7900 + }, + { + "epoch": 0.5937922741620322, + "grad_norm": 1.3737686018513071, + "learning_rate": 1.4953053807689671e-06, + "loss": 0.8755, + "num_input_tokens_seen": 122925420, + "step": 7901 + }, + { + "epoch": 0.5938674282278671, + "grad_norm": 1.5773144897346811, + "learning_rate": 1.4948343093426656e-06, + "loss": 0.8949, + "num_input_tokens_seen": 122952595, + "step": 7902 + }, + { + "epoch": 0.5939425822937021, + "grad_norm": 1.7430911607217168, + "learning_rate": 1.4943632678493668e-06, + "loss": 0.9235, + "num_input_tokens_seen": 122978270, + "step": 7903 + }, + { + "epoch": 0.594017736359537, + "grad_norm": 1.5060494348738402, + "learning_rate": 1.4938922563169801e-06, + "loss": 0.8773, + "num_input_tokens_seen": 123000675, + "step": 7904 + }, + { + "epoch": 0.594092890425372, + "grad_norm": 1.5203765636091997, + "learning_rate": 1.4934212747734153e-06, + "loss": 0.8853, + "num_input_tokens_seen": 123022990, + "step": 7905 + }, + { + "epoch": 0.594168044491207, + "grad_norm": 1.664461114985553, + "learning_rate": 1.49295032324658e-06, + "loss": 0.9508, + "num_input_tokens_seen": 123046930, + "step": 7906 + }, + { + "epoch": 0.5942431985570419, + "grad_norm": 1.4003162656910921, + "learning_rate": 1.492479401764379e-06, + "loss": 0.8845, + "num_input_tokens_seen": 123072220, + "step": 7907 + }, + { + "epoch": 0.5943183526228769, + "grad_norm": 1.474157888250844, + "learning_rate": 1.4920085103547177e-06, + "loss": 0.9278, + "num_input_tokens_seen": 123098165, + "step": 7908 + }, + { + "epoch": 0.5943935066887118, + "grad_norm": 2.151967093574196, + "learning_rate": 1.491537649045497e-06, + "loss": 1.0276, + "num_input_tokens_seen": 123119260, + "step": 7909 + }, + { + "epoch": 0.5944686607545469, + "grad_norm": 1.3779301200665959, + "learning_rate": 1.4910668178646178e-06, + "loss": 0.8665, + "num_input_tokens_seen": 123145240, + "step": 7910 + }, + { + "epoch": 0.5945438148203818, + "grad_norm": 1.3439727421164924, + "learning_rate": 1.4905960168399783e-06, + "loss": 1.0091, + "num_input_tokens_seen": 123170550, + "step": 7911 + }, + { + "epoch": 0.5946189688862168, + "grad_norm": 1.9472438309807225, + "learning_rate": 1.4901252459994753e-06, + "loss": 0.899, + "num_input_tokens_seen": 123193850, + "step": 7912 + }, + { + "epoch": 0.5946941229520517, + "grad_norm": 1.2373621565216493, + "learning_rate": 1.4896545053710044e-06, + "loss": 0.8401, + "num_input_tokens_seen": 123224075, + "step": 7913 + }, + { + "epoch": 0.5947692770178866, + "grad_norm": 1.5655592914124903, + "learning_rate": 1.4891837949824578e-06, + "loss": 1.0017, + "num_input_tokens_seen": 123246530, + "step": 7914 + }, + { + "epoch": 0.5948444310837216, + "grad_norm": 1.615319024927234, + "learning_rate": 1.4887131148617279e-06, + "loss": 1.0114, + "num_input_tokens_seen": 123269090, + "step": 7915 + }, + { + "epoch": 0.5949195851495566, + "grad_norm": 1.5458621938364245, + "learning_rate": 1.4882424650367034e-06, + "loss": 0.9313, + "num_input_tokens_seen": 123297145, + "step": 7916 + }, + { + "epoch": 0.5949947392153916, + "grad_norm": 1.8073299598118278, + "learning_rate": 1.4877718455352723e-06, + "loss": 1.0026, + "num_input_tokens_seen": 123318945, + "step": 7917 + }, + { + "epoch": 0.5950698932812265, + "grad_norm": 1.7634912611699147, + "learning_rate": 1.4873012563853213e-06, + "loss": 1.0, + "num_input_tokens_seen": 123344080, + "step": 7918 + }, + { + "epoch": 0.5951450473470615, + "grad_norm": 1.605961930385504, + "learning_rate": 1.4868306976147337e-06, + "loss": 0.924, + "num_input_tokens_seen": 123367050, + "step": 7919 + }, + { + "epoch": 0.5952202014128964, + "grad_norm": 1.8836122452576642, + "learning_rate": 1.4863601692513927e-06, + "loss": 0.9516, + "num_input_tokens_seen": 123389185, + "step": 7920 + }, + { + "epoch": 0.5952953554787314, + "grad_norm": 1.3823382324025846, + "learning_rate": 1.4858896713231786e-06, + "loss": 1.001, + "num_input_tokens_seen": 123411145, + "step": 7921 + }, + { + "epoch": 0.5953705095445664, + "grad_norm": 1.3945205645325474, + "learning_rate": 1.485419203857969e-06, + "loss": 0.9729, + "num_input_tokens_seen": 123436260, + "step": 7922 + }, + { + "epoch": 0.5954456636104013, + "grad_norm": 1.468619320514219, + "learning_rate": 1.4849487668836439e-06, + "loss": 0.9948, + "num_input_tokens_seen": 123457265, + "step": 7923 + }, + { + "epoch": 0.5955208176762363, + "grad_norm": 1.505399680817399, + "learning_rate": 1.4844783604280746e-06, + "loss": 0.8796, + "num_input_tokens_seen": 123480955, + "step": 7924 + }, + { + "epoch": 0.5955959717420712, + "grad_norm": 1.3593542163180843, + "learning_rate": 1.4840079845191379e-06, + "loss": 0.8806, + "num_input_tokens_seen": 123507570, + "step": 7925 + }, + { + "epoch": 0.5956711258079062, + "grad_norm": 1.583368108601735, + "learning_rate": 1.483537639184704e-06, + "loss": 0.9903, + "num_input_tokens_seen": 123527875, + "step": 7926 + }, + { + "epoch": 0.5957462798737412, + "grad_norm": 1.425296732475118, + "learning_rate": 1.4830673244526418e-06, + "loss": 1.0222, + "num_input_tokens_seen": 123548805, + "step": 7927 + }, + { + "epoch": 0.5958214339395761, + "grad_norm": 1.692910034682459, + "learning_rate": 1.4825970403508208e-06, + "loss": 0.939, + "num_input_tokens_seen": 123572145, + "step": 7928 + }, + { + "epoch": 0.5958965880054111, + "grad_norm": 1.4901338995863054, + "learning_rate": 1.482126786907106e-06, + "loss": 0.9353, + "num_input_tokens_seen": 123593555, + "step": 7929 + }, + { + "epoch": 0.595971742071246, + "grad_norm": 1.2564024989244444, + "learning_rate": 1.4816565641493623e-06, + "loss": 0.8852, + "num_input_tokens_seen": 123619830, + "step": 7930 + }, + { + "epoch": 0.5960468961370811, + "grad_norm": 1.729332958023356, + "learning_rate": 1.481186372105452e-06, + "loss": 0.8965, + "num_input_tokens_seen": 123643210, + "step": 7931 + }, + { + "epoch": 0.596122050202916, + "grad_norm": 1.4971793197889283, + "learning_rate": 1.4807162108032363e-06, + "loss": 1.0043, + "num_input_tokens_seen": 123668320, + "step": 7932 + }, + { + "epoch": 0.5961972042687509, + "grad_norm": 1.5093057778480583, + "learning_rate": 1.4802460802705731e-06, + "loss": 0.9331, + "num_input_tokens_seen": 123693300, + "step": 7933 + }, + { + "epoch": 0.5962723583345859, + "grad_norm": 0.8003219776121161, + "learning_rate": 1.4797759805353199e-06, + "loss": 0.8114, + "num_input_tokens_seen": 123773345, + "step": 7934 + }, + { + "epoch": 0.5963475124004208, + "grad_norm": 1.4358325907475962, + "learning_rate": 1.4793059116253322e-06, + "loss": 0.9975, + "num_input_tokens_seen": 123796460, + "step": 7935 + }, + { + "epoch": 0.5964226664662559, + "grad_norm": 1.737431899458257, + "learning_rate": 1.4788358735684626e-06, + "loss": 0.9407, + "num_input_tokens_seen": 123821370, + "step": 7936 + }, + { + "epoch": 0.5964978205320908, + "grad_norm": 1.377796147700016, + "learning_rate": 1.4783658663925637e-06, + "loss": 0.9687, + "num_input_tokens_seen": 123843075, + "step": 7937 + }, + { + "epoch": 0.5965729745979258, + "grad_norm": 0.8904742063056302, + "learning_rate": 1.4778958901254847e-06, + "loss": 0.8801, + "num_input_tokens_seen": 123905575, + "step": 7938 + }, + { + "epoch": 0.5966481286637607, + "grad_norm": 1.5124307380139523, + "learning_rate": 1.477425944795073e-06, + "loss": 0.8794, + "num_input_tokens_seen": 123928070, + "step": 7939 + }, + { + "epoch": 0.5967232827295956, + "grad_norm": 1.6590818990412861, + "learning_rate": 1.4769560304291755e-06, + "loss": 0.8974, + "num_input_tokens_seen": 123951860, + "step": 7940 + }, + { + "epoch": 0.5967984367954307, + "grad_norm": 1.2783059045782532, + "learning_rate": 1.4764861470556357e-06, + "loss": 0.9317, + "num_input_tokens_seen": 123976845, + "step": 7941 + }, + { + "epoch": 0.5968735908612656, + "grad_norm": 1.3803407377636934, + "learning_rate": 1.4760162947022972e-06, + "loss": 1.0627, + "num_input_tokens_seen": 124003060, + "step": 7942 + }, + { + "epoch": 0.5969487449271006, + "grad_norm": 1.4492634974560679, + "learning_rate": 1.475546473396999e-06, + "loss": 1.0108, + "num_input_tokens_seen": 124026025, + "step": 7943 + }, + { + "epoch": 0.5970238989929355, + "grad_norm": 1.9179489227948248, + "learning_rate": 1.47507668316758e-06, + "loss": 0.8595, + "num_input_tokens_seen": 124049390, + "step": 7944 + }, + { + "epoch": 0.5970990530587705, + "grad_norm": 1.7770924820241483, + "learning_rate": 1.4746069240418785e-06, + "loss": 0.9533, + "num_input_tokens_seen": 124071425, + "step": 7945 + }, + { + "epoch": 0.5971742071246054, + "grad_norm": 1.7677228499539366, + "learning_rate": 1.474137196047728e-06, + "loss": 0.9029, + "num_input_tokens_seen": 124091210, + "step": 7946 + }, + { + "epoch": 0.5972493611904404, + "grad_norm": 1.7075523436581372, + "learning_rate": 1.473667499212963e-06, + "loss": 0.8566, + "num_input_tokens_seen": 124113720, + "step": 7947 + }, + { + "epoch": 0.5973245152562754, + "grad_norm": 1.7297736638359904, + "learning_rate": 1.4731978335654138e-06, + "loss": 0.9051, + "num_input_tokens_seen": 124137690, + "step": 7948 + }, + { + "epoch": 0.5973996693221103, + "grad_norm": 1.3964978624239417, + "learning_rate": 1.47272819913291e-06, + "loss": 1.0336, + "num_input_tokens_seen": 124162130, + "step": 7949 + }, + { + "epoch": 0.5974748233879453, + "grad_norm": 2.2162773430993794, + "learning_rate": 1.4722585959432802e-06, + "loss": 0.935, + "num_input_tokens_seen": 124184640, + "step": 7950 + }, + { + "epoch": 0.5975499774537802, + "grad_norm": 1.7425233662597066, + "learning_rate": 1.4717890240243484e-06, + "loss": 0.9811, + "num_input_tokens_seen": 124208750, + "step": 7951 + }, + { + "epoch": 0.5976251315196152, + "grad_norm": 1.2479770025952612, + "learning_rate": 1.4713194834039401e-06, + "loss": 0.9859, + "num_input_tokens_seen": 124233790, + "step": 7952 + }, + { + "epoch": 0.5977002855854502, + "grad_norm": 1.4283791480771557, + "learning_rate": 1.470849974109877e-06, + "loss": 0.9474, + "num_input_tokens_seen": 124256000, + "step": 7953 + }, + { + "epoch": 0.5977754396512851, + "grad_norm": 1.3921345864461094, + "learning_rate": 1.470380496169979e-06, + "loss": 0.9013, + "num_input_tokens_seen": 124282120, + "step": 7954 + }, + { + "epoch": 0.5978505937171201, + "grad_norm": 1.8868671619496356, + "learning_rate": 1.4699110496120648e-06, + "loss": 0.9324, + "num_input_tokens_seen": 124303790, + "step": 7955 + }, + { + "epoch": 0.597925747782955, + "grad_norm": 1.4645177045997708, + "learning_rate": 1.4694416344639503e-06, + "loss": 0.885, + "num_input_tokens_seen": 124330530, + "step": 7956 + }, + { + "epoch": 0.5980009018487901, + "grad_norm": 1.4941017694345216, + "learning_rate": 1.4689722507534514e-06, + "loss": 0.8938, + "num_input_tokens_seen": 124357795, + "step": 7957 + }, + { + "epoch": 0.598076055914625, + "grad_norm": 1.4398878416781458, + "learning_rate": 1.4685028985083794e-06, + "loss": 0.9968, + "num_input_tokens_seen": 124380935, + "step": 7958 + }, + { + "epoch": 0.5981512099804599, + "grad_norm": 1.4971317047251838, + "learning_rate": 1.4680335777565462e-06, + "loss": 0.9945, + "num_input_tokens_seen": 124404070, + "step": 7959 + }, + { + "epoch": 0.5982263640462949, + "grad_norm": 1.3127529717932798, + "learning_rate": 1.467564288525761e-06, + "loss": 1.0256, + "num_input_tokens_seen": 124429875, + "step": 7960 + }, + { + "epoch": 0.5983015181121298, + "grad_norm": 1.2946708056059293, + "learning_rate": 1.4670950308438298e-06, + "loss": 1.0198, + "num_input_tokens_seen": 124455475, + "step": 7961 + }, + { + "epoch": 0.5983766721779649, + "grad_norm": 0.8322979853218735, + "learning_rate": 1.4666258047385588e-06, + "loss": 0.8184, + "num_input_tokens_seen": 124518375, + "step": 7962 + }, + { + "epoch": 0.5984518262437998, + "grad_norm": 1.3731969802874358, + "learning_rate": 1.4661566102377507e-06, + "loss": 0.914, + "num_input_tokens_seen": 124540630, + "step": 7963 + }, + { + "epoch": 0.5985269803096348, + "grad_norm": 1.5510400512905518, + "learning_rate": 1.465687447369209e-06, + "loss": 0.9893, + "num_input_tokens_seen": 124565815, + "step": 7964 + }, + { + "epoch": 0.5986021343754697, + "grad_norm": 1.3949532604932957, + "learning_rate": 1.4652183161607314e-06, + "loss": 0.8118, + "num_input_tokens_seen": 124595415, + "step": 7965 + }, + { + "epoch": 0.5986772884413046, + "grad_norm": 1.8057253276856395, + "learning_rate": 1.4647492166401159e-06, + "loss": 0.9693, + "num_input_tokens_seen": 124615415, + "step": 7966 + }, + { + "epoch": 0.5987524425071397, + "grad_norm": 1.2823135334859863, + "learning_rate": 1.4642801488351598e-06, + "loss": 0.8801, + "num_input_tokens_seen": 124641640, + "step": 7967 + }, + { + "epoch": 0.5988275965729746, + "grad_norm": 1.6831759401269861, + "learning_rate": 1.4638111127736555e-06, + "loss": 0.9517, + "num_input_tokens_seen": 124665355, + "step": 7968 + }, + { + "epoch": 0.5989027506388096, + "grad_norm": 1.3659385425962485, + "learning_rate": 1.4633421084833965e-06, + "loss": 1.0258, + "num_input_tokens_seen": 124694430, + "step": 7969 + }, + { + "epoch": 0.5989779047046445, + "grad_norm": 1.6356500673687209, + "learning_rate": 1.4628731359921727e-06, + "loss": 0.874, + "num_input_tokens_seen": 124717260, + "step": 7970 + }, + { + "epoch": 0.5990530587704794, + "grad_norm": 0.7745787167697985, + "learning_rate": 1.462404195327772e-06, + "loss": 0.8316, + "num_input_tokens_seen": 124790465, + "step": 7971 + }, + { + "epoch": 0.5991282128363145, + "grad_norm": 1.67773359340876, + "learning_rate": 1.4619352865179814e-06, + "loss": 0.8096, + "num_input_tokens_seen": 124815670, + "step": 7972 + }, + { + "epoch": 0.5992033669021494, + "grad_norm": 1.645261480078346, + "learning_rate": 1.4614664095905856e-06, + "loss": 0.941, + "num_input_tokens_seen": 124836900, + "step": 7973 + }, + { + "epoch": 0.5992785209679844, + "grad_norm": 1.468601868646891, + "learning_rate": 1.460997564573367e-06, + "loss": 0.9755, + "num_input_tokens_seen": 124864065, + "step": 7974 + }, + { + "epoch": 0.5993536750338193, + "grad_norm": 1.844691731097348, + "learning_rate": 1.4605287514941068e-06, + "loss": 1.0244, + "num_input_tokens_seen": 124887705, + "step": 7975 + }, + { + "epoch": 0.5994288290996543, + "grad_norm": 1.566236986256914, + "learning_rate": 1.460059970380584e-06, + "loss": 0.9995, + "num_input_tokens_seen": 124913210, + "step": 7976 + }, + { + "epoch": 0.5995039831654893, + "grad_norm": 1.4864222813783037, + "learning_rate": 1.4595912212605755e-06, + "loss": 0.922, + "num_input_tokens_seen": 124938920, + "step": 7977 + }, + { + "epoch": 0.5995791372313242, + "grad_norm": 1.671471876938807, + "learning_rate": 1.459122504161856e-06, + "loss": 1.0699, + "num_input_tokens_seen": 124959720, + "step": 7978 + }, + { + "epoch": 0.5996542912971592, + "grad_norm": 1.7003337139873804, + "learning_rate": 1.4586538191121999e-06, + "loss": 0.9387, + "num_input_tokens_seen": 124977555, + "step": 7979 + }, + { + "epoch": 0.5997294453629941, + "grad_norm": 1.6158198964014558, + "learning_rate": 1.4581851661393776e-06, + "loss": 0.9182, + "num_input_tokens_seen": 125001660, + "step": 7980 + }, + { + "epoch": 0.5998045994288291, + "grad_norm": 2.966991023930471, + "learning_rate": 1.4577165452711592e-06, + "loss": 0.9874, + "num_input_tokens_seen": 125023770, + "step": 7981 + }, + { + "epoch": 0.599879753494664, + "grad_norm": 1.600975859594057, + "learning_rate": 1.4572479565353122e-06, + "loss": 0.8739, + "num_input_tokens_seen": 125046050, + "step": 7982 + }, + { + "epoch": 0.5999549075604991, + "grad_norm": 1.4801797307915572, + "learning_rate": 1.4567793999596014e-06, + "loss": 0.9909, + "num_input_tokens_seen": 125067350, + "step": 7983 + }, + { + "epoch": 0.600030061626334, + "grad_norm": 1.7433104039587388, + "learning_rate": 1.456310875571792e-06, + "loss": 0.944, + "num_input_tokens_seen": 125089755, + "step": 7984 + }, + { + "epoch": 0.6001052156921689, + "grad_norm": 1.4903701980309976, + "learning_rate": 1.4558423833996443e-06, + "loss": 0.8458, + "num_input_tokens_seen": 125114750, + "step": 7985 + }, + { + "epoch": 0.6001803697580039, + "grad_norm": 1.4999491365074267, + "learning_rate": 1.4553739234709199e-06, + "loss": 0.9158, + "num_input_tokens_seen": 125142005, + "step": 7986 + }, + { + "epoch": 0.6002555238238388, + "grad_norm": 1.9350433772545486, + "learning_rate": 1.4549054958133765e-06, + "loss": 0.9093, + "num_input_tokens_seen": 125166415, + "step": 7987 + }, + { + "epoch": 0.6003306778896739, + "grad_norm": 1.5988217605102526, + "learning_rate": 1.4544371004547685e-06, + "loss": 0.8353, + "num_input_tokens_seen": 125190995, + "step": 7988 + }, + { + "epoch": 0.6004058319555088, + "grad_norm": 1.7735070706472023, + "learning_rate": 1.453968737422852e-06, + "loss": 0.899, + "num_input_tokens_seen": 125214750, + "step": 7989 + }, + { + "epoch": 0.6004809860213438, + "grad_norm": 1.75367643554138, + "learning_rate": 1.4535004067453785e-06, + "loss": 1.0121, + "num_input_tokens_seen": 125237210, + "step": 7990 + }, + { + "epoch": 0.6005561400871787, + "grad_norm": 1.7165377858617437, + "learning_rate": 1.453032108450099e-06, + "loss": 0.8796, + "num_input_tokens_seen": 125263360, + "step": 7991 + }, + { + "epoch": 0.6006312941530136, + "grad_norm": 1.7047227836614731, + "learning_rate": 1.4525638425647615e-06, + "loss": 0.9525, + "num_input_tokens_seen": 125290520, + "step": 7992 + }, + { + "epoch": 0.6007064482188487, + "grad_norm": 1.5855086174198094, + "learning_rate": 1.4520956091171121e-06, + "loss": 1.0089, + "num_input_tokens_seen": 125315460, + "step": 7993 + }, + { + "epoch": 0.6007816022846836, + "grad_norm": 4.559323184664768, + "learning_rate": 1.4516274081348965e-06, + "loss": 0.8885, + "num_input_tokens_seen": 125342360, + "step": 7994 + }, + { + "epoch": 0.6008567563505186, + "grad_norm": 1.8590810847992614, + "learning_rate": 1.4511592396458565e-06, + "loss": 1.0303, + "num_input_tokens_seen": 125363665, + "step": 7995 + }, + { + "epoch": 0.6009319104163535, + "grad_norm": 1.9661910365552542, + "learning_rate": 1.4506911036777335e-06, + "loss": 0.9534, + "num_input_tokens_seen": 125388130, + "step": 7996 + }, + { + "epoch": 0.6010070644821884, + "grad_norm": 1.644856907728592, + "learning_rate": 1.4502230002582655e-06, + "loss": 1.0443, + "num_input_tokens_seen": 125412240, + "step": 7997 + }, + { + "epoch": 0.6010822185480235, + "grad_norm": 1.5898136070449465, + "learning_rate": 1.4497549294151905e-06, + "loss": 0.9794, + "num_input_tokens_seen": 125435080, + "step": 7998 + }, + { + "epoch": 0.6011573726138584, + "grad_norm": 2.209541122471657, + "learning_rate": 1.4492868911762428e-06, + "loss": 0.9425, + "num_input_tokens_seen": 125458265, + "step": 7999 + }, + { + "epoch": 0.6012325266796934, + "grad_norm": 1.5108910791155827, + "learning_rate": 1.4488188855691555e-06, + "loss": 0.9095, + "num_input_tokens_seen": 125481965, + "step": 8000 + }, + { + "epoch": 0.6013076807455283, + "grad_norm": 1.6205464002068526, + "learning_rate": 1.44835091262166e-06, + "loss": 0.9469, + "num_input_tokens_seen": 125506620, + "step": 8001 + }, + { + "epoch": 0.6013828348113633, + "grad_norm": 1.4702252223251246, + "learning_rate": 1.447882972361485e-06, + "loss": 0.9828, + "num_input_tokens_seen": 125529410, + "step": 8002 + }, + { + "epoch": 0.6014579888771983, + "grad_norm": 1.3074311059193382, + "learning_rate": 1.4474150648163588e-06, + "loss": 0.9679, + "num_input_tokens_seen": 125556055, + "step": 8003 + }, + { + "epoch": 0.6015331429430332, + "grad_norm": 1.7351063054302462, + "learning_rate": 1.4469471900140056e-06, + "loss": 0.9369, + "num_input_tokens_seen": 125579145, + "step": 8004 + }, + { + "epoch": 0.6016082970088682, + "grad_norm": 1.5511545649580698, + "learning_rate": 1.4464793479821489e-06, + "loss": 0.9655, + "num_input_tokens_seen": 125602615, + "step": 8005 + }, + { + "epoch": 0.6016834510747031, + "grad_norm": 1.7667317930249367, + "learning_rate": 1.446011538748511e-06, + "loss": 0.8998, + "num_input_tokens_seen": 125625790, + "step": 8006 + }, + { + "epoch": 0.6017586051405381, + "grad_norm": 1.3029359492839245, + "learning_rate": 1.4455437623408097e-06, + "loss": 0.9158, + "num_input_tokens_seen": 125649405, + "step": 8007 + }, + { + "epoch": 0.601833759206373, + "grad_norm": 1.6360526174650551, + "learning_rate": 1.4450760187867644e-06, + "loss": 0.9201, + "num_input_tokens_seen": 125674060, + "step": 8008 + }, + { + "epoch": 0.6019089132722081, + "grad_norm": 1.6378762606412363, + "learning_rate": 1.4446083081140904e-06, + "loss": 0.9091, + "num_input_tokens_seen": 125695515, + "step": 8009 + }, + { + "epoch": 0.601984067338043, + "grad_norm": 1.829648613781209, + "learning_rate": 1.4441406303504998e-06, + "loss": 0.8714, + "num_input_tokens_seen": 125717710, + "step": 8010 + }, + { + "epoch": 0.6020592214038779, + "grad_norm": 2.168021096637177, + "learning_rate": 1.4436729855237063e-06, + "loss": 0.8525, + "num_input_tokens_seen": 125738680, + "step": 8011 + }, + { + "epoch": 0.6021343754697129, + "grad_norm": 1.752046818604083, + "learning_rate": 1.443205373661418e-06, + "loss": 0.836, + "num_input_tokens_seen": 125764745, + "step": 8012 + }, + { + "epoch": 0.6022095295355478, + "grad_norm": 0.8613383666660096, + "learning_rate": 1.442737794791344e-06, + "loss": 0.8896, + "num_input_tokens_seen": 125837225, + "step": 8013 + }, + { + "epoch": 0.6022846836013829, + "grad_norm": 2.838910354795212, + "learning_rate": 1.4422702489411896e-06, + "loss": 0.8594, + "num_input_tokens_seen": 125860840, + "step": 8014 + }, + { + "epoch": 0.6023598376672178, + "grad_norm": 1.4701583278794168, + "learning_rate": 1.441802736138658e-06, + "loss": 0.9232, + "num_input_tokens_seen": 125884830, + "step": 8015 + }, + { + "epoch": 0.6024349917330527, + "grad_norm": 1.7736969476552589, + "learning_rate": 1.4413352564114525e-06, + "loss": 0.874, + "num_input_tokens_seen": 125907005, + "step": 8016 + }, + { + "epoch": 0.6025101457988877, + "grad_norm": 0.7573963871209156, + "learning_rate": 1.4408678097872717e-06, + "loss": 0.8201, + "num_input_tokens_seen": 125985600, + "step": 8017 + }, + { + "epoch": 0.6025852998647226, + "grad_norm": 1.5187634879561178, + "learning_rate": 1.440400396293815e-06, + "loss": 0.7847, + "num_input_tokens_seen": 126055975, + "step": 8018 + }, + { + "epoch": 0.6026604539305577, + "grad_norm": 1.399492856540691, + "learning_rate": 1.439933015958777e-06, + "loss": 0.9267, + "num_input_tokens_seen": 126081675, + "step": 8019 + }, + { + "epoch": 0.6027356079963926, + "grad_norm": 1.7006711869029574, + "learning_rate": 1.4394656688098526e-06, + "loss": 0.9761, + "num_input_tokens_seen": 126103910, + "step": 8020 + }, + { + "epoch": 0.6028107620622276, + "grad_norm": 1.8085965842436957, + "learning_rate": 1.4389983548747337e-06, + "loss": 1.0425, + "num_input_tokens_seen": 126126410, + "step": 8021 + }, + { + "epoch": 0.6028859161280625, + "grad_norm": 1.2749613905651989, + "learning_rate": 1.4385310741811106e-06, + "loss": 0.8874, + "num_input_tokens_seen": 126151285, + "step": 8022 + }, + { + "epoch": 0.6029610701938974, + "grad_norm": 1.5300731418305453, + "learning_rate": 1.4380638267566716e-06, + "loss": 0.9408, + "num_input_tokens_seen": 126176675, + "step": 8023 + }, + { + "epoch": 0.6030362242597325, + "grad_norm": 1.469665728859369, + "learning_rate": 1.4375966126291022e-06, + "loss": 0.9809, + "num_input_tokens_seen": 126201175, + "step": 8024 + }, + { + "epoch": 0.6031113783255674, + "grad_norm": 1.4691070771894221, + "learning_rate": 1.4371294318260874e-06, + "loss": 0.934, + "num_input_tokens_seen": 126227750, + "step": 8025 + }, + { + "epoch": 0.6031865323914024, + "grad_norm": 1.3614528604332747, + "learning_rate": 1.4366622843753092e-06, + "loss": 0.9657, + "num_input_tokens_seen": 126252945, + "step": 8026 + }, + { + "epoch": 0.6032616864572373, + "grad_norm": 1.4203949655695531, + "learning_rate": 1.4361951703044475e-06, + "loss": 0.9574, + "num_input_tokens_seen": 126274665, + "step": 8027 + }, + { + "epoch": 0.6033368405230723, + "grad_norm": 2.042100185637907, + "learning_rate": 1.4357280896411813e-06, + "loss": 0.9676, + "num_input_tokens_seen": 126296465, + "step": 8028 + }, + { + "epoch": 0.6034119945889073, + "grad_norm": 1.6902638865874997, + "learning_rate": 1.4352610424131854e-06, + "loss": 0.9821, + "num_input_tokens_seen": 126319980, + "step": 8029 + }, + { + "epoch": 0.6034871486547422, + "grad_norm": 1.8495558153915381, + "learning_rate": 1.4347940286481364e-06, + "loss": 0.841, + "num_input_tokens_seen": 126341370, + "step": 8030 + }, + { + "epoch": 0.6035623027205772, + "grad_norm": 1.3604029572493712, + "learning_rate": 1.434327048373706e-06, + "loss": 0.9875, + "num_input_tokens_seen": 126366635, + "step": 8031 + }, + { + "epoch": 0.6036374567864121, + "grad_norm": 1.7622293548565244, + "learning_rate": 1.4338601016175624e-06, + "loss": 0.8493, + "num_input_tokens_seen": 126391205, + "step": 8032 + }, + { + "epoch": 0.6037126108522471, + "grad_norm": 1.5755491480108932, + "learning_rate": 1.4333931884073769e-06, + "loss": 0.9116, + "num_input_tokens_seen": 126416315, + "step": 8033 + }, + { + "epoch": 0.6037877649180821, + "grad_norm": 1.2326871220758744, + "learning_rate": 1.4329263087708144e-06, + "loss": 0.9947, + "num_input_tokens_seen": 126440950, + "step": 8034 + }, + { + "epoch": 0.6038629189839171, + "grad_norm": 1.6520687081778793, + "learning_rate": 1.4324594627355397e-06, + "loss": 0.8916, + "num_input_tokens_seen": 126462930, + "step": 8035 + }, + { + "epoch": 0.603938073049752, + "grad_norm": 1.334229297410853, + "learning_rate": 1.431992650329215e-06, + "loss": 0.9772, + "num_input_tokens_seen": 126486050, + "step": 8036 + }, + { + "epoch": 0.6040132271155869, + "grad_norm": 1.5787541108066563, + "learning_rate": 1.4315258715795007e-06, + "loss": 1.0709, + "num_input_tokens_seen": 126511350, + "step": 8037 + }, + { + "epoch": 0.6040883811814219, + "grad_norm": 1.2763894625692331, + "learning_rate": 1.4310591265140555e-06, + "loss": 0.9554, + "num_input_tokens_seen": 126532555, + "step": 8038 + }, + { + "epoch": 0.6041635352472569, + "grad_norm": 1.8840379960830977, + "learning_rate": 1.4305924151605354e-06, + "loss": 0.907, + "num_input_tokens_seen": 126557855, + "step": 8039 + }, + { + "epoch": 0.6042386893130919, + "grad_norm": 2.1437330862551383, + "learning_rate": 1.4301257375465956e-06, + "loss": 1.078, + "num_input_tokens_seen": 126575625, + "step": 8040 + }, + { + "epoch": 0.6043138433789268, + "grad_norm": 1.4793785731438056, + "learning_rate": 1.4296590936998874e-06, + "loss": 0.9005, + "num_input_tokens_seen": 126599625, + "step": 8041 + }, + { + "epoch": 0.6043889974447617, + "grad_norm": 1.8388673171551335, + "learning_rate": 1.4291924836480625e-06, + "loss": 0.8819, + "num_input_tokens_seen": 126625050, + "step": 8042 + }, + { + "epoch": 0.6044641515105967, + "grad_norm": 1.9391983341220902, + "learning_rate": 1.4287259074187685e-06, + "loss": 0.8993, + "num_input_tokens_seen": 126647790, + "step": 8043 + }, + { + "epoch": 0.6045393055764317, + "grad_norm": 1.4405448841648278, + "learning_rate": 1.428259365039652e-06, + "loss": 0.9309, + "num_input_tokens_seen": 126672045, + "step": 8044 + }, + { + "epoch": 0.6046144596422667, + "grad_norm": 2.353345834772695, + "learning_rate": 1.4277928565383577e-06, + "loss": 0.8381, + "num_input_tokens_seen": 126693900, + "step": 8045 + }, + { + "epoch": 0.6046896137081016, + "grad_norm": 1.362257608673527, + "learning_rate": 1.4273263819425272e-06, + "loss": 0.9794, + "num_input_tokens_seen": 126717775, + "step": 8046 + }, + { + "epoch": 0.6047647677739366, + "grad_norm": 1.5326299871729598, + "learning_rate": 1.426859941279802e-06, + "loss": 0.8936, + "num_input_tokens_seen": 126743615, + "step": 8047 + }, + { + "epoch": 0.6048399218397715, + "grad_norm": 1.6666780074051621, + "learning_rate": 1.42639353457782e-06, + "loss": 0.8903, + "num_input_tokens_seen": 126766645, + "step": 8048 + }, + { + "epoch": 0.6049150759056064, + "grad_norm": 1.401814735535653, + "learning_rate": 1.4259271618642166e-06, + "loss": 0.9556, + "num_input_tokens_seen": 126793145, + "step": 8049 + }, + { + "epoch": 0.6049902299714415, + "grad_norm": 1.4420955839447787, + "learning_rate": 1.4254608231666286e-06, + "loss": 0.8972, + "num_input_tokens_seen": 126817915, + "step": 8050 + }, + { + "epoch": 0.6050653840372764, + "grad_norm": 1.3629811128691718, + "learning_rate": 1.4249945185126855e-06, + "loss": 0.9416, + "num_input_tokens_seen": 126842835, + "step": 8051 + }, + { + "epoch": 0.6051405381031114, + "grad_norm": 1.4604510145007348, + "learning_rate": 1.4245282479300199e-06, + "loss": 0.9234, + "num_input_tokens_seen": 126865360, + "step": 8052 + }, + { + "epoch": 0.6052156921689463, + "grad_norm": 1.2581396210640967, + "learning_rate": 1.424062011446259e-06, + "loss": 0.9882, + "num_input_tokens_seen": 126893240, + "step": 8053 + }, + { + "epoch": 0.6052908462347814, + "grad_norm": 1.7490836878057727, + "learning_rate": 1.4235958090890293e-06, + "loss": 0.8594, + "num_input_tokens_seen": 126913860, + "step": 8054 + }, + { + "epoch": 0.6053660003006163, + "grad_norm": 1.7558308600124808, + "learning_rate": 1.4231296408859553e-06, + "loss": 1.036, + "num_input_tokens_seen": 126936605, + "step": 8055 + }, + { + "epoch": 0.6054411543664512, + "grad_norm": 1.3604200884010236, + "learning_rate": 1.4226635068646586e-06, + "loss": 0.9572, + "num_input_tokens_seen": 126961715, + "step": 8056 + }, + { + "epoch": 0.6055163084322862, + "grad_norm": 1.454037615597768, + "learning_rate": 1.4221974070527606e-06, + "loss": 0.8852, + "num_input_tokens_seen": 126987100, + "step": 8057 + }, + { + "epoch": 0.6055914624981211, + "grad_norm": 1.387846324416552, + "learning_rate": 1.4217313414778786e-06, + "loss": 0.9157, + "num_input_tokens_seen": 127011580, + "step": 8058 + }, + { + "epoch": 0.6056666165639562, + "grad_norm": 1.5511615584718947, + "learning_rate": 1.4212653101676285e-06, + "loss": 0.8845, + "num_input_tokens_seen": 127035760, + "step": 8059 + }, + { + "epoch": 0.6057417706297911, + "grad_norm": 1.6108006209229002, + "learning_rate": 1.4207993131496254e-06, + "loss": 0.8914, + "num_input_tokens_seen": 127058075, + "step": 8060 + }, + { + "epoch": 0.605816924695626, + "grad_norm": 1.4089719919327366, + "learning_rate": 1.4203333504514805e-06, + "loss": 0.9221, + "num_input_tokens_seen": 127082660, + "step": 8061 + }, + { + "epoch": 0.605892078761461, + "grad_norm": 1.4690248759729758, + "learning_rate": 1.4198674221008045e-06, + "loss": 0.9143, + "num_input_tokens_seen": 127106035, + "step": 8062 + }, + { + "epoch": 0.6059672328272959, + "grad_norm": 1.2554764942625865, + "learning_rate": 1.419401528125205e-06, + "loss": 0.8703, + "num_input_tokens_seen": 127130480, + "step": 8063 + }, + { + "epoch": 0.606042386893131, + "grad_norm": 1.5002743152287379, + "learning_rate": 1.4189356685522884e-06, + "loss": 1.0368, + "num_input_tokens_seen": 127154235, + "step": 8064 + }, + { + "epoch": 0.6061175409589659, + "grad_norm": 1.4301972861348382, + "learning_rate": 1.4184698434096586e-06, + "loss": 0.824, + "num_input_tokens_seen": 127182335, + "step": 8065 + }, + { + "epoch": 0.6061926950248009, + "grad_norm": 1.4920211619392536, + "learning_rate": 1.4180040527249172e-06, + "loss": 1.0081, + "num_input_tokens_seen": 127206410, + "step": 8066 + }, + { + "epoch": 0.6062678490906358, + "grad_norm": 1.3606450516673672, + "learning_rate": 1.4175382965256644e-06, + "loss": 1.0065, + "num_input_tokens_seen": 127230090, + "step": 8067 + }, + { + "epoch": 0.6063430031564707, + "grad_norm": 1.7617199003296202, + "learning_rate": 1.4170725748394977e-06, + "loss": 0.866, + "num_input_tokens_seen": 127254575, + "step": 8068 + }, + { + "epoch": 0.6064181572223057, + "grad_norm": 1.6084169665846033, + "learning_rate": 1.4166068876940135e-06, + "loss": 1.0001, + "num_input_tokens_seen": 127276685, + "step": 8069 + }, + { + "epoch": 0.6064933112881407, + "grad_norm": 1.7569594284175911, + "learning_rate": 1.4161412351168053e-06, + "loss": 0.7929, + "num_input_tokens_seen": 127300520, + "step": 8070 + }, + { + "epoch": 0.6065684653539757, + "grad_norm": 1.682920670995303, + "learning_rate": 1.4156756171354637e-06, + "loss": 0.9595, + "num_input_tokens_seen": 127321690, + "step": 8071 + }, + { + "epoch": 0.6066436194198106, + "grad_norm": 1.8200526891145086, + "learning_rate": 1.4152100337775804e-06, + "loss": 1.063, + "num_input_tokens_seen": 127343510, + "step": 8072 + }, + { + "epoch": 0.6067187734856456, + "grad_norm": 1.525118169740018, + "learning_rate": 1.414744485070741e-06, + "loss": 1.0327, + "num_input_tokens_seen": 127365935, + "step": 8073 + }, + { + "epoch": 0.6067939275514805, + "grad_norm": 1.3767489234528039, + "learning_rate": 1.4142789710425325e-06, + "loss": 1.0387, + "num_input_tokens_seen": 127388310, + "step": 8074 + }, + { + "epoch": 0.6068690816173155, + "grad_norm": 1.6308645648074236, + "learning_rate": 1.4138134917205377e-06, + "loss": 0.9082, + "num_input_tokens_seen": 127412130, + "step": 8075 + }, + { + "epoch": 0.6069442356831505, + "grad_norm": 1.311612374023464, + "learning_rate": 1.413348047132338e-06, + "loss": 0.9353, + "num_input_tokens_seen": 127436470, + "step": 8076 + }, + { + "epoch": 0.6070193897489854, + "grad_norm": 1.8084665340230646, + "learning_rate": 1.4128826373055134e-06, + "loss": 0.9473, + "num_input_tokens_seen": 127454705, + "step": 8077 + }, + { + "epoch": 0.6070945438148204, + "grad_norm": 1.8460172517079887, + "learning_rate": 1.4124172622676406e-06, + "loss": 0.9678, + "num_input_tokens_seen": 127472785, + "step": 8078 + }, + { + "epoch": 0.6071696978806553, + "grad_norm": 1.5007957096386244, + "learning_rate": 1.411951922046295e-06, + "loss": 1.0796, + "num_input_tokens_seen": 127494775, + "step": 8079 + }, + { + "epoch": 0.6072448519464904, + "grad_norm": 1.4656177154555194, + "learning_rate": 1.4114866166690498e-06, + "loss": 0.9525, + "num_input_tokens_seen": 127518085, + "step": 8080 + }, + { + "epoch": 0.6073200060123253, + "grad_norm": 1.6749521618744432, + "learning_rate": 1.411021346163476e-06, + "loss": 0.8666, + "num_input_tokens_seen": 127543710, + "step": 8081 + }, + { + "epoch": 0.6073951600781602, + "grad_norm": 1.4088702056968356, + "learning_rate": 1.4105561105571428e-06, + "loss": 1.0006, + "num_input_tokens_seen": 127568215, + "step": 8082 + }, + { + "epoch": 0.6074703141439952, + "grad_norm": 1.4341808644814216, + "learning_rate": 1.410090909877617e-06, + "loss": 0.9014, + "num_input_tokens_seen": 127593955, + "step": 8083 + }, + { + "epoch": 0.6075454682098301, + "grad_norm": 1.680650940422154, + "learning_rate": 1.4096257441524643e-06, + "loss": 0.9166, + "num_input_tokens_seen": 127616570, + "step": 8084 + }, + { + "epoch": 0.6076206222756652, + "grad_norm": 1.5890110093798273, + "learning_rate": 1.4091606134092465e-06, + "loss": 0.9118, + "num_input_tokens_seen": 127636935, + "step": 8085 + }, + { + "epoch": 0.6076957763415001, + "grad_norm": 1.8288547982821006, + "learning_rate": 1.4086955176755248e-06, + "loss": 0.8617, + "num_input_tokens_seen": 127661365, + "step": 8086 + }, + { + "epoch": 0.607770930407335, + "grad_norm": 1.7669435826887363, + "learning_rate": 1.4082304569788582e-06, + "loss": 0.9056, + "num_input_tokens_seen": 127685385, + "step": 8087 + }, + { + "epoch": 0.60784608447317, + "grad_norm": 1.6605389322267254, + "learning_rate": 1.407765431346803e-06, + "loss": 0.94, + "num_input_tokens_seen": 127711235, + "step": 8088 + }, + { + "epoch": 0.6079212385390049, + "grad_norm": 0.8295958789877586, + "learning_rate": 1.4073004408069138e-06, + "loss": 0.9132, + "num_input_tokens_seen": 127778120, + "step": 8089 + }, + { + "epoch": 0.60799639260484, + "grad_norm": 1.6420281359205338, + "learning_rate": 1.4068354853867429e-06, + "loss": 0.9952, + "num_input_tokens_seen": 127799160, + "step": 8090 + }, + { + "epoch": 0.6080715466706749, + "grad_norm": 1.4407131110199458, + "learning_rate": 1.406370565113841e-06, + "loss": 0.9392, + "num_input_tokens_seen": 127822375, + "step": 8091 + }, + { + "epoch": 0.6081467007365099, + "grad_norm": 1.4920036642065384, + "learning_rate": 1.4059056800157567e-06, + "loss": 1.0383, + "num_input_tokens_seen": 127844075, + "step": 8092 + }, + { + "epoch": 0.6082218548023448, + "grad_norm": 1.830790085048052, + "learning_rate": 1.4054408301200345e-06, + "loss": 0.9871, + "num_input_tokens_seen": 127869075, + "step": 8093 + }, + { + "epoch": 0.6082970088681797, + "grad_norm": 1.6617741140191618, + "learning_rate": 1.4049760154542214e-06, + "loss": 0.9393, + "num_input_tokens_seen": 127893720, + "step": 8094 + }, + { + "epoch": 0.6083721629340147, + "grad_norm": 1.4359261563157075, + "learning_rate": 1.4045112360458564e-06, + "loss": 0.9395, + "num_input_tokens_seen": 127922030, + "step": 8095 + }, + { + "epoch": 0.6084473169998497, + "grad_norm": 1.76723683471352, + "learning_rate": 1.404046491922482e-06, + "loss": 0.9061, + "num_input_tokens_seen": 127943335, + "step": 8096 + }, + { + "epoch": 0.6085224710656847, + "grad_norm": 1.3928225437869464, + "learning_rate": 1.403581783111635e-06, + "loss": 1.0009, + "num_input_tokens_seen": 127967500, + "step": 8097 + }, + { + "epoch": 0.6085976251315196, + "grad_norm": 1.5443499214115284, + "learning_rate": 1.4031171096408506e-06, + "loss": 0.9376, + "num_input_tokens_seen": 127992125, + "step": 8098 + }, + { + "epoch": 0.6086727791973546, + "grad_norm": 1.9064188397788073, + "learning_rate": 1.4026524715376637e-06, + "loss": 0.9257, + "num_input_tokens_seen": 128013485, + "step": 8099 + }, + { + "epoch": 0.6087479332631895, + "grad_norm": 1.674053804546971, + "learning_rate": 1.4021878688296047e-06, + "loss": 0.97, + "num_input_tokens_seen": 128036850, + "step": 8100 + }, + { + "epoch": 0.6088230873290245, + "grad_norm": 1.7904202835203904, + "learning_rate": 1.401723301544204e-06, + "loss": 0.9709, + "num_input_tokens_seen": 128061075, + "step": 8101 + }, + { + "epoch": 0.6088982413948595, + "grad_norm": 1.6270829569107512, + "learning_rate": 1.4012587697089885e-06, + "loss": 0.9828, + "num_input_tokens_seen": 128084770, + "step": 8102 + }, + { + "epoch": 0.6089733954606944, + "grad_norm": 1.507261343516748, + "learning_rate": 1.4007942733514836e-06, + "loss": 0.8789, + "num_input_tokens_seen": 128108125, + "step": 8103 + }, + { + "epoch": 0.6090485495265294, + "grad_norm": 1.3116885810549865, + "learning_rate": 1.400329812499213e-06, + "loss": 1.0469, + "num_input_tokens_seen": 128132910, + "step": 8104 + }, + { + "epoch": 0.6091237035923643, + "grad_norm": 1.5244472189699756, + "learning_rate": 1.3998653871796964e-06, + "loss": 1.051, + "num_input_tokens_seen": 128155260, + "step": 8105 + }, + { + "epoch": 0.6091988576581993, + "grad_norm": 1.6887988990237985, + "learning_rate": 1.3994009974204547e-06, + "loss": 0.9099, + "num_input_tokens_seen": 128178680, + "step": 8106 + }, + { + "epoch": 0.6092740117240343, + "grad_norm": 1.5499353672211194, + "learning_rate": 1.3989366432490028e-06, + "loss": 0.9915, + "num_input_tokens_seen": 128201825, + "step": 8107 + }, + { + "epoch": 0.6093491657898692, + "grad_norm": 1.4319324482216826, + "learning_rate": 1.3984723246928569e-06, + "loss": 0.9823, + "num_input_tokens_seen": 128225820, + "step": 8108 + }, + { + "epoch": 0.6094243198557042, + "grad_norm": 1.803003822102571, + "learning_rate": 1.3980080417795296e-06, + "loss": 0.9189, + "num_input_tokens_seen": 128246540, + "step": 8109 + }, + { + "epoch": 0.6094994739215391, + "grad_norm": 1.594205604206649, + "learning_rate": 1.39754379453653e-06, + "loss": 1.0371, + "num_input_tokens_seen": 128267200, + "step": 8110 + }, + { + "epoch": 0.6095746279873742, + "grad_norm": 1.5403434728324346, + "learning_rate": 1.3970795829913682e-06, + "loss": 0.9562, + "num_input_tokens_seen": 128291715, + "step": 8111 + }, + { + "epoch": 0.6096497820532091, + "grad_norm": 1.7062969941874266, + "learning_rate": 1.396615407171549e-06, + "loss": 0.9765, + "num_input_tokens_seen": 128313760, + "step": 8112 + }, + { + "epoch": 0.609724936119044, + "grad_norm": 0.7337855550414618, + "learning_rate": 1.3961512671045787e-06, + "loss": 0.8161, + "num_input_tokens_seen": 128397260, + "step": 8113 + }, + { + "epoch": 0.609800090184879, + "grad_norm": 1.4592701582242535, + "learning_rate": 1.3956871628179577e-06, + "loss": 1.0279, + "num_input_tokens_seen": 128423410, + "step": 8114 + }, + { + "epoch": 0.6098752442507139, + "grad_norm": 1.320449551122392, + "learning_rate": 1.3952230943391856e-06, + "loss": 0.977, + "num_input_tokens_seen": 128451640, + "step": 8115 + }, + { + "epoch": 0.609950398316549, + "grad_norm": 1.677090220308873, + "learning_rate": 1.3947590616957618e-06, + "loss": 0.8096, + "num_input_tokens_seen": 128475495, + "step": 8116 + }, + { + "epoch": 0.6100255523823839, + "grad_norm": 1.4664474457130088, + "learning_rate": 1.3942950649151808e-06, + "loss": 0.9641, + "num_input_tokens_seen": 128499265, + "step": 8117 + }, + { + "epoch": 0.6101007064482189, + "grad_norm": 1.9714054183311187, + "learning_rate": 1.3938311040249371e-06, + "loss": 0.9566, + "num_input_tokens_seen": 128521170, + "step": 8118 + }, + { + "epoch": 0.6101758605140538, + "grad_norm": 1.6323608065700204, + "learning_rate": 1.3933671790525215e-06, + "loss": 0.987, + "num_input_tokens_seen": 128547435, + "step": 8119 + }, + { + "epoch": 0.6102510145798887, + "grad_norm": 1.2231529616257626, + "learning_rate": 1.3929032900254232e-06, + "loss": 0.9685, + "num_input_tokens_seen": 128573530, + "step": 8120 + }, + { + "epoch": 0.6103261686457238, + "grad_norm": 1.2849996639503605, + "learning_rate": 1.39243943697113e-06, + "loss": 0.9491, + "num_input_tokens_seen": 128597610, + "step": 8121 + }, + { + "epoch": 0.6104013227115587, + "grad_norm": 1.8258527017828652, + "learning_rate": 1.3919756199171263e-06, + "loss": 0.9689, + "num_input_tokens_seen": 128621710, + "step": 8122 + }, + { + "epoch": 0.6104764767773937, + "grad_norm": 1.7979674625698767, + "learning_rate": 1.3915118388908958e-06, + "loss": 0.9828, + "num_input_tokens_seen": 128642720, + "step": 8123 + }, + { + "epoch": 0.6105516308432286, + "grad_norm": 0.7497942960296794, + "learning_rate": 1.3910480939199184e-06, + "loss": 0.7996, + "num_input_tokens_seen": 128725060, + "step": 8124 + }, + { + "epoch": 0.6106267849090636, + "grad_norm": 1.6166426620798702, + "learning_rate": 1.3905843850316738e-06, + "loss": 0.8096, + "num_input_tokens_seen": 128747235, + "step": 8125 + }, + { + "epoch": 0.6107019389748986, + "grad_norm": 1.5621635074684603, + "learning_rate": 1.3901207122536383e-06, + "loss": 0.9885, + "num_input_tokens_seen": 128770390, + "step": 8126 + }, + { + "epoch": 0.6107770930407335, + "grad_norm": 1.3947531472837442, + "learning_rate": 1.3896570756132851e-06, + "loss": 0.9398, + "num_input_tokens_seen": 128794335, + "step": 8127 + }, + { + "epoch": 0.6108522471065685, + "grad_norm": 1.3710587108514662, + "learning_rate": 1.3891934751380879e-06, + "loss": 0.9325, + "num_input_tokens_seen": 128824890, + "step": 8128 + }, + { + "epoch": 0.6109274011724034, + "grad_norm": 1.2726081534629727, + "learning_rate": 1.3887299108555158e-06, + "loss": 0.9522, + "num_input_tokens_seen": 128849980, + "step": 8129 + }, + { + "epoch": 0.6110025552382384, + "grad_norm": 1.2629013896451502, + "learning_rate": 1.3882663827930375e-06, + "loss": 1.01, + "num_input_tokens_seen": 128875365, + "step": 8130 + }, + { + "epoch": 0.6110777093040733, + "grad_norm": 1.292534885046392, + "learning_rate": 1.3878028909781187e-06, + "loss": 0.9175, + "num_input_tokens_seen": 128899530, + "step": 8131 + }, + { + "epoch": 0.6111528633699083, + "grad_norm": 1.31884535132909, + "learning_rate": 1.3873394354382225e-06, + "loss": 0.9588, + "num_input_tokens_seen": 128924275, + "step": 8132 + }, + { + "epoch": 0.6112280174357433, + "grad_norm": 1.6060804700631781, + "learning_rate": 1.3868760162008108e-06, + "loss": 0.9426, + "num_input_tokens_seen": 128947405, + "step": 8133 + }, + { + "epoch": 0.6113031715015782, + "grad_norm": 1.2590485653800876, + "learning_rate": 1.3864126332933425e-06, + "loss": 0.9326, + "num_input_tokens_seen": 128972300, + "step": 8134 + }, + { + "epoch": 0.6113783255674132, + "grad_norm": 2.3808432524513075, + "learning_rate": 1.3859492867432765e-06, + "loss": 0.9681, + "num_input_tokens_seen": 128993890, + "step": 8135 + }, + { + "epoch": 0.6114534796332481, + "grad_norm": 1.7827877300390553, + "learning_rate": 1.385485976578066e-06, + "loss": 0.9092, + "num_input_tokens_seen": 129018340, + "step": 8136 + }, + { + "epoch": 0.6115286336990832, + "grad_norm": 1.3630713708429647, + "learning_rate": 1.3850227028251639e-06, + "loss": 0.977, + "num_input_tokens_seen": 129043375, + "step": 8137 + }, + { + "epoch": 0.6116037877649181, + "grad_norm": 1.4128272631134808, + "learning_rate": 1.3845594655120224e-06, + "loss": 0.9437, + "num_input_tokens_seen": 129066935, + "step": 8138 + }, + { + "epoch": 0.611678941830753, + "grad_norm": 1.6517489469700073, + "learning_rate": 1.3840962646660885e-06, + "loss": 0.8574, + "num_input_tokens_seen": 129089865, + "step": 8139 + }, + { + "epoch": 0.611754095896588, + "grad_norm": 1.4791487392718166, + "learning_rate": 1.3836331003148101e-06, + "loss": 0.9912, + "num_input_tokens_seen": 129110755, + "step": 8140 + }, + { + "epoch": 0.6118292499624229, + "grad_norm": 1.8046647107555631, + "learning_rate": 1.3831699724856307e-06, + "loss": 0.9198, + "num_input_tokens_seen": 129130305, + "step": 8141 + }, + { + "epoch": 0.611904404028258, + "grad_norm": 1.4707550809682657, + "learning_rate": 1.3827068812059918e-06, + "loss": 0.8708, + "num_input_tokens_seen": 129153930, + "step": 8142 + }, + { + "epoch": 0.6119795580940929, + "grad_norm": 1.3523861354486142, + "learning_rate": 1.3822438265033345e-06, + "loss": 0.9424, + "num_input_tokens_seen": 129176885, + "step": 8143 + }, + { + "epoch": 0.6120547121599279, + "grad_norm": 1.6016338797689091, + "learning_rate": 1.3817808084050957e-06, + "loss": 0.8677, + "num_input_tokens_seen": 129202045, + "step": 8144 + }, + { + "epoch": 0.6121298662257628, + "grad_norm": 1.5014682577451024, + "learning_rate": 1.3813178269387119e-06, + "loss": 0.9796, + "num_input_tokens_seen": 129227490, + "step": 8145 + }, + { + "epoch": 0.6122050202915977, + "grad_norm": 1.5253787461525308, + "learning_rate": 1.380854882131615e-06, + "loss": 0.9021, + "num_input_tokens_seen": 129250845, + "step": 8146 + }, + { + "epoch": 0.6122801743574328, + "grad_norm": 1.6778696558526975, + "learning_rate": 1.3803919740112383e-06, + "loss": 0.8328, + "num_input_tokens_seen": 129275405, + "step": 8147 + }, + { + "epoch": 0.6123553284232677, + "grad_norm": 2.2283229360097736, + "learning_rate": 1.379929102605009e-06, + "loss": 1.0065, + "num_input_tokens_seen": 129296520, + "step": 8148 + }, + { + "epoch": 0.6124304824891027, + "grad_norm": 1.6025328883041101, + "learning_rate": 1.379466267940355e-06, + "loss": 0.8825, + "num_input_tokens_seen": 129318300, + "step": 8149 + }, + { + "epoch": 0.6125056365549376, + "grad_norm": 1.4648662921442603, + "learning_rate": 1.3790034700447008e-06, + "loss": 0.9354, + "num_input_tokens_seen": 129344040, + "step": 8150 + }, + { + "epoch": 0.6125807906207725, + "grad_norm": 1.3682717084230689, + "learning_rate": 1.378540708945469e-06, + "loss": 1.0231, + "num_input_tokens_seen": 129369220, + "step": 8151 + }, + { + "epoch": 0.6126559446866076, + "grad_norm": 1.3571677429742741, + "learning_rate": 1.3780779846700799e-06, + "loss": 0.9867, + "num_input_tokens_seen": 129394730, + "step": 8152 + }, + { + "epoch": 0.6127310987524425, + "grad_norm": 1.5721750932301382, + "learning_rate": 1.3776152972459517e-06, + "loss": 0.8708, + "num_input_tokens_seen": 129420205, + "step": 8153 + }, + { + "epoch": 0.6128062528182775, + "grad_norm": 1.864169630759686, + "learning_rate": 1.3771526467005004e-06, + "loss": 0.9385, + "num_input_tokens_seen": 129442730, + "step": 8154 + }, + { + "epoch": 0.6128814068841124, + "grad_norm": 1.3656584990907925, + "learning_rate": 1.37669003306114e-06, + "loss": 1.0267, + "num_input_tokens_seen": 129468060, + "step": 8155 + }, + { + "epoch": 0.6129565609499474, + "grad_norm": 1.481363715462551, + "learning_rate": 1.3762274563552811e-06, + "loss": 0.945, + "num_input_tokens_seen": 129491615, + "step": 8156 + }, + { + "epoch": 0.6130317150157824, + "grad_norm": 1.4863740009140571, + "learning_rate": 1.375764916610335e-06, + "loss": 0.8903, + "num_input_tokens_seen": 129519550, + "step": 8157 + }, + { + "epoch": 0.6131068690816173, + "grad_norm": 0.7690852465437518, + "learning_rate": 1.3753024138537082e-06, + "loss": 0.7872, + "num_input_tokens_seen": 129589670, + "step": 8158 + }, + { + "epoch": 0.6131820231474523, + "grad_norm": 1.6638470720708955, + "learning_rate": 1.3748399481128043e-06, + "loss": 0.8906, + "num_input_tokens_seen": 129610770, + "step": 8159 + }, + { + "epoch": 0.6132571772132872, + "grad_norm": 1.541066061337967, + "learning_rate": 1.3743775194150281e-06, + "loss": 1.0116, + "num_input_tokens_seen": 129631960, + "step": 8160 + }, + { + "epoch": 0.6133323312791222, + "grad_norm": 1.6090366044797315, + "learning_rate": 1.3739151277877792e-06, + "loss": 0.9115, + "num_input_tokens_seen": 129659285, + "step": 8161 + }, + { + "epoch": 0.6134074853449571, + "grad_norm": 1.6795689030851253, + "learning_rate": 1.3734527732584568e-06, + "loss": 0.9554, + "num_input_tokens_seen": 129682305, + "step": 8162 + }, + { + "epoch": 0.6134826394107922, + "grad_norm": 1.5173602837687123, + "learning_rate": 1.372990455854457e-06, + "loss": 0.9226, + "num_input_tokens_seen": 129705275, + "step": 8163 + }, + { + "epoch": 0.6135577934766271, + "grad_norm": 1.421742862807595, + "learning_rate": 1.372528175603173e-06, + "loss": 0.9112, + "num_input_tokens_seen": 129730480, + "step": 8164 + }, + { + "epoch": 0.613632947542462, + "grad_norm": 1.6603381966013748, + "learning_rate": 1.372065932531998e-06, + "loss": 0.9258, + "num_input_tokens_seen": 129756630, + "step": 8165 + }, + { + "epoch": 0.613708101608297, + "grad_norm": 1.9651683743971748, + "learning_rate": 1.3716037266683203e-06, + "loss": 0.9316, + "num_input_tokens_seen": 129780140, + "step": 8166 + }, + { + "epoch": 0.613783255674132, + "grad_norm": 1.9438394998141173, + "learning_rate": 1.3711415580395288e-06, + "loss": 0.8388, + "num_input_tokens_seen": 129804430, + "step": 8167 + }, + { + "epoch": 0.613858409739967, + "grad_norm": 1.6970337598518854, + "learning_rate": 1.3706794266730072e-06, + "loss": 0.9853, + "num_input_tokens_seen": 129828590, + "step": 8168 + }, + { + "epoch": 0.6139335638058019, + "grad_norm": 1.6495115568664447, + "learning_rate": 1.37021733259614e-06, + "loss": 0.8794, + "num_input_tokens_seen": 129852450, + "step": 8169 + }, + { + "epoch": 0.6140087178716369, + "grad_norm": 1.9958910456564487, + "learning_rate": 1.3697552758363079e-06, + "loss": 0.7781, + "num_input_tokens_seen": 129881010, + "step": 8170 + }, + { + "epoch": 0.6140838719374718, + "grad_norm": 1.4846984510874999, + "learning_rate": 1.3692932564208884e-06, + "loss": 0.9426, + "num_input_tokens_seen": 129906140, + "step": 8171 + }, + { + "epoch": 0.6141590260033067, + "grad_norm": 1.2714085241332336, + "learning_rate": 1.3688312743772588e-06, + "loss": 0.951, + "num_input_tokens_seen": 129932320, + "step": 8172 + }, + { + "epoch": 0.6142341800691418, + "grad_norm": 1.4568626163043084, + "learning_rate": 1.3683693297327927e-06, + "loss": 0.9703, + "num_input_tokens_seen": 129956275, + "step": 8173 + }, + { + "epoch": 0.6143093341349767, + "grad_norm": 1.6768195930167993, + "learning_rate": 1.367907422514863e-06, + "loss": 0.9139, + "num_input_tokens_seen": 129977200, + "step": 8174 + }, + { + "epoch": 0.6143844882008117, + "grad_norm": 5.326618417864307, + "learning_rate": 1.367445552750839e-06, + "loss": 0.9646, + "num_input_tokens_seen": 129999755, + "step": 8175 + }, + { + "epoch": 0.6144596422666466, + "grad_norm": 1.7306424090390473, + "learning_rate": 1.3669837204680876e-06, + "loss": 0.8248, + "num_input_tokens_seen": 130021690, + "step": 8176 + }, + { + "epoch": 0.6145347963324815, + "grad_norm": 3.0837773355633953, + "learning_rate": 1.3665219256939753e-06, + "loss": 0.9621, + "num_input_tokens_seen": 130047715, + "step": 8177 + }, + { + "epoch": 0.6146099503983166, + "grad_norm": 1.344852882053753, + "learning_rate": 1.3660601684558639e-06, + "loss": 0.8274, + "num_input_tokens_seen": 130075610, + "step": 8178 + }, + { + "epoch": 0.6146851044641515, + "grad_norm": 1.7409752703931096, + "learning_rate": 1.3655984487811158e-06, + "loss": 0.9768, + "num_input_tokens_seen": 130100815, + "step": 8179 + }, + { + "epoch": 0.6147602585299865, + "grad_norm": 1.4459940360224603, + "learning_rate": 1.3651367666970895e-06, + "loss": 1.0143, + "num_input_tokens_seen": 130122670, + "step": 8180 + }, + { + "epoch": 0.6148354125958214, + "grad_norm": 1.437921462263351, + "learning_rate": 1.3646751222311392e-06, + "loss": 0.8958, + "num_input_tokens_seen": 130149740, + "step": 8181 + }, + { + "epoch": 0.6149105666616564, + "grad_norm": 1.4397520585652317, + "learning_rate": 1.3642135154106219e-06, + "loss": 0.9523, + "num_input_tokens_seen": 130176960, + "step": 8182 + }, + { + "epoch": 0.6149857207274914, + "grad_norm": 0.6876391573434801, + "learning_rate": 1.3637519462628876e-06, + "loss": 0.7461, + "num_input_tokens_seen": 130249210, + "step": 8183 + }, + { + "epoch": 0.6150608747933263, + "grad_norm": 1.8995940527669635, + "learning_rate": 1.3632904148152877e-06, + "loss": 0.9056, + "num_input_tokens_seen": 130276715, + "step": 8184 + }, + { + "epoch": 0.6151360288591613, + "grad_norm": 1.60979396958958, + "learning_rate": 1.3628289210951687e-06, + "loss": 0.9592, + "num_input_tokens_seen": 130299455, + "step": 8185 + }, + { + "epoch": 0.6152111829249962, + "grad_norm": 1.358193289474584, + "learning_rate": 1.3623674651298752e-06, + "loss": 0.8286, + "num_input_tokens_seen": 130325700, + "step": 8186 + }, + { + "epoch": 0.6152863369908312, + "grad_norm": 1.689273749735131, + "learning_rate": 1.361906046946752e-06, + "loss": 0.9633, + "num_input_tokens_seen": 130349090, + "step": 8187 + }, + { + "epoch": 0.6153614910566662, + "grad_norm": 1.472495720408621, + "learning_rate": 1.3614446665731385e-06, + "loss": 0.8262, + "num_input_tokens_seen": 130372185, + "step": 8188 + }, + { + "epoch": 0.6154366451225012, + "grad_norm": 1.3515844618936865, + "learning_rate": 1.3609833240363738e-06, + "loss": 0.9559, + "num_input_tokens_seen": 130397060, + "step": 8189 + }, + { + "epoch": 0.6155117991883361, + "grad_norm": 1.9260131201133799, + "learning_rate": 1.3605220193637942e-06, + "loss": 0.8936, + "num_input_tokens_seen": 130418575, + "step": 8190 + }, + { + "epoch": 0.615586953254171, + "grad_norm": 1.682709711844178, + "learning_rate": 1.3600607525827335e-06, + "loss": 0.9247, + "num_input_tokens_seen": 130442330, + "step": 8191 + }, + { + "epoch": 0.615662107320006, + "grad_norm": 0.8731234728090532, + "learning_rate": 1.359599523720524e-06, + "loss": 0.8766, + "num_input_tokens_seen": 130515445, + "step": 8192 + }, + { + "epoch": 0.615737261385841, + "grad_norm": 1.71022220796315, + "learning_rate": 1.3591383328044943e-06, + "loss": 0.8729, + "num_input_tokens_seen": 130537660, + "step": 8193 + }, + { + "epoch": 0.615812415451676, + "grad_norm": 1.5183863702940135, + "learning_rate": 1.358677179861973e-06, + "loss": 1.0691, + "num_input_tokens_seen": 130559805, + "step": 8194 + }, + { + "epoch": 0.6158875695175109, + "grad_norm": 1.7305476943494544, + "learning_rate": 1.3582160649202844e-06, + "loss": 1.0432, + "num_input_tokens_seen": 130577960, + "step": 8195 + }, + { + "epoch": 0.6159627235833458, + "grad_norm": 1.6585358734023292, + "learning_rate": 1.3577549880067516e-06, + "loss": 0.9987, + "num_input_tokens_seen": 130599445, + "step": 8196 + }, + { + "epoch": 0.6160378776491808, + "grad_norm": 2.0140049064220378, + "learning_rate": 1.3572939491486952e-06, + "loss": 1.0105, + "num_input_tokens_seen": 130621825, + "step": 8197 + }, + { + "epoch": 0.6161130317150157, + "grad_norm": 1.4528714850996203, + "learning_rate": 1.3568329483734329e-06, + "loss": 0.9951, + "num_input_tokens_seen": 130644830, + "step": 8198 + }, + { + "epoch": 0.6161881857808508, + "grad_norm": 1.464358725040144, + "learning_rate": 1.3563719857082817e-06, + "loss": 0.944, + "num_input_tokens_seen": 130669265, + "step": 8199 + }, + { + "epoch": 0.6162633398466857, + "grad_norm": 1.5144939331507457, + "learning_rate": 1.3559110611805542e-06, + "loss": 0.9724, + "num_input_tokens_seen": 130692250, + "step": 8200 + }, + { + "epoch": 0.6163384939125207, + "grad_norm": 1.4499467116462486, + "learning_rate": 1.3554501748175637e-06, + "loss": 0.8977, + "num_input_tokens_seen": 130717415, + "step": 8201 + }, + { + "epoch": 0.6164136479783556, + "grad_norm": 1.2106173184318039, + "learning_rate": 1.3549893266466188e-06, + "loss": 0.9386, + "num_input_tokens_seen": 130744200, + "step": 8202 + }, + { + "epoch": 0.6164888020441905, + "grad_norm": 1.7925471722342647, + "learning_rate": 1.3545285166950246e-06, + "loss": 0.9151, + "num_input_tokens_seen": 130765155, + "step": 8203 + }, + { + "epoch": 0.6165639561100256, + "grad_norm": 1.614849372193285, + "learning_rate": 1.3540677449900887e-06, + "loss": 0.9669, + "num_input_tokens_seen": 130787920, + "step": 8204 + }, + { + "epoch": 0.6166391101758605, + "grad_norm": 1.8572318925579474, + "learning_rate": 1.3536070115591118e-06, + "loss": 0.9263, + "num_input_tokens_seen": 130809975, + "step": 8205 + }, + { + "epoch": 0.6167142642416955, + "grad_norm": 1.51195687875364, + "learning_rate": 1.3531463164293952e-06, + "loss": 1.0429, + "num_input_tokens_seen": 130831605, + "step": 8206 + }, + { + "epoch": 0.6167894183075304, + "grad_norm": 1.3975873526089908, + "learning_rate": 1.352685659628236e-06, + "loss": 0.85, + "num_input_tokens_seen": 130859640, + "step": 8207 + }, + { + "epoch": 0.6168645723733654, + "grad_norm": 1.2895233255371894, + "learning_rate": 1.3522250411829301e-06, + "loss": 0.9475, + "num_input_tokens_seen": 130885810, + "step": 8208 + }, + { + "epoch": 0.6169397264392004, + "grad_norm": 1.8139350077458811, + "learning_rate": 1.3517644611207715e-06, + "loss": 0.9701, + "num_input_tokens_seen": 130907480, + "step": 8209 + }, + { + "epoch": 0.6170148805050353, + "grad_norm": 1.5961480144998754, + "learning_rate": 1.35130391946905e-06, + "loss": 0.9096, + "num_input_tokens_seen": 130929715, + "step": 8210 + }, + { + "epoch": 0.6170900345708703, + "grad_norm": 1.4215661017271022, + "learning_rate": 1.350843416255056e-06, + "loss": 0.9978, + "num_input_tokens_seen": 130955500, + "step": 8211 + }, + { + "epoch": 0.6171651886367052, + "grad_norm": 1.3258431972581903, + "learning_rate": 1.350382951506075e-06, + "loss": 0.8975, + "num_input_tokens_seen": 130979370, + "step": 8212 + }, + { + "epoch": 0.6172403427025402, + "grad_norm": 0.8102355359089092, + "learning_rate": 1.3499225252493918e-06, + "loss": 0.8351, + "num_input_tokens_seen": 131048735, + "step": 8213 + }, + { + "epoch": 0.6173154967683752, + "grad_norm": 1.5363978056111613, + "learning_rate": 1.3494621375122886e-06, + "loss": 1.0558, + "num_input_tokens_seen": 131067320, + "step": 8214 + }, + { + "epoch": 0.6173906508342102, + "grad_norm": 2.3313944344866386, + "learning_rate": 1.3490017883220443e-06, + "loss": 0.9832, + "num_input_tokens_seen": 131090150, + "step": 8215 + }, + { + "epoch": 0.6174658049000451, + "grad_norm": 0.7393758016986645, + "learning_rate": 1.3485414777059375e-06, + "loss": 0.7928, + "num_input_tokens_seen": 131162405, + "step": 8216 + }, + { + "epoch": 0.61754095896588, + "grad_norm": 2.2240587579594657, + "learning_rate": 1.3480812056912417e-06, + "loss": 0.8888, + "num_input_tokens_seen": 131183670, + "step": 8217 + }, + { + "epoch": 0.617616113031715, + "grad_norm": 1.934559220768935, + "learning_rate": 1.3476209723052318e-06, + "loss": 0.8729, + "num_input_tokens_seen": 131205320, + "step": 8218 + }, + { + "epoch": 0.61769126709755, + "grad_norm": 2.658766956499458, + "learning_rate": 1.3471607775751774e-06, + "loss": 0.8956, + "num_input_tokens_seen": 131227670, + "step": 8219 + }, + { + "epoch": 0.617766421163385, + "grad_norm": 2.133761352079795, + "learning_rate": 1.3467006215283459e-06, + "loss": 0.8717, + "num_input_tokens_seen": 131250650, + "step": 8220 + }, + { + "epoch": 0.6178415752292199, + "grad_norm": 0.7835933255530109, + "learning_rate": 1.3462405041920053e-06, + "loss": 0.8111, + "num_input_tokens_seen": 131317600, + "step": 8221 + }, + { + "epoch": 0.6179167292950548, + "grad_norm": 1.7158399047775128, + "learning_rate": 1.3457804255934172e-06, + "loss": 0.9279, + "num_input_tokens_seen": 131338650, + "step": 8222 + }, + { + "epoch": 0.6179918833608898, + "grad_norm": 1.843773275971988, + "learning_rate": 1.3453203857598449e-06, + "loss": 0.8511, + "num_input_tokens_seen": 131361125, + "step": 8223 + }, + { + "epoch": 0.6180670374267248, + "grad_norm": 1.7008923937150047, + "learning_rate": 1.3448603847185464e-06, + "loss": 0.8679, + "num_input_tokens_seen": 131384390, + "step": 8224 + }, + { + "epoch": 0.6181421914925598, + "grad_norm": 1.3661779981872852, + "learning_rate": 1.3444004224967787e-06, + "loss": 0.8532, + "num_input_tokens_seen": 131408150, + "step": 8225 + }, + { + "epoch": 0.6182173455583947, + "grad_norm": 1.5231700955971594, + "learning_rate": 1.3439404991217968e-06, + "loss": 0.8781, + "num_input_tokens_seen": 131433525, + "step": 8226 + }, + { + "epoch": 0.6182924996242297, + "grad_norm": 1.6483331439355962, + "learning_rate": 1.343480614620852e-06, + "loss": 0.9386, + "num_input_tokens_seen": 131455280, + "step": 8227 + }, + { + "epoch": 0.6183676536900646, + "grad_norm": 0.7677194402159542, + "learning_rate": 1.3430207690211953e-06, + "loss": 0.743, + "num_input_tokens_seen": 131534390, + "step": 8228 + }, + { + "epoch": 0.6184428077558995, + "grad_norm": 1.7447325587343088, + "learning_rate": 1.3425609623500738e-06, + "loss": 0.8218, + "num_input_tokens_seen": 131558600, + "step": 8229 + }, + { + "epoch": 0.6185179618217346, + "grad_norm": 1.5216630039957901, + "learning_rate": 1.3421011946347323e-06, + "loss": 0.8852, + "num_input_tokens_seen": 131579840, + "step": 8230 + }, + { + "epoch": 0.6185931158875695, + "grad_norm": 2.122598188306078, + "learning_rate": 1.3416414659024147e-06, + "loss": 0.9924, + "num_input_tokens_seen": 131602175, + "step": 8231 + }, + { + "epoch": 0.6186682699534045, + "grad_norm": 1.5381002758508262, + "learning_rate": 1.3411817761803608e-06, + "loss": 0.9568, + "num_input_tokens_seen": 131628165, + "step": 8232 + }, + { + "epoch": 0.6187434240192394, + "grad_norm": 1.940507216265421, + "learning_rate": 1.34072212549581e-06, + "loss": 1.0399, + "num_input_tokens_seen": 131649430, + "step": 8233 + }, + { + "epoch": 0.6188185780850745, + "grad_norm": 1.5500564226525344, + "learning_rate": 1.3402625138759972e-06, + "loss": 0.9347, + "num_input_tokens_seen": 131672250, + "step": 8234 + }, + { + "epoch": 0.6188937321509094, + "grad_norm": 0.8837126809513444, + "learning_rate": 1.3398029413481573e-06, + "loss": 0.7443, + "num_input_tokens_seen": 131754620, + "step": 8235 + }, + { + "epoch": 0.6189688862167443, + "grad_norm": 1.9454155706959628, + "learning_rate": 1.3393434079395212e-06, + "loss": 0.8719, + "num_input_tokens_seen": 131779060, + "step": 8236 + }, + { + "epoch": 0.6190440402825793, + "grad_norm": 1.4622878091886888, + "learning_rate": 1.3388839136773174e-06, + "loss": 0.9787, + "num_input_tokens_seen": 131802180, + "step": 8237 + }, + { + "epoch": 0.6191191943484142, + "grad_norm": 1.8774755030835835, + "learning_rate": 1.3384244585887738e-06, + "loss": 0.9911, + "num_input_tokens_seen": 131823620, + "step": 8238 + }, + { + "epoch": 0.6191943484142493, + "grad_norm": 0.8227468790269261, + "learning_rate": 1.3379650427011141e-06, + "loss": 0.7962, + "num_input_tokens_seen": 131899085, + "step": 8239 + }, + { + "epoch": 0.6192695024800842, + "grad_norm": 1.4725585419395342, + "learning_rate": 1.337505666041561e-06, + "loss": 1.0476, + "num_input_tokens_seen": 131922910, + "step": 8240 + }, + { + "epoch": 0.6193446565459191, + "grad_norm": 1.8091985955826013, + "learning_rate": 1.337046328637334e-06, + "loss": 0.9042, + "num_input_tokens_seen": 131945405, + "step": 8241 + }, + { + "epoch": 0.6194198106117541, + "grad_norm": 1.9733527738455547, + "learning_rate": 1.3365870305156502e-06, + "loss": 0.9391, + "num_input_tokens_seen": 131972330, + "step": 8242 + }, + { + "epoch": 0.619494964677589, + "grad_norm": 1.956581292420869, + "learning_rate": 1.336127771703726e-06, + "loss": 0.9562, + "num_input_tokens_seen": 131993120, + "step": 8243 + }, + { + "epoch": 0.619570118743424, + "grad_norm": 1.7640509551813395, + "learning_rate": 1.3356685522287724e-06, + "loss": 0.968, + "num_input_tokens_seen": 132014580, + "step": 8244 + }, + { + "epoch": 0.619645272809259, + "grad_norm": 0.7672094959594202, + "learning_rate": 1.3352093721180017e-06, + "loss": 0.7494, + "num_input_tokens_seen": 132086115, + "step": 8245 + }, + { + "epoch": 0.619720426875094, + "grad_norm": 1.6236536610885899, + "learning_rate": 1.3347502313986216e-06, + "loss": 0.774, + "num_input_tokens_seen": 132107630, + "step": 8246 + }, + { + "epoch": 0.6197955809409289, + "grad_norm": 1.5212718731977586, + "learning_rate": 1.3342911300978373e-06, + "loss": 0.9158, + "num_input_tokens_seen": 132131705, + "step": 8247 + }, + { + "epoch": 0.6198707350067638, + "grad_norm": 1.8062321381527215, + "learning_rate": 1.3338320682428527e-06, + "loss": 0.964, + "num_input_tokens_seen": 132157020, + "step": 8248 + }, + { + "epoch": 0.6199458890725988, + "grad_norm": 1.6988833743207499, + "learning_rate": 1.3333730458608688e-06, + "loss": 0.8975, + "num_input_tokens_seen": 132180145, + "step": 8249 + }, + { + "epoch": 0.6200210431384338, + "grad_norm": 1.536074298343188, + "learning_rate": 1.3329140629790851e-06, + "loss": 1.0008, + "num_input_tokens_seen": 132201440, + "step": 8250 + }, + { + "epoch": 0.6200961972042688, + "grad_norm": 1.4382309920913046, + "learning_rate": 1.3324551196246977e-06, + "loss": 0.939, + "num_input_tokens_seen": 132224130, + "step": 8251 + }, + { + "epoch": 0.6201713512701037, + "grad_norm": 1.310557517450501, + "learning_rate": 1.3319962158249e-06, + "loss": 0.9654, + "num_input_tokens_seen": 132248840, + "step": 8252 + }, + { + "epoch": 0.6202465053359387, + "grad_norm": 1.4775765786762665, + "learning_rate": 1.331537351606885e-06, + "loss": 0.9989, + "num_input_tokens_seen": 132271355, + "step": 8253 + }, + { + "epoch": 0.6203216594017736, + "grad_norm": 1.7018064241746, + "learning_rate": 1.3310785269978413e-06, + "loss": 0.8135, + "num_input_tokens_seen": 132293725, + "step": 8254 + }, + { + "epoch": 0.6203968134676086, + "grad_norm": 1.6014119891643457, + "learning_rate": 1.3306197420249566e-06, + "loss": 0.9381, + "num_input_tokens_seen": 132317605, + "step": 8255 + }, + { + "epoch": 0.6204719675334436, + "grad_norm": 1.7735208500168065, + "learning_rate": 1.3301609967154152e-06, + "loss": 1.0065, + "num_input_tokens_seen": 132337895, + "step": 8256 + }, + { + "epoch": 0.6205471215992785, + "grad_norm": 1.4349258428944414, + "learning_rate": 1.3297022910964e-06, + "loss": 0.9229, + "num_input_tokens_seen": 132362805, + "step": 8257 + }, + { + "epoch": 0.6206222756651135, + "grad_norm": 1.7645268358766348, + "learning_rate": 1.3292436251950906e-06, + "loss": 0.9762, + "num_input_tokens_seen": 132385995, + "step": 8258 + }, + { + "epoch": 0.6206974297309484, + "grad_norm": 1.5798154030174172, + "learning_rate": 1.3287849990386647e-06, + "loss": 0.97, + "num_input_tokens_seen": 132411880, + "step": 8259 + }, + { + "epoch": 0.6207725837967835, + "grad_norm": 1.1419427653866656, + "learning_rate": 1.3283264126542986e-06, + "loss": 0.8786, + "num_input_tokens_seen": 132439315, + "step": 8260 + }, + { + "epoch": 0.6208477378626184, + "grad_norm": 1.4568302128346013, + "learning_rate": 1.3278678660691638e-06, + "loss": 0.9331, + "num_input_tokens_seen": 132464480, + "step": 8261 + }, + { + "epoch": 0.6209228919284533, + "grad_norm": 1.5603422906905495, + "learning_rate": 1.327409359310432e-06, + "loss": 0.8802, + "num_input_tokens_seen": 132486840, + "step": 8262 + }, + { + "epoch": 0.6209980459942883, + "grad_norm": 2.6722020177766903, + "learning_rate": 1.3269508924052715e-06, + "loss": 0.9041, + "num_input_tokens_seen": 132509420, + "step": 8263 + }, + { + "epoch": 0.6210732000601232, + "grad_norm": 1.4594188283346912, + "learning_rate": 1.326492465380847e-06, + "loss": 0.9117, + "num_input_tokens_seen": 132533095, + "step": 8264 + }, + { + "epoch": 0.6211483541259583, + "grad_norm": 1.631665936055213, + "learning_rate": 1.326034078264324e-06, + "loss": 0.9606, + "num_input_tokens_seen": 132556555, + "step": 8265 + }, + { + "epoch": 0.6212235081917932, + "grad_norm": 1.746339102282931, + "learning_rate": 1.3255757310828614e-06, + "loss": 0.9347, + "num_input_tokens_seen": 132579840, + "step": 8266 + }, + { + "epoch": 0.6212986622576281, + "grad_norm": 1.0252868599840936, + "learning_rate": 1.3251174238636202e-06, + "loss": 0.8437, + "num_input_tokens_seen": 132634875, + "step": 8267 + }, + { + "epoch": 0.6213738163234631, + "grad_norm": 1.4799675004844648, + "learning_rate": 1.3246591566337563e-06, + "loss": 0.9587, + "num_input_tokens_seen": 132657575, + "step": 8268 + }, + { + "epoch": 0.621448970389298, + "grad_norm": 1.860393597786254, + "learning_rate": 1.3242009294204223e-06, + "loss": 0.8448, + "num_input_tokens_seen": 132680345, + "step": 8269 + }, + { + "epoch": 0.621524124455133, + "grad_norm": 2.6368950678895167, + "learning_rate": 1.3237427422507721e-06, + "loss": 0.9755, + "num_input_tokens_seen": 132702950, + "step": 8270 + }, + { + "epoch": 0.621599278520968, + "grad_norm": 1.605887773689996, + "learning_rate": 1.323284595151953e-06, + "loss": 1.0491, + "num_input_tokens_seen": 132726360, + "step": 8271 + }, + { + "epoch": 0.621674432586803, + "grad_norm": 1.3467222653768094, + "learning_rate": 1.3228264881511137e-06, + "loss": 0.9303, + "num_input_tokens_seen": 132748200, + "step": 8272 + }, + { + "epoch": 0.6217495866526379, + "grad_norm": 2.1251750200787694, + "learning_rate": 1.322368421275398e-06, + "loss": 0.8457, + "num_input_tokens_seen": 132770055, + "step": 8273 + }, + { + "epoch": 0.6218247407184728, + "grad_norm": 1.6178695051478211, + "learning_rate": 1.3219103945519479e-06, + "loss": 0.8962, + "num_input_tokens_seen": 132794090, + "step": 8274 + }, + { + "epoch": 0.6218998947843078, + "grad_norm": 1.4404772734459683, + "learning_rate": 1.3214524080079038e-06, + "loss": 0.9101, + "num_input_tokens_seen": 132817690, + "step": 8275 + }, + { + "epoch": 0.6219750488501428, + "grad_norm": 1.5335470603485604, + "learning_rate": 1.3209944616704023e-06, + "loss": 0.8602, + "num_input_tokens_seen": 132840835, + "step": 8276 + }, + { + "epoch": 0.6220502029159778, + "grad_norm": 1.5796895345038982, + "learning_rate": 1.3205365555665795e-06, + "loss": 0.9747, + "num_input_tokens_seen": 132862840, + "step": 8277 + }, + { + "epoch": 0.6221253569818127, + "grad_norm": 1.2648112188612575, + "learning_rate": 1.3200786897235675e-06, + "loss": 0.9197, + "num_input_tokens_seen": 132886305, + "step": 8278 + }, + { + "epoch": 0.6222005110476477, + "grad_norm": 1.7217633275202642, + "learning_rate": 1.3196208641684968e-06, + "loss": 0.9656, + "num_input_tokens_seen": 132909250, + "step": 8279 + }, + { + "epoch": 0.6222756651134826, + "grad_norm": 1.456321973209046, + "learning_rate": 1.3191630789284954e-06, + "loss": 0.9871, + "num_input_tokens_seen": 132931695, + "step": 8280 + }, + { + "epoch": 0.6223508191793176, + "grad_norm": 1.4087012226152777, + "learning_rate": 1.318705334030688e-06, + "loss": 0.9633, + "num_input_tokens_seen": 132955465, + "step": 8281 + }, + { + "epoch": 0.6224259732451526, + "grad_norm": 1.5689212082528836, + "learning_rate": 1.318247629502199e-06, + "loss": 1.0022, + "num_input_tokens_seen": 132977770, + "step": 8282 + }, + { + "epoch": 0.6225011273109875, + "grad_norm": 1.8185848829687603, + "learning_rate": 1.317789965370148e-06, + "loss": 0.9135, + "num_input_tokens_seen": 132998380, + "step": 8283 + }, + { + "epoch": 0.6225762813768225, + "grad_norm": 1.7137297398589006, + "learning_rate": 1.3173323416616549e-06, + "loss": 0.9351, + "num_input_tokens_seen": 133020265, + "step": 8284 + }, + { + "epoch": 0.6226514354426574, + "grad_norm": 1.4436878042324377, + "learning_rate": 1.3168747584038341e-06, + "loss": 0.8951, + "num_input_tokens_seen": 133047815, + "step": 8285 + }, + { + "epoch": 0.6227265895084924, + "grad_norm": 0.6871783848080341, + "learning_rate": 1.3164172156237992e-06, + "loss": 0.7724, + "num_input_tokens_seen": 133130100, + "step": 8286 + }, + { + "epoch": 0.6228017435743274, + "grad_norm": 1.504810012155593, + "learning_rate": 1.3159597133486625e-06, + "loss": 0.921, + "num_input_tokens_seen": 133152205, + "step": 8287 + }, + { + "epoch": 0.6228768976401623, + "grad_norm": 1.5193824989450766, + "learning_rate": 1.315502251605532e-06, + "loss": 0.956, + "num_input_tokens_seen": 133176150, + "step": 8288 + }, + { + "epoch": 0.6229520517059973, + "grad_norm": 1.3113621820658434, + "learning_rate": 1.3150448304215142e-06, + "loss": 0.9283, + "num_input_tokens_seen": 133199910, + "step": 8289 + }, + { + "epoch": 0.6230272057718322, + "grad_norm": 0.7890418021631763, + "learning_rate": 1.3145874498237133e-06, + "loss": 0.7867, + "num_input_tokens_seen": 133274080, + "step": 8290 + }, + { + "epoch": 0.6231023598376673, + "grad_norm": 1.7075609306310644, + "learning_rate": 1.3141301098392302e-06, + "loss": 0.8926, + "num_input_tokens_seen": 133295265, + "step": 8291 + }, + { + "epoch": 0.6231775139035022, + "grad_norm": 1.5158671549413734, + "learning_rate": 1.3136728104951652e-06, + "loss": 0.9733, + "num_input_tokens_seen": 133321995, + "step": 8292 + }, + { + "epoch": 0.6232526679693371, + "grad_norm": 1.475209499287379, + "learning_rate": 1.3132155518186135e-06, + "loss": 0.9509, + "num_input_tokens_seen": 133345135, + "step": 8293 + }, + { + "epoch": 0.6233278220351721, + "grad_norm": 1.9225842167498426, + "learning_rate": 1.3127583338366707e-06, + "loss": 0.9347, + "num_input_tokens_seen": 133369110, + "step": 8294 + }, + { + "epoch": 0.623402976101007, + "grad_norm": 1.3571196953869311, + "learning_rate": 1.312301156576428e-06, + "loss": 0.9522, + "num_input_tokens_seen": 133396190, + "step": 8295 + }, + { + "epoch": 0.6234781301668421, + "grad_norm": 1.5216725616229474, + "learning_rate": 1.3118440200649752e-06, + "loss": 0.9545, + "num_input_tokens_seen": 133421060, + "step": 8296 + }, + { + "epoch": 0.623553284232677, + "grad_norm": 1.4008457183895544, + "learning_rate": 1.3113869243293993e-06, + "loss": 1.0363, + "num_input_tokens_seen": 133445470, + "step": 8297 + }, + { + "epoch": 0.623628438298512, + "grad_norm": 2.2451542489940692, + "learning_rate": 1.310929869396785e-06, + "loss": 0.791, + "num_input_tokens_seen": 133466960, + "step": 8298 + }, + { + "epoch": 0.6237035923643469, + "grad_norm": 2.239686641232547, + "learning_rate": 1.3104728552942149e-06, + "loss": 0.8076, + "num_input_tokens_seen": 133489280, + "step": 8299 + }, + { + "epoch": 0.6237787464301818, + "grad_norm": 1.4287645873545758, + "learning_rate": 1.3100158820487679e-06, + "loss": 0.9894, + "num_input_tokens_seen": 133511530, + "step": 8300 + }, + { + "epoch": 0.6238539004960169, + "grad_norm": 1.7723281011905367, + "learning_rate": 1.3095589496875224e-06, + "loss": 0.9998, + "num_input_tokens_seen": 133530950, + "step": 8301 + }, + { + "epoch": 0.6239290545618518, + "grad_norm": 3.441945894258285, + "learning_rate": 1.309102058237553e-06, + "loss": 0.8524, + "num_input_tokens_seen": 133552625, + "step": 8302 + }, + { + "epoch": 0.6240042086276868, + "grad_norm": 1.7722968243582595, + "learning_rate": 1.3086452077259323e-06, + "loss": 0.9407, + "num_input_tokens_seen": 133574210, + "step": 8303 + }, + { + "epoch": 0.6240793626935217, + "grad_norm": 1.765596102587776, + "learning_rate": 1.3081883981797303e-06, + "loss": 0.9139, + "num_input_tokens_seen": 133596295, + "step": 8304 + }, + { + "epoch": 0.6241545167593567, + "grad_norm": 1.8417274890827278, + "learning_rate": 1.3077316296260144e-06, + "loss": 0.9399, + "num_input_tokens_seen": 133621145, + "step": 8305 + }, + { + "epoch": 0.6242296708251917, + "grad_norm": 1.8376243471383586, + "learning_rate": 1.3072749020918514e-06, + "loss": 0.8519, + "num_input_tokens_seen": 133647220, + "step": 8306 + }, + { + "epoch": 0.6243048248910266, + "grad_norm": 1.4450894570296884, + "learning_rate": 1.3068182156043026e-06, + "loss": 0.906, + "num_input_tokens_seen": 133673065, + "step": 8307 + }, + { + "epoch": 0.6243799789568616, + "grad_norm": 1.4878299058786606, + "learning_rate": 1.306361570190428e-06, + "loss": 0.9905, + "num_input_tokens_seen": 133696485, + "step": 8308 + }, + { + "epoch": 0.6244551330226965, + "grad_norm": 1.727053197115394, + "learning_rate": 1.3059049658772875e-06, + "loss": 0.9159, + "num_input_tokens_seen": 133720395, + "step": 8309 + }, + { + "epoch": 0.6245302870885315, + "grad_norm": 1.552357600341812, + "learning_rate": 1.305448402691935e-06, + "loss": 0.9442, + "num_input_tokens_seen": 133744800, + "step": 8310 + }, + { + "epoch": 0.6246054411543664, + "grad_norm": 1.5182137158645301, + "learning_rate": 1.304991880661425e-06, + "loss": 0.8923, + "num_input_tokens_seen": 133765870, + "step": 8311 + }, + { + "epoch": 0.6246805952202014, + "grad_norm": 0.7112345651249087, + "learning_rate": 1.3045353998128073e-06, + "loss": 0.7817, + "num_input_tokens_seen": 133845650, + "step": 8312 + }, + { + "epoch": 0.6247557492860364, + "grad_norm": 1.6732971702457375, + "learning_rate": 1.30407896017313e-06, + "loss": 0.9491, + "num_input_tokens_seen": 133869115, + "step": 8313 + }, + { + "epoch": 0.6248309033518713, + "grad_norm": 1.6466047756569233, + "learning_rate": 1.3036225617694391e-06, + "loss": 0.9813, + "num_input_tokens_seen": 133892825, + "step": 8314 + }, + { + "epoch": 0.6249060574177063, + "grad_norm": 1.7141178480338632, + "learning_rate": 1.3031662046287778e-06, + "loss": 0.9354, + "num_input_tokens_seen": 133915780, + "step": 8315 + }, + { + "epoch": 0.6249812114835412, + "grad_norm": 1.7628626862840875, + "learning_rate": 1.302709888778188e-06, + "loss": 0.8667, + "num_input_tokens_seen": 133939285, + "step": 8316 + }, + { + "epoch": 0.6250563655493763, + "grad_norm": 1.430415567423021, + "learning_rate": 1.3022536142447069e-06, + "loss": 0.8948, + "num_input_tokens_seen": 133964405, + "step": 8317 + }, + { + "epoch": 0.6251315196152112, + "grad_norm": 1.59106061325324, + "learning_rate": 1.3017973810553709e-06, + "loss": 0.8509, + "num_input_tokens_seen": 133987300, + "step": 8318 + }, + { + "epoch": 0.6252066736810461, + "grad_norm": 1.577116388993264, + "learning_rate": 1.301341189237214e-06, + "loss": 0.9955, + "num_input_tokens_seen": 134013535, + "step": 8319 + }, + { + "epoch": 0.6252818277468811, + "grad_norm": 1.3671707588260738, + "learning_rate": 1.3008850388172668e-06, + "loss": 0.9398, + "num_input_tokens_seen": 134039450, + "step": 8320 + }, + { + "epoch": 0.625356981812716, + "grad_norm": 1.4447176147200353, + "learning_rate": 1.3004289298225582e-06, + "loss": 0.8656, + "num_input_tokens_seen": 134065140, + "step": 8321 + }, + { + "epoch": 0.6254321358785511, + "grad_norm": 1.427005023153509, + "learning_rate": 1.299972862280114e-06, + "loss": 0.9147, + "num_input_tokens_seen": 134087250, + "step": 8322 + }, + { + "epoch": 0.625507289944386, + "grad_norm": 1.8455501109481893, + "learning_rate": 1.299516836216959e-06, + "loss": 0.9472, + "num_input_tokens_seen": 134109820, + "step": 8323 + }, + { + "epoch": 0.625582444010221, + "grad_norm": 1.4635498028548264, + "learning_rate": 1.2990608516601133e-06, + "loss": 0.9445, + "num_input_tokens_seen": 134133020, + "step": 8324 + }, + { + "epoch": 0.6256575980760559, + "grad_norm": 1.344427869281232, + "learning_rate": 1.2986049086365963e-06, + "loss": 0.8769, + "num_input_tokens_seen": 134160280, + "step": 8325 + }, + { + "epoch": 0.6257327521418908, + "grad_norm": 1.5846111595618835, + "learning_rate": 1.2981490071734244e-06, + "loss": 0.9262, + "num_input_tokens_seen": 134183645, + "step": 8326 + }, + { + "epoch": 0.6258079062077259, + "grad_norm": 1.687257890640734, + "learning_rate": 1.2976931472976106e-06, + "loss": 0.8354, + "num_input_tokens_seen": 134208165, + "step": 8327 + }, + { + "epoch": 0.6258830602735608, + "grad_norm": 1.6208529006182224, + "learning_rate": 1.2972373290361683e-06, + "loss": 0.8522, + "num_input_tokens_seen": 134231705, + "step": 8328 + }, + { + "epoch": 0.6259582143393958, + "grad_norm": 1.2397889786793006, + "learning_rate": 1.296781552416105e-06, + "loss": 0.9547, + "num_input_tokens_seen": 134256760, + "step": 8329 + }, + { + "epoch": 0.6260333684052307, + "grad_norm": 1.426010322458537, + "learning_rate": 1.2963258174644266e-06, + "loss": 0.9355, + "num_input_tokens_seen": 134283250, + "step": 8330 + }, + { + "epoch": 0.6261085224710656, + "grad_norm": 1.3528677750942357, + "learning_rate": 1.295870124208139e-06, + "loss": 0.9561, + "num_input_tokens_seen": 134307070, + "step": 8331 + }, + { + "epoch": 0.6261836765369007, + "grad_norm": 1.9377205015655639, + "learning_rate": 1.2954144726742424e-06, + "loss": 0.8246, + "num_input_tokens_seen": 134329635, + "step": 8332 + }, + { + "epoch": 0.6262588306027356, + "grad_norm": 1.8287793805988306, + "learning_rate": 1.2949588628897367e-06, + "loss": 0.9543, + "num_input_tokens_seen": 134349875, + "step": 8333 + }, + { + "epoch": 0.6263339846685706, + "grad_norm": 1.850231158790069, + "learning_rate": 1.2945032948816183e-06, + "loss": 0.8357, + "num_input_tokens_seen": 134372790, + "step": 8334 + }, + { + "epoch": 0.6264091387344055, + "grad_norm": 1.5726498341701904, + "learning_rate": 1.2940477686768806e-06, + "loss": 1.0023, + "num_input_tokens_seen": 134396900, + "step": 8335 + }, + { + "epoch": 0.6264842928002405, + "grad_norm": 1.8269220290393373, + "learning_rate": 1.2935922843025165e-06, + "loss": 0.9198, + "num_input_tokens_seen": 134417460, + "step": 8336 + }, + { + "epoch": 0.6265594468660755, + "grad_norm": 2.622069039513192, + "learning_rate": 1.293136841785514e-06, + "loss": 0.9675, + "num_input_tokens_seen": 134440290, + "step": 8337 + }, + { + "epoch": 0.6266346009319104, + "grad_norm": 1.4974510470176017, + "learning_rate": 1.292681441152861e-06, + "loss": 0.8766, + "num_input_tokens_seen": 134463340, + "step": 8338 + }, + { + "epoch": 0.6267097549977454, + "grad_norm": 1.9000854247113632, + "learning_rate": 1.2922260824315409e-06, + "loss": 0.9424, + "num_input_tokens_seen": 134487605, + "step": 8339 + }, + { + "epoch": 0.6267849090635803, + "grad_norm": 1.6509632448972287, + "learning_rate": 1.2917707656485352e-06, + "loss": 0.987, + "num_input_tokens_seen": 134509495, + "step": 8340 + }, + { + "epoch": 0.6268600631294153, + "grad_norm": 1.5550604138183828, + "learning_rate": 1.2913154908308244e-06, + "loss": 0.9272, + "num_input_tokens_seen": 134532370, + "step": 8341 + }, + { + "epoch": 0.6269352171952502, + "grad_norm": 1.9484560282666747, + "learning_rate": 1.2908602580053836e-06, + "loss": 0.8682, + "num_input_tokens_seen": 134554685, + "step": 8342 + }, + { + "epoch": 0.6270103712610853, + "grad_norm": 2.04787589492131, + "learning_rate": 1.2904050671991887e-06, + "loss": 0.8401, + "num_input_tokens_seen": 134578945, + "step": 8343 + }, + { + "epoch": 0.6270855253269202, + "grad_norm": 1.5472816934499996, + "learning_rate": 1.2899499184392105e-06, + "loss": 0.8437, + "num_input_tokens_seen": 134602010, + "step": 8344 + }, + { + "epoch": 0.6271606793927551, + "grad_norm": 1.8143935342617985, + "learning_rate": 1.2894948117524188e-06, + "loss": 0.7878, + "num_input_tokens_seen": 134623700, + "step": 8345 + }, + { + "epoch": 0.6272358334585901, + "grad_norm": 1.3935093488047716, + "learning_rate": 1.2890397471657802e-06, + "loss": 1.0013, + "num_input_tokens_seen": 134649240, + "step": 8346 + }, + { + "epoch": 0.627310987524425, + "grad_norm": 1.5323984160367048, + "learning_rate": 1.2885847247062587e-06, + "loss": 1.022, + "num_input_tokens_seen": 134674260, + "step": 8347 + }, + { + "epoch": 0.6273861415902601, + "grad_norm": 1.2043591894310655, + "learning_rate": 1.2881297444008165e-06, + "loss": 0.9726, + "num_input_tokens_seen": 134701580, + "step": 8348 + }, + { + "epoch": 0.627461295656095, + "grad_norm": 2.2242585693482426, + "learning_rate": 1.2876748062764127e-06, + "loss": 0.9185, + "num_input_tokens_seen": 134725105, + "step": 8349 + }, + { + "epoch": 0.6275364497219299, + "grad_norm": 1.5434276043392594, + "learning_rate": 1.2872199103600046e-06, + "loss": 0.9352, + "num_input_tokens_seen": 134749100, + "step": 8350 + }, + { + "epoch": 0.6276116037877649, + "grad_norm": 1.4920957047752876, + "learning_rate": 1.286765056678547e-06, + "loss": 0.9999, + "num_input_tokens_seen": 134769990, + "step": 8351 + }, + { + "epoch": 0.6276867578535998, + "grad_norm": 1.3381571037934907, + "learning_rate": 1.2863102452589893e-06, + "loss": 0.9547, + "num_input_tokens_seen": 134794280, + "step": 8352 + }, + { + "epoch": 0.6277619119194349, + "grad_norm": 1.6209276229119152, + "learning_rate": 1.2858554761282837e-06, + "loss": 0.9643, + "num_input_tokens_seen": 134816470, + "step": 8353 + }, + { + "epoch": 0.6278370659852698, + "grad_norm": 1.3678588091005266, + "learning_rate": 1.2854007493133754e-06, + "loss": 0.997, + "num_input_tokens_seen": 134841965, + "step": 8354 + }, + { + "epoch": 0.6279122200511048, + "grad_norm": 1.2544660893065107, + "learning_rate": 1.2849460648412092e-06, + "loss": 0.9141, + "num_input_tokens_seen": 134866520, + "step": 8355 + }, + { + "epoch": 0.6279873741169397, + "grad_norm": 1.8612893131528694, + "learning_rate": 1.2844914227387266e-06, + "loss": 0.9305, + "num_input_tokens_seen": 134889420, + "step": 8356 + }, + { + "epoch": 0.6280625281827746, + "grad_norm": 2.0557974366619725, + "learning_rate": 1.2840368230328672e-06, + "loss": 0.8344, + "num_input_tokens_seen": 134911485, + "step": 8357 + }, + { + "epoch": 0.6281376822486097, + "grad_norm": 1.520985434603501, + "learning_rate": 1.2835822657505678e-06, + "loss": 1.0348, + "num_input_tokens_seen": 134936225, + "step": 8358 + }, + { + "epoch": 0.6282128363144446, + "grad_norm": 3.2029627316939107, + "learning_rate": 1.2831277509187622e-06, + "loss": 0.9379, + "num_input_tokens_seen": 134957690, + "step": 8359 + }, + { + "epoch": 0.6282879903802796, + "grad_norm": 0.8774952731317652, + "learning_rate": 1.2826732785643826e-06, + "loss": 0.8001, + "num_input_tokens_seen": 135033490, + "step": 8360 + }, + { + "epoch": 0.6283631444461145, + "grad_norm": 1.3860176570725418, + "learning_rate": 1.2822188487143581e-06, + "loss": 0.8824, + "num_input_tokens_seen": 135062065, + "step": 8361 + }, + { + "epoch": 0.6284382985119495, + "grad_norm": 1.7430278992031403, + "learning_rate": 1.2817644613956153e-06, + "loss": 0.9733, + "num_input_tokens_seen": 135083880, + "step": 8362 + }, + { + "epoch": 0.6285134525777845, + "grad_norm": 1.4577059985935337, + "learning_rate": 1.2813101166350786e-06, + "loss": 0.9846, + "num_input_tokens_seen": 135107520, + "step": 8363 + }, + { + "epoch": 0.6285886066436194, + "grad_norm": 1.3304898021239804, + "learning_rate": 1.2808558144596692e-06, + "loss": 0.9052, + "num_input_tokens_seen": 135134860, + "step": 8364 + }, + { + "epoch": 0.6286637607094544, + "grad_norm": 1.572052252845003, + "learning_rate": 1.280401554896307e-06, + "loss": 0.9975, + "num_input_tokens_seen": 135157685, + "step": 8365 + }, + { + "epoch": 0.6287389147752893, + "grad_norm": 1.9080996060673023, + "learning_rate": 1.2799473379719077e-06, + "loss": 1.0242, + "num_input_tokens_seen": 135179270, + "step": 8366 + }, + { + "epoch": 0.6288140688411243, + "grad_norm": 1.535075721861151, + "learning_rate": 1.2794931637133863e-06, + "loss": 0.9359, + "num_input_tokens_seen": 135201025, + "step": 8367 + }, + { + "epoch": 0.6288892229069593, + "grad_norm": 1.5538014709834937, + "learning_rate": 1.2790390321476542e-06, + "loss": 0.9777, + "num_input_tokens_seen": 135226295, + "step": 8368 + }, + { + "epoch": 0.6289643769727943, + "grad_norm": 1.8017148723499365, + "learning_rate": 1.2785849433016198e-06, + "loss": 0.9569, + "num_input_tokens_seen": 135249435, + "step": 8369 + }, + { + "epoch": 0.6290395310386292, + "grad_norm": 1.5561239995849787, + "learning_rate": 1.27813089720219e-06, + "loss": 1.0104, + "num_input_tokens_seen": 135269830, + "step": 8370 + }, + { + "epoch": 0.6291146851044641, + "grad_norm": 1.4449203371574508, + "learning_rate": 1.277676893876268e-06, + "loss": 1.0026, + "num_input_tokens_seen": 135292965, + "step": 8371 + }, + { + "epoch": 0.6291898391702991, + "grad_norm": 1.8257225095728704, + "learning_rate": 1.277222933350757e-06, + "loss": 0.8955, + "num_input_tokens_seen": 135315505, + "step": 8372 + }, + { + "epoch": 0.629264993236134, + "grad_norm": 1.5211101262680988, + "learning_rate": 1.2767690156525554e-06, + "loss": 0.8549, + "num_input_tokens_seen": 135336720, + "step": 8373 + }, + { + "epoch": 0.6293401473019691, + "grad_norm": 1.2485090901231426, + "learning_rate": 1.276315140808558e-06, + "loss": 0.9783, + "num_input_tokens_seen": 135362090, + "step": 8374 + }, + { + "epoch": 0.629415301367804, + "grad_norm": 1.4253980147676437, + "learning_rate": 1.27586130884566e-06, + "loss": 0.9234, + "num_input_tokens_seen": 135386115, + "step": 8375 + }, + { + "epoch": 0.6294904554336389, + "grad_norm": 1.9474128506789155, + "learning_rate": 1.275407519790752e-06, + "loss": 1.038, + "num_input_tokens_seen": 135403770, + "step": 8376 + }, + { + "epoch": 0.6295656094994739, + "grad_norm": 1.4535074038479376, + "learning_rate": 1.2749537736707239e-06, + "loss": 0.9062, + "num_input_tokens_seen": 135426295, + "step": 8377 + }, + { + "epoch": 0.6296407635653088, + "grad_norm": 1.7401760104375208, + "learning_rate": 1.274500070512461e-06, + "loss": 1.0598, + "num_input_tokens_seen": 135450515, + "step": 8378 + }, + { + "epoch": 0.6297159176311439, + "grad_norm": 1.3707344475572993, + "learning_rate": 1.2740464103428463e-06, + "loss": 0.934, + "num_input_tokens_seen": 135472500, + "step": 8379 + }, + { + "epoch": 0.6297910716969788, + "grad_norm": 1.5411950068628815, + "learning_rate": 1.2735927931887625e-06, + "loss": 0.9458, + "num_input_tokens_seen": 135498050, + "step": 8380 + }, + { + "epoch": 0.6298662257628138, + "grad_norm": 1.3707793655130358, + "learning_rate": 1.2731392190770866e-06, + "loss": 0.8833, + "num_input_tokens_seen": 135522845, + "step": 8381 + }, + { + "epoch": 0.6299413798286487, + "grad_norm": 1.4311620893507742, + "learning_rate": 1.2726856880346956e-06, + "loss": 0.9687, + "num_input_tokens_seen": 135547205, + "step": 8382 + }, + { + "epoch": 0.6300165338944836, + "grad_norm": 1.7304670966766933, + "learning_rate": 1.2722322000884628e-06, + "loss": 1.0336, + "num_input_tokens_seen": 135571025, + "step": 8383 + }, + { + "epoch": 0.6300916879603187, + "grad_norm": 1.4305772357931537, + "learning_rate": 1.2717787552652585e-06, + "loss": 0.9204, + "num_input_tokens_seen": 135595830, + "step": 8384 + }, + { + "epoch": 0.6301668420261536, + "grad_norm": 1.7799746314394806, + "learning_rate": 1.2713253535919521e-06, + "loss": 0.8972, + "num_input_tokens_seen": 135617695, + "step": 8385 + }, + { + "epoch": 0.6302419960919886, + "grad_norm": 1.642177464883706, + "learning_rate": 1.2708719950954082e-06, + "loss": 0.9866, + "num_input_tokens_seen": 135643295, + "step": 8386 + }, + { + "epoch": 0.6303171501578235, + "grad_norm": 1.4020407516716997, + "learning_rate": 1.2704186798024913e-06, + "loss": 0.971, + "num_input_tokens_seen": 135667275, + "step": 8387 + }, + { + "epoch": 0.6303923042236586, + "grad_norm": 0.7435545127473651, + "learning_rate": 1.2699654077400608e-06, + "loss": 0.8021, + "num_input_tokens_seen": 135736065, + "step": 8388 + }, + { + "epoch": 0.6304674582894935, + "grad_norm": 1.8481490463947488, + "learning_rate": 1.2695121789349757e-06, + "loss": 0.9437, + "num_input_tokens_seen": 135758160, + "step": 8389 + }, + { + "epoch": 0.6305426123553284, + "grad_norm": 1.7577939859051046, + "learning_rate": 1.2690589934140912e-06, + "loss": 0.8001, + "num_input_tokens_seen": 135781870, + "step": 8390 + }, + { + "epoch": 0.6306177664211634, + "grad_norm": 1.5909970760786138, + "learning_rate": 1.2686058512042594e-06, + "loss": 0.9066, + "num_input_tokens_seen": 135807000, + "step": 8391 + }, + { + "epoch": 0.6306929204869983, + "grad_norm": 2.746900892949368, + "learning_rate": 1.268152752332333e-06, + "loss": 0.9594, + "num_input_tokens_seen": 135830840, + "step": 8392 + }, + { + "epoch": 0.6307680745528333, + "grad_norm": 1.8419375805794105, + "learning_rate": 1.2676996968251574e-06, + "loss": 0.962, + "num_input_tokens_seen": 135852335, + "step": 8393 + }, + { + "epoch": 0.6308432286186683, + "grad_norm": 1.7646170245500605, + "learning_rate": 1.2672466847095793e-06, + "loss": 1.0242, + "num_input_tokens_seen": 135872790, + "step": 8394 + }, + { + "epoch": 0.6309183826845032, + "grad_norm": 1.7412120329146863, + "learning_rate": 1.2667937160124416e-06, + "loss": 0.9533, + "num_input_tokens_seen": 135892300, + "step": 8395 + }, + { + "epoch": 0.6309935367503382, + "grad_norm": 1.7118585624337397, + "learning_rate": 1.266340790760583e-06, + "loss": 0.907, + "num_input_tokens_seen": 135912105, + "step": 8396 + }, + { + "epoch": 0.6310686908161731, + "grad_norm": 1.4932275313266528, + "learning_rate": 1.2658879089808423e-06, + "loss": 1.0398, + "num_input_tokens_seen": 135935710, + "step": 8397 + }, + { + "epoch": 0.6311438448820081, + "grad_norm": 1.5874841974628326, + "learning_rate": 1.2654350707000538e-06, + "loss": 0.9859, + "num_input_tokens_seen": 135960430, + "step": 8398 + }, + { + "epoch": 0.6312189989478431, + "grad_norm": 1.4258377691066524, + "learning_rate": 1.264982275945051e-06, + "loss": 0.9435, + "num_input_tokens_seen": 135982925, + "step": 8399 + }, + { + "epoch": 0.6312941530136781, + "grad_norm": 1.5489199443986392, + "learning_rate": 1.2645295247426625e-06, + "loss": 0.9918, + "num_input_tokens_seen": 136007385, + "step": 8400 + }, + { + "epoch": 0.631369307079513, + "grad_norm": 2.129674706312166, + "learning_rate": 1.2640768171197156e-06, + "loss": 0.9454, + "num_input_tokens_seen": 136031135, + "step": 8401 + }, + { + "epoch": 0.6314444611453479, + "grad_norm": 1.3301835655509462, + "learning_rate": 1.2636241531030355e-06, + "loss": 0.9803, + "num_input_tokens_seen": 136055460, + "step": 8402 + }, + { + "epoch": 0.6315196152111829, + "grad_norm": 1.7689974150134704, + "learning_rate": 1.263171532719444e-06, + "loss": 0.9274, + "num_input_tokens_seen": 136082840, + "step": 8403 + }, + { + "epoch": 0.6315947692770179, + "grad_norm": 0.6870363579448044, + "learning_rate": 1.2627189559957612e-06, + "loss": 0.8297, + "num_input_tokens_seen": 136168455, + "step": 8404 + }, + { + "epoch": 0.6316699233428529, + "grad_norm": 1.4155969040126744, + "learning_rate": 1.2622664229588033e-06, + "loss": 0.833, + "num_input_tokens_seen": 136192935, + "step": 8405 + }, + { + "epoch": 0.6317450774086878, + "grad_norm": 1.4639986983069382, + "learning_rate": 1.2618139336353846e-06, + "loss": 0.9764, + "num_input_tokens_seen": 136217805, + "step": 8406 + }, + { + "epoch": 0.6318202314745228, + "grad_norm": 1.5728728265105365, + "learning_rate": 1.2613614880523172e-06, + "loss": 0.874, + "num_input_tokens_seen": 136241415, + "step": 8407 + }, + { + "epoch": 0.6318953855403577, + "grad_norm": 1.720957985058694, + "learning_rate": 1.2609090862364099e-06, + "loss": 0.8852, + "num_input_tokens_seen": 136262855, + "step": 8408 + }, + { + "epoch": 0.6319705396061926, + "grad_norm": 1.4524466459156313, + "learning_rate": 1.2604567282144696e-06, + "loss": 0.9355, + "num_input_tokens_seen": 136285925, + "step": 8409 + }, + { + "epoch": 0.6320456936720277, + "grad_norm": 1.2492813905776503, + "learning_rate": 1.2600044140132994e-06, + "loss": 0.9433, + "num_input_tokens_seen": 136312675, + "step": 8410 + }, + { + "epoch": 0.6321208477378626, + "grad_norm": 1.3257997690232541, + "learning_rate": 1.259552143659702e-06, + "loss": 1.0268, + "num_input_tokens_seen": 136336065, + "step": 8411 + }, + { + "epoch": 0.6321960018036976, + "grad_norm": 0.7341451894048491, + "learning_rate": 1.2590999171804758e-06, + "loss": 0.7941, + "num_input_tokens_seen": 136423300, + "step": 8412 + }, + { + "epoch": 0.6322711558695325, + "grad_norm": 1.4031233466803448, + "learning_rate": 1.2586477346024154e-06, + "loss": 0.798, + "num_input_tokens_seen": 136448785, + "step": 8413 + }, + { + "epoch": 0.6323463099353676, + "grad_norm": 1.3066814057011942, + "learning_rate": 1.258195595952317e-06, + "loss": 0.9144, + "num_input_tokens_seen": 136475705, + "step": 8414 + }, + { + "epoch": 0.6324214640012025, + "grad_norm": 1.661166756455256, + "learning_rate": 1.2577435012569684e-06, + "loss": 1.0057, + "num_input_tokens_seen": 136496015, + "step": 8415 + }, + { + "epoch": 0.6324966180670374, + "grad_norm": 1.586700406290138, + "learning_rate": 1.2572914505431609e-06, + "loss": 1.0337, + "num_input_tokens_seen": 136517640, + "step": 8416 + }, + { + "epoch": 0.6325717721328724, + "grad_norm": 1.5018662922416657, + "learning_rate": 1.2568394438376788e-06, + "loss": 0.8964, + "num_input_tokens_seen": 136540385, + "step": 8417 + }, + { + "epoch": 0.6326469261987073, + "grad_norm": 1.797625907234041, + "learning_rate": 1.2563874811673053e-06, + "loss": 0.951, + "num_input_tokens_seen": 136563255, + "step": 8418 + }, + { + "epoch": 0.6327220802645424, + "grad_norm": 1.665651449311978, + "learning_rate": 1.2559355625588208e-06, + "loss": 0.9274, + "num_input_tokens_seen": 136589325, + "step": 8419 + }, + { + "epoch": 0.6327972343303773, + "grad_norm": 1.595038472850626, + "learning_rate": 1.2554836880390033e-06, + "loss": 0.9353, + "num_input_tokens_seen": 136611605, + "step": 8420 + }, + { + "epoch": 0.6328723883962122, + "grad_norm": 1.4845245286014264, + "learning_rate": 1.2550318576346287e-06, + "loss": 0.916, + "num_input_tokens_seen": 136638545, + "step": 8421 + }, + { + "epoch": 0.6329475424620472, + "grad_norm": 1.4124057400740468, + "learning_rate": 1.2545800713724694e-06, + "loss": 1.0078, + "num_input_tokens_seen": 136663720, + "step": 8422 + }, + { + "epoch": 0.6330226965278821, + "grad_norm": 1.4241067328582733, + "learning_rate": 1.2541283292792949e-06, + "loss": 0.8699, + "num_input_tokens_seen": 136691800, + "step": 8423 + }, + { + "epoch": 0.6330978505937171, + "grad_norm": 1.4358304321088537, + "learning_rate": 1.2536766313818732e-06, + "loss": 0.9649, + "num_input_tokens_seen": 136719680, + "step": 8424 + }, + { + "epoch": 0.6331730046595521, + "grad_norm": 1.639681517566653, + "learning_rate": 1.2532249777069686e-06, + "loss": 0.9634, + "num_input_tokens_seen": 136741840, + "step": 8425 + }, + { + "epoch": 0.6332481587253871, + "grad_norm": 1.7746029020103649, + "learning_rate": 1.252773368281344e-06, + "loss": 1.0791, + "num_input_tokens_seen": 136762505, + "step": 8426 + }, + { + "epoch": 0.633323312791222, + "grad_norm": 1.5723847338181611, + "learning_rate": 1.2523218031317586e-06, + "loss": 1.0724, + "num_input_tokens_seen": 136785790, + "step": 8427 + }, + { + "epoch": 0.6333984668570569, + "grad_norm": 1.313221324526975, + "learning_rate": 1.2518702822849694e-06, + "loss": 0.9132, + "num_input_tokens_seen": 136809740, + "step": 8428 + }, + { + "epoch": 0.633473620922892, + "grad_norm": 2.159167430744554, + "learning_rate": 1.2514188057677309e-06, + "loss": 0.8652, + "num_input_tokens_seen": 136835365, + "step": 8429 + }, + { + "epoch": 0.6335487749887269, + "grad_norm": 1.399896595746505, + "learning_rate": 1.250967373606794e-06, + "loss": 0.8964, + "num_input_tokens_seen": 136858455, + "step": 8430 + }, + { + "epoch": 0.6336239290545619, + "grad_norm": 1.4083140166874675, + "learning_rate": 1.2505159858289092e-06, + "loss": 0.9718, + "num_input_tokens_seen": 136881270, + "step": 8431 + }, + { + "epoch": 0.6336990831203968, + "grad_norm": 1.6826595537959284, + "learning_rate": 1.2500646424608217e-06, + "loss": 0.9046, + "num_input_tokens_seen": 136905810, + "step": 8432 + }, + { + "epoch": 0.6337742371862318, + "grad_norm": 1.4468846199259622, + "learning_rate": 1.2496133435292762e-06, + "loss": 0.8959, + "num_input_tokens_seen": 136931265, + "step": 8433 + }, + { + "epoch": 0.6338493912520667, + "grad_norm": 0.7790274476980691, + "learning_rate": 1.2491620890610135e-06, + "loss": 0.7763, + "num_input_tokens_seen": 137005765, + "step": 8434 + }, + { + "epoch": 0.6339245453179017, + "grad_norm": 1.5165745228594045, + "learning_rate": 1.2487108790827714e-06, + "loss": 0.9132, + "num_input_tokens_seen": 137028435, + "step": 8435 + }, + { + "epoch": 0.6339996993837367, + "grad_norm": 2.2912540295487935, + "learning_rate": 1.2482597136212877e-06, + "loss": 0.9509, + "num_input_tokens_seen": 137050040, + "step": 8436 + }, + { + "epoch": 0.6340748534495716, + "grad_norm": 1.554108785142903, + "learning_rate": 1.2478085927032935e-06, + "loss": 0.9796, + "num_input_tokens_seen": 137073395, + "step": 8437 + }, + { + "epoch": 0.6341500075154066, + "grad_norm": 1.6472439173570055, + "learning_rate": 1.2473575163555215e-06, + "loss": 0.9641, + "num_input_tokens_seen": 137095565, + "step": 8438 + }, + { + "epoch": 0.6342251615812415, + "grad_norm": 0.8589878597423026, + "learning_rate": 1.2469064846046986e-06, + "loss": 0.8821, + "num_input_tokens_seen": 137171420, + "step": 8439 + }, + { + "epoch": 0.6343003156470765, + "grad_norm": 1.6386894798015146, + "learning_rate": 1.2464554974775496e-06, + "loss": 0.9439, + "num_input_tokens_seen": 137193755, + "step": 8440 + }, + { + "epoch": 0.6343754697129115, + "grad_norm": 1.8624042870423045, + "learning_rate": 1.2460045550007985e-06, + "loss": 0.9406, + "num_input_tokens_seen": 137216570, + "step": 8441 + }, + { + "epoch": 0.6344506237787464, + "grad_norm": 1.5385384998878666, + "learning_rate": 1.2455536572011643e-06, + "loss": 0.9566, + "num_input_tokens_seen": 137237670, + "step": 8442 + }, + { + "epoch": 0.6345257778445814, + "grad_norm": 2.270229002538842, + "learning_rate": 1.2451028041053656e-06, + "loss": 0.9643, + "num_input_tokens_seen": 137259305, + "step": 8443 + }, + { + "epoch": 0.6346009319104163, + "grad_norm": 1.5432353501631053, + "learning_rate": 1.2446519957401157e-06, + "loss": 0.8843, + "num_input_tokens_seen": 137285200, + "step": 8444 + }, + { + "epoch": 0.6346760859762514, + "grad_norm": 1.8152856138618514, + "learning_rate": 1.2442012321321277e-06, + "loss": 0.9116, + "num_input_tokens_seen": 137309465, + "step": 8445 + }, + { + "epoch": 0.6347512400420863, + "grad_norm": 1.5304120164747428, + "learning_rate": 1.2437505133081112e-06, + "loss": 0.9208, + "num_input_tokens_seen": 137332650, + "step": 8446 + }, + { + "epoch": 0.6348263941079212, + "grad_norm": 0.8486845046828471, + "learning_rate": 1.2432998392947723e-06, + "loss": 0.7968, + "num_input_tokens_seen": 137406550, + "step": 8447 + }, + { + "epoch": 0.6349015481737562, + "grad_norm": 4.613716589456082, + "learning_rate": 1.2428492101188156e-06, + "loss": 1.0601, + "num_input_tokens_seen": 137425355, + "step": 8448 + }, + { + "epoch": 0.6349767022395911, + "grad_norm": 1.4499476982418085, + "learning_rate": 1.2423986258069428e-06, + "loss": 0.9063, + "num_input_tokens_seen": 137449250, + "step": 8449 + }, + { + "epoch": 0.6350518563054262, + "grad_norm": 1.6213956620685939, + "learning_rate": 1.241948086385852e-06, + "loss": 0.9722, + "num_input_tokens_seen": 137470670, + "step": 8450 + }, + { + "epoch": 0.6351270103712611, + "grad_norm": 0.7506888722759811, + "learning_rate": 1.24149759188224e-06, + "loss": 0.8111, + "num_input_tokens_seen": 137552480, + "step": 8451 + }, + { + "epoch": 0.6352021644370961, + "grad_norm": 0.7777737142441578, + "learning_rate": 1.2410471423228002e-06, + "loss": 0.8112, + "num_input_tokens_seen": 137636525, + "step": 8452 + }, + { + "epoch": 0.635277318502931, + "grad_norm": 1.4653405732996192, + "learning_rate": 1.2405967377342236e-06, + "loss": 0.8128, + "num_input_tokens_seen": 137661520, + "step": 8453 + }, + { + "epoch": 0.6353524725687659, + "grad_norm": 1.599448693185045, + "learning_rate": 1.2401463781431974e-06, + "loss": 0.9574, + "num_input_tokens_seen": 137682860, + "step": 8454 + }, + { + "epoch": 0.635427626634601, + "grad_norm": 1.5951277817390317, + "learning_rate": 1.2396960635764093e-06, + "loss": 1.0009, + "num_input_tokens_seen": 137707025, + "step": 8455 + }, + { + "epoch": 0.6355027807004359, + "grad_norm": 1.3797754244583018, + "learning_rate": 1.2392457940605402e-06, + "loss": 0.8664, + "num_input_tokens_seen": 137734775, + "step": 8456 + }, + { + "epoch": 0.6355779347662709, + "grad_norm": 1.3451714648887207, + "learning_rate": 1.2387955696222702e-06, + "loss": 0.9697, + "num_input_tokens_seen": 137757075, + "step": 8457 + }, + { + "epoch": 0.6356530888321058, + "grad_norm": 1.516792162569336, + "learning_rate": 1.2383453902882787e-06, + "loss": 1.0546, + "num_input_tokens_seen": 137780840, + "step": 8458 + }, + { + "epoch": 0.6357282428979408, + "grad_norm": 1.562456817030227, + "learning_rate": 1.2378952560852386e-06, + "loss": 0.998, + "num_input_tokens_seen": 137803100, + "step": 8459 + }, + { + "epoch": 0.6358033969637757, + "grad_norm": 1.4384145729811864, + "learning_rate": 1.2374451670398233e-06, + "loss": 0.9032, + "num_input_tokens_seen": 137824400, + "step": 8460 + }, + { + "epoch": 0.6358785510296107, + "grad_norm": 1.6205627307043529, + "learning_rate": 1.236995123178702e-06, + "loss": 0.9685, + "num_input_tokens_seen": 137844705, + "step": 8461 + }, + { + "epoch": 0.6359537050954457, + "grad_norm": 1.5219857365545153, + "learning_rate": 1.2365451245285413e-06, + "loss": 0.9166, + "num_input_tokens_seen": 137869240, + "step": 8462 + }, + { + "epoch": 0.6360288591612806, + "grad_norm": 1.3556748818857627, + "learning_rate": 1.2360951711160055e-06, + "loss": 0.945, + "num_input_tokens_seen": 137894215, + "step": 8463 + }, + { + "epoch": 0.6361040132271156, + "grad_norm": 1.6872781678376632, + "learning_rate": 1.2356452629677554e-06, + "loss": 0.966, + "num_input_tokens_seen": 137917300, + "step": 8464 + }, + { + "epoch": 0.6361791672929505, + "grad_norm": 1.359766805024198, + "learning_rate": 1.235195400110451e-06, + "loss": 1.0, + "num_input_tokens_seen": 137942215, + "step": 8465 + }, + { + "epoch": 0.6362543213587855, + "grad_norm": 1.5565508693344023, + "learning_rate": 1.2347455825707477e-06, + "loss": 0.7837, + "num_input_tokens_seen": 137966060, + "step": 8466 + }, + { + "epoch": 0.6363294754246205, + "grad_norm": 1.5263144690841945, + "learning_rate": 1.2342958103752987e-06, + "loss": 0.8757, + "num_input_tokens_seen": 137990600, + "step": 8467 + }, + { + "epoch": 0.6364046294904554, + "grad_norm": 1.220133265424022, + "learning_rate": 1.2338460835507554e-06, + "loss": 0.9069, + "num_input_tokens_seen": 138017705, + "step": 8468 + }, + { + "epoch": 0.6364797835562904, + "grad_norm": 1.5360402287291475, + "learning_rate": 1.233396402123765e-06, + "loss": 0.9894, + "num_input_tokens_seen": 138040565, + "step": 8469 + }, + { + "epoch": 0.6365549376221253, + "grad_norm": 1.4578701191888899, + "learning_rate": 1.2329467661209738e-06, + "loss": 0.9804, + "num_input_tokens_seen": 138062625, + "step": 8470 + }, + { + "epoch": 0.6366300916879604, + "grad_norm": 1.410595981162515, + "learning_rate": 1.232497175569024e-06, + "loss": 0.885, + "num_input_tokens_seen": 138088145, + "step": 8471 + }, + { + "epoch": 0.6367052457537953, + "grad_norm": 1.607839351685627, + "learning_rate": 1.2320476304945548e-06, + "loss": 0.9216, + "num_input_tokens_seen": 138111665, + "step": 8472 + }, + { + "epoch": 0.6367803998196302, + "grad_norm": 1.4221411655013885, + "learning_rate": 1.2315981309242046e-06, + "loss": 0.8864, + "num_input_tokens_seen": 138140310, + "step": 8473 + }, + { + "epoch": 0.6368555538854652, + "grad_norm": 1.3923191509229245, + "learning_rate": 1.2311486768846075e-06, + "loss": 0.9208, + "num_input_tokens_seen": 138166145, + "step": 8474 + }, + { + "epoch": 0.6369307079513001, + "grad_norm": 1.5987813480933628, + "learning_rate": 1.2306992684023955e-06, + "loss": 0.8882, + "num_input_tokens_seen": 138192165, + "step": 8475 + }, + { + "epoch": 0.6370058620171352, + "grad_norm": 0.7728438362774618, + "learning_rate": 1.230249905504197e-06, + "loss": 0.7642, + "num_input_tokens_seen": 138272400, + "step": 8476 + }, + { + "epoch": 0.6370810160829701, + "grad_norm": 1.7140989315365454, + "learning_rate": 1.2298005882166406e-06, + "loss": 0.9528, + "num_input_tokens_seen": 138296975, + "step": 8477 + }, + { + "epoch": 0.6371561701488051, + "grad_norm": 1.3712466336733202, + "learning_rate": 1.229351316566348e-06, + "loss": 0.9613, + "num_input_tokens_seen": 138321150, + "step": 8478 + }, + { + "epoch": 0.63723132421464, + "grad_norm": 1.445346976204093, + "learning_rate": 1.2289020905799401e-06, + "loss": 0.9093, + "num_input_tokens_seen": 138345185, + "step": 8479 + }, + { + "epoch": 0.6373064782804749, + "grad_norm": 1.7143181275527757, + "learning_rate": 1.2284529102840369e-06, + "loss": 1.0013, + "num_input_tokens_seen": 138367020, + "step": 8480 + }, + { + "epoch": 0.63738163234631, + "grad_norm": 0.8180602262076127, + "learning_rate": 1.2280037757052527e-06, + "loss": 0.8372, + "num_input_tokens_seen": 138435990, + "step": 8481 + }, + { + "epoch": 0.6374567864121449, + "grad_norm": 1.645832697550333, + "learning_rate": 1.2275546868702017e-06, + "loss": 0.9336, + "num_input_tokens_seen": 138457480, + "step": 8482 + }, + { + "epoch": 0.6375319404779799, + "grad_norm": 1.8375239883206447, + "learning_rate": 1.2271056438054933e-06, + "loss": 0.8745, + "num_input_tokens_seen": 138480405, + "step": 8483 + }, + { + "epoch": 0.6376070945438148, + "grad_norm": 1.303473404522667, + "learning_rate": 1.2266566465377343e-06, + "loss": 0.9077, + "num_input_tokens_seen": 138504370, + "step": 8484 + }, + { + "epoch": 0.6376822486096497, + "grad_norm": 1.5242047847418922, + "learning_rate": 1.2262076950935311e-06, + "loss": 0.8902, + "num_input_tokens_seen": 138526940, + "step": 8485 + }, + { + "epoch": 0.6377574026754848, + "grad_norm": 1.6451533719716136, + "learning_rate": 1.2257587894994842e-06, + "loss": 0.9057, + "num_input_tokens_seen": 138547795, + "step": 8486 + }, + { + "epoch": 0.6378325567413197, + "grad_norm": 1.1919138343057454, + "learning_rate": 1.2253099297821948e-06, + "loss": 0.9964, + "num_input_tokens_seen": 138572455, + "step": 8487 + }, + { + "epoch": 0.6379077108071547, + "grad_norm": 1.295718171173497, + "learning_rate": 1.2248611159682582e-06, + "loss": 0.9027, + "num_input_tokens_seen": 138596490, + "step": 8488 + }, + { + "epoch": 0.6379828648729896, + "grad_norm": 2.264882880376491, + "learning_rate": 1.2244123480842685e-06, + "loss": 0.9912, + "num_input_tokens_seen": 138619125, + "step": 8489 + }, + { + "epoch": 0.6380580189388246, + "grad_norm": 2.086990262230463, + "learning_rate": 1.2239636261568174e-06, + "loss": 0.8952, + "num_input_tokens_seen": 138638555, + "step": 8490 + }, + { + "epoch": 0.6381331730046595, + "grad_norm": 1.7568942914329055, + "learning_rate": 1.2235149502124924e-06, + "loss": 0.9572, + "num_input_tokens_seen": 138659975, + "step": 8491 + }, + { + "epoch": 0.6382083270704945, + "grad_norm": 1.2743586853389928, + "learning_rate": 1.2230663202778806e-06, + "loss": 0.9385, + "num_input_tokens_seen": 138686375, + "step": 8492 + }, + { + "epoch": 0.6382834811363295, + "grad_norm": 1.9651237879775398, + "learning_rate": 1.2226177363795645e-06, + "loss": 0.9968, + "num_input_tokens_seen": 138710780, + "step": 8493 + }, + { + "epoch": 0.6383586352021644, + "grad_norm": 1.4326057875686087, + "learning_rate": 1.2221691985441238e-06, + "loss": 0.8896, + "num_input_tokens_seen": 138737350, + "step": 8494 + }, + { + "epoch": 0.6384337892679994, + "grad_norm": 1.4168898836691042, + "learning_rate": 1.221720706798137e-06, + "loss": 0.9709, + "num_input_tokens_seen": 138762080, + "step": 8495 + }, + { + "epoch": 0.6385089433338343, + "grad_norm": 0.873247776342095, + "learning_rate": 1.221272261168178e-06, + "loss": 0.8319, + "num_input_tokens_seen": 138832445, + "step": 8496 + }, + { + "epoch": 0.6385840973996694, + "grad_norm": 1.4142278922289884, + "learning_rate": 1.2208238616808202e-06, + "loss": 0.8958, + "num_input_tokens_seen": 138857920, + "step": 8497 + }, + { + "epoch": 0.6386592514655043, + "grad_norm": 1.5371097627709565, + "learning_rate": 1.2203755083626312e-06, + "loss": 0.8619, + "num_input_tokens_seen": 138882535, + "step": 8498 + }, + { + "epoch": 0.6387344055313392, + "grad_norm": 1.6791529314916602, + "learning_rate": 1.21992720124018e-06, + "loss": 0.9879, + "num_input_tokens_seen": 138904295, + "step": 8499 + }, + { + "epoch": 0.6388095595971742, + "grad_norm": 1.6186696507606044, + "learning_rate": 1.2194789403400289e-06, + "loss": 0.8658, + "num_input_tokens_seen": 138928220, + "step": 8500 + }, + { + "epoch": 0.6388847136630091, + "grad_norm": 1.5305564243816503, + "learning_rate": 1.2190307256887384e-06, + "loss": 0.936, + "num_input_tokens_seen": 138952955, + "step": 8501 + }, + { + "epoch": 0.6389598677288442, + "grad_norm": 1.8817869693737541, + "learning_rate": 1.218582557312869e-06, + "loss": 0.8694, + "num_input_tokens_seen": 138977985, + "step": 8502 + }, + { + "epoch": 0.6390350217946791, + "grad_norm": 1.4570683124318144, + "learning_rate": 1.2181344352389746e-06, + "loss": 0.9909, + "num_input_tokens_seen": 139002690, + "step": 8503 + }, + { + "epoch": 0.6391101758605141, + "grad_norm": 1.3794531669644778, + "learning_rate": 1.2176863594936095e-06, + "loss": 0.8749, + "num_input_tokens_seen": 139026680, + "step": 8504 + }, + { + "epoch": 0.639185329926349, + "grad_norm": 1.403676029746314, + "learning_rate": 1.2172383301033233e-06, + "loss": 0.9886, + "num_input_tokens_seen": 139050230, + "step": 8505 + }, + { + "epoch": 0.6392604839921839, + "grad_norm": 1.5421336557869396, + "learning_rate": 1.216790347094663e-06, + "loss": 1.0148, + "num_input_tokens_seen": 139072010, + "step": 8506 + }, + { + "epoch": 0.639335638058019, + "grad_norm": 2.003559283284561, + "learning_rate": 1.2163424104941743e-06, + "loss": 0.9114, + "num_input_tokens_seen": 139094680, + "step": 8507 + }, + { + "epoch": 0.6394107921238539, + "grad_norm": 1.7836860596090822, + "learning_rate": 1.215894520328398e-06, + "loss": 0.9178, + "num_input_tokens_seen": 139118850, + "step": 8508 + }, + { + "epoch": 0.6394859461896889, + "grad_norm": 1.5297871043865268, + "learning_rate": 1.2154466766238742e-06, + "loss": 0.9733, + "num_input_tokens_seen": 139141295, + "step": 8509 + }, + { + "epoch": 0.6395611002555238, + "grad_norm": 1.5010339828954595, + "learning_rate": 1.2149988794071392e-06, + "loss": 0.9034, + "num_input_tokens_seen": 139167395, + "step": 8510 + }, + { + "epoch": 0.6396362543213587, + "grad_norm": 1.719040655355485, + "learning_rate": 1.214551128704726e-06, + "loss": 0.9897, + "num_input_tokens_seen": 139188985, + "step": 8511 + }, + { + "epoch": 0.6397114083871938, + "grad_norm": 1.670669120779668, + "learning_rate": 1.214103424543167e-06, + "loss": 0.951, + "num_input_tokens_seen": 139211065, + "step": 8512 + }, + { + "epoch": 0.6397865624530287, + "grad_norm": 1.4560879269240021, + "learning_rate": 1.2136557669489886e-06, + "loss": 0.94, + "num_input_tokens_seen": 139235490, + "step": 8513 + }, + { + "epoch": 0.6398617165188637, + "grad_norm": 1.4793199898500646, + "learning_rate": 1.2132081559487177e-06, + "loss": 0.8772, + "num_input_tokens_seen": 139258425, + "step": 8514 + }, + { + "epoch": 0.6399368705846986, + "grad_norm": 2.019578826166812, + "learning_rate": 1.2127605915688764e-06, + "loss": 0.9816, + "num_input_tokens_seen": 139281055, + "step": 8515 + }, + { + "epoch": 0.6400120246505336, + "grad_norm": 0.8025356148752265, + "learning_rate": 1.2123130738359842e-06, + "loss": 0.834, + "num_input_tokens_seen": 139351080, + "step": 8516 + }, + { + "epoch": 0.6400871787163686, + "grad_norm": 1.24834666584129, + "learning_rate": 1.2118656027765591e-06, + "loss": 0.9208, + "num_input_tokens_seen": 139375120, + "step": 8517 + }, + { + "epoch": 0.6401623327822035, + "grad_norm": 1.6145332410691748, + "learning_rate": 1.2114181784171144e-06, + "loss": 0.9062, + "num_input_tokens_seen": 139401850, + "step": 8518 + }, + { + "epoch": 0.6402374868480385, + "grad_norm": 1.7579360579404966, + "learning_rate": 1.2109708007841629e-06, + "loss": 0.9618, + "num_input_tokens_seen": 139426585, + "step": 8519 + }, + { + "epoch": 0.6403126409138734, + "grad_norm": 1.5305818929712023, + "learning_rate": 1.2105234699042117e-06, + "loss": 0.9366, + "num_input_tokens_seen": 139449960, + "step": 8520 + }, + { + "epoch": 0.6403877949797084, + "grad_norm": 1.9993366691156187, + "learning_rate": 1.2100761858037692e-06, + "loss": 0.9919, + "num_input_tokens_seen": 139471805, + "step": 8521 + }, + { + "epoch": 0.6404629490455434, + "grad_norm": 1.65338715059357, + "learning_rate": 1.2096289485093379e-06, + "loss": 0.9503, + "num_input_tokens_seen": 139496295, + "step": 8522 + }, + { + "epoch": 0.6405381031113784, + "grad_norm": 1.489714485565821, + "learning_rate": 1.2091817580474164e-06, + "loss": 0.8831, + "num_input_tokens_seen": 139525650, + "step": 8523 + }, + { + "epoch": 0.6406132571772133, + "grad_norm": 1.539135616521469, + "learning_rate": 1.2087346144445053e-06, + "loss": 0.9299, + "num_input_tokens_seen": 139548660, + "step": 8524 + }, + { + "epoch": 0.6406884112430482, + "grad_norm": 1.6088486338436203, + "learning_rate": 1.2082875177270974e-06, + "loss": 0.8712, + "num_input_tokens_seen": 139573175, + "step": 8525 + }, + { + "epoch": 0.6407635653088832, + "grad_norm": 1.5837848504739975, + "learning_rate": 1.2078404679216862e-06, + "loss": 0.9383, + "num_input_tokens_seen": 139596430, + "step": 8526 + }, + { + "epoch": 0.6408387193747181, + "grad_norm": 2.7751332431456177, + "learning_rate": 1.207393465054761e-06, + "loss": 1.0088, + "num_input_tokens_seen": 139618860, + "step": 8527 + }, + { + "epoch": 0.6409138734405532, + "grad_norm": 2.0209434190896145, + "learning_rate": 1.2069465091528074e-06, + "loss": 0.9001, + "num_input_tokens_seen": 139643350, + "step": 8528 + }, + { + "epoch": 0.6409890275063881, + "grad_norm": 1.7257558051427104, + "learning_rate": 1.2064996002423105e-06, + "loss": 0.9448, + "num_input_tokens_seen": 139667120, + "step": 8529 + }, + { + "epoch": 0.641064181572223, + "grad_norm": 1.554476009533466, + "learning_rate": 1.2060527383497501e-06, + "loss": 0.9404, + "num_input_tokens_seen": 139689065, + "step": 8530 + }, + { + "epoch": 0.641139335638058, + "grad_norm": 1.7012291952258287, + "learning_rate": 1.2056059235016056e-06, + "loss": 0.9723, + "num_input_tokens_seen": 139711125, + "step": 8531 + }, + { + "epoch": 0.6412144897038929, + "grad_norm": 1.560255498503874, + "learning_rate": 1.2051591557243526e-06, + "loss": 0.961, + "num_input_tokens_seen": 139735795, + "step": 8532 + }, + { + "epoch": 0.641289643769728, + "grad_norm": 1.866272706227742, + "learning_rate": 1.2047124350444624e-06, + "loss": 0.8888, + "num_input_tokens_seen": 139759675, + "step": 8533 + }, + { + "epoch": 0.6413647978355629, + "grad_norm": 0.7550974709706018, + "learning_rate": 1.2042657614884062e-06, + "loss": 0.8416, + "num_input_tokens_seen": 139841875, + "step": 8534 + }, + { + "epoch": 0.6414399519013979, + "grad_norm": 1.558043724660936, + "learning_rate": 1.2038191350826506e-06, + "loss": 0.9193, + "num_input_tokens_seen": 139865955, + "step": 8535 + }, + { + "epoch": 0.6415151059672328, + "grad_norm": 1.6782299726391652, + "learning_rate": 1.20337255585366e-06, + "loss": 0.8025, + "num_input_tokens_seen": 139890420, + "step": 8536 + }, + { + "epoch": 0.6415902600330677, + "grad_norm": 1.514496766790627, + "learning_rate": 1.2029260238278962e-06, + "loss": 0.9277, + "num_input_tokens_seen": 139914820, + "step": 8537 + }, + { + "epoch": 0.6416654140989028, + "grad_norm": 1.2369354829269168, + "learning_rate": 1.2024795390318172e-06, + "loss": 0.8879, + "num_input_tokens_seen": 139939220, + "step": 8538 + }, + { + "epoch": 0.6417405681647377, + "grad_norm": 1.81196928475728, + "learning_rate": 1.2020331014918799e-06, + "loss": 0.9544, + "num_input_tokens_seen": 139958655, + "step": 8539 + }, + { + "epoch": 0.6418157222305727, + "grad_norm": 1.4031870651961473, + "learning_rate": 1.2015867112345367e-06, + "loss": 0.9634, + "num_input_tokens_seen": 139982085, + "step": 8540 + }, + { + "epoch": 0.6418908762964076, + "grad_norm": 1.542636574830574, + "learning_rate": 1.2011403682862384e-06, + "loss": 0.9592, + "num_input_tokens_seen": 140003635, + "step": 8541 + }, + { + "epoch": 0.6419660303622426, + "grad_norm": 1.7603627333947398, + "learning_rate": 1.2006940726734315e-06, + "loss": 0.9903, + "num_input_tokens_seen": 140027725, + "step": 8542 + }, + { + "epoch": 0.6420411844280776, + "grad_norm": 1.4558563813756282, + "learning_rate": 1.2002478244225623e-06, + "loss": 0.8679, + "num_input_tokens_seen": 140051290, + "step": 8543 + }, + { + "epoch": 0.6421163384939125, + "grad_norm": 1.533881825671033, + "learning_rate": 1.1998016235600726e-06, + "loss": 0.9784, + "num_input_tokens_seen": 140074355, + "step": 8544 + }, + { + "epoch": 0.6421914925597475, + "grad_norm": 1.3336566393995777, + "learning_rate": 1.1993554701123993e-06, + "loss": 0.8627, + "num_input_tokens_seen": 140104355, + "step": 8545 + }, + { + "epoch": 0.6422666466255824, + "grad_norm": 1.5643434712322415, + "learning_rate": 1.1989093641059813e-06, + "loss": 1.0277, + "num_input_tokens_seen": 140128975, + "step": 8546 + }, + { + "epoch": 0.6423418006914174, + "grad_norm": 1.9993154427093731, + "learning_rate": 1.1984633055672508e-06, + "loss": 0.8818, + "num_input_tokens_seen": 140153775, + "step": 8547 + }, + { + "epoch": 0.6424169547572524, + "grad_norm": 1.5227917207254849, + "learning_rate": 1.1980172945226389e-06, + "loss": 0.9906, + "num_input_tokens_seen": 140178925, + "step": 8548 + }, + { + "epoch": 0.6424921088230874, + "grad_norm": 1.517853033236542, + "learning_rate": 1.1975713309985732e-06, + "loss": 1.037, + "num_input_tokens_seen": 140200840, + "step": 8549 + }, + { + "epoch": 0.6425672628889223, + "grad_norm": 1.5879189276430998, + "learning_rate": 1.1971254150214788e-06, + "loss": 0.8016, + "num_input_tokens_seen": 140224510, + "step": 8550 + }, + { + "epoch": 0.6426424169547572, + "grad_norm": 1.5383571034527486, + "learning_rate": 1.1966795466177782e-06, + "loss": 1.0339, + "num_input_tokens_seen": 140248785, + "step": 8551 + }, + { + "epoch": 0.6427175710205922, + "grad_norm": 1.8964655174944285, + "learning_rate": 1.1962337258138902e-06, + "loss": 0.9137, + "num_input_tokens_seen": 140273760, + "step": 8552 + }, + { + "epoch": 0.6427927250864272, + "grad_norm": 1.3432137506060604, + "learning_rate": 1.1957879526362323e-06, + "loss": 0.8729, + "num_input_tokens_seen": 140298300, + "step": 8553 + }, + { + "epoch": 0.6428678791522622, + "grad_norm": 1.299847471018859, + "learning_rate": 1.1953422271112175e-06, + "loss": 0.9345, + "num_input_tokens_seen": 140325970, + "step": 8554 + }, + { + "epoch": 0.6429430332180971, + "grad_norm": 1.913459900957431, + "learning_rate": 1.1948965492652565e-06, + "loss": 0.8354, + "num_input_tokens_seen": 140350210, + "step": 8555 + }, + { + "epoch": 0.643018187283932, + "grad_norm": 1.6248625917293262, + "learning_rate": 1.1944509191247585e-06, + "loss": 0.9903, + "num_input_tokens_seen": 140370150, + "step": 8556 + }, + { + "epoch": 0.643093341349767, + "grad_norm": 1.5636574082440327, + "learning_rate": 1.1940053367161278e-06, + "loss": 0.9231, + "num_input_tokens_seen": 140394010, + "step": 8557 + }, + { + "epoch": 0.643168495415602, + "grad_norm": 1.9153739951442967, + "learning_rate": 1.1935598020657676e-06, + "loss": 0.8621, + "num_input_tokens_seen": 140417100, + "step": 8558 + }, + { + "epoch": 0.643243649481437, + "grad_norm": 2.0135242248926972, + "learning_rate": 1.193114315200077e-06, + "loss": 0.9452, + "num_input_tokens_seen": 140437850, + "step": 8559 + }, + { + "epoch": 0.6433188035472719, + "grad_norm": 1.552800859671369, + "learning_rate": 1.1926688761454531e-06, + "loss": 0.8785, + "num_input_tokens_seen": 140460155, + "step": 8560 + }, + { + "epoch": 0.6433939576131069, + "grad_norm": 1.7050389725602224, + "learning_rate": 1.1922234849282897e-06, + "loss": 0.9168, + "num_input_tokens_seen": 140485600, + "step": 8561 + }, + { + "epoch": 0.6434691116789418, + "grad_norm": 1.6987782576051422, + "learning_rate": 1.1917781415749774e-06, + "loss": 1.0855, + "num_input_tokens_seen": 140507755, + "step": 8562 + }, + { + "epoch": 0.6435442657447767, + "grad_norm": 1.7074126422032887, + "learning_rate": 1.1913328461119062e-06, + "loss": 0.9862, + "num_input_tokens_seen": 140529505, + "step": 8563 + }, + { + "epoch": 0.6436194198106118, + "grad_norm": 1.7255077333545077, + "learning_rate": 1.1908875985654593e-06, + "loss": 0.8918, + "num_input_tokens_seen": 140552845, + "step": 8564 + }, + { + "epoch": 0.6436945738764467, + "grad_norm": 1.4028436731319487, + "learning_rate": 1.1904423989620216e-06, + "loss": 0.9686, + "num_input_tokens_seen": 140578960, + "step": 8565 + }, + { + "epoch": 0.6437697279422817, + "grad_norm": 1.5320481536259514, + "learning_rate": 1.1899972473279713e-06, + "loss": 0.9208, + "num_input_tokens_seen": 140601935, + "step": 8566 + }, + { + "epoch": 0.6438448820081166, + "grad_norm": 1.9480873753889116, + "learning_rate": 1.1895521436896857e-06, + "loss": 0.9355, + "num_input_tokens_seen": 140624690, + "step": 8567 + }, + { + "epoch": 0.6439200360739517, + "grad_norm": 1.5260633482277055, + "learning_rate": 1.1891070880735395e-06, + "loss": 1.032, + "num_input_tokens_seen": 140644110, + "step": 8568 + }, + { + "epoch": 0.6439951901397866, + "grad_norm": 1.7714746847171567, + "learning_rate": 1.1886620805059027e-06, + "loss": 0.8896, + "num_input_tokens_seen": 140665255, + "step": 8569 + }, + { + "epoch": 0.6440703442056215, + "grad_norm": 1.4599149664846391, + "learning_rate": 1.1882171210131452e-06, + "loss": 0.9279, + "num_input_tokens_seen": 140691000, + "step": 8570 + }, + { + "epoch": 0.6441454982714565, + "grad_norm": 1.5830065323049651, + "learning_rate": 1.1877722096216313e-06, + "loss": 0.8592, + "num_input_tokens_seen": 140716355, + "step": 8571 + }, + { + "epoch": 0.6442206523372914, + "grad_norm": 1.4999318107364517, + "learning_rate": 1.187327346357724e-06, + "loss": 0.889, + "num_input_tokens_seen": 140740735, + "step": 8572 + }, + { + "epoch": 0.6442958064031264, + "grad_norm": 1.5390160258290975, + "learning_rate": 1.186882531247784e-06, + "loss": 0.9076, + "num_input_tokens_seen": 140764415, + "step": 8573 + }, + { + "epoch": 0.6443709604689614, + "grad_norm": 1.1777817446164038, + "learning_rate": 1.1864377643181671e-06, + "loss": 0.8822, + "num_input_tokens_seen": 140792325, + "step": 8574 + }, + { + "epoch": 0.6444461145347963, + "grad_norm": 2.0658532828777925, + "learning_rate": 1.1859930455952283e-06, + "loss": 0.8632, + "num_input_tokens_seen": 140816585, + "step": 8575 + }, + { + "epoch": 0.6445212686006313, + "grad_norm": 2.1635805550467193, + "learning_rate": 1.185548375105319e-06, + "loss": 1.0398, + "num_input_tokens_seen": 140837425, + "step": 8576 + }, + { + "epoch": 0.6445964226664662, + "grad_norm": 1.4496910785582364, + "learning_rate": 1.1851037528747863e-06, + "loss": 0.9083, + "num_input_tokens_seen": 140859405, + "step": 8577 + }, + { + "epoch": 0.6446715767323012, + "grad_norm": 1.7094347202392373, + "learning_rate": 1.1846591789299774e-06, + "loss": 0.967, + "num_input_tokens_seen": 140880635, + "step": 8578 + }, + { + "epoch": 0.6447467307981362, + "grad_norm": 1.4087452684812114, + "learning_rate": 1.184214653297234e-06, + "loss": 0.9725, + "num_input_tokens_seen": 140907485, + "step": 8579 + }, + { + "epoch": 0.6448218848639712, + "grad_norm": 1.2333654790417834, + "learning_rate": 1.1837701760028962e-06, + "loss": 0.9614, + "num_input_tokens_seen": 140931935, + "step": 8580 + }, + { + "epoch": 0.6448970389298061, + "grad_norm": 2.5438112893558933, + "learning_rate": 1.1833257470733013e-06, + "loss": 0.9297, + "num_input_tokens_seen": 140955990, + "step": 8581 + }, + { + "epoch": 0.644972192995641, + "grad_norm": 1.6242320373490737, + "learning_rate": 1.1828813665347828e-06, + "loss": 0.9063, + "num_input_tokens_seen": 140978790, + "step": 8582 + }, + { + "epoch": 0.645047347061476, + "grad_norm": 1.8718216342356064, + "learning_rate": 1.1824370344136724e-06, + "loss": 0.9095, + "num_input_tokens_seen": 141000255, + "step": 8583 + }, + { + "epoch": 0.645122501127311, + "grad_norm": 1.8806721404973077, + "learning_rate": 1.181992750736298e-06, + "loss": 0.9741, + "num_input_tokens_seen": 141021220, + "step": 8584 + }, + { + "epoch": 0.645197655193146, + "grad_norm": 0.8274878804105894, + "learning_rate": 1.1815485155289864e-06, + "loss": 0.9075, + "num_input_tokens_seen": 141094980, + "step": 8585 + }, + { + "epoch": 0.6452728092589809, + "grad_norm": 0.7825595942098197, + "learning_rate": 1.1811043288180583e-06, + "loss": 0.7766, + "num_input_tokens_seen": 141166700, + "step": 8586 + }, + { + "epoch": 0.6453479633248159, + "grad_norm": 2.4219312107577156, + "learning_rate": 1.180660190629835e-06, + "loss": 0.915, + "num_input_tokens_seen": 141187105, + "step": 8587 + }, + { + "epoch": 0.6454231173906508, + "grad_norm": 1.6026992857232338, + "learning_rate": 1.180216100990633e-06, + "loss": 1.0043, + "num_input_tokens_seen": 141209375, + "step": 8588 + }, + { + "epoch": 0.6454982714564858, + "grad_norm": 1.7238471311294676, + "learning_rate": 1.179772059926766e-06, + "loss": 0.9252, + "num_input_tokens_seen": 141232500, + "step": 8589 + }, + { + "epoch": 0.6455734255223208, + "grad_norm": 2.0370057940094037, + "learning_rate": 1.1793280674645454e-06, + "loss": 0.957, + "num_input_tokens_seen": 141255555, + "step": 8590 + }, + { + "epoch": 0.6456485795881557, + "grad_norm": 2.125102882138668, + "learning_rate": 1.1788841236302789e-06, + "loss": 0.893, + "num_input_tokens_seen": 141277060, + "step": 8591 + }, + { + "epoch": 0.6457237336539907, + "grad_norm": 1.7881990039689124, + "learning_rate": 1.178440228450273e-06, + "loss": 0.8512, + "num_input_tokens_seen": 141297155, + "step": 8592 + }, + { + "epoch": 0.6457988877198256, + "grad_norm": 1.5840349483295173, + "learning_rate": 1.1779963819508293e-06, + "loss": 1.0557, + "num_input_tokens_seen": 141317545, + "step": 8593 + }, + { + "epoch": 0.6458740417856607, + "grad_norm": 1.9100888480247804, + "learning_rate": 1.1775525841582475e-06, + "loss": 0.9584, + "num_input_tokens_seen": 141336305, + "step": 8594 + }, + { + "epoch": 0.6459491958514956, + "grad_norm": 1.4984088246582101, + "learning_rate": 1.1771088350988247e-06, + "loss": 0.9803, + "num_input_tokens_seen": 141358045, + "step": 8595 + }, + { + "epoch": 0.6460243499173305, + "grad_norm": 1.4860196288443808, + "learning_rate": 1.1766651347988542e-06, + "loss": 0.9813, + "num_input_tokens_seen": 141382015, + "step": 8596 + }, + { + "epoch": 0.6460995039831655, + "grad_norm": 1.4962954870717708, + "learning_rate": 1.1762214832846274e-06, + "loss": 0.9637, + "num_input_tokens_seen": 141407590, + "step": 8597 + }, + { + "epoch": 0.6461746580490004, + "grad_norm": 1.405719529870682, + "learning_rate": 1.1757778805824324e-06, + "loss": 0.9843, + "num_input_tokens_seen": 141431335, + "step": 8598 + }, + { + "epoch": 0.6462498121148355, + "grad_norm": 1.877506615110241, + "learning_rate": 1.1753343267185535e-06, + "loss": 0.9013, + "num_input_tokens_seen": 141455005, + "step": 8599 + }, + { + "epoch": 0.6463249661806704, + "grad_norm": 1.3832267097761388, + "learning_rate": 1.1748908217192744e-06, + "loss": 1.0056, + "num_input_tokens_seen": 141477170, + "step": 8600 + }, + { + "epoch": 0.6464001202465053, + "grad_norm": 1.7685362858607476, + "learning_rate": 1.1744473656108729e-06, + "loss": 0.8794, + "num_input_tokens_seen": 141499220, + "step": 8601 + }, + { + "epoch": 0.6464752743123403, + "grad_norm": 1.8158265860392662, + "learning_rate": 1.1740039584196265e-06, + "loss": 0.9307, + "num_input_tokens_seen": 141520195, + "step": 8602 + }, + { + "epoch": 0.6465504283781752, + "grad_norm": 1.3352778183539598, + "learning_rate": 1.1735606001718087e-06, + "loss": 0.9694, + "num_input_tokens_seen": 141543375, + "step": 8603 + }, + { + "epoch": 0.6466255824440102, + "grad_norm": 1.6547093693456028, + "learning_rate": 1.17311729089369e-06, + "loss": 0.9552, + "num_input_tokens_seen": 141570715, + "step": 8604 + }, + { + "epoch": 0.6467007365098452, + "grad_norm": 1.8678624676917919, + "learning_rate": 1.172674030611538e-06, + "loss": 0.9751, + "num_input_tokens_seen": 141593435, + "step": 8605 + }, + { + "epoch": 0.6467758905756802, + "grad_norm": 5.953020550469854, + "learning_rate": 1.172230819351617e-06, + "loss": 0.9953, + "num_input_tokens_seen": 141618520, + "step": 8606 + }, + { + "epoch": 0.6468510446415151, + "grad_norm": 0.7996830550827158, + "learning_rate": 1.1717876571401913e-06, + "loss": 0.7972, + "num_input_tokens_seen": 141688850, + "step": 8607 + }, + { + "epoch": 0.64692619870735, + "grad_norm": 1.3140868630345004, + "learning_rate": 1.1713445440035168e-06, + "loss": 0.8798, + "num_input_tokens_seen": 141715425, + "step": 8608 + }, + { + "epoch": 0.647001352773185, + "grad_norm": 1.4703840540339868, + "learning_rate": 1.170901479967852e-06, + "loss": 1.0009, + "num_input_tokens_seen": 141737460, + "step": 8609 + }, + { + "epoch": 0.64707650683902, + "grad_norm": 1.6351426577379036, + "learning_rate": 1.1704584650594495e-06, + "loss": 0.9477, + "num_input_tokens_seen": 141761000, + "step": 8610 + }, + { + "epoch": 0.647151660904855, + "grad_norm": 1.4712762412532094, + "learning_rate": 1.1700154993045588e-06, + "loss": 0.8657, + "num_input_tokens_seen": 141788305, + "step": 8611 + }, + { + "epoch": 0.6472268149706899, + "grad_norm": 1.9345116488455278, + "learning_rate": 1.1695725827294286e-06, + "loss": 0.9617, + "num_input_tokens_seen": 141809235, + "step": 8612 + }, + { + "epoch": 0.6473019690365249, + "grad_norm": 1.5163801219483564, + "learning_rate": 1.1691297153603023e-06, + "loss": 0.9586, + "num_input_tokens_seen": 141830705, + "step": 8613 + }, + { + "epoch": 0.6473771231023598, + "grad_norm": 1.4571192819627623, + "learning_rate": 1.1686868972234227e-06, + "loss": 0.9561, + "num_input_tokens_seen": 141853105, + "step": 8614 + }, + { + "epoch": 0.6474522771681948, + "grad_norm": 1.3864520608983166, + "learning_rate": 1.1682441283450275e-06, + "loss": 0.9452, + "num_input_tokens_seen": 141878365, + "step": 8615 + }, + { + "epoch": 0.6475274312340298, + "grad_norm": 1.339328499973425, + "learning_rate": 1.1678014087513522e-06, + "loss": 0.8684, + "num_input_tokens_seen": 141901045, + "step": 8616 + }, + { + "epoch": 0.6476025852998647, + "grad_norm": 1.6636445142865621, + "learning_rate": 1.1673587384686308e-06, + "loss": 0.9549, + "num_input_tokens_seen": 141923600, + "step": 8617 + }, + { + "epoch": 0.6476777393656997, + "grad_norm": 1.408717300996835, + "learning_rate": 1.1669161175230913e-06, + "loss": 0.8892, + "num_input_tokens_seen": 141949725, + "step": 8618 + }, + { + "epoch": 0.6477528934315346, + "grad_norm": 1.4156636820175532, + "learning_rate": 1.1664735459409632e-06, + "loss": 0.8533, + "num_input_tokens_seen": 141976580, + "step": 8619 + }, + { + "epoch": 0.6478280474973696, + "grad_norm": 1.4132480277927146, + "learning_rate": 1.1660310237484691e-06, + "loss": 0.907, + "num_input_tokens_seen": 142001340, + "step": 8620 + }, + { + "epoch": 0.6479032015632046, + "grad_norm": 1.6355050259509718, + "learning_rate": 1.1655885509718304e-06, + "loss": 0.9016, + "num_input_tokens_seen": 142025745, + "step": 8621 + }, + { + "epoch": 0.6479783556290395, + "grad_norm": 0.7428968553224669, + "learning_rate": 1.165146127637265e-06, + "loss": 0.8281, + "num_input_tokens_seen": 142110005, + "step": 8622 + }, + { + "epoch": 0.6480535096948745, + "grad_norm": 1.7437829892992796, + "learning_rate": 1.1647037537709876e-06, + "loss": 0.9341, + "num_input_tokens_seen": 142131080, + "step": 8623 + }, + { + "epoch": 0.6481286637607094, + "grad_norm": 1.5005572396597335, + "learning_rate": 1.1642614293992123e-06, + "loss": 0.8657, + "num_input_tokens_seen": 142152945, + "step": 8624 + }, + { + "epoch": 0.6482038178265445, + "grad_norm": 1.8277681442568574, + "learning_rate": 1.1638191545481476e-06, + "loss": 0.9166, + "num_input_tokens_seen": 142173590, + "step": 8625 + }, + { + "epoch": 0.6482789718923794, + "grad_norm": 2.233181401204303, + "learning_rate": 1.163376929244e-06, + "loss": 0.9597, + "num_input_tokens_seen": 142197460, + "step": 8626 + }, + { + "epoch": 0.6483541259582143, + "grad_norm": 1.4569095018671747, + "learning_rate": 1.1629347535129728e-06, + "loss": 0.842, + "num_input_tokens_seen": 142225480, + "step": 8627 + }, + { + "epoch": 0.6484292800240493, + "grad_norm": 1.7364740128048581, + "learning_rate": 1.1624926273812664e-06, + "loss": 0.9614, + "num_input_tokens_seen": 142244915, + "step": 8628 + }, + { + "epoch": 0.6485044340898842, + "grad_norm": 0.7502377848382343, + "learning_rate": 1.162050550875079e-06, + "loss": 0.8115, + "num_input_tokens_seen": 142321665, + "step": 8629 + }, + { + "epoch": 0.6485795881557193, + "grad_norm": 1.6862818948657505, + "learning_rate": 1.1616085240206058e-06, + "loss": 0.9182, + "num_input_tokens_seen": 142344670, + "step": 8630 + }, + { + "epoch": 0.6486547422215542, + "grad_norm": 1.8976036092744761, + "learning_rate": 1.1611665468440376e-06, + "loss": 0.9978, + "num_input_tokens_seen": 142370235, + "step": 8631 + }, + { + "epoch": 0.6487298962873892, + "grad_norm": 1.6620455414721613, + "learning_rate": 1.1607246193715637e-06, + "loss": 0.9296, + "num_input_tokens_seen": 142390485, + "step": 8632 + }, + { + "epoch": 0.6488050503532241, + "grad_norm": 1.6227869222716287, + "learning_rate": 1.160282741629369e-06, + "loss": 1.0024, + "num_input_tokens_seen": 142412150, + "step": 8633 + }, + { + "epoch": 0.648880204419059, + "grad_norm": 2.45391551115905, + "learning_rate": 1.1598409136436385e-06, + "loss": 0.9112, + "num_input_tokens_seen": 142437840, + "step": 8634 + }, + { + "epoch": 0.648955358484894, + "grad_norm": 1.5162173028173742, + "learning_rate": 1.1593991354405505e-06, + "loss": 0.9466, + "num_input_tokens_seen": 142461290, + "step": 8635 + }, + { + "epoch": 0.649030512550729, + "grad_norm": 0.8088339481090717, + "learning_rate": 1.158957407046283e-06, + "loss": 0.7772, + "num_input_tokens_seen": 142539030, + "step": 8636 + }, + { + "epoch": 0.649105666616564, + "grad_norm": 1.473662257927322, + "learning_rate": 1.1585157284870097e-06, + "loss": 0.9898, + "num_input_tokens_seen": 142563195, + "step": 8637 + }, + { + "epoch": 0.6491808206823989, + "grad_norm": 1.6561898274557392, + "learning_rate": 1.1580740997889008e-06, + "loss": 0.9613, + "num_input_tokens_seen": 142582885, + "step": 8638 + }, + { + "epoch": 0.6492559747482339, + "grad_norm": 1.2479105174953844, + "learning_rate": 1.1576325209781263e-06, + "loss": 1.0095, + "num_input_tokens_seen": 142607695, + "step": 8639 + }, + { + "epoch": 0.6493311288140688, + "grad_norm": 1.8214854771142186, + "learning_rate": 1.1571909920808498e-06, + "loss": 0.9034, + "num_input_tokens_seen": 142632675, + "step": 8640 + }, + { + "epoch": 0.6494062828799038, + "grad_norm": 0.7163596001831741, + "learning_rate": 1.156749513123235e-06, + "loss": 0.8188, + "num_input_tokens_seen": 142716425, + "step": 8641 + }, + { + "epoch": 0.6494814369457388, + "grad_norm": 0.7147049482489997, + "learning_rate": 1.1563080841314408e-06, + "loss": 0.751, + "num_input_tokens_seen": 142791720, + "step": 8642 + }, + { + "epoch": 0.6495565910115737, + "grad_norm": 1.8697825000364687, + "learning_rate": 1.155866705131623e-06, + "loss": 0.8671, + "num_input_tokens_seen": 142814985, + "step": 8643 + }, + { + "epoch": 0.6496317450774087, + "grad_norm": 1.6591214893320878, + "learning_rate": 1.1554253761499358e-06, + "loss": 0.8725, + "num_input_tokens_seen": 142838125, + "step": 8644 + }, + { + "epoch": 0.6497068991432436, + "grad_norm": 1.5524330854507855, + "learning_rate": 1.154984097212528e-06, + "loss": 0.9439, + "num_input_tokens_seen": 142858005, + "step": 8645 + }, + { + "epoch": 0.6497820532090786, + "grad_norm": 1.7593147470784527, + "learning_rate": 1.154542868345549e-06, + "loss": 0.965, + "num_input_tokens_seen": 142876695, + "step": 8646 + }, + { + "epoch": 0.6498572072749136, + "grad_norm": 2.573331410781195, + "learning_rate": 1.1541016895751425e-06, + "loss": 0.8728, + "num_input_tokens_seen": 142905845, + "step": 8647 + }, + { + "epoch": 0.6499323613407485, + "grad_norm": 1.5946180186383856, + "learning_rate": 1.1536605609274504e-06, + "loss": 0.8123, + "num_input_tokens_seen": 142930765, + "step": 8648 + }, + { + "epoch": 0.6500075154065835, + "grad_norm": 1.4900235768187389, + "learning_rate": 1.1532194824286107e-06, + "loss": 0.8384, + "num_input_tokens_seen": 142955365, + "step": 8649 + }, + { + "epoch": 0.6500826694724184, + "grad_norm": 1.4353640070069626, + "learning_rate": 1.1527784541047583e-06, + "loss": 0.9782, + "num_input_tokens_seen": 142979955, + "step": 8650 + }, + { + "epoch": 0.6501578235382535, + "grad_norm": 1.6747722300061887, + "learning_rate": 1.1523374759820276e-06, + "loss": 0.9462, + "num_input_tokens_seen": 143005555, + "step": 8651 + }, + { + "epoch": 0.6502329776040884, + "grad_norm": 1.657794663947224, + "learning_rate": 1.1518965480865474e-06, + "loss": 0.9501, + "num_input_tokens_seen": 143026895, + "step": 8652 + }, + { + "epoch": 0.6503081316699233, + "grad_norm": 1.3681773934541919, + "learning_rate": 1.1514556704444446e-06, + "loss": 0.9653, + "num_input_tokens_seen": 143051105, + "step": 8653 + }, + { + "epoch": 0.6503832857357583, + "grad_norm": 1.5098640834193586, + "learning_rate": 1.151014843081842e-06, + "loss": 0.9587, + "num_input_tokens_seen": 143074265, + "step": 8654 + }, + { + "epoch": 0.6504584398015932, + "grad_norm": 1.6310983813918316, + "learning_rate": 1.1505740660248606e-06, + "loss": 0.9533, + "num_input_tokens_seen": 143097695, + "step": 8655 + }, + { + "epoch": 0.6505335938674283, + "grad_norm": 1.4770408541763513, + "learning_rate": 1.1501333392996194e-06, + "loss": 0.8888, + "num_input_tokens_seen": 143123660, + "step": 8656 + }, + { + "epoch": 0.6506087479332632, + "grad_norm": 1.7241051910021334, + "learning_rate": 1.1496926629322316e-06, + "loss": 1.0207, + "num_input_tokens_seen": 143146430, + "step": 8657 + }, + { + "epoch": 0.6506839019990982, + "grad_norm": 0.8189727378308823, + "learning_rate": 1.14925203694881e-06, + "loss": 0.8125, + "num_input_tokens_seen": 143212255, + "step": 8658 + }, + { + "epoch": 0.6507590560649331, + "grad_norm": 1.4479424545223738, + "learning_rate": 1.148811461375463e-06, + "loss": 0.9822, + "num_input_tokens_seen": 143237360, + "step": 8659 + }, + { + "epoch": 0.650834210130768, + "grad_norm": 1.487016756436638, + "learning_rate": 1.1483709362382953e-06, + "loss": 1.0265, + "num_input_tokens_seen": 143262225, + "step": 8660 + }, + { + "epoch": 0.6509093641966031, + "grad_norm": 1.719564834259826, + "learning_rate": 1.1479304615634115e-06, + "loss": 0.7629, + "num_input_tokens_seen": 143285200, + "step": 8661 + }, + { + "epoch": 0.650984518262438, + "grad_norm": 1.361369894473301, + "learning_rate": 1.14749003737691e-06, + "loss": 0.9529, + "num_input_tokens_seen": 143311435, + "step": 8662 + }, + { + "epoch": 0.651059672328273, + "grad_norm": 1.5529718949132276, + "learning_rate": 1.147049663704889e-06, + "loss": 0.9627, + "num_input_tokens_seen": 143334020, + "step": 8663 + }, + { + "epoch": 0.6511348263941079, + "grad_norm": 1.9872874716123157, + "learning_rate": 1.1466093405734417e-06, + "loss": 1.0256, + "num_input_tokens_seen": 143352320, + "step": 8664 + }, + { + "epoch": 0.6512099804599428, + "grad_norm": 1.6947503630191354, + "learning_rate": 1.1461690680086587e-06, + "loss": 0.9896, + "num_input_tokens_seen": 143375590, + "step": 8665 + }, + { + "epoch": 0.6512851345257779, + "grad_norm": 1.6178681788553826, + "learning_rate": 1.145728846036628e-06, + "loss": 1.0106, + "num_input_tokens_seen": 143399975, + "step": 8666 + }, + { + "epoch": 0.6513602885916128, + "grad_norm": 1.4392215537314252, + "learning_rate": 1.1452886746834335e-06, + "loss": 0.8909, + "num_input_tokens_seen": 143423480, + "step": 8667 + }, + { + "epoch": 0.6514354426574478, + "grad_norm": 1.6092042554629822, + "learning_rate": 1.1448485539751586e-06, + "loss": 0.9051, + "num_input_tokens_seen": 143447365, + "step": 8668 + }, + { + "epoch": 0.6515105967232827, + "grad_norm": 1.4768027458322444, + "learning_rate": 1.144408483937882e-06, + "loss": 0.8475, + "num_input_tokens_seen": 143471765, + "step": 8669 + }, + { + "epoch": 0.6515857507891177, + "grad_norm": 1.6392464795420236, + "learning_rate": 1.1439684645976787e-06, + "loss": 0.8883, + "num_input_tokens_seen": 143494715, + "step": 8670 + }, + { + "epoch": 0.6516609048549526, + "grad_norm": 1.6199523584401248, + "learning_rate": 1.1435284959806218e-06, + "loss": 0.8184, + "num_input_tokens_seen": 143517625, + "step": 8671 + }, + { + "epoch": 0.6517360589207876, + "grad_norm": 1.3682964949734902, + "learning_rate": 1.1430885781127803e-06, + "loss": 0.9053, + "num_input_tokens_seen": 143542805, + "step": 8672 + }, + { + "epoch": 0.6518112129866226, + "grad_norm": 0.9120015115286197, + "learning_rate": 1.1426487110202228e-06, + "loss": 0.8824, + "num_input_tokens_seen": 143621945, + "step": 8673 + }, + { + "epoch": 0.6518863670524575, + "grad_norm": 1.586152912412279, + "learning_rate": 1.142208894729012e-06, + "loss": 0.8868, + "num_input_tokens_seen": 143646980, + "step": 8674 + }, + { + "epoch": 0.6519615211182925, + "grad_norm": 1.5479362007370083, + "learning_rate": 1.1417691292652091e-06, + "loss": 0.8823, + "num_input_tokens_seen": 143671080, + "step": 8675 + }, + { + "epoch": 0.6520366751841274, + "grad_norm": 1.8284774831710735, + "learning_rate": 1.1413294146548716e-06, + "loss": 0.9661, + "num_input_tokens_seen": 143693005, + "step": 8676 + }, + { + "epoch": 0.6521118292499625, + "grad_norm": 1.384649920236638, + "learning_rate": 1.1408897509240537e-06, + "loss": 0.9551, + "num_input_tokens_seen": 143717480, + "step": 8677 + }, + { + "epoch": 0.6521869833157974, + "grad_norm": 1.5384832541143592, + "learning_rate": 1.1404501380988084e-06, + "loss": 0.863, + "num_input_tokens_seen": 143739705, + "step": 8678 + }, + { + "epoch": 0.6522621373816323, + "grad_norm": 1.4352698233929335, + "learning_rate": 1.1400105762051833e-06, + "loss": 0.8735, + "num_input_tokens_seen": 143761430, + "step": 8679 + }, + { + "epoch": 0.6523372914474673, + "grad_norm": 1.5807304403267453, + "learning_rate": 1.139571065269226e-06, + "loss": 0.8686, + "num_input_tokens_seen": 143784165, + "step": 8680 + }, + { + "epoch": 0.6524124455133022, + "grad_norm": 1.3351550571532302, + "learning_rate": 1.1391316053169773e-06, + "loss": 0.9404, + "num_input_tokens_seen": 143807405, + "step": 8681 + }, + { + "epoch": 0.6524875995791373, + "grad_norm": 1.661063343638523, + "learning_rate": 1.1386921963744765e-06, + "loss": 0.8709, + "num_input_tokens_seen": 143831340, + "step": 8682 + }, + { + "epoch": 0.6525627536449722, + "grad_norm": 1.2746765100333155, + "learning_rate": 1.1382528384677619e-06, + "loss": 1.0379, + "num_input_tokens_seen": 143857560, + "step": 8683 + }, + { + "epoch": 0.6526379077108072, + "grad_norm": 1.623759896921445, + "learning_rate": 1.137813531622866e-06, + "loss": 0.9715, + "num_input_tokens_seen": 143876260, + "step": 8684 + }, + { + "epoch": 0.6527130617766421, + "grad_norm": 1.66824945950996, + "learning_rate": 1.1373742758658206e-06, + "loss": 0.9026, + "num_input_tokens_seen": 143898840, + "step": 8685 + }, + { + "epoch": 0.652788215842477, + "grad_norm": 1.4946801102998108, + "learning_rate": 1.1369350712226525e-06, + "loss": 1.0275, + "num_input_tokens_seen": 143921590, + "step": 8686 + }, + { + "epoch": 0.6528633699083121, + "grad_norm": 0.8651422662553895, + "learning_rate": 1.1364959177193863e-06, + "loss": 0.8159, + "num_input_tokens_seen": 143999705, + "step": 8687 + }, + { + "epoch": 0.652938523974147, + "grad_norm": 1.8440978158589003, + "learning_rate": 1.1360568153820436e-06, + "loss": 1.0023, + "num_input_tokens_seen": 144020975, + "step": 8688 + }, + { + "epoch": 0.653013678039982, + "grad_norm": 0.7512821523225321, + "learning_rate": 1.1356177642366422e-06, + "loss": 0.8589, + "num_input_tokens_seen": 144094160, + "step": 8689 + }, + { + "epoch": 0.6530888321058169, + "grad_norm": 1.490176618464244, + "learning_rate": 1.1351787643091988e-06, + "loss": 0.9091, + "num_input_tokens_seen": 144119995, + "step": 8690 + }, + { + "epoch": 0.6531639861716518, + "grad_norm": 1.3630309216664471, + "learning_rate": 1.1347398156257253e-06, + "loss": 0.9802, + "num_input_tokens_seen": 144146095, + "step": 8691 + }, + { + "epoch": 0.6532391402374869, + "grad_norm": 2.0262490552915318, + "learning_rate": 1.134300918212231e-06, + "loss": 1.0026, + "num_input_tokens_seen": 144165255, + "step": 8692 + }, + { + "epoch": 0.6533142943033218, + "grad_norm": 1.2457722216175764, + "learning_rate": 1.1338620720947223e-06, + "loss": 0.8175, + "num_input_tokens_seen": 144193905, + "step": 8693 + }, + { + "epoch": 0.6533894483691568, + "grad_norm": 1.38115705116957, + "learning_rate": 1.1334232772992018e-06, + "loss": 0.931, + "num_input_tokens_seen": 144217915, + "step": 8694 + }, + { + "epoch": 0.6534646024349917, + "grad_norm": 1.732602287454602, + "learning_rate": 1.132984533851671e-06, + "loss": 1.0506, + "num_input_tokens_seen": 144240135, + "step": 8695 + }, + { + "epoch": 0.6535397565008267, + "grad_norm": 1.494815928121608, + "learning_rate": 1.132545841778127e-06, + "loss": 0.9599, + "num_input_tokens_seen": 144262055, + "step": 8696 + }, + { + "epoch": 0.6536149105666617, + "grad_norm": 1.2906049805351618, + "learning_rate": 1.1321072011045631e-06, + "loss": 0.9467, + "num_input_tokens_seen": 144287305, + "step": 8697 + }, + { + "epoch": 0.6536900646324966, + "grad_norm": 1.823071921866991, + "learning_rate": 1.1316686118569712e-06, + "loss": 0.9896, + "num_input_tokens_seen": 144310185, + "step": 8698 + }, + { + "epoch": 0.6537652186983316, + "grad_norm": 1.8948672910939024, + "learning_rate": 1.1312300740613382e-06, + "loss": 0.9508, + "num_input_tokens_seen": 144330030, + "step": 8699 + }, + { + "epoch": 0.6538403727641665, + "grad_norm": 1.6613333086031141, + "learning_rate": 1.130791587743651e-06, + "loss": 0.9491, + "num_input_tokens_seen": 144355570, + "step": 8700 + }, + { + "epoch": 0.6539155268300015, + "grad_norm": 1.5092142339351142, + "learning_rate": 1.1303531529298898e-06, + "loss": 0.8614, + "num_input_tokens_seen": 144379545, + "step": 8701 + }, + { + "epoch": 0.6539906808958365, + "grad_norm": 0.7983804208318521, + "learning_rate": 1.1299147696460361e-06, + "loss": 0.7968, + "num_input_tokens_seen": 144461800, + "step": 8702 + }, + { + "epoch": 0.6540658349616715, + "grad_norm": 0.8166926123591239, + "learning_rate": 1.129476437918063e-06, + "loss": 0.8219, + "num_input_tokens_seen": 144532770, + "step": 8703 + }, + { + "epoch": 0.6541409890275064, + "grad_norm": 1.5441542308544738, + "learning_rate": 1.1290381577719436e-06, + "loss": 1.0064, + "num_input_tokens_seen": 144555050, + "step": 8704 + }, + { + "epoch": 0.6542161430933413, + "grad_norm": 2.0112027886197614, + "learning_rate": 1.1285999292336495e-06, + "loss": 0.8803, + "num_input_tokens_seen": 144580195, + "step": 8705 + }, + { + "epoch": 0.6542912971591763, + "grad_norm": 1.5103528064074956, + "learning_rate": 1.1281617523291456e-06, + "loss": 1.0236, + "num_input_tokens_seen": 144603195, + "step": 8706 + }, + { + "epoch": 0.6543664512250112, + "grad_norm": 1.6229529324920655, + "learning_rate": 1.127723627084397e-06, + "loss": 0.8194, + "num_input_tokens_seen": 144627825, + "step": 8707 + }, + { + "epoch": 0.6544416052908463, + "grad_norm": 1.5869001652035708, + "learning_rate": 1.1272855535253637e-06, + "loss": 0.9379, + "num_input_tokens_seen": 144652575, + "step": 8708 + }, + { + "epoch": 0.6545167593566812, + "grad_norm": 2.9821075293779167, + "learning_rate": 1.1268475316780036e-06, + "loss": 0.9631, + "num_input_tokens_seen": 144674700, + "step": 8709 + }, + { + "epoch": 0.6545919134225161, + "grad_norm": 1.3562609201861815, + "learning_rate": 1.1264095615682704e-06, + "loss": 0.9108, + "num_input_tokens_seen": 144699000, + "step": 8710 + }, + { + "epoch": 0.6546670674883511, + "grad_norm": 1.5363619585576984, + "learning_rate": 1.125971643222115e-06, + "loss": 0.866, + "num_input_tokens_seen": 144728430, + "step": 8711 + }, + { + "epoch": 0.654742221554186, + "grad_norm": 1.4129745766417172, + "learning_rate": 1.1255337766654873e-06, + "loss": 0.9689, + "num_input_tokens_seen": 144752810, + "step": 8712 + }, + { + "epoch": 0.6548173756200211, + "grad_norm": 1.4315273765746894, + "learning_rate": 1.1250959619243322e-06, + "loss": 0.9303, + "num_input_tokens_seen": 144777100, + "step": 8713 + }, + { + "epoch": 0.654892529685856, + "grad_norm": 1.5751191018500212, + "learning_rate": 1.1246581990245916e-06, + "loss": 0.923, + "num_input_tokens_seen": 144799805, + "step": 8714 + }, + { + "epoch": 0.654967683751691, + "grad_norm": 1.6359440465466404, + "learning_rate": 1.1242204879922045e-06, + "loss": 1.0032, + "num_input_tokens_seen": 144822310, + "step": 8715 + }, + { + "epoch": 0.6550428378175259, + "grad_norm": 1.5042793900933007, + "learning_rate": 1.1237828288531063e-06, + "loss": 0.9307, + "num_input_tokens_seen": 144849070, + "step": 8716 + }, + { + "epoch": 0.6551179918833608, + "grad_norm": 1.585100300620738, + "learning_rate": 1.1233452216332316e-06, + "loss": 0.8424, + "num_input_tokens_seen": 144869710, + "step": 8717 + }, + { + "epoch": 0.6551931459491959, + "grad_norm": 1.7392239532516245, + "learning_rate": 1.1229076663585094e-06, + "loss": 0.936, + "num_input_tokens_seen": 144893250, + "step": 8718 + }, + { + "epoch": 0.6552683000150308, + "grad_norm": 1.5553840342069758, + "learning_rate": 1.1224701630548665e-06, + "loss": 0.9185, + "num_input_tokens_seen": 144916755, + "step": 8719 + }, + { + "epoch": 0.6553434540808658, + "grad_norm": 3.1623376736803315, + "learning_rate": 1.122032711748227e-06, + "loss": 0.9225, + "num_input_tokens_seen": 144942405, + "step": 8720 + }, + { + "epoch": 0.6554186081467007, + "grad_norm": 2.4834006456132043, + "learning_rate": 1.12159531246451e-06, + "loss": 0.9566, + "num_input_tokens_seen": 144966795, + "step": 8721 + }, + { + "epoch": 0.6554937622125357, + "grad_norm": 1.9295368830785278, + "learning_rate": 1.1211579652296355e-06, + "loss": 1.003, + "num_input_tokens_seen": 144985290, + "step": 8722 + }, + { + "epoch": 0.6555689162783707, + "grad_norm": 1.690522207354026, + "learning_rate": 1.1207206700695161e-06, + "loss": 0.9537, + "num_input_tokens_seen": 145011100, + "step": 8723 + }, + { + "epoch": 0.6556440703442056, + "grad_norm": 1.9191116066821068, + "learning_rate": 1.1202834270100655e-06, + "loss": 0.9061, + "num_input_tokens_seen": 145034740, + "step": 8724 + }, + { + "epoch": 0.6557192244100406, + "grad_norm": 1.4699153740845814, + "learning_rate": 1.1198462360771895e-06, + "loss": 0.9446, + "num_input_tokens_seen": 145059585, + "step": 8725 + }, + { + "epoch": 0.6557943784758755, + "grad_norm": 1.486044657440506, + "learning_rate": 1.1194090972967943e-06, + "loss": 0.917, + "num_input_tokens_seen": 145083490, + "step": 8726 + }, + { + "epoch": 0.6558695325417105, + "grad_norm": 1.5165479543708777, + "learning_rate": 1.1189720106947823e-06, + "loss": 0.9811, + "num_input_tokens_seen": 145106350, + "step": 8727 + }, + { + "epoch": 0.6559446866075455, + "grad_norm": 1.7882345357637155, + "learning_rate": 1.1185349762970515e-06, + "loss": 0.868, + "num_input_tokens_seen": 145131565, + "step": 8728 + }, + { + "epoch": 0.6560198406733805, + "grad_norm": 1.5538450479321928, + "learning_rate": 1.1180979941294998e-06, + "loss": 0.908, + "num_input_tokens_seen": 145156020, + "step": 8729 + }, + { + "epoch": 0.6560949947392154, + "grad_norm": 1.6701659977810717, + "learning_rate": 1.1176610642180184e-06, + "loss": 0.836, + "num_input_tokens_seen": 145178255, + "step": 8730 + }, + { + "epoch": 0.6561701488050503, + "grad_norm": 1.4288175677378203, + "learning_rate": 1.117224186588498e-06, + "loss": 1.0349, + "num_input_tokens_seen": 145201065, + "step": 8731 + }, + { + "epoch": 0.6562453028708853, + "grad_norm": 1.8074266528051381, + "learning_rate": 1.1167873612668252e-06, + "loss": 1.0783, + "num_input_tokens_seen": 145223175, + "step": 8732 + }, + { + "epoch": 0.6563204569367203, + "grad_norm": 1.4608712410818483, + "learning_rate": 1.1163505882788821e-06, + "loss": 0.9338, + "num_input_tokens_seen": 145247120, + "step": 8733 + }, + { + "epoch": 0.6563956110025553, + "grad_norm": 1.5829806269594646, + "learning_rate": 1.1159138676505516e-06, + "loss": 0.8862, + "num_input_tokens_seen": 145269805, + "step": 8734 + }, + { + "epoch": 0.6564707650683902, + "grad_norm": 2.2185907441057133, + "learning_rate": 1.1154771994077095e-06, + "loss": 0.9858, + "num_input_tokens_seen": 145295380, + "step": 8735 + }, + { + "epoch": 0.6565459191342251, + "grad_norm": 0.7627961974880858, + "learning_rate": 1.1150405835762304e-06, + "loss": 0.7887, + "num_input_tokens_seen": 145380075, + "step": 8736 + }, + { + "epoch": 0.6566210732000601, + "grad_norm": 1.530994238221572, + "learning_rate": 1.1146040201819855e-06, + "loss": 0.9665, + "num_input_tokens_seen": 145402800, + "step": 8737 + }, + { + "epoch": 0.656696227265895, + "grad_norm": 1.4173739948815829, + "learning_rate": 1.114167509250842e-06, + "loss": 0.9727, + "num_input_tokens_seen": 145425395, + "step": 8738 + }, + { + "epoch": 0.6567713813317301, + "grad_norm": 1.5704999973087013, + "learning_rate": 1.1137310508086666e-06, + "loss": 0.9598, + "num_input_tokens_seen": 145446850, + "step": 8739 + }, + { + "epoch": 0.656846535397565, + "grad_norm": 1.5842626588257573, + "learning_rate": 1.11329464488132e-06, + "loss": 0.9017, + "num_input_tokens_seen": 145469370, + "step": 8740 + }, + { + "epoch": 0.6569216894634, + "grad_norm": 1.338105210964615, + "learning_rate": 1.112858291494661e-06, + "loss": 0.9571, + "num_input_tokens_seen": 145494830, + "step": 8741 + }, + { + "epoch": 0.6569968435292349, + "grad_norm": 1.7354173019914811, + "learning_rate": 1.1124219906745458e-06, + "loss": 0.8718, + "num_input_tokens_seen": 145516400, + "step": 8742 + }, + { + "epoch": 0.6570719975950698, + "grad_norm": 0.8327098818998339, + "learning_rate": 1.1119857424468252e-06, + "loss": 0.8345, + "num_input_tokens_seen": 145591745, + "step": 8743 + }, + { + "epoch": 0.6571471516609049, + "grad_norm": 1.5686843554375842, + "learning_rate": 1.1115495468373505e-06, + "loss": 0.9461, + "num_input_tokens_seen": 145615470, + "step": 8744 + }, + { + "epoch": 0.6572223057267398, + "grad_norm": 1.5723702532151265, + "learning_rate": 1.111113403871967e-06, + "loss": 1.0279, + "num_input_tokens_seen": 145638250, + "step": 8745 + }, + { + "epoch": 0.6572974597925748, + "grad_norm": 1.6027126741253412, + "learning_rate": 1.1106773135765183e-06, + "loss": 0.9195, + "num_input_tokens_seen": 145663015, + "step": 8746 + }, + { + "epoch": 0.6573726138584097, + "grad_norm": 1.4538697825946523, + "learning_rate": 1.1102412759768455e-06, + "loss": 0.9604, + "num_input_tokens_seen": 145690155, + "step": 8747 + }, + { + "epoch": 0.6574477679242448, + "grad_norm": 2.2000103300025478, + "learning_rate": 1.1098052910987824e-06, + "loss": 0.8121, + "num_input_tokens_seen": 145714035, + "step": 8748 + }, + { + "epoch": 0.6575229219900797, + "grad_norm": 1.8367494776081066, + "learning_rate": 1.1093693589681654e-06, + "loss": 0.9065, + "num_input_tokens_seen": 145736250, + "step": 8749 + }, + { + "epoch": 0.6575980760559146, + "grad_norm": 1.2803877743805598, + "learning_rate": 1.1089334796108235e-06, + "loss": 0.897, + "num_input_tokens_seen": 145758940, + "step": 8750 + }, + { + "epoch": 0.6576732301217496, + "grad_norm": 1.393551522393699, + "learning_rate": 1.1084976530525858e-06, + "loss": 0.9615, + "num_input_tokens_seen": 145786850, + "step": 8751 + }, + { + "epoch": 0.6577483841875845, + "grad_norm": 1.5007482093140738, + "learning_rate": 1.108061879319276e-06, + "loss": 0.9484, + "num_input_tokens_seen": 145810510, + "step": 8752 + }, + { + "epoch": 0.6578235382534195, + "grad_norm": 1.5223008830837927, + "learning_rate": 1.1076261584367154e-06, + "loss": 0.9844, + "num_input_tokens_seen": 145838430, + "step": 8753 + }, + { + "epoch": 0.6578986923192545, + "grad_norm": 1.5741992125706816, + "learning_rate": 1.107190490430722e-06, + "loss": 0.9807, + "num_input_tokens_seen": 145861100, + "step": 8754 + }, + { + "epoch": 0.6579738463850894, + "grad_norm": 2.9997027567791315, + "learning_rate": 1.10675487532711e-06, + "loss": 1.008, + "num_input_tokens_seen": 145886040, + "step": 8755 + }, + { + "epoch": 0.6580490004509244, + "grad_norm": 1.4084432664359174, + "learning_rate": 1.1063193131516928e-06, + "loss": 0.9894, + "num_input_tokens_seen": 145909545, + "step": 8756 + }, + { + "epoch": 0.6581241545167593, + "grad_norm": 1.5633282564293602, + "learning_rate": 1.1058838039302788e-06, + "loss": 0.867, + "num_input_tokens_seen": 145934635, + "step": 8757 + }, + { + "epoch": 0.6581993085825943, + "grad_norm": 1.5832709835558774, + "learning_rate": 1.1054483476886727e-06, + "loss": 0.868, + "num_input_tokens_seen": 145960340, + "step": 8758 + }, + { + "epoch": 0.6582744626484293, + "grad_norm": 1.5714879736881644, + "learning_rate": 1.1050129444526777e-06, + "loss": 0.9323, + "num_input_tokens_seen": 145984550, + "step": 8759 + }, + { + "epoch": 0.6583496167142643, + "grad_norm": 1.5231134314522312, + "learning_rate": 1.104577594248092e-06, + "loss": 0.9127, + "num_input_tokens_seen": 146011130, + "step": 8760 + }, + { + "epoch": 0.6584247707800992, + "grad_norm": 1.564178328364776, + "learning_rate": 1.1041422971007137e-06, + "loss": 0.8984, + "num_input_tokens_seen": 146035890, + "step": 8761 + }, + { + "epoch": 0.6584999248459341, + "grad_norm": 1.5219500200083182, + "learning_rate": 1.1037070530363343e-06, + "loss": 0.8901, + "num_input_tokens_seen": 146059870, + "step": 8762 + }, + { + "epoch": 0.6585750789117691, + "grad_norm": 1.8317912581014344, + "learning_rate": 1.1032718620807446e-06, + "loss": 0.8549, + "num_input_tokens_seen": 146080535, + "step": 8763 + }, + { + "epoch": 0.658650232977604, + "grad_norm": 1.6288089860377848, + "learning_rate": 1.1028367242597307e-06, + "loss": 0.8982, + "num_input_tokens_seen": 146103290, + "step": 8764 + }, + { + "epoch": 0.6587253870434391, + "grad_norm": 1.7655739565872621, + "learning_rate": 1.1024016395990756e-06, + "loss": 0.968, + "num_input_tokens_seen": 146126985, + "step": 8765 + }, + { + "epoch": 0.658800541109274, + "grad_norm": 1.493184500555023, + "learning_rate": 1.1019666081245613e-06, + "loss": 0.9754, + "num_input_tokens_seen": 146149555, + "step": 8766 + }, + { + "epoch": 0.658875695175109, + "grad_norm": 1.4028208566519993, + "learning_rate": 1.1015316298619634e-06, + "loss": 0.9351, + "num_input_tokens_seen": 146174445, + "step": 8767 + }, + { + "epoch": 0.6589508492409439, + "grad_norm": 1.4824654594115805, + "learning_rate": 1.1010967048370577e-06, + "loss": 0.9559, + "num_input_tokens_seen": 146198465, + "step": 8768 + }, + { + "epoch": 0.6590260033067789, + "grad_norm": 1.4489049854600091, + "learning_rate": 1.1006618330756153e-06, + "loss": 0.8442, + "num_input_tokens_seen": 146227805, + "step": 8769 + }, + { + "epoch": 0.6591011573726139, + "grad_norm": 1.2958367109510227, + "learning_rate": 1.1002270146034013e-06, + "loss": 0.9799, + "num_input_tokens_seen": 146252285, + "step": 8770 + }, + { + "epoch": 0.6591763114384488, + "grad_norm": 1.560823609375624, + "learning_rate": 1.099792249446183e-06, + "loss": 0.8998, + "num_input_tokens_seen": 146274980, + "step": 8771 + }, + { + "epoch": 0.6592514655042838, + "grad_norm": 0.7617023172806794, + "learning_rate": 1.0993575376297201e-06, + "loss": 0.8093, + "num_input_tokens_seen": 146353380, + "step": 8772 + }, + { + "epoch": 0.6593266195701187, + "grad_norm": 1.809087302834762, + "learning_rate": 1.0989228791797729e-06, + "loss": 1.0283, + "num_input_tokens_seen": 146376920, + "step": 8773 + }, + { + "epoch": 0.6594017736359538, + "grad_norm": 6.229246243209921, + "learning_rate": 1.0984882741220957e-06, + "loss": 1.0069, + "num_input_tokens_seen": 146399245, + "step": 8774 + }, + { + "epoch": 0.6594769277017887, + "grad_norm": 1.5336059818567056, + "learning_rate": 1.0980537224824403e-06, + "loss": 0.9475, + "num_input_tokens_seen": 146422790, + "step": 8775 + }, + { + "epoch": 0.6595520817676236, + "grad_norm": 2.14907106247783, + "learning_rate": 1.0976192242865554e-06, + "loss": 0.8977, + "num_input_tokens_seen": 146446575, + "step": 8776 + }, + { + "epoch": 0.6596272358334586, + "grad_norm": 1.3710381042839173, + "learning_rate": 1.097184779560186e-06, + "loss": 0.9463, + "num_input_tokens_seen": 146472355, + "step": 8777 + }, + { + "epoch": 0.6597023898992935, + "grad_norm": 1.7759278411038726, + "learning_rate": 1.0967503883290768e-06, + "loss": 1.0117, + "num_input_tokens_seen": 146493415, + "step": 8778 + }, + { + "epoch": 0.6597775439651286, + "grad_norm": 1.718790435315375, + "learning_rate": 1.0963160506189655e-06, + "loss": 0.9415, + "num_input_tokens_seen": 146513275, + "step": 8779 + }, + { + "epoch": 0.6598526980309635, + "grad_norm": 1.3857930707024133, + "learning_rate": 1.0958817664555886e-06, + "loss": 0.9608, + "num_input_tokens_seen": 146539005, + "step": 8780 + }, + { + "epoch": 0.6599278520967984, + "grad_norm": 1.6589654938027782, + "learning_rate": 1.0954475358646793e-06, + "loss": 0.8866, + "num_input_tokens_seen": 146561220, + "step": 8781 + }, + { + "epoch": 0.6600030061626334, + "grad_norm": 1.5501400945786914, + "learning_rate": 1.0950133588719665e-06, + "loss": 0.9054, + "num_input_tokens_seen": 146585325, + "step": 8782 + }, + { + "epoch": 0.6600781602284683, + "grad_norm": 1.4100565782653198, + "learning_rate": 1.0945792355031785e-06, + "loss": 0.9534, + "num_input_tokens_seen": 146608440, + "step": 8783 + }, + { + "epoch": 0.6601533142943034, + "grad_norm": 1.5716396056940527, + "learning_rate": 1.0941451657840377e-06, + "loss": 0.917, + "num_input_tokens_seen": 146631300, + "step": 8784 + }, + { + "epoch": 0.6602284683601383, + "grad_norm": 1.4624102507872632, + "learning_rate": 1.0937111497402648e-06, + "loss": 0.9532, + "num_input_tokens_seen": 146654580, + "step": 8785 + }, + { + "epoch": 0.6603036224259733, + "grad_norm": 1.6610713815066616, + "learning_rate": 1.0932771873975764e-06, + "loss": 1.0022, + "num_input_tokens_seen": 146677315, + "step": 8786 + }, + { + "epoch": 0.6603787764918082, + "grad_norm": 0.7892501154776566, + "learning_rate": 1.0928432787816859e-06, + "loss": 0.8131, + "num_input_tokens_seen": 146752410, + "step": 8787 + }, + { + "epoch": 0.6604539305576431, + "grad_norm": 1.666302243763724, + "learning_rate": 1.092409423918306e-06, + "loss": 0.8745, + "num_input_tokens_seen": 146776060, + "step": 8788 + }, + { + "epoch": 0.6605290846234781, + "grad_norm": 0.8118568589295909, + "learning_rate": 1.091975622833142e-06, + "loss": 0.7801, + "num_input_tokens_seen": 146847335, + "step": 8789 + }, + { + "epoch": 0.6606042386893131, + "grad_norm": 1.4188865024590063, + "learning_rate": 1.0915418755519004e-06, + "loss": 0.8703, + "num_input_tokens_seen": 146872160, + "step": 8790 + }, + { + "epoch": 0.6606793927551481, + "grad_norm": 0.7399163682821521, + "learning_rate": 1.0911081821002811e-06, + "loss": 0.8269, + "num_input_tokens_seen": 146952310, + "step": 8791 + }, + { + "epoch": 0.660754546820983, + "grad_norm": 1.439347862421808, + "learning_rate": 1.0906745425039829e-06, + "loss": 0.9461, + "num_input_tokens_seen": 146977440, + "step": 8792 + }, + { + "epoch": 0.660829700886818, + "grad_norm": 1.6178174841960418, + "learning_rate": 1.0902409567886996e-06, + "loss": 0.9753, + "num_input_tokens_seen": 147001970, + "step": 8793 + }, + { + "epoch": 0.6609048549526529, + "grad_norm": 1.5868276719273229, + "learning_rate": 1.0898074249801227e-06, + "loss": 0.9116, + "num_input_tokens_seen": 147024405, + "step": 8794 + }, + { + "epoch": 0.6609800090184879, + "grad_norm": 1.769217423321709, + "learning_rate": 1.089373947103942e-06, + "loss": 0.9772, + "num_input_tokens_seen": 147047140, + "step": 8795 + }, + { + "epoch": 0.6610551630843229, + "grad_norm": 1.5777013228851011, + "learning_rate": 1.0889405231858422e-06, + "loss": 0.8404, + "num_input_tokens_seen": 147072895, + "step": 8796 + }, + { + "epoch": 0.6611303171501578, + "grad_norm": 1.5830551036851848, + "learning_rate": 1.0885071532515049e-06, + "loss": 0.9741, + "num_input_tokens_seen": 147096315, + "step": 8797 + }, + { + "epoch": 0.6612054712159928, + "grad_norm": 1.6306870055335172, + "learning_rate": 1.088073837326609e-06, + "loss": 0.9359, + "num_input_tokens_seen": 147118055, + "step": 8798 + }, + { + "epoch": 0.6612806252818277, + "grad_norm": 1.3924876393671886, + "learning_rate": 1.0876405754368296e-06, + "loss": 0.8856, + "num_input_tokens_seen": 147140875, + "step": 8799 + }, + { + "epoch": 0.6613557793476627, + "grad_norm": 1.4069709625121682, + "learning_rate": 1.0872073676078405e-06, + "loss": 0.9312, + "num_input_tokens_seen": 147163125, + "step": 8800 + }, + { + "epoch": 0.6614309334134977, + "grad_norm": 1.7533161895455311, + "learning_rate": 1.0867742138653103e-06, + "loss": 1.0117, + "num_input_tokens_seen": 147183765, + "step": 8801 + }, + { + "epoch": 0.6615060874793326, + "grad_norm": 1.4577362563954561, + "learning_rate": 1.0863411142349046e-06, + "loss": 0.9784, + "num_input_tokens_seen": 147204880, + "step": 8802 + }, + { + "epoch": 0.6615812415451676, + "grad_norm": 1.6652919185753186, + "learning_rate": 1.0859080687422868e-06, + "loss": 0.9388, + "num_input_tokens_seen": 147228460, + "step": 8803 + }, + { + "epoch": 0.6616563956110025, + "grad_norm": 1.5565170947446416, + "learning_rate": 1.0854750774131153e-06, + "loss": 0.873, + "num_input_tokens_seen": 147250705, + "step": 8804 + }, + { + "epoch": 0.6617315496768376, + "grad_norm": 1.5167336096329942, + "learning_rate": 1.0850421402730482e-06, + "loss": 1.0117, + "num_input_tokens_seen": 147273040, + "step": 8805 + }, + { + "epoch": 0.6618067037426725, + "grad_norm": 1.4270673412160186, + "learning_rate": 1.084609257347738e-06, + "loss": 0.8664, + "num_input_tokens_seen": 147300365, + "step": 8806 + }, + { + "epoch": 0.6618818578085074, + "grad_norm": 1.6745669360218998, + "learning_rate": 1.0841764286628344e-06, + "loss": 0.9648, + "num_input_tokens_seen": 147323850, + "step": 8807 + }, + { + "epoch": 0.6619570118743424, + "grad_norm": 1.4972072670558683, + "learning_rate": 1.0837436542439843e-06, + "loss": 1.0193, + "num_input_tokens_seen": 147347300, + "step": 8808 + }, + { + "epoch": 0.6620321659401773, + "grad_norm": 1.428832502010603, + "learning_rate": 1.0833109341168308e-06, + "loss": 0.9197, + "num_input_tokens_seen": 147369550, + "step": 8809 + }, + { + "epoch": 0.6621073200060124, + "grad_norm": 1.768404300726575, + "learning_rate": 1.0828782683070153e-06, + "loss": 0.9666, + "num_input_tokens_seen": 147391445, + "step": 8810 + }, + { + "epoch": 0.6621824740718473, + "grad_norm": 1.774850223830693, + "learning_rate": 1.0824456568401735e-06, + "loss": 0.9691, + "num_input_tokens_seen": 147417290, + "step": 8811 + }, + { + "epoch": 0.6622576281376823, + "grad_norm": 1.5067057921103515, + "learning_rate": 1.0820130997419407e-06, + "loss": 0.986, + "num_input_tokens_seen": 147442620, + "step": 8812 + }, + { + "epoch": 0.6623327822035172, + "grad_norm": 1.412695967321931, + "learning_rate": 1.0815805970379473e-06, + "loss": 0.9492, + "num_input_tokens_seen": 147469300, + "step": 8813 + }, + { + "epoch": 0.6624079362693521, + "grad_norm": 1.7284874175620395, + "learning_rate": 1.08114814875382e-06, + "loss": 1.048, + "num_input_tokens_seen": 147489020, + "step": 8814 + }, + { + "epoch": 0.6624830903351872, + "grad_norm": 1.8244452979098353, + "learning_rate": 1.0807157549151838e-06, + "loss": 0.8812, + "num_input_tokens_seen": 147513180, + "step": 8815 + }, + { + "epoch": 0.6625582444010221, + "grad_norm": 1.5588986664995068, + "learning_rate": 1.0802834155476582e-06, + "loss": 0.8173, + "num_input_tokens_seen": 147536180, + "step": 8816 + }, + { + "epoch": 0.6626333984668571, + "grad_norm": 0.7054241314166511, + "learning_rate": 1.0798511306768628e-06, + "loss": 0.8014, + "num_input_tokens_seen": 147616470, + "step": 8817 + }, + { + "epoch": 0.662708552532692, + "grad_norm": 1.3667994357180748, + "learning_rate": 1.0794189003284118e-06, + "loss": 0.9624, + "num_input_tokens_seen": 147639450, + "step": 8818 + }, + { + "epoch": 0.662783706598527, + "grad_norm": 1.6002526858592845, + "learning_rate": 1.0789867245279157e-06, + "loss": 0.9031, + "num_input_tokens_seen": 147661485, + "step": 8819 + }, + { + "epoch": 0.662858860664362, + "grad_norm": 1.5904799179732885, + "learning_rate": 1.0785546033009829e-06, + "loss": 0.9778, + "num_input_tokens_seen": 147685760, + "step": 8820 + }, + { + "epoch": 0.6629340147301969, + "grad_norm": 1.9333717912378614, + "learning_rate": 1.0781225366732179e-06, + "loss": 0.9087, + "num_input_tokens_seen": 147708390, + "step": 8821 + }, + { + "epoch": 0.6630091687960319, + "grad_norm": 2.382211478056489, + "learning_rate": 1.0776905246702233e-06, + "loss": 0.9835, + "num_input_tokens_seen": 147732445, + "step": 8822 + }, + { + "epoch": 0.6630843228618668, + "grad_norm": 1.7637979962951027, + "learning_rate": 1.077258567317597e-06, + "loss": 1.0233, + "num_input_tokens_seen": 147753735, + "step": 8823 + }, + { + "epoch": 0.6631594769277018, + "grad_norm": 2.4665427662258055, + "learning_rate": 1.076826664640934e-06, + "loss": 0.9318, + "num_input_tokens_seen": 147778165, + "step": 8824 + }, + { + "epoch": 0.6632346309935367, + "grad_norm": 1.4146283951081775, + "learning_rate": 1.076394816665826e-06, + "loss": 0.9501, + "num_input_tokens_seen": 147803455, + "step": 8825 + }, + { + "epoch": 0.6633097850593717, + "grad_norm": 1.7989840527941434, + "learning_rate": 1.075963023417861e-06, + "loss": 0.9205, + "num_input_tokens_seen": 147824410, + "step": 8826 + }, + { + "epoch": 0.6633849391252067, + "grad_norm": 1.2599239752555578, + "learning_rate": 1.075531284922626e-06, + "loss": 1.0091, + "num_input_tokens_seen": 147850810, + "step": 8827 + }, + { + "epoch": 0.6634600931910416, + "grad_norm": 1.4500478341992944, + "learning_rate": 1.0750996012057028e-06, + "loss": 0.9188, + "num_input_tokens_seen": 147873870, + "step": 8828 + }, + { + "epoch": 0.6635352472568766, + "grad_norm": 0.7311282113536759, + "learning_rate": 1.0746679722926695e-06, + "loss": 0.8055, + "num_input_tokens_seen": 147944750, + "step": 8829 + }, + { + "epoch": 0.6636104013227115, + "grad_norm": 1.544542422758298, + "learning_rate": 1.0742363982091023e-06, + "loss": 0.8363, + "num_input_tokens_seen": 147974085, + "step": 8830 + }, + { + "epoch": 0.6636855553885466, + "grad_norm": 1.4742097224619677, + "learning_rate": 1.0738048789805727e-06, + "loss": 0.9565, + "num_input_tokens_seen": 147997970, + "step": 8831 + }, + { + "epoch": 0.6637607094543815, + "grad_norm": 2.0638514628153803, + "learning_rate": 1.0733734146326513e-06, + "loss": 0.91, + "num_input_tokens_seen": 148019675, + "step": 8832 + }, + { + "epoch": 0.6638358635202164, + "grad_norm": 1.4565568821620083, + "learning_rate": 1.072942005190903e-06, + "loss": 1.0256, + "num_input_tokens_seen": 148042995, + "step": 8833 + }, + { + "epoch": 0.6639110175860514, + "grad_norm": 1.7663771116138676, + "learning_rate": 1.0725106506808912e-06, + "loss": 0.9032, + "num_input_tokens_seen": 148067405, + "step": 8834 + }, + { + "epoch": 0.6639861716518863, + "grad_norm": 1.5467267206196473, + "learning_rate": 1.0720793511281754e-06, + "loss": 0.9492, + "num_input_tokens_seen": 148088495, + "step": 8835 + }, + { + "epoch": 0.6640613257177214, + "grad_norm": 1.3139513256748425, + "learning_rate": 1.0716481065583108e-06, + "loss": 0.9226, + "num_input_tokens_seen": 148113860, + "step": 8836 + }, + { + "epoch": 0.6641364797835563, + "grad_norm": 1.7973758165440017, + "learning_rate": 1.071216916996851e-06, + "loss": 1.0076, + "num_input_tokens_seen": 148138465, + "step": 8837 + }, + { + "epoch": 0.6642116338493913, + "grad_norm": 1.5308271330221774, + "learning_rate": 1.0707857824693446e-06, + "loss": 0.8573, + "num_input_tokens_seen": 148160595, + "step": 8838 + }, + { + "epoch": 0.6642867879152262, + "grad_norm": 1.4990321851841666, + "learning_rate": 1.0703547030013399e-06, + "loss": 0.9195, + "num_input_tokens_seen": 148183140, + "step": 8839 + }, + { + "epoch": 0.6643619419810611, + "grad_norm": 1.6941967645921594, + "learning_rate": 1.0699236786183786e-06, + "loss": 0.9333, + "num_input_tokens_seen": 148205895, + "step": 8840 + }, + { + "epoch": 0.6644370960468962, + "grad_norm": 1.6321979443053258, + "learning_rate": 1.0694927093460007e-06, + "loss": 0.9316, + "num_input_tokens_seen": 148229995, + "step": 8841 + }, + { + "epoch": 0.6645122501127311, + "grad_norm": 1.5226261423656442, + "learning_rate": 1.069061795209743e-06, + "loss": 0.7855, + "num_input_tokens_seen": 148257090, + "step": 8842 + }, + { + "epoch": 0.6645874041785661, + "grad_norm": 1.5071709409433287, + "learning_rate": 1.068630936235138e-06, + "loss": 0.9794, + "num_input_tokens_seen": 148280040, + "step": 8843 + }, + { + "epoch": 0.664662558244401, + "grad_norm": 1.6365472890406674, + "learning_rate": 1.0682001324477173e-06, + "loss": 0.9203, + "num_input_tokens_seen": 148306160, + "step": 8844 + }, + { + "epoch": 0.6647377123102359, + "grad_norm": 1.3608360331536395, + "learning_rate": 1.0677693838730068e-06, + "loss": 0.9119, + "num_input_tokens_seen": 148329240, + "step": 8845 + }, + { + "epoch": 0.664812866376071, + "grad_norm": 0.7589036224175233, + "learning_rate": 1.06733869053653e-06, + "loss": 0.8271, + "num_input_tokens_seen": 148412300, + "step": 8846 + }, + { + "epoch": 0.6648880204419059, + "grad_norm": 1.4592717920440386, + "learning_rate": 1.0669080524638072e-06, + "loss": 0.9091, + "num_input_tokens_seen": 148436900, + "step": 8847 + }, + { + "epoch": 0.6649631745077409, + "grad_norm": 1.5956058542587437, + "learning_rate": 1.0664774696803548e-06, + "loss": 0.9215, + "num_input_tokens_seen": 148460715, + "step": 8848 + }, + { + "epoch": 0.6650383285735758, + "grad_norm": 1.365865275110546, + "learning_rate": 1.0660469422116876e-06, + "loss": 1.0045, + "num_input_tokens_seen": 148484430, + "step": 8849 + }, + { + "epoch": 0.6651134826394108, + "grad_norm": 1.6457990167959524, + "learning_rate": 1.0656164700833148e-06, + "loss": 0.9193, + "num_input_tokens_seen": 148505410, + "step": 8850 + }, + { + "epoch": 0.6651886367052458, + "grad_norm": 1.3616491128139445, + "learning_rate": 1.0651860533207452e-06, + "loss": 0.9556, + "num_input_tokens_seen": 148528890, + "step": 8851 + }, + { + "epoch": 0.6652637907710807, + "grad_norm": 1.7016104166869725, + "learning_rate": 1.0647556919494814e-06, + "loss": 0.8971, + "num_input_tokens_seen": 148551905, + "step": 8852 + }, + { + "epoch": 0.6653389448369157, + "grad_norm": 1.6757843800050898, + "learning_rate": 1.0643253859950231e-06, + "loss": 0.92, + "num_input_tokens_seen": 148575535, + "step": 8853 + }, + { + "epoch": 0.6654140989027506, + "grad_norm": 1.6667318649255212, + "learning_rate": 1.0638951354828693e-06, + "loss": 0.9229, + "num_input_tokens_seen": 148599055, + "step": 8854 + }, + { + "epoch": 0.6654892529685856, + "grad_norm": 1.463096288426207, + "learning_rate": 1.0634649404385127e-06, + "loss": 1.0181, + "num_input_tokens_seen": 148621035, + "step": 8855 + }, + { + "epoch": 0.6655644070344205, + "grad_norm": 1.396442460718236, + "learning_rate": 1.0630348008874452e-06, + "loss": 0.8952, + "num_input_tokens_seen": 148645565, + "step": 8856 + }, + { + "epoch": 0.6656395611002556, + "grad_norm": 1.5367749248994806, + "learning_rate": 1.062604716855154e-06, + "loss": 0.9737, + "num_input_tokens_seen": 148665830, + "step": 8857 + }, + { + "epoch": 0.6657147151660905, + "grad_norm": 1.4851171144228443, + "learning_rate": 1.0621746883671226e-06, + "loss": 0.9474, + "num_input_tokens_seen": 148690235, + "step": 8858 + }, + { + "epoch": 0.6657898692319254, + "grad_norm": 1.4924464774958255, + "learning_rate": 1.0617447154488322e-06, + "loss": 0.9208, + "num_input_tokens_seen": 148712925, + "step": 8859 + }, + { + "epoch": 0.6658650232977604, + "grad_norm": 1.616960739970805, + "learning_rate": 1.061314798125759e-06, + "loss": 1.0036, + "num_input_tokens_seen": 148735180, + "step": 8860 + }, + { + "epoch": 0.6659401773635953, + "grad_norm": 1.404798521549103, + "learning_rate": 1.0608849364233798e-06, + "loss": 0.8404, + "num_input_tokens_seen": 148763550, + "step": 8861 + }, + { + "epoch": 0.6660153314294304, + "grad_norm": 1.7924981590484999, + "learning_rate": 1.0604551303671641e-06, + "loss": 0.9252, + "num_input_tokens_seen": 148786325, + "step": 8862 + }, + { + "epoch": 0.6660904854952653, + "grad_norm": 1.3925760705191716, + "learning_rate": 1.0600253799825797e-06, + "loss": 0.9709, + "num_input_tokens_seen": 148808460, + "step": 8863 + }, + { + "epoch": 0.6661656395611003, + "grad_norm": 1.8870702064727491, + "learning_rate": 1.0595956852950907e-06, + "loss": 0.8868, + "num_input_tokens_seen": 148831140, + "step": 8864 + }, + { + "epoch": 0.6662407936269352, + "grad_norm": 1.785976163190318, + "learning_rate": 1.0591660463301578e-06, + "loss": 0.9154, + "num_input_tokens_seen": 148854605, + "step": 8865 + }, + { + "epoch": 0.6663159476927701, + "grad_norm": 0.8367865697906235, + "learning_rate": 1.0587364631132402e-06, + "loss": 0.8119, + "num_input_tokens_seen": 148928835, + "step": 8866 + }, + { + "epoch": 0.6663911017586052, + "grad_norm": 1.7193628085670343, + "learning_rate": 1.0583069356697913e-06, + "loss": 0.9141, + "num_input_tokens_seen": 148948495, + "step": 8867 + }, + { + "epoch": 0.6664662558244401, + "grad_norm": 1.7807908386577473, + "learning_rate": 1.0578774640252626e-06, + "loss": 1.0205, + "num_input_tokens_seen": 148969890, + "step": 8868 + }, + { + "epoch": 0.6665414098902751, + "grad_norm": 1.5303507129898477, + "learning_rate": 1.0574480482051017e-06, + "loss": 0.9995, + "num_input_tokens_seen": 148994665, + "step": 8869 + }, + { + "epoch": 0.66661656395611, + "grad_norm": 1.6479338106918575, + "learning_rate": 1.0570186882347525e-06, + "loss": 1.0137, + "num_input_tokens_seen": 149016685, + "step": 8870 + }, + { + "epoch": 0.6666917180219449, + "grad_norm": 1.7899068647268668, + "learning_rate": 1.0565893841396575e-06, + "loss": 0.9346, + "num_input_tokens_seen": 149040445, + "step": 8871 + }, + { + "epoch": 0.66676687208778, + "grad_norm": 1.4408583181114976, + "learning_rate": 1.0561601359452537e-06, + "loss": 0.9549, + "num_input_tokens_seen": 149063410, + "step": 8872 + } + ], + "logging_steps": 1.0, + "max_steps": 13306, + "num_input_tokens_seen": 149063410, + "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": false + }, + "attributes": {} + } + }, + "total_flos": 8.957825207280599e+17, + "train_batch_size": 5, + "trial_name": null, + "trial_params": null +}