| { | |
| "best_metric": 0.11645537614822388, | |
| "best_model_checkpoint": "miner_id_24/checkpoint-50", | |
| "epoch": 2.012802275960171, | |
| "eval_steps": 25, | |
| "global_step": 88, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02275960170697013, | |
| "grad_norm": 1.7705003023147583, | |
| "learning_rate": 0.00015, | |
| "loss": 3.3287, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02275960170697013, | |
| "eval_loss": 3.4932775497436523, | |
| "eval_runtime": 0.2913, | |
| "eval_samples_per_second": 171.665, | |
| "eval_steps_per_second": 44.633, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04551920341394026, | |
| "grad_norm": 1.6582751274108887, | |
| "learning_rate": 0.0003, | |
| "loss": 3.318, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06827880512091039, | |
| "grad_norm": 1.748265266418457, | |
| "learning_rate": 0.00029990993452998227, | |
| "loss": 3.3114, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.09103840682788052, | |
| "grad_norm": 1.5232975482940674, | |
| "learning_rate": 0.00029963985829457943, | |
| "loss": 2.9468, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11379800853485064, | |
| "grad_norm": 1.5051178932189941, | |
| "learning_rate": 0.0002991901316573927, | |
| "loss": 2.7052, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13655761024182078, | |
| "grad_norm": 1.422306776046753, | |
| "learning_rate": 0.00029856135469013987, | |
| "loss": 2.3768, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1593172119487909, | |
| "grad_norm": 1.5374380350112915, | |
| "learning_rate": 0.0002977543663719779, | |
| "loss": 2.1616, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.18207681365576103, | |
| "grad_norm": 1.8372466564178467, | |
| "learning_rate": 0.00029677024347005013, | |
| "loss": 2.0086, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.20483641536273114, | |
| "grad_norm": 2.2954390048980713, | |
| "learning_rate": 0.0002956102991027524, | |
| "loss": 1.7581, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.22759601706970128, | |
| "grad_norm": 2.7270781993865967, | |
| "learning_rate": 0.0002942760809876348, | |
| "loss": 1.5204, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2503556187766714, | |
| "grad_norm": 2.2713541984558105, | |
| "learning_rate": 0.00029276936937627725, | |
| "loss": 1.2835, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.27311522048364156, | |
| "grad_norm": 2.338096857070923, | |
| "learning_rate": 0.000291092174678894, | |
| "loss": 1.1385, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2958748221906117, | |
| "grad_norm": 1.553781270980835, | |
| "learning_rate": 0.00028924673478183645, | |
| "loss": 0.9631, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3186344238975818, | |
| "grad_norm": 1.5467333793640137, | |
| "learning_rate": 0.0002872355120615748, | |
| "loss": 0.8275, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3413940256045519, | |
| "grad_norm": 1.656901240348816, | |
| "learning_rate": 0.00028506119009914, | |
| "loss": 0.7097, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.36415362731152207, | |
| "grad_norm": 1.358717679977417, | |
| "learning_rate": 0.000282726670099414, | |
| "loss": 0.6673, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3869132290184922, | |
| "grad_norm": 0.8717578649520874, | |
| "learning_rate": 0.00028023506702004174, | |
| "loss": 0.4671, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.4096728307254623, | |
| "grad_norm": 0.977169930934906, | |
| "learning_rate": 0.0002775897054151335, | |
| "loss": 0.4218, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.43243243243243246, | |
| "grad_norm": 0.9587895274162292, | |
| "learning_rate": 0.00027479411499930134, | |
| "loss": 0.4066, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.45519203413940257, | |
| "grad_norm": 0.6970980167388916, | |
| "learning_rate": 0.00027185202593794927, | |
| "loss": 0.3582, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4779516358463727, | |
| "grad_norm": 0.7278751730918884, | |
| "learning_rate": 0.0002687673638701018, | |
| "loss": 0.3208, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5007112375533428, | |
| "grad_norm": 0.7912285327911377, | |
| "learning_rate": 0.00026554424467041055, | |
| "loss": 0.4071, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5234708392603129, | |
| "grad_norm": 0.8546118140220642, | |
| "learning_rate": 0.00026218696895732944, | |
| "loss": 0.3329, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5462304409672831, | |
| "grad_norm": 0.4830358326435089, | |
| "learning_rate": 0.0002587000163547856, | |
| "loss": 0.3032, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5689900426742532, | |
| "grad_norm": 0.7155347466468811, | |
| "learning_rate": 0.0002550880395150023, | |
| "loss": 0.2946, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5689900426742532, | |
| "eval_loss": 0.2797185778617859, | |
| "eval_runtime": 0.2962, | |
| "eval_samples_per_second": 168.833, | |
| "eval_steps_per_second": 43.897, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5917496443812233, | |
| "grad_norm": 0.5971561670303345, | |
| "learning_rate": 0.0002513558579104503, | |
| "loss": 0.2866, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.6145092460881935, | |
| "grad_norm": 0.5399057269096375, | |
| "learning_rate": 0.00024750845140320964, | |
| "loss": 0.3, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6372688477951636, | |
| "grad_norm": 0.5159599781036377, | |
| "learning_rate": 0.00024355095360032364, | |
| "loss": 0.2229, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6600284495021337, | |
| "grad_norm": 0.36613932251930237, | |
| "learning_rate": 0.00023948864500401016, | |
| "loss": 0.2183, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6827880512091038, | |
| "grad_norm": 0.4476757049560547, | |
| "learning_rate": 0.00023532694596587055, | |
| "loss": 0.1993, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.705547652916074, | |
| "grad_norm": 0.47915318608283997, | |
| "learning_rate": 0.00023107140945449652, | |
| "loss": 0.177, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.7283072546230441, | |
| "grad_norm": 0.47024720907211304, | |
| "learning_rate": 0.0002267277136461262, | |
| "loss": 0.1281, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.7510668563300142, | |
| "grad_norm": 0.43254679441452026, | |
| "learning_rate": 0.00022230165434823502, | |
| "loss": 0.2339, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7738264580369844, | |
| "grad_norm": 0.6018503904342651, | |
| "learning_rate": 0.00021779913726617102, | |
| "loss": 0.2309, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7965860597439545, | |
| "grad_norm": 0.42272430658340454, | |
| "learning_rate": 0.00021322617012315288, | |
| "loss": 0.1736, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.8193456614509246, | |
| "grad_norm": 0.4929526746273041, | |
| "learning_rate": 0.00020858885464414522, | |
| "loss": 0.1818, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 0.5257923007011414, | |
| "learning_rate": 0.00020389337841430707, | |
| "loss": 0.1759, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.8648648648648649, | |
| "grad_norm": 0.32733604311943054, | |
| "learning_rate": 0.00019914600662287684, | |
| "loss": 0.1765, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.887624466571835, | |
| "grad_norm": 0.3499005138874054, | |
| "learning_rate": 0.00019435307370351017, | |
| "loss": 0.1172, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9103840682788051, | |
| "grad_norm": 0.32600706815719604, | |
| "learning_rate": 0.0001895209748822239, | |
| "loss": 0.1107, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9331436699857752, | |
| "grad_norm": 0.3189937174320221, | |
| "learning_rate": 0.00018465615764422566, | |
| "loss": 0.134, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.9559032716927454, | |
| "grad_norm": 0.26007068157196045, | |
| "learning_rate": 0.00017976511313101307, | |
| "loss": 0.1037, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9786628733997155, | |
| "grad_norm": 0.3448164463043213, | |
| "learning_rate": 0.00017485436747922248, | |
| "loss": 0.0882, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.0064011379800855, | |
| "grad_norm": 0.28717777132987976, | |
| "learning_rate": 0.00016993047311278397, | |
| "loss": 0.2189, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.0291607396870555, | |
| "grad_norm": 0.2990550994873047, | |
| "learning_rate": 0.000165, | |
| "loss": 0.1578, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0519203413940257, | |
| "grad_norm": 0.3317011594772339, | |
| "learning_rate": 0.00016006952688721603, | |
| "loss": 0.1449, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.0746799431009957, | |
| "grad_norm": 0.31591519713401794, | |
| "learning_rate": 0.0001551456325207775, | |
| "loss": 0.1431, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.097439544807966, | |
| "grad_norm": 0.26354071497917175, | |
| "learning_rate": 0.00015023488686898698, | |
| "loss": 0.1444, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.120199146514936, | |
| "grad_norm": 0.2392982542514801, | |
| "learning_rate": 0.00014534384235577436, | |
| "loss": 0.1312, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.1429587482219061, | |
| "grad_norm": 0.3284483551979065, | |
| "learning_rate": 0.0001404790251177761, | |
| "loss": 0.1079, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1429587482219061, | |
| "eval_loss": 0.11645537614822388, | |
| "eval_runtime": 0.2863, | |
| "eval_samples_per_second": 174.617, | |
| "eval_steps_per_second": 45.4, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1657183499288761, | |
| "grad_norm": 0.35609501600265503, | |
| "learning_rate": 0.00013564692629648982, | |
| "loss": 0.0858, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.1884779516358464, | |
| "grad_norm": 0.1918356865644455, | |
| "learning_rate": 0.00013085399337712307, | |
| "loss": 0.0959, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.2112375533428166, | |
| "grad_norm": 0.2323223352432251, | |
| "learning_rate": 0.00012610662158569293, | |
| "loss": 0.0919, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.2339971550497866, | |
| "grad_norm": 0.2522870600223541, | |
| "learning_rate": 0.0001214111453558548, | |
| "loss": 0.0706, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "grad_norm": 0.29997286200523376, | |
| "learning_rate": 0.00011677382987684708, | |
| "loss": 0.1814, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.2795163584637268, | |
| "grad_norm": 0.30940982699394226, | |
| "learning_rate": 0.00011220086273382896, | |
| "loss": 0.1421, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.302275960170697, | |
| "grad_norm": 0.37868553400039673, | |
| "learning_rate": 0.00010769834565176498, | |
| "loss": 0.1265, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.3250355618776672, | |
| "grad_norm": 0.23006294667720795, | |
| "learning_rate": 0.0001032722863538738, | |
| "loss": 0.1177, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.3477951635846372, | |
| "grad_norm": 0.26485735177993774, | |
| "learning_rate": 9.892859054550347e-05, | |
| "loss": 0.1212, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.3705547652916075, | |
| "grad_norm": 0.31136396527290344, | |
| "learning_rate": 9.467305403412942e-05, | |
| "loss": 0.1419, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3933143669985775, | |
| "grad_norm": 0.2314603179693222, | |
| "learning_rate": 9.05113549959898e-05, | |
| "loss": 0.0957, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.4160739687055477, | |
| "grad_norm": 0.19551394879817963, | |
| "learning_rate": 8.644904639967639e-05, | |
| "loss": 0.0842, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.438833570412518, | |
| "grad_norm": 0.28223028779029846, | |
| "learning_rate": 8.249154859679033e-05, | |
| "loss": 0.0833, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.461593172119488, | |
| "grad_norm": 0.2497485727071762, | |
| "learning_rate": 7.864414208954971e-05, | |
| "loss": 0.0865, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.484352773826458, | |
| "grad_norm": 0.20309872925281525, | |
| "learning_rate": 7.491196048499769e-05, | |
| "loss": 0.0712, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.5071123755334281, | |
| "grad_norm": 0.33666595816612244, | |
| "learning_rate": 7.12999836452144e-05, | |
| "loss": 0.135, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.5298719772403984, | |
| "grad_norm": 0.24082504212856293, | |
| "learning_rate": 6.781303104267059e-05, | |
| "loss": 0.1203, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.5526315789473686, | |
| "grad_norm": 0.1645207554101944, | |
| "learning_rate": 6.445575532958945e-05, | |
| "loss": 0.0944, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.5753911806543386, | |
| "grad_norm": 0.1779041886329651, | |
| "learning_rate": 6.123263612989815e-05, | |
| "loss": 0.1227, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.5981507823613086, | |
| "grad_norm": 0.21718856692314148, | |
| "learning_rate": 5.81479740620507e-05, | |
| "loss": 0.1008, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6209103840682788, | |
| "grad_norm": 0.23834489285945892, | |
| "learning_rate": 5.520588500069867e-05, | |
| "loss": 0.124, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.643669985775249, | |
| "grad_norm": 0.22460603713989258, | |
| "learning_rate": 5.241029458486649e-05, | |
| "loss": 0.0758, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.666429587482219, | |
| "grad_norm": 0.13075967133045197, | |
| "learning_rate": 4.976493297995823e-05, | |
| "loss": 0.0748, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 0.1683950275182724, | |
| "learning_rate": 4.7273329900585954e-05, | |
| "loss": 0.0806, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.7119487908961593, | |
| "grad_norm": 0.13967812061309814, | |
| "learning_rate": 4.4938809900859955e-05, | |
| "loss": 0.0732, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.7119487908961593, | |
| "eval_loss": 0.09383056312799454, | |
| "eval_runtime": 0.284, | |
| "eval_samples_per_second": 176.043, | |
| "eval_steps_per_second": 45.771, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.7347083926031295, | |
| "grad_norm": 0.1622483730316162, | |
| "learning_rate": 4.2764487938425205e-05, | |
| "loss": 0.0564, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.7574679943100997, | |
| "grad_norm": 0.26513513922691345, | |
| "learning_rate": 4.0753265218163486e-05, | |
| "loss": 0.1552, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.7802275960170697, | |
| "grad_norm": 0.25839948654174805, | |
| "learning_rate": 3.890782532110603e-05, | |
| "loss": 0.1107, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.8029871977240397, | |
| "grad_norm": 0.24427573382854462, | |
| "learning_rate": 3.7230630623722724e-05, | |
| "loss": 0.1011, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.82574679943101, | |
| "grad_norm": 0.22686628997325897, | |
| "learning_rate": 3.572391901236521e-05, | |
| "loss": 0.1001, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.8485064011379801, | |
| "grad_norm": 0.15434470772743225, | |
| "learning_rate": 3.4389700897247595e-05, | |
| "loss": 0.0999, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.8712660028449504, | |
| "grad_norm": 0.15275193750858307, | |
| "learning_rate": 3.322975652994985e-05, | |
| "loss": 0.0832, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.8940256045519204, | |
| "grad_norm": 0.28379786014556885, | |
| "learning_rate": 3.2245633628022074e-05, | |
| "loss": 0.0678, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.9167852062588904, | |
| "grad_norm": 0.11858035624027252, | |
| "learning_rate": 3.143864530986012e-05, | |
| "loss": 0.071, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.9395448079658606, | |
| "grad_norm": 0.1222662478685379, | |
| "learning_rate": 3.08098683426073e-05, | |
| "loss": 0.0855, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.9623044096728308, | |
| "grad_norm": 0.2606646418571472, | |
| "learning_rate": 3.0360141705420527e-05, | |
| "loss": 0.0727, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.9850640113798008, | |
| "grad_norm": 0.21433259546756744, | |
| "learning_rate": 3.009006547001768e-05, | |
| "loss": 0.0841, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.012802275960171, | |
| "grad_norm": 0.21917670965194702, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 0.1182, | |
| "step": 88 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 88, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 1, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.909666424120934e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |