| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.2, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 0.3297976851463318, | |
| "learning_rate": 0.0002990322580645161, | |
| "loss": 1.0389, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.4069916307926178, | |
| "learning_rate": 0.0002980645161290322, | |
| "loss": 1.3377, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.42084500193595886, | |
| "learning_rate": 0.00029709677419354836, | |
| "loss": 0.9366, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.4641948938369751, | |
| "learning_rate": 0.0002961290322580645, | |
| "loss": 1.0086, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.3840750455856323, | |
| "learning_rate": 0.00029516129032258065, | |
| "loss": 0.8333, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.4263865053653717, | |
| "learning_rate": 0.00029419354838709674, | |
| "loss": 0.854, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 0.48615148663520813, | |
| "learning_rate": 0.0002932258064516129, | |
| "loss": 0.9548, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.44419369101524353, | |
| "learning_rate": 0.00029225806451612903, | |
| "loss": 0.8482, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.5317733883857727, | |
| "learning_rate": 0.0002912903225806451, | |
| "loss": 0.9426, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.47260937094688416, | |
| "learning_rate": 0.00029032258064516127, | |
| "loss": 0.9816, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 0.39063283801078796, | |
| "learning_rate": 0.00028935483870967736, | |
| "loss": 0.84, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.39234670996665955, | |
| "learning_rate": 0.0002883870967741935, | |
| "loss": 0.7476, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 0.40661805868148804, | |
| "learning_rate": 0.00028741935483870965, | |
| "loss": 0.9282, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.42970865964889526, | |
| "learning_rate": 0.0002864516129032258, | |
| "loss": 0.7858, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.3780193626880646, | |
| "learning_rate": 0.00028548387096774194, | |
| "loss": 0.7968, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.37006014585494995, | |
| "learning_rate": 0.00028451612903225803, | |
| "loss": 0.6801, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 0.3660840392112732, | |
| "learning_rate": 0.0002835483870967742, | |
| "loss": 0.5914, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.3270975351333618, | |
| "learning_rate": 0.00028258064516129027, | |
| "loss": 0.6449, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 0.3859024941921234, | |
| "learning_rate": 0.0002816129032258064, | |
| "loss": 0.8144, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.37092071771621704, | |
| "learning_rate": 0.00028064516129032256, | |
| "loss": 0.7667, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 0.37667015194892883, | |
| "learning_rate": 0.0002796774193548387, | |
| "loss": 0.7751, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 0.3832458555698395, | |
| "learning_rate": 0.0002787096774193548, | |
| "loss": 0.755, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 0.327288419008255, | |
| "learning_rate": 0.00027774193548387095, | |
| "loss": 0.7178, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.34552687406539917, | |
| "learning_rate": 0.0002767741935483871, | |
| "loss": 0.7057, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.3611259460449219, | |
| "learning_rate": 0.0002758064516129032, | |
| "loss": 0.8159, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 0.3345054090023041, | |
| "learning_rate": 0.00027483870967741933, | |
| "loss": 0.7208, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.3697254955768585, | |
| "learning_rate": 0.0002738709677419355, | |
| "loss": 0.8964, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.3905017375946045, | |
| "learning_rate": 0.00027290322580645157, | |
| "loss": 0.7794, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 0.3715725243091583, | |
| "learning_rate": 0.0002719354838709677, | |
| "loss": 0.6966, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.3650343120098114, | |
| "learning_rate": 0.00027096774193548386, | |
| "loss": 0.5761, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 0.33932459354400635, | |
| "learning_rate": 0.00027, | |
| "loss": 0.556, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 0.6371742486953735, | |
| "learning_rate": 0.0002690322580645161, | |
| "loss": 0.847, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.056, | |
| "grad_norm": 0.37499895691871643, | |
| "learning_rate": 0.00026806451612903224, | |
| "loss": 0.8419, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.088, | |
| "grad_norm": 0.33221954107284546, | |
| "learning_rate": 0.0002670967741935484, | |
| "loss": 0.6011, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.344096839427948, | |
| "learning_rate": 0.0002661290322580645, | |
| "loss": 0.6501, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 0.38429391384124756, | |
| "learning_rate": 0.0002651612903225806, | |
| "loss": 0.8091, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.184, | |
| "grad_norm": 0.38014867901802063, | |
| "learning_rate": 0.00026419354838709677, | |
| "loss": 0.7668, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.216, | |
| "grad_norm": 0.3352573812007904, | |
| "learning_rate": 0.00026322580645161286, | |
| "loss": 0.5444, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.248, | |
| "grad_norm": 0.33811062574386597, | |
| "learning_rate": 0.000262258064516129, | |
| "loss": 0.512, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.3998416066169739, | |
| "learning_rate": 0.00026129032258064515, | |
| "loss": 0.6315, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.312, | |
| "grad_norm": 0.3983341157436371, | |
| "learning_rate": 0.0002603225806451613, | |
| "loss": 0.5882, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.4585898816585541, | |
| "learning_rate": 0.0002593548387096774, | |
| "loss": 0.761, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.376, | |
| "grad_norm": 0.4080730080604553, | |
| "learning_rate": 0.00025838709677419354, | |
| "loss": 0.6716, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 0.4068273901939392, | |
| "learning_rate": 0.0002574193548387096, | |
| "loss": 0.6376, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.4406949579715729, | |
| "learning_rate": 0.00025645161290322577, | |
| "loss": 0.4594, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.472, | |
| "grad_norm": 0.34500986337661743, | |
| "learning_rate": 0.0002554838709677419, | |
| "loss": 0.3672, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.504, | |
| "grad_norm": 0.4760681390762329, | |
| "learning_rate": 0.00025451612903225806, | |
| "loss": 0.6331, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.39281558990478516, | |
| "learning_rate": 0.0002535483870967742, | |
| "loss": 0.5845, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.568, | |
| "grad_norm": 0.4265002906322479, | |
| "learning_rate": 0.0002525806451612903, | |
| "loss": 0.4461, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.40967294573783875, | |
| "learning_rate": 0.00025161290322580645, | |
| "loss": 0.7011, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6320000000000001, | |
| "grad_norm": 0.4288088381290436, | |
| "learning_rate": 0.00025064516129032254, | |
| "loss": 0.6928, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 0.4356289803981781, | |
| "learning_rate": 0.0002496774193548387, | |
| "loss": 0.7972, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.696, | |
| "grad_norm": 0.3827487826347351, | |
| "learning_rate": 0.0002487096774193548, | |
| "loss": 0.2991, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.40093398094177246, | |
| "learning_rate": 0.0002477419354838709, | |
| "loss": 0.416, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.41548973321914673, | |
| "learning_rate": 0.00024677419354838707, | |
| "loss": 0.5501, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 0.4093388617038727, | |
| "learning_rate": 0.0002458064516129032, | |
| "loss": 0.5557, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.8239999999999998, | |
| "grad_norm": 0.3934040665626526, | |
| "learning_rate": 0.00024483870967741936, | |
| "loss": 0.602, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.8559999999999999, | |
| "grad_norm": 0.42221033573150635, | |
| "learning_rate": 0.00024387096774193545, | |
| "loss": 0.6421, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.888, | |
| "grad_norm": 0.4351339340209961, | |
| "learning_rate": 0.0002429032258064516, | |
| "loss": 0.5615, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.4319838881492615, | |
| "learning_rate": 0.00024193548387096771, | |
| "loss": 0.6804, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.952, | |
| "grad_norm": 0.40016525983810425, | |
| "learning_rate": 0.00024096774193548386, | |
| "loss": 0.5432, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.984, | |
| "grad_norm": 0.3905942440032959, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 0.4187, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.8056382536888123, | |
| "learning_rate": 0.0002390322580645161, | |
| "loss": 1.0174, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 0.3835236430168152, | |
| "learning_rate": 0.00023806451612903224, | |
| "loss": 0.5992, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.41092216968536377, | |
| "learning_rate": 0.00023709677419354836, | |
| "loss": 0.4746, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.39536622166633606, | |
| "learning_rate": 0.0002361290322580645, | |
| "loss": 0.3946, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.144, | |
| "grad_norm": 0.3927665948867798, | |
| "learning_rate": 0.0002351612903225806, | |
| "loss": 0.5187, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 0.39792704582214355, | |
| "learning_rate": 0.00023419354838709674, | |
| "loss": 0.4568, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.5023652911186218, | |
| "learning_rate": 0.0002332258064516129, | |
| "loss": 0.6166, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.425017774105072, | |
| "learning_rate": 0.000232258064516129, | |
| "loss": 0.42, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2720000000000002, | |
| "grad_norm": 0.46458110213279724, | |
| "learning_rate": 0.00023129032258064516, | |
| "loss": 0.4613, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.49037960171699524, | |
| "learning_rate": 0.00023032258064516125, | |
| "loss": 0.5509, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.336, | |
| "grad_norm": 0.5233697891235352, | |
| "learning_rate": 0.0002293548387096774, | |
| "loss": 0.6396, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.368, | |
| "grad_norm": 0.4720582962036133, | |
| "learning_rate": 0.0002283870967741935, | |
| "loss": 0.5076, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.4900650382041931, | |
| "learning_rate": 0.00022741935483870966, | |
| "loss": 0.4794, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 0.6321704983711243, | |
| "learning_rate": 0.0002264516129032258, | |
| "loss": 0.6677, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.464, | |
| "grad_norm": 0.5305324792861938, | |
| "learning_rate": 0.00022548387096774192, | |
| "loss": 0.5102, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.5799248218536377, | |
| "learning_rate": 0.00022451612903225804, | |
| "loss": 0.5274, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.528, | |
| "grad_norm": 0.4990101456642151, | |
| "learning_rate": 0.00022354838709677416, | |
| "loss": 0.5407, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.4779827296733856, | |
| "learning_rate": 0.0002225806451612903, | |
| "loss": 0.5166, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.5140111446380615, | |
| "learning_rate": 0.00022161290322580645, | |
| "loss": 0.3288, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.624, | |
| "grad_norm": 0.5674853920936584, | |
| "learning_rate": 0.00022064516129032257, | |
| "loss": 0.666, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.656, | |
| "grad_norm": 0.5277597308158875, | |
| "learning_rate": 0.00021967741935483871, | |
| "loss": 0.5335, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.6029439568519592, | |
| "learning_rate": 0.0002187096774193548, | |
| "loss": 0.693, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.5039327144622803, | |
| "learning_rate": 0.00021774193548387095, | |
| "loss": 0.5728, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.752, | |
| "grad_norm": 0.5564692616462708, | |
| "learning_rate": 0.00021677419354838707, | |
| "loss": 0.4734, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.5278319120407104, | |
| "learning_rate": 0.00021580645161290322, | |
| "loss": 0.5834, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.816, | |
| "grad_norm": 0.5445135831832886, | |
| "learning_rate": 0.00021483870967741936, | |
| "loss": 0.4642, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.848, | |
| "grad_norm": 0.5394749045372009, | |
| "learning_rate": 0.00021387096774193545, | |
| "loss": 0.4779, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.5756134390830994, | |
| "learning_rate": 0.0002129032258064516, | |
| "loss": 0.5607, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.912, | |
| "grad_norm": 0.48361241817474365, | |
| "learning_rate": 0.00021193548387096772, | |
| "loss": 0.4278, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.944, | |
| "grad_norm": 0.5017121434211731, | |
| "learning_rate": 0.00021096774193548386, | |
| "loss": 0.4834, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.4741989076137543, | |
| "learning_rate": 0.00020999999999999998, | |
| "loss": 0.468, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.008, | |
| "grad_norm": 1.003368854522705, | |
| "learning_rate": 0.0002090322580645161, | |
| "loss": 0.8614, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.4782228469848633, | |
| "learning_rate": 0.00020806451612903225, | |
| "loss": 0.4111, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.072, | |
| "grad_norm": 0.4558674395084381, | |
| "learning_rate": 0.00020709677419354836, | |
| "loss": 0.3463, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.104, | |
| "grad_norm": 0.4409371316432953, | |
| "learning_rate": 0.0002061290322580645, | |
| "loss": 0.2571, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.136, | |
| "grad_norm": 0.5415034890174866, | |
| "learning_rate": 0.00020516129032258063, | |
| "loss": 0.5707, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.168, | |
| "grad_norm": 0.6157724857330322, | |
| "learning_rate": 0.00020419354838709677, | |
| "loss": 0.5692, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.4855688810348511, | |
| "learning_rate": 0.00020322580645161287, | |
| "loss": 0.3311, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 310, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.3700930822144e+16, | |
| "train_batch_size": 3, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |