| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9775689803480447, | |
| "global_step": 45000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 8e-05, | |
| "loss": 1.5093, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.91079194451259e-05, | |
| "loss": 1.2959, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 7.82158388902518e-05, | |
| "loss": 1.2604, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 7.732375833537769e-05, | |
| "loss": 1.2426, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 7.643167778050358e-05, | |
| "loss": 1.2368, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7.553959722562947e-05, | |
| "loss": 1.2252, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.464751667075538e-05, | |
| "loss": 1.2204, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.375543611588127e-05, | |
| "loss": 1.2146, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.286335556100716e-05, | |
| "loss": 1.2081, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.197127500613307e-05, | |
| "loss": 1.2069, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.107919445125896e-05, | |
| "loss": 1.2017, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 7.018711389638485e-05, | |
| "loss": 1.2015, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 6.929503334151074e-05, | |
| "loss": 1.1906, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 6.840295278663664e-05, | |
| "loss": 1.1962, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6.751087223176253e-05, | |
| "loss": 1.1939, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 6.661879167688842e-05, | |
| "loss": 1.1844, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 6.572671112201433e-05, | |
| "loss": 1.1923, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 6.483463056714022e-05, | |
| "loss": 1.1864, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 6.394255001226611e-05, | |
| "loss": 1.1801, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 6.305046945739202e-05, | |
| "loss": 1.1832, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.5757644176483154, | |
| "eval_loss": 1.1787368059158325, | |
| "eval_runtime": 214.7637, | |
| "eval_samples_per_second": 4547.505, | |
| "eval_steps_per_second": 15.161, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 6.215838890251789e-05, | |
| "loss": 1.1804, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 6.12663083476438e-05, | |
| "loss": 1.1813, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 6.037422779276969e-05, | |
| "loss": 1.1795, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 5.9482147237895586e-05, | |
| "loss": 1.178, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 5.8590066683021484e-05, | |
| "loss": 1.1815, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 5.769798612814738e-05, | |
| "loss": 1.1745, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 5.680590557327327e-05, | |
| "loss": 1.1744, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.591382501839917e-05, | |
| "loss": 1.1745, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 5.5021744463525054e-05, | |
| "loss": 1.1712, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 5.412966390865095e-05, | |
| "loss": 1.1686, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.323758335377685e-05, | |
| "loss": 1.1572, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.234550279890275e-05, | |
| "loss": 1.1503, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.145342224402864e-05, | |
| "loss": 1.1519, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.0561341689154536e-05, | |
| "loss": 1.1502, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.9669261134280434e-05, | |
| "loss": 1.1465, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.877718057940632e-05, | |
| "loss": 1.1534, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.7885100024532215e-05, | |
| "loss": 1.1505, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 4.699301946965811e-05, | |
| "loss": 1.1492, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.610093891478401e-05, | |
| "loss": 1.1518, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.52088583599099e-05, | |
| "loss": 1.1492, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_accuracy": 0.5826605558395386, | |
| "eval_loss": 1.1605184078216553, | |
| "eval_runtime": 207.3392, | |
| "eval_samples_per_second": 4710.344, | |
| "eval_steps_per_second": 15.704, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 4.43167778050358e-05, | |
| "loss": 1.1465, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.34246972501617e-05, | |
| "loss": 1.1472, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 4.2532616695287595e-05, | |
| "loss": 1.1507, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 4.164053614041348e-05, | |
| "loss": 1.1466, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.074845558553938e-05, | |
| "loss": 1.1479, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 3.985637503066527e-05, | |
| "loss": 1.1425, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 3.8964294475791165e-05, | |
| "loss": 1.1497, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 3.807221392091706e-05, | |
| "loss": 1.1394, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 3.718013336604296e-05, | |
| "loss": 1.1472, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.628805281116885e-05, | |
| "loss": 1.1444, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.539597225629475e-05, | |
| "loss": 1.1439, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 3.450389170142064e-05, | |
| "loss": 1.1468, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 3.361181114654654e-05, | |
| "loss": 1.1467, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.271973059167243e-05, | |
| "loss": 1.1417, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 3.182765003679833e-05, | |
| "loss": 1.1423, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 3.0935569481924224e-05, | |
| "loss": 1.1444, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.0043488927050112e-05, | |
| "loss": 1.1382, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 2.915140837217601e-05, | |
| "loss": 1.1392, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 2.8259327817301904e-05, | |
| "loss": 1.1407, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 2.73672472624278e-05, | |
| "loss": 1.143, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.5862544775009155, | |
| "eval_loss": 1.1498275995254517, | |
| "eval_runtime": 205.0061, | |
| "eval_samples_per_second": 4763.951, | |
| "eval_steps_per_second": 15.882, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 2.6475166707553693e-05, | |
| "loss": 1.1246, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 2.5583086152679587e-05, | |
| "loss": 1.1176, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.4691005597805485e-05, | |
| "loss": 1.1131, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.379892504293138e-05, | |
| "loss": 1.1106, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.2906844488057273e-05, | |
| "loss": 1.1104, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 2.2014763933183168e-05, | |
| "loss": 1.1127, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 2.1122683378309062e-05, | |
| "loss": 1.1113, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 2.023060282343496e-05, | |
| "loss": 1.1151, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1.9338522268560854e-05, | |
| "loss": 1.1177, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1.8446441713686748e-05, | |
| "loss": 1.113, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 1.7554361158812642e-05, | |
| "loss": 1.11, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.6662280603938537e-05, | |
| "loss": 1.1172, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1.577020004906443e-05, | |
| "loss": 1.1127, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.4878119494190325e-05, | |
| "loss": 1.1116, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 1.3986038939316221e-05, | |
| "loss": 1.1093, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 1.3093958384442116e-05, | |
| "loss": 1.1152, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.2201877829568012e-05, | |
| "loss": 1.1084, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 1.1309797274693906e-05, | |
| "loss": 1.1095, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.0417716719819802e-05, | |
| "loss": 1.1071, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 9.525636164945695e-06, | |
| "loss": 1.1088, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "eval_accuracy": 0.5878129005432129, | |
| "eval_loss": 1.149354338645935, | |
| "eval_runtime": 211.165, | |
| "eval_samples_per_second": 4625.004, | |
| "eval_steps_per_second": 15.419, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 8.63355561007159e-06, | |
| "loss": 1.1103, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 7.741475055197485e-06, | |
| "loss": 1.113, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 6.849394500323379e-06, | |
| "loss": 1.108, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 5.9573139454492745e-06, | |
| "loss": 1.1082, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 5.06523339057517e-06, | |
| "loss": 1.1121, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.173152835701064e-06, | |
| "loss": 1.1051, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 3.281072280826959e-06, | |
| "loss": 1.1057, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 2.3889917259528535e-06, | |
| "loss": 1.1155, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.4969111710787484e-06, | |
| "loss": 1.1102, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 6.048306162046434e-07, | |
| "loss": 1.1091, | |
| "step": 45000 | |
| } | |
| ], | |
| "max_steps": 45339, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.1008030583503237e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |