| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3436426116838488, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003436426116838488, | |
| "eval_loss": 1.0753878355026245, | |
| "eval_runtime": 70.8712, | |
| "eval_samples_per_second": 6.914, | |
| "eval_steps_per_second": 0.875, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010309278350515464, | |
| "grad_norm": 9.508790016174316, | |
| "learning_rate": 3e-05, | |
| "loss": 4.2728, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.020618556701030927, | |
| "grad_norm": 4.23081111907959, | |
| "learning_rate": 6e-05, | |
| "loss": 4.2308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "grad_norm": 2.994974136352539, | |
| "learning_rate": 9e-05, | |
| "loss": 4.0505, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.030927835051546393, | |
| "eval_loss": 0.9270894527435303, | |
| "eval_runtime": 72.0129, | |
| "eval_samples_per_second": 6.804, | |
| "eval_steps_per_second": 0.861, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041237113402061855, | |
| "grad_norm": 3.4861974716186523, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 3.3622, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.05154639175257732, | |
| "grad_norm": 2.4739320278167725, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 3.4222, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "grad_norm": 2.656764507293701, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 3.3192, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.061855670103092786, | |
| "eval_loss": 0.781949520111084, | |
| "eval_runtime": 72.0068, | |
| "eval_samples_per_second": 6.805, | |
| "eval_steps_per_second": 0.861, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07216494845360824, | |
| "grad_norm": 2.133312702178955, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 2.9592, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08247422680412371, | |
| "grad_norm": 2.0654103755950928, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 3.2154, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "grad_norm": 1.8565651178359985, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 2.9275, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09278350515463918, | |
| "eval_loss": 0.7440634965896606, | |
| "eval_runtime": 72.0175, | |
| "eval_samples_per_second": 6.804, | |
| "eval_steps_per_second": 0.861, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10309278350515463, | |
| "grad_norm": 2.0615859031677246, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 2.9618, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1134020618556701, | |
| "grad_norm": 2.0000863075256348, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 3.0231, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "grad_norm": 1.8480544090270996, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 2.8036, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12371134020618557, | |
| "eval_loss": 0.727074146270752, | |
| "eval_runtime": 72.0101, | |
| "eval_samples_per_second": 6.805, | |
| "eval_steps_per_second": 0.861, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13402061855670103, | |
| "grad_norm": 2.198150634765625, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 2.795, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14432989690721648, | |
| "grad_norm": 1.8142603635787964, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 2.7169, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "grad_norm": 1.9104666709899902, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 2.8454, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.15463917525773196, | |
| "eval_loss": 0.720156729221344, | |
| "eval_runtime": 72.0724, | |
| "eval_samples_per_second": 6.799, | |
| "eval_steps_per_second": 0.86, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 3.0694377422332764, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 3.0856, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.17525773195876287, | |
| "grad_norm": 1.9787322282791138, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 2.83, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "grad_norm": 1.9755208492279053, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 2.765, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.18556701030927836, | |
| "eval_loss": 0.7139042615890503, | |
| "eval_runtime": 72.0348, | |
| "eval_samples_per_second": 6.802, | |
| "eval_steps_per_second": 0.861, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.1958762886597938, | |
| "grad_norm": 1.9179588556289673, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 2.9697, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.20618556701030927, | |
| "grad_norm": 2.107278823852539, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.0271, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "grad_norm": 1.985304832458496, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 2.799, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.21649484536082475, | |
| "eval_loss": 0.7117459177970886, | |
| "eval_runtime": 71.9674, | |
| "eval_samples_per_second": 6.809, | |
| "eval_steps_per_second": 0.862, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2268041237113402, | |
| "grad_norm": 2.0945565700531006, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 2.8955, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.23711340206185566, | |
| "grad_norm": 2.2720227241516113, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 2.9467, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "grad_norm": 1.7193487882614136, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 2.9671, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.24742268041237114, | |
| "eval_loss": 0.7079923748970032, | |
| "eval_runtime": 72.1273, | |
| "eval_samples_per_second": 6.794, | |
| "eval_steps_per_second": 0.86, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.25773195876288657, | |
| "grad_norm": 1.9435456991195679, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 3.1961, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.26804123711340205, | |
| "grad_norm": 3.1002867221832275, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 3.0302, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "grad_norm": 1.6273797750473022, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 2.8564, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.27835051546391754, | |
| "eval_loss": 0.7072886228561401, | |
| "eval_runtime": 72.0658, | |
| "eval_samples_per_second": 6.799, | |
| "eval_steps_per_second": 0.86, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.28865979381443296, | |
| "grad_norm": 1.9313831329345703, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 2.6863, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.29896907216494845, | |
| "grad_norm": 1.7214744091033936, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 2.9589, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "grad_norm": 1.789853811264038, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 3.0606, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.30927835051546393, | |
| "eval_loss": 0.7060263752937317, | |
| "eval_runtime": 72.1222, | |
| "eval_samples_per_second": 6.794, | |
| "eval_steps_per_second": 0.86, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.31958762886597936, | |
| "grad_norm": 1.5414116382598877, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 2.7664, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 1.7637780904769897, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 2.8141, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "grad_norm": 1.83016836643219, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 2.8253, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3402061855670103, | |
| "eval_loss": 0.7058959603309631, | |
| "eval_runtime": 71.8781, | |
| "eval_samples_per_second": 6.817, | |
| "eval_steps_per_second": 0.863, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.5224119259693056e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |