| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06749915626054674, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006749915626054674, | |
| "eval_loss": 2.0244810581207275, | |
| "eval_runtime": 109.4925, | |
| "eval_samples_per_second": 22.787, | |
| "eval_steps_per_second": 2.85, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0020249746878164025, | |
| "grad_norm": 1.2031842470169067, | |
| "learning_rate": 3e-05, | |
| "loss": 2.002, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004049949375632805, | |
| "grad_norm": 1.3379194736480713, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0061, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.006074924063449207, | |
| "grad_norm": 1.1463463306427002, | |
| "learning_rate": 9e-05, | |
| "loss": 1.7661, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006074924063449207, | |
| "eval_loss": 1.702438473701477, | |
| "eval_runtime": 110.6716, | |
| "eval_samples_per_second": 22.544, | |
| "eval_steps_per_second": 2.819, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00809989875126561, | |
| "grad_norm": 1.4622615575790405, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.6239, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010124873439082012, | |
| "grad_norm": 1.6571316719055176, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.3919, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012149848126898413, | |
| "grad_norm": 0.7293809056282043, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.1924, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012149848126898413, | |
| "eval_loss": 1.1869244575500488, | |
| "eval_runtime": 110.6832, | |
| "eval_samples_per_second": 22.542, | |
| "eval_steps_per_second": 2.819, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014174822814714817, | |
| "grad_norm": 0.7512520551681519, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.1928, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.01619979750253122, | |
| "grad_norm": 0.5832200646400452, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.1469, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01822477219034762, | |
| "grad_norm": 0.5950210690498352, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.0966, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01822477219034762, | |
| "eval_loss": 1.077985167503357, | |
| "eval_runtime": 110.6972, | |
| "eval_samples_per_second": 22.539, | |
| "eval_steps_per_second": 2.818, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.020249746878164023, | |
| "grad_norm": 0.591693639755249, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.0773, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.022274721565980427, | |
| "grad_norm": 0.5439223051071167, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.015, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.024299696253796826, | |
| "grad_norm": 0.7373484969139099, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.0284, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.024299696253796826, | |
| "eval_loss": 1.0242911577224731, | |
| "eval_runtime": 110.7565, | |
| "eval_samples_per_second": 22.527, | |
| "eval_steps_per_second": 2.817, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02632467094161323, | |
| "grad_norm": 0.5430842638015747, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.0203, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.028349645629429633, | |
| "grad_norm": 0.5873537063598633, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.9888, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.030374620317246033, | |
| "grad_norm": 0.5597456693649292, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.9699, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.030374620317246033, | |
| "eval_loss": 0.9986519813537598, | |
| "eval_runtime": 110.809, | |
| "eval_samples_per_second": 22.516, | |
| "eval_steps_per_second": 2.816, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03239959500506244, | |
| "grad_norm": 0.5352265238761902, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.9991, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03442456969287884, | |
| "grad_norm": 0.5468347668647766, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.9773, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.03644954438069524, | |
| "grad_norm": 0.5996096730232239, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.9651, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03644954438069524, | |
| "eval_loss": 0.9821661114692688, | |
| "eval_runtime": 110.6058, | |
| "eval_samples_per_second": 22.558, | |
| "eval_steps_per_second": 2.821, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.038474519068511646, | |
| "grad_norm": 0.5582756996154785, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.0048, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.040499493756328046, | |
| "grad_norm": 0.6157979369163513, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.9963, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.042524468444144446, | |
| "grad_norm": 0.5799768567085266, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.9534, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.042524468444144446, | |
| "eval_loss": 0.9695780277252197, | |
| "eval_runtime": 110.7486, | |
| "eval_samples_per_second": 22.529, | |
| "eval_steps_per_second": 2.817, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04454944313196085, | |
| "grad_norm": 0.5786073803901672, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.9286, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.04657441781977725, | |
| "grad_norm": 0.5951570868492126, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.0054, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.04859939250759365, | |
| "grad_norm": 0.61005699634552, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.9484, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04859939250759365, | |
| "eval_loss": 0.9628006219863892, | |
| "eval_runtime": 110.7184, | |
| "eval_samples_per_second": 22.535, | |
| "eval_steps_per_second": 2.818, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.05062436719541006, | |
| "grad_norm": 0.6022469997406006, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.9346, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05264934188322646, | |
| "grad_norm": 0.5777412056922913, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.9572, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.05467431657104286, | |
| "grad_norm": 0.5540062785148621, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.9708, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05467431657104286, | |
| "eval_loss": 0.9590880274772644, | |
| "eval_runtime": 110.6871, | |
| "eval_samples_per_second": 22.541, | |
| "eval_steps_per_second": 2.819, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.056699291258859266, | |
| "grad_norm": 0.5456353425979614, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.9559, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.058724265946675666, | |
| "grad_norm": 0.5648323893547058, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.9798, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.060749240634492066, | |
| "grad_norm": 0.562163770198822, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.9847, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.060749240634492066, | |
| "eval_loss": 0.9570688009262085, | |
| "eval_runtime": 110.7882, | |
| "eval_samples_per_second": 22.52, | |
| "eval_steps_per_second": 2.816, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06277421532230847, | |
| "grad_norm": 0.6110678911209106, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.9614, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.06479919001012488, | |
| "grad_norm": 0.5960271954536438, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.9421, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.06682416469794128, | |
| "grad_norm": 0.5943837761878967, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.9477, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.06682416469794128, | |
| "eval_loss": 0.956795334815979, | |
| "eval_runtime": 110.7795, | |
| "eval_samples_per_second": 22.522, | |
| "eval_steps_per_second": 2.816, | |
| "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": 5.51427898343424e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |