| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.01633186346562143, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006532745386248571, | |
| "grad_norm": 0.5963796377182007, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1508, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006532745386248571, | |
| "eval_loss": 1.2967846393585205, | |
| "eval_runtime": 400.7493, | |
| "eval_samples_per_second": 6.433, | |
| "eval_steps_per_second": 1.609, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013065490772497142, | |
| "grad_norm": 0.6392633318901062, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2868, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0019598236158745713, | |
| "grad_norm": 0.6073038578033447, | |
| "learning_rate": 9.953429730181653e-05, | |
| "loss": 1.1726, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0019598236158745713, | |
| "eval_loss": 1.2707507610321045, | |
| "eval_runtime": 402.7641, | |
| "eval_samples_per_second": 6.401, | |
| "eval_steps_per_second": 1.601, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0026130981544994283, | |
| "grad_norm": 0.6297516822814941, | |
| "learning_rate": 9.814586436738998e-05, | |
| "loss": 1.2031, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0032663726931242854, | |
| "grad_norm": 0.5256909728050232, | |
| "learning_rate": 9.586056507527266e-05, | |
| "loss": 1.1517, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0039196472317491425, | |
| "grad_norm": 0.5755878686904907, | |
| "learning_rate": 9.272097022732443e-05, | |
| "loss": 1.328, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0039196472317491425, | |
| "eval_loss": 1.2350541353225708, | |
| "eval_runtime": 402.0728, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 1.604, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004572921770374, | |
| "grad_norm": 0.5528045892715454, | |
| "learning_rate": 8.8785564535221e-05, | |
| "loss": 1.2442, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005226196308998857, | |
| "grad_norm": 0.5592995882034302, | |
| "learning_rate": 8.412765716093272e-05, | |
| "loss": 1.1386, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005879470847623714, | |
| "grad_norm": 0.626818060874939, | |
| "learning_rate": 7.883401610574336e-05, | |
| "loss": 1.3114, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005879470847623714, | |
| "eval_loss": 1.2249033451080322, | |
| "eval_runtime": 402.0303, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 1.604, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006532745386248571, | |
| "grad_norm": 0.5507422089576721, | |
| "learning_rate": 7.300325188655761e-05, | |
| "loss": 1.1543, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007186019924873428, | |
| "grad_norm": 0.6158396005630493, | |
| "learning_rate": 6.674398060854931e-05, | |
| "loss": 1.2007, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007839294463498285, | |
| "grad_norm": 0.582711935043335, | |
| "learning_rate": 6.01728006526317e-05, | |
| "loss": 1.2875, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.007839294463498285, | |
| "eval_loss": 1.2190592288970947, | |
| "eval_runtime": 401.9323, | |
| "eval_samples_per_second": 6.414, | |
| "eval_steps_per_second": 1.605, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008492569002123142, | |
| "grad_norm": 0.7021968960762024, | |
| "learning_rate": 5.341212066823355e-05, | |
| "loss": 1.3059, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009145843540748, | |
| "grad_norm": 0.5167236328125, | |
| "learning_rate": 4.658787933176646e-05, | |
| "loss": 1.1782, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009799118079372856, | |
| "grad_norm": 0.5842359662055969, | |
| "learning_rate": 3.982719934736832e-05, | |
| "loss": 1.2259, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009799118079372856, | |
| "eval_loss": 1.2140125036239624, | |
| "eval_runtime": 402.6801, | |
| "eval_samples_per_second": 6.402, | |
| "eval_steps_per_second": 1.602, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010452392617997713, | |
| "grad_norm": 0.6493592858314514, | |
| "learning_rate": 3.325601939145069e-05, | |
| "loss": 1.1797, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01110566715662257, | |
| "grad_norm": 0.5524556636810303, | |
| "learning_rate": 2.6996748113442394e-05, | |
| "loss": 1.2013, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011758941695247428, | |
| "grad_norm": 0.6769681572914124, | |
| "learning_rate": 2.1165983894256647e-05, | |
| "loss": 1.2394, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011758941695247428, | |
| "eval_loss": 1.2114737033843994, | |
| "eval_runtime": 403.2749, | |
| "eval_samples_per_second": 6.393, | |
| "eval_steps_per_second": 1.599, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012412216233872285, | |
| "grad_norm": 0.5104557871818542, | |
| "learning_rate": 1.5872342839067306e-05, | |
| "loss": 1.2101, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.013065490772497142, | |
| "grad_norm": 0.4540850818157196, | |
| "learning_rate": 1.1214435464779006e-05, | |
| "loss": 1.171, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013718765311121999, | |
| "grad_norm": 0.483037531375885, | |
| "learning_rate": 7.2790297726755716e-06, | |
| "loss": 1.1566, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.013718765311121999, | |
| "eval_loss": 1.2103384733200073, | |
| "eval_runtime": 402.9848, | |
| "eval_samples_per_second": 6.397, | |
| "eval_steps_per_second": 1.601, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014372039849746856, | |
| "grad_norm": 0.5860415697097778, | |
| "learning_rate": 4.139434924727359e-06, | |
| "loss": 1.2214, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.015025314388371713, | |
| "grad_norm": 0.5485131144523621, | |
| "learning_rate": 1.8541356326100433e-06, | |
| "loss": 1.1605, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.01567858892699657, | |
| "grad_norm": 0.45531585812568665, | |
| "learning_rate": 4.6570269818346224e-07, | |
| "loss": 1.1747, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01567858892699657, | |
| "eval_loss": 1.2099705934524536, | |
| "eval_runtime": 402.0889, | |
| "eval_samples_per_second": 6.412, | |
| "eval_steps_per_second": 1.604, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01633186346562143, | |
| "grad_norm": 0.5713936686515808, | |
| "learning_rate": 0.0, | |
| "loss": 1.2352, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.13314785247232e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |