| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 198, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.010129787907565685, | |
| "grad_norm": 5.634124615502245, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3589, | |
| "mean_token_accuracy": 0.9127573072910309, | |
| "num_tokens": 586686.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.10129787907565685, | |
| "grad_norm": 0.34120253622850233, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2653, | |
| "mean_token_accuracy": 0.9179378294696411, | |
| "num_tokens": 5928997.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2025957581513137, | |
| "grad_norm": 0.15880751179799044, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1607, | |
| "mean_token_accuracy": 0.9390914304181933, | |
| "num_tokens": 11864766.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.30389363722697055, | |
| "grad_norm": 0.10858768982732472, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1389, | |
| "mean_token_accuracy": 0.9470763374119997, | |
| "num_tokens": 17788976.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4051915163026274, | |
| "grad_norm": 0.07843910135917725, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1254, | |
| "mean_token_accuracy": 0.9517961731180549, | |
| "num_tokens": 23762431.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5064893953782843, | |
| "grad_norm": 0.09149076132992681, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1173, | |
| "mean_token_accuracy": 0.9551124922931195, | |
| "num_tokens": 29691182.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6077872744539411, | |
| "grad_norm": 0.0988060674250123, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1126, | |
| "mean_token_accuracy": 0.9569881336763502, | |
| "num_tokens": 35615925.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.709085153529598, | |
| "grad_norm": 0.0747939087236935, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1078, | |
| "mean_token_accuracy": 0.9588578680530191, | |
| "num_tokens": 41567014.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8103830326052548, | |
| "grad_norm": 0.09381859172196332, | |
| "learning_rate": 0.0001, | |
| "loss": 0.107, | |
| "mean_token_accuracy": 0.9592851245775819, | |
| "num_tokens": 47465256.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9116809116809117, | |
| "grad_norm": 0.07959394726227396, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1045, | |
| "mean_token_accuracy": 0.9601302666589617, | |
| "num_tokens": 53394634.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0101297879075657, | |
| "grad_norm": 0.08937759002808963, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1018, | |
| "mean_token_accuracy": 0.9612304681367047, | |
| "num_tokens": 59195734.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1114276669832226, | |
| "grad_norm": 0.07683886672466819, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0887, | |
| "mean_token_accuracy": 0.9659375650808215, | |
| "num_tokens": 65134645.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2127255460588793, | |
| "grad_norm": 0.08142019069177424, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0885, | |
| "mean_token_accuracy": 0.965754240937531, | |
| "num_tokens": 71072294.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3140234251345362, | |
| "grad_norm": 0.07420029680099594, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0892, | |
| "mean_token_accuracy": 0.9654547093436122, | |
| "num_tokens": 77036690.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.415321304210193, | |
| "grad_norm": 0.082199072710958, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0908, | |
| "mean_token_accuracy": 0.9649530470371246, | |
| "num_tokens": 82943314.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.51661918328585, | |
| "grad_norm": 0.07212494518614486, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0906, | |
| "mean_token_accuracy": 0.9651403101161122, | |
| "num_tokens": 88875871.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.617917062361507, | |
| "grad_norm": 0.10352344172757606, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0909, | |
| "mean_token_accuracy": 0.9648977383971215, | |
| "num_tokens": 94820130.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7192149414371638, | |
| "grad_norm": 0.07698503957302266, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0915, | |
| "mean_token_accuracy": 0.9645912747830152, | |
| "num_tokens": 100733124.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8205128205128205, | |
| "grad_norm": 0.08462828819382234, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0911, | |
| "mean_token_accuracy": 0.9649000752717256, | |
| "num_tokens": 106690216.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9218106995884774, | |
| "grad_norm": 0.06837628398080642, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0915, | |
| "mean_token_accuracy": 0.9646391872316599, | |
| "num_tokens": 112618930.0, | |
| "step": 190 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 392, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 309419476910080.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |