| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.000930135218407376, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.65067609203688e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.65067609203688e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2778.97, | |
| "eval_samples_per_second": 13.032, | |
| "eval_steps_per_second": 6.516, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.30135218407376e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001395202827611064, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0001860270436814752, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.000232533804601844, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.000232533804601844, | |
| "eval_loss": NaN, | |
| "eval_runtime": 3014.8915, | |
| "eval_samples_per_second": 12.012, | |
| "eval_steps_per_second": 6.006, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0002790405655222128, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 52.1349, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003255473264425816, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003720540873629504, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 7.8335, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0004185608482833192, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.000465067609203688, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.000465067609203688, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2859.22, | |
| "eval_samples_per_second": 12.666, | |
| "eval_steps_per_second": 6.333, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005115743701240568, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0005580811310444256, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006045878919647944, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 18.9963, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0006510946528851632, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.000697601413805532, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0366, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.000697601413805532, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1744.4969, | |
| "eval_samples_per_second": 20.76, | |
| "eval_steps_per_second": 10.38, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007441081747259008, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0007906149356462696, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0253, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0008371216965666384, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0008836284574870072, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 11.7524, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.000930135218407376, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000930135218407376, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1964.7488, | |
| "eval_samples_per_second": 18.432, | |
| "eval_steps_per_second": 9.216, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.422742037037056e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |