| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7476635514018691, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.014953271028037384, | |
| "eval_loss": 3.3571724891662598, | |
| "eval_runtime": 28.1099, | |
| "eval_samples_per_second": 16.044, | |
| "eval_steps_per_second": 2.028, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14953271028037382, | |
| "grad_norm": 2.5893006324768066, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.7703, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14953271028037382, | |
| "eval_loss": 2.0607998371124268, | |
| "eval_runtime": 28.8738, | |
| "eval_samples_per_second": 15.62, | |
| "eval_steps_per_second": 1.974, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.29906542056074764, | |
| "grad_norm": 1.9176541566848755, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.8657, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.29906542056074764, | |
| "eval_loss": 1.6928550004959106, | |
| "eval_runtime": 28.7448, | |
| "eval_samples_per_second": 15.69, | |
| "eval_steps_per_second": 1.983, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4485981308411215, | |
| "grad_norm": 1.3622472286224365, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.6464, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4485981308411215, | |
| "eval_loss": 1.5652191638946533, | |
| "eval_runtime": 28.9251, | |
| "eval_samples_per_second": 15.592, | |
| "eval_steps_per_second": 1.971, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5981308411214953, | |
| "grad_norm": 1.2955882549285889, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.5296, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5981308411214953, | |
| "eval_loss": 1.519897222518921, | |
| "eval_runtime": 28.852, | |
| "eval_samples_per_second": 15.631, | |
| "eval_steps_per_second": 1.976, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7476635514018691, | |
| "grad_norm": 1.2415874004364014, | |
| "learning_rate": 0.0, | |
| "loss": 1.549, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7476635514018691, | |
| "eval_loss": 1.514955759048462, | |
| "eval_runtime": 28.7576, | |
| "eval_samples_per_second": 15.683, | |
| "eval_steps_per_second": 1.982, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.52887313301504e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |