| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0379542651105418, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000759085302210836, | |
| "eval_loss": 3.139888286590576, | |
| "eval_runtime": 52.3233, | |
| "eval_samples_per_second": 169.618, | |
| "eval_steps_per_second": 21.214, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007590853022108359, | |
| "grad_norm": 0.27882295846939087, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.8562, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007590853022108359, | |
| "eval_loss": 3.068080425262451, | |
| "eval_runtime": 51.5948, | |
| "eval_samples_per_second": 172.013, | |
| "eval_steps_per_second": 21.514, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015181706044216719, | |
| "grad_norm": 0.32615259289741516, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.7322, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015181706044216719, | |
| "eval_loss": 2.9421498775482178, | |
| "eval_runtime": 52.2462, | |
| "eval_samples_per_second": 169.869, | |
| "eval_steps_per_second": 21.246, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02277255906632508, | |
| "grad_norm": 0.3428688645362854, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 2.6651, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02277255906632508, | |
| "eval_loss": 2.8514456748962402, | |
| "eval_runtime": 52.0319, | |
| "eval_samples_per_second": 170.569, | |
| "eval_steps_per_second": 21.333, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.030363412088433437, | |
| "grad_norm": 0.3088821470737457, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 2.5504, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.030363412088433437, | |
| "eval_loss": 2.807931900024414, | |
| "eval_runtime": 52.0481, | |
| "eval_samples_per_second": 170.515, | |
| "eval_steps_per_second": 21.326, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0379542651105418, | |
| "grad_norm": 0.34101369976997375, | |
| "learning_rate": 0.0, | |
| "loss": 2.5387, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0379542651105418, | |
| "eval_loss": 2.800560474395752, | |
| "eval_runtime": 51.8735, | |
| "eval_samples_per_second": 171.089, | |
| "eval_steps_per_second": 21.398, | |
| "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": 3806484202782720.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |