| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.526315789473684, | |
| "eval_steps": 500, | |
| "global_step": 12, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 7.686098770544664, | |
| "learning_rate": 5e-07, | |
| "loss": 1.4116, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_loss": 1.558640480041504, | |
| "eval_runtime": 5.3808, | |
| "eval_samples_per_second": 11.894, | |
| "eval_steps_per_second": 0.743, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 7.297067521077281, | |
| "learning_rate": 7.938926261462365e-07, | |
| "loss": 1.4074, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.5179415941238403, | |
| "eval_runtime": 5.36, | |
| "eval_samples_per_second": 11.94, | |
| "eval_steps_per_second": 0.746, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 6.501168641091188, | |
| "learning_rate": 9.549150281252632e-08, | |
| "loss": 1.3759, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 1.5165581703186035, | |
| "eval_runtime": 5.5377, | |
| "eval_samples_per_second": 11.557, | |
| "eval_steps_per_second": 0.722, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "step": 12, | |
| "total_flos": 2460210954240.0, | |
| "train_loss": 1.3832024335861206, | |
| "train_runtime": 214.4516, | |
| "train_samples_per_second": 4.127, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 12, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2460210954240.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |