| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0004938068392247233, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.23011398707872e-05, | |
| "grad_norm": 0.7503308057785034, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5917, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.23011398707872e-05, | |
| "eval_loss": 1.959477186203003, | |
| "eval_runtime": 133.9988, | |
| "eval_samples_per_second": 19.09, | |
| "eval_steps_per_second": 19.09, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001646022797415744, | |
| "grad_norm": 0.8759475350379944, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8288, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00024690341961236165, | |
| "grad_norm": 0.6373544931411743, | |
| "learning_rate": 6e-05, | |
| "loss": 1.0265, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00024690341961236165, | |
| "eval_loss": 1.9565452337265015, | |
| "eval_runtime": 160.8486, | |
| "eval_samples_per_second": 15.903, | |
| "eval_steps_per_second": 15.903, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0003292045594831488, | |
| "grad_norm": 0.9935461282730103, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6082, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00041150569935393603, | |
| "grad_norm": 0.8985377550125122, | |
| "learning_rate": 0.0001, | |
| "loss": 1.332, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0004938068392247233, | |
| "grad_norm": 0.804458498954773, | |
| "learning_rate": 0.00012, | |
| "loss": 1.2777, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0004938068392247233, | |
| "eval_loss": 1.9385048151016235, | |
| "eval_runtime": 172.7761, | |
| "eval_samples_per_second": 14.805, | |
| "eval_steps_per_second": 14.805, | |
| "step": 6 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 144327309262848.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |