| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.076923076923077, | |
| "eval_steps": 1, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3076923076923077, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.7104, | |
| "eval_samples_per_second": 8.446, | |
| "eval_steps_per_second": 1.408, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1686, | |
| "eval_samples_per_second": 35.585, | |
| "eval_steps_per_second": 5.931, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1721, | |
| "eval_samples_per_second": 34.857, | |
| "eval_steps_per_second": 5.809, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1687, | |
| "eval_samples_per_second": 35.562, | |
| "eval_steps_per_second": 5.927, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.17, | |
| "eval_samples_per_second": 35.288, | |
| "eval_steps_per_second": 5.881, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1715, | |
| "eval_samples_per_second": 34.989, | |
| "eval_steps_per_second": 5.831, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1688, | |
| "eval_samples_per_second": 35.549, | |
| "eval_steps_per_second": 5.925, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1729, | |
| "eval_samples_per_second": 34.694, | |
| "eval_steps_per_second": 5.782, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1715, | |
| "eval_samples_per_second": 34.979, | |
| "eval_steps_per_second": 5.83, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1688, | |
| "eval_samples_per_second": 35.538, | |
| "eval_steps_per_second": 5.923, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2593879464345600.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |