| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.013736263736263736, | |
| "eval_steps": 2, | |
| "global_step": 5, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027472527472527475, | |
| "grad_norm": 9.20065689086914, | |
| "learning_rate": 1e-05, | |
| "loss": 12.4868, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0027472527472527475, | |
| "eval_loss": 12.483174324035645, | |
| "eval_runtime": 41.2422, | |
| "eval_samples_per_second": 14.888, | |
| "eval_steps_per_second": 7.444, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005494505494505495, | |
| "grad_norm": 9.102940559387207, | |
| "learning_rate": 2e-05, | |
| "loss": 12.6001, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.005494505494505495, | |
| "eval_loss": 12.152963638305664, | |
| "eval_runtime": 5.0637, | |
| "eval_samples_per_second": 121.255, | |
| "eval_steps_per_second": 60.627, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008241758241758242, | |
| "grad_norm": 9.035112380981445, | |
| "learning_rate": 3e-05, | |
| "loss": 11.8393, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01098901098901099, | |
| "grad_norm": 9.177264213562012, | |
| "learning_rate": 4e-05, | |
| "loss": 11.8988, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01098901098901099, | |
| "eval_loss": 11.74571704864502, | |
| "eval_runtime": 5.1616, | |
| "eval_samples_per_second": 118.956, | |
| "eval_steps_per_second": 59.478, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013736263736263736, | |
| "grad_norm": 9.4981107711792, | |
| "learning_rate": 5e-05, | |
| "loss": 11.7482, | |
| "step": 5 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 5, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 2, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 649047740252160.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |