| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.002240751513584556, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.618275052248293e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 180.462, | |
| "eval_samples_per_second": 27.075, | |
| "eval_steps_per_second": 13.537, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004309137526124146, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0008618275052248292, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0008618275052248292, | |
| "eval_loss": NaN, | |
| "eval_runtime": 181.9637, | |
| "eval_samples_per_second": 26.852, | |
| "eval_steps_per_second": 13.426, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0012927412578372438, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0017236550104496584, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0017236550104496584, | |
| "eval_loss": NaN, | |
| "eval_runtime": 183.07, | |
| "eval_samples_per_second": 26.689, | |
| "eval_steps_per_second": 13.345, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002154568763062073, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 981002232004608.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |