| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05083088954056696, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0019550342130987292, | |
| "eval_loss": 2.0243232250213623, | |
| "eval_runtime": 13.8701, | |
| "eval_samples_per_second": 15.573, | |
| "eval_steps_per_second": 7.787, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009775171065493646, | |
| "grad_norm": 1.1785598993301392, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6526, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.019550342130987292, | |
| "grad_norm": 1.4754868745803833, | |
| "learning_rate": 0.0001, | |
| "loss": 1.735, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.019550342130987292, | |
| "eval_loss": 1.6014662981033325, | |
| "eval_runtime": 13.4717, | |
| "eval_samples_per_second": 16.034, | |
| "eval_steps_per_second": 8.017, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02932551319648094, | |
| "grad_norm": 0.9464597105979919, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.0844, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.039100684261974585, | |
| "grad_norm": 0.9503406286239624, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7427, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.039100684261974585, | |
| "eval_loss": 0.7665548920631409, | |
| "eval_runtime": 13.6189, | |
| "eval_samples_per_second": 15.86, | |
| "eval_steps_per_second": 7.93, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04887585532746823, | |
| "grad_norm": 1.5756646394729614, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.7232, | |
| "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": 4530853719834624.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |