| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00019259061120770361, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.139895680085596e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.139895680085596e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1038.7907, | |
| "eval_samples_per_second": 18.942, | |
| "eval_steps_per_second": 9.472, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.279791360171192e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 6.419687040256787e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 6.419687040256787e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1037.6203, | |
| "eval_samples_per_second": 18.964, | |
| "eval_steps_per_second": 9.482, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 8.559582720342384e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00010699478400427979, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00012839374080513574, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00012839374080513574, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1037.9167, | |
| "eval_samples_per_second": 18.958, | |
| "eval_steps_per_second": 9.48, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0001497926976059917, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00017119165440684768, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00019259061120770361, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00019259061120770361, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1038.3217, | |
| "eval_samples_per_second": 18.951, | |
| "eval_steps_per_second": 9.476, | |
| "step": 9 | |
| } | |
| ], | |
| "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": 2914875063926784.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |