| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005421278518183872, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 6.02364279798208e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.02364279798208e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 141.3767, | |
| "eval_samples_per_second": 49.442, | |
| "eval_steps_per_second": 24.721, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001204728559596416, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001807092839394624, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 4.1641, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0001807092839394624, | |
| "eval_loss": NaN, | |
| "eval_runtime": 140.7466, | |
| "eval_samples_per_second": 49.664, | |
| "eval_steps_per_second": 24.832, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002409457119192832, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00030118213989910396, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0003614185678789248, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 6.308, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003614185678789248, | |
| "eval_loss": NaN, | |
| "eval_runtime": 140.5913, | |
| "eval_samples_per_second": 49.719, | |
| "eval_steps_per_second": 24.859, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00042165499585874555, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0004818914238385664, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0005421278518183872, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 3.1081, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005421278518183872, | |
| "eval_loss": NaN, | |
| "eval_runtime": 139.9544, | |
| "eval_samples_per_second": 49.945, | |
| "eval_steps_per_second": 24.972, | |
| "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": 507541154955264.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |