| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00014433138485963774, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6036820539959746e-05, | |
| "grad_norm": 1.1646537780761719, | |
| "learning_rate": 2e-05, | |
| "loss": 5.8615, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.6036820539959746e-05, | |
| "eval_loss": 5.424288749694824, | |
| "eval_runtime": 228.5715, | |
| "eval_samples_per_second": 114.87, | |
| "eval_steps_per_second": 57.435, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.207364107991949e-05, | |
| "grad_norm": 1.042622685432434, | |
| "learning_rate": 4e-05, | |
| "loss": 5.3043, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 4.811046161987924e-05, | |
| "grad_norm": 1.0880378484725952, | |
| "learning_rate": 6e-05, | |
| "loss": 5.5592, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.811046161987924e-05, | |
| "eval_loss": 5.421998977661133, | |
| "eval_runtime": 236.3537, | |
| "eval_samples_per_second": 111.088, | |
| "eval_steps_per_second": 55.544, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 6.414728215983899e-05, | |
| "grad_norm": 1.2153558731079102, | |
| "learning_rate": 8e-05, | |
| "loss": 5.6178, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 8.018410269979874e-05, | |
| "grad_norm": 0.9212684631347656, | |
| "learning_rate": 0.0001, | |
| "loss": 4.9863, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 9.622092323975848e-05, | |
| "grad_norm": 1.0262424945831299, | |
| "learning_rate": 0.00012, | |
| "loss": 5.5765, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 9.622092323975848e-05, | |
| "eval_loss": 5.394266605377197, | |
| "eval_runtime": 225.3453, | |
| "eval_samples_per_second": 116.515, | |
| "eval_steps_per_second": 58.257, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00011225774377971823, | |
| "grad_norm": 1.0572900772094727, | |
| "learning_rate": 0.00014, | |
| "loss": 5.1598, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00012829456431967797, | |
| "grad_norm": 1.0930063724517822, | |
| "learning_rate": 0.00016, | |
| "loss": 5.2336, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00014433138485963774, | |
| "grad_norm": 1.3030791282653809, | |
| "learning_rate": 0.00018, | |
| "loss": 5.5289, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00014433138485963774, | |
| "eval_loss": 5.312864780426025, | |
| "eval_runtime": 221.8218, | |
| "eval_samples_per_second": 118.365, | |
| "eval_steps_per_second": 59.183, | |
| "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": 30863385427968.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |