| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.7809744348607595, | |
| "learning_rate": 3.9702372106487333e-05, | |
| "loss": 0.4575, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.496724878092226, | |
| "learning_rate": 3.8072567212170113e-05, | |
| "loss": 0.3252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.3560218616246147, | |
| "learning_rate": 3.513425631635391e-05, | |
| "loss": 0.2945, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5682843699995858, | |
| "learning_rate": 3.11001816042694e-05, | |
| "loss": 0.2756, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.7145585075943379, | |
| "learning_rate": 2.6262421722480054e-05, | |
| "loss": 0.2619, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.31576670128072, | |
| "eval_loss": 0.14805860817432404, | |
| "eval_runtime": 576.5922, | |
| "eval_samples_per_second": 20.042, | |
| "eval_steps_per_second": 2.506, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.6547999966414457, | |
| "learning_rate": 2.097124444228897e-05, | |
| "loss": 0.2491, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.3173571306465527, | |
| "learning_rate": 1.560974626265191e-05, | |
| "loss": 0.2389, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.34350674964303235, | |
| "learning_rate": 1.0566115118504683e-05, | |
| "loss": 0.2305, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.36539189432192776, | |
| "learning_rate": 6.2055244544379145e-06, | |
| "loss": 0.2245, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.36512046775231743, | |
| "learning_rate": 2.8436936138645153e-06, | |
| "loss": 0.2206, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4080131533402561, | |
| "eval_loss": 0.13542868196964264, | |
| "eval_runtime": 577.1982, | |
| "eval_samples_per_second": 20.021, | |
| "eval_steps_per_second": 2.503, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.3525044796093815, | |
| "learning_rate": 7.240288477764412e-07, | |
| "loss": 0.2187, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.29607233783759995, | |
| "learning_rate": 0.0, | |
| "loss": 0.2179, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 1200, | |
| "total_flos": 0.0, | |
| "train_loss": 0.2679197629292806, | |
| "train_runtime": 27665.2644, | |
| "train_samples_per_second": 45.109, | |
| "train_steps_per_second": 0.043 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 12, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 26, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |