| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 625, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.7108097076416016, | |
| "learning_rate": 4.608e-05, | |
| "loss": 0.3769, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.3901269435882568, | |
| "learning_rate": 4.2080000000000004e-05, | |
| "loss": 0.2247, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.5942708253860474, | |
| "learning_rate": 3.808e-05, | |
| "loss": 0.1805, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.3408385515213013, | |
| "learning_rate": 3.408e-05, | |
| "loss": 0.1612, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.718839168548584, | |
| "learning_rate": 3.0080000000000003e-05, | |
| "loss": 0.1457, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.317421317100525, | |
| "learning_rate": 2.6079999999999998e-05, | |
| "loss": 0.1073, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.8703721761703491, | |
| "learning_rate": 2.2080000000000002e-05, | |
| "loss": 0.1069, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.669245719909668, | |
| "learning_rate": 1.808e-05, | |
| "loss": 0.1054, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 3.690030336380005, | |
| "learning_rate": 1.408e-05, | |
| "loss": 0.0716, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.9665364027023315, | |
| "learning_rate": 1.008e-05, | |
| "loss": 0.0798, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 0.6615438461303711, | |
| "learning_rate": 6.08e-06, | |
| "loss": 0.0549, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 1.9341555833816528, | |
| "learning_rate": 2.08e-06, | |
| "loss": 0.0547, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 625, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 653253703680000.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |