| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.010158914447427617, | |
| "eval_steps": 500, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007256367462448298, | |
| "grad_norm": 12.0, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0014512734924896596, | |
| "grad_norm": 10.4375, | |
| "learning_rate": 0.0001, | |
| "loss": 2.208, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0021769102387344894, | |
| "grad_norm": 6.5625, | |
| "learning_rate": 0.00015, | |
| "loss": 1.3285, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0029025469849793192, | |
| "grad_norm": 6.21875, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6895, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.003628183731224149, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 0.00025, | |
| "loss": 0.714, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.004353820477468979, | |
| "grad_norm": 4.71875, | |
| "learning_rate": 0.0003, | |
| "loss": 0.4849, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005079457223713809, | |
| "grad_norm": 3.828125, | |
| "learning_rate": 0.00035, | |
| "loss": 0.3671, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.0058050939699586385, | |
| "grad_norm": 4.03125, | |
| "learning_rate": 0.0004, | |
| "loss": 0.5693, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.006530730716203468, | |
| "grad_norm": 3.296875, | |
| "learning_rate": 0.00045000000000000004, | |
| "loss": 0.4133, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.007256367462448298, | |
| "grad_norm": 2.640625, | |
| "learning_rate": 0.0005, | |
| "loss": 0.2668, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.007982004208693128, | |
| "grad_norm": 1.5859375, | |
| "learning_rate": 0.000499999340865746, | |
| "loss": 0.2922, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.008707640954937958, | |
| "grad_norm": 3.8125, | |
| "learning_rate": 0.0004999973634664594, | |
| "loss": 0.3996, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.009433277701182788, | |
| "grad_norm": 2.0625, | |
| "learning_rate": 0.0004999940678125673, | |
| "loss": 0.2841, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.010158914447427617, | |
| "grad_norm": 3.390625, | |
| "learning_rate": 0.000499989453921448, | |
| "loss": 0.3003, | |
| "step": 140 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 13781, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "total_flos": 0.0, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |