| { |
| "best_global_step": 50, |
| "best_metric": 0.458571195602417, |
| "best_model_checkpoint": "rewrite_plan_results/checkpoint-50", |
| "epoch": 4.0, |
| "eval_steps": 25, |
| "global_step": 56, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.35714285714285715, |
| "grad_norm": 0.17697976082751796, |
| "learning_rate": 0.0004, |
| "loss": 1.071, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.7142857142857143, |
| "grad_norm": 0.2008396622881431, |
| "learning_rate": 0.00036862745098039214, |
| "loss": 0.7765, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.0714285714285714, |
| "grad_norm": 1.0337814287154459, |
| "learning_rate": 0.0003294117647058824, |
| "loss": 0.6828, |
| "step": 15 |
| }, |
| { |
| "epoch": 1.4285714285714286, |
| "grad_norm": 0.10447206497531818, |
| "learning_rate": 0.00029019607843137256, |
| "loss": 0.5171, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.7857142857142856, |
| "grad_norm": 0.1408100367921861, |
| "learning_rate": 0.00025098039215686274, |
| "loss": 0.4798, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.7857142857142856, |
| "eval_accuracy": 0.8643008461342605, |
| "eval_loss": 0.48145776987075806, |
| "eval_runtime": 284.1203, |
| "eval_samples_per_second": 0.567, |
| "eval_steps_per_second": 0.074, |
| "step": 25 |
| }, |
| { |
| "epoch": 2.142857142857143, |
| "grad_norm": 0.11436954179128357, |
| "learning_rate": 0.00021176470588235295, |
| "loss": 0.42, |
| "step": 30 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 0.14359528940424995, |
| "learning_rate": 0.00017254901960784316, |
| "loss": 0.3501, |
| "step": 35 |
| }, |
| { |
| "epoch": 2.857142857142857, |
| "grad_norm": 0.11140560490077989, |
| "learning_rate": 0.00013333333333333334, |
| "loss": 0.3359, |
| "step": 40 |
| }, |
| { |
| "epoch": 3.2142857142857144, |
| "grad_norm": 0.13609807295506027, |
| "learning_rate": 9.411764705882353e-05, |
| "loss": 0.2673, |
| "step": 45 |
| }, |
| { |
| "epoch": 3.571428571428571, |
| "grad_norm": 0.12397837803610176, |
| "learning_rate": 5.490196078431373e-05, |
| "loss": 0.2373, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.571428571428571, |
| "eval_accuracy": 0.8817340902302905, |
| "eval_loss": 0.458571195602417, |
| "eval_runtime": 308.9695, |
| "eval_samples_per_second": 0.521, |
| "eval_steps_per_second": 0.068, |
| "step": 50 |
| }, |
| { |
| "epoch": 3.928571428571429, |
| "grad_norm": 0.12856059584554186, |
| "learning_rate": 1.568627450980392e-05, |
| "loss": 0.25, |
| "step": 55 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 56, |
| "total_flos": 86439169622016.0, |
| "train_loss": 0.4839397243091038, |
| "train_runtime": 7475.7363, |
| "train_samples_per_second": 0.233, |
| "train_steps_per_second": 0.007 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 56, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 4, |
| "save_steps": 25, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 86439169622016.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
