| { |
| "best_global_step": 1036, |
| "best_metric": 0.3279893696308136, |
| "best_model_checkpoint": "./results/checkpoint-1036", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 3108, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.4826254826254826, |
| "grad_norm": 14.053184509277344, |
| "learning_rate": 1.680823680823681e-05, |
| "loss": 0.5034, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.9652509652509652, |
| "grad_norm": 6.887276649475098, |
| "learning_rate": 1.3590733590733592e-05, |
| "loss": 0.3842, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8750678978815861, |
| "eval_loss": 0.3279893696308136, |
| "eval_runtime": 14.343, |
| "eval_samples_per_second": 128.355, |
| "eval_steps_per_second": 4.044, |
| "step": 1036 |
| }, |
| { |
| "epoch": 1.4478764478764479, |
| "grad_norm": 6.9067792892456055, |
| "learning_rate": 1.0373230373230373e-05, |
| "loss": 0.2572, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.9305019305019306, |
| "grad_norm": 13.993288040161133, |
| "learning_rate": 7.155727155727157e-06, |
| "loss": 0.2339, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8750678978815861, |
| "eval_loss": 0.3709438443183899, |
| "eval_runtime": 14.2627, |
| "eval_samples_per_second": 129.078, |
| "eval_steps_per_second": 4.067, |
| "step": 2072 |
| }, |
| { |
| "epoch": 2.413127413127413, |
| "grad_norm": 0.07056383788585663, |
| "learning_rate": 3.938223938223939e-06, |
| "loss": 0.1442, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.8957528957528957, |
| "grad_norm": 22.20907211303711, |
| "learning_rate": 7.335907335907337e-07, |
| "loss": 0.1488, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8804997284084737, |
| "eval_loss": 0.46850883960723877, |
| "eval_runtime": 14.2613, |
| "eval_samples_per_second": 129.09, |
| "eval_steps_per_second": 4.067, |
| "step": 3108 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 3108, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "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": 1.3076999974499328e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
