| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6674193548387096, | |
| "eval_loss": 0.416048139333725, | |
| "eval_runtime": 1.3187, | |
| "eval_samples_per_second": 2350.748, | |
| "eval_steps_per_second": 49.29, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.9905994534492493, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.6477, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8332258064516129, | |
| "eval_loss": 0.15088918805122375, | |
| "eval_runtime": 1.3272, | |
| "eval_samples_per_second": 2335.773, | |
| "eval_steps_per_second": 48.976, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.886774193548387, | |
| "eval_loss": 0.07951628416776657, | |
| "eval_runtime": 1.3591, | |
| "eval_samples_per_second": 2280.965, | |
| "eval_steps_per_second": 47.827, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.7344270348548889, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1749, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9090322580645162, | |
| "eval_loss": 0.05769029259681702, | |
| "eval_runtime": 1.3435, | |
| "eval_samples_per_second": 2307.35, | |
| "eval_steps_per_second": 48.38, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.4701627790927887, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0902, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9193548387096774, | |
| "eval_loss": 0.04808399826288223, | |
| "eval_runtime": 1.3493, | |
| "eval_samples_per_second": 2297.566, | |
| "eval_steps_per_second": 48.175, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9235483870967742, | |
| "eval_loss": 0.0440458320081234, | |
| "eval_runtime": 1.3465, | |
| "eval_samples_per_second": 2302.187, | |
| "eval_steps_per_second": 48.272, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.4379720985889435, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0704, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 520991326672152.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.5999229797501479, | |
| "num_train_epochs": 7, | |
| "temperature": 2 | |
| } | |
| } | |