| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5896774193548387, | |
| "eval_loss": 0.19703345000743866, | |
| "eval_runtime": 5.3285, | |
| "eval_samples_per_second": 581.779, | |
| "eval_steps_per_second": 12.199, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.4503379762172699, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3169, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8264516129032258, | |
| "eval_loss": 0.09507494419813156, | |
| "eval_runtime": 5.785, | |
| "eval_samples_per_second": 535.865, | |
| "eval_steps_per_second": 11.236, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8848387096774194, | |
| "eval_loss": 0.0635090321302414, | |
| "eval_runtime": 5.5393, | |
| "eval_samples_per_second": 559.633, | |
| "eval_steps_per_second": 11.734, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.5165953636169434, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1106, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.896774193548387, | |
| "eval_loss": 0.04883500561118126, | |
| "eval_runtime": 5.6259, | |
| "eval_samples_per_second": 551.023, | |
| "eval_steps_per_second": 11.554, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3079015016555786, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0705, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.91, | |
| "eval_loss": 0.040311940014362335, | |
| "eval_runtime": 5.5435, | |
| "eval_samples_per_second": 559.215, | |
| "eval_steps_per_second": 11.725, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9145161290322581, | |
| "eval_loss": 0.03531214967370033, | |
| "eval_runtime": 5.293, | |
| "eval_samples_per_second": 585.677, | |
| "eval_steps_per_second": 12.28, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.24189795553684235, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0556, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9187096774193548, | |
| "eval_loss": 0.032718539237976074, | |
| "eval_runtime": 5.5371, | |
| "eval_samples_per_second": 559.856, | |
| "eval_steps_per_second": 11.739, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.2118339240550995, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0488, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "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": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.48240196387166423, | |
| "num_train_epochs": 9, | |
| "temperature": 12 | |
| } | |
| } | |