| { | |
| "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": 0.6670967741935484, | |
| "eval_loss": 0.4093398451805115, | |
| "eval_runtime": 5.4719, | |
| "eval_samples_per_second": 566.527, | |
| "eval_steps_per_second": 11.879, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.9353984594345093, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.6434, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8422580645161291, | |
| "eval_loss": 0.13946937024593353, | |
| "eval_runtime": 6.3606, | |
| "eval_samples_per_second": 487.373, | |
| "eval_steps_per_second": 10.219, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9, | |
| "eval_loss": 0.07272056490182877, | |
| "eval_runtime": 5.361, | |
| "eval_samples_per_second": 578.251, | |
| "eval_steps_per_second": 12.125, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.7816033363342285, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1635, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9145161290322581, | |
| "eval_loss": 0.05225358158349991, | |
| "eval_runtime": 6.011, | |
| "eval_samples_per_second": 515.717, | |
| "eval_steps_per_second": 10.813, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.4276784658432007, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.0817, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9270967741935484, | |
| "eval_loss": 0.04296453669667244, | |
| "eval_runtime": 6.1809, | |
| "eval_samples_per_second": 501.545, | |
| "eval_steps_per_second": 10.516, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9293548387096774, | |
| "eval_loss": 0.038294676691293716, | |
| "eval_runtime": 5.6762, | |
| "eval_samples_per_second": 546.14, | |
| "eval_steps_per_second": 11.451, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.31978192925453186, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0617, | |
| "step": 2000 | |
| } | |
| ], | |
| "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": 520991326672152.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.06945743007212535, | |
| "num_train_epochs": 9, | |
| "temperature": 2 | |
| } | |
| } | |