| { | |
| "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": 0.594516129032258, | |
| "eval_loss": 0.20276567339897156, | |
| "eval_runtime": 5.4162, | |
| "eval_samples_per_second": 572.36, | |
| "eval_steps_per_second": 12.001, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.535661518573761, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.3272, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8270967741935484, | |
| "eval_loss": 0.09640489518642426, | |
| "eval_runtime": 6.1013, | |
| "eval_samples_per_second": 508.09, | |
| "eval_steps_per_second": 10.653, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8841935483870967, | |
| "eval_loss": 0.06340694427490234, | |
| "eval_runtime": 5.5585, | |
| "eval_samples_per_second": 557.704, | |
| "eval_steps_per_second": 11.694, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.4495888650417328, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1121, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9006451612903226, | |
| "eval_loss": 0.048330824822187424, | |
| "eval_runtime": 5.5377, | |
| "eval_samples_per_second": 559.795, | |
| "eval_steps_per_second": 11.738, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.30502042174339294, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0703, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9106451612903226, | |
| "eval_loss": 0.03915739059448242, | |
| "eval_runtime": 5.8885, | |
| "eval_samples_per_second": 526.453, | |
| "eval_steps_per_second": 11.039, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9151612903225806, | |
| "eval_loss": 0.03403216972947121, | |
| "eval_runtime": 5.6759, | |
| "eval_samples_per_second": 546.164, | |
| "eval_steps_per_second": 11.452, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.2584752142429352, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0544, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9245161290322581, | |
| "eval_loss": 0.03124151937663555, | |
| "eval_runtime": 5.5585, | |
| "eval_samples_per_second": 557.7, | |
| "eval_steps_per_second": 11.694, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.2786354124546051, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.047, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 0.07863039618542955, | |
| "num_train_epochs": 10, | |
| "temperature": 9 | |
| } | |
| } | |