| { | |
| "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": 0.6187096774193548, | |
| "eval_loss": 0.2273363471031189, | |
| "eval_runtime": 5.4212, | |
| "eval_samples_per_second": 571.832, | |
| "eval_steps_per_second": 11.99, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5702099204063416, | |
| "learning_rate": 1.7141223556317898e-05, | |
| "loss": 0.3672, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8416129032258064, | |
| "eval_loss": 0.10179296880960464, | |
| "eval_runtime": 5.7619, | |
| "eval_samples_per_second": 538.015, | |
| "eval_steps_per_second": 11.281, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8906451612903226, | |
| "eval_loss": 0.06358441710472107, | |
| "eval_runtime": 5.4704, | |
| "eval_samples_per_second": 566.684, | |
| "eval_steps_per_second": 11.882, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.48276859521865845, | |
| "learning_rate": 1.4282447112635793e-05, | |
| "loss": 0.1191, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.905483870967742, | |
| "eval_loss": 0.04695529490709305, | |
| "eval_runtime": 5.4589, | |
| "eval_samples_per_second": 567.879, | |
| "eval_steps_per_second": 11.907, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3197612166404724, | |
| "learning_rate": 1.1423670668953687e-05, | |
| "loss": 0.0712, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9167741935483871, | |
| "eval_loss": 0.03740057349205017, | |
| "eval_runtime": 5.721, | |
| "eval_samples_per_second": 541.86, | |
| "eval_steps_per_second": 11.362, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9241935483870968, | |
| "eval_loss": 0.032310228794813156, | |
| "eval_runtime": 5.4558, | |
| "eval_samples_per_second": 568.208, | |
| "eval_steps_per_second": 11.914, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.2629503309726715, | |
| "learning_rate": 8.564894225271586e-06, | |
| "loss": 0.0539, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9283870967741935, | |
| "eval_loss": 0.029569072648882866, | |
| "eval_runtime": 5.5263, | |
| "eval_samples_per_second": 560.95, | |
| "eval_steps_per_second": 11.762, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.29348886013031006, | |
| "learning_rate": 5.706117781589479e-06, | |
| "loss": 0.046, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3498, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "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": 0.5264187954467557, | |
| "num_train_epochs": 11, | |
| "temperature": 5 | |
| } | |
| } | |