| { | |
| "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": 0.6212903225806452, | |
| "eval_loss": 0.2071644514799118, | |
| "eval_runtime": 5.7935, | |
| "eval_samples_per_second": 535.086, | |
| "eval_steps_per_second": 11.22, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.3314, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.82, | |
| "eval_loss": 0.1023462787270546, | |
| "eval_runtime": 5.8254, | |
| "eval_samples_per_second": 532.155, | |
| "eval_steps_per_second": 11.158, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8774193548387097, | |
| "eval_loss": 0.06782012432813644, | |
| "eval_runtime": 5.7323, | |
| "eval_samples_per_second": 540.794, | |
| "eval_steps_per_second": 11.339, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1175, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8970967741935484, | |
| "eval_loss": 0.053100328892469406, | |
| "eval_runtime": 5.7605, | |
| "eval_samples_per_second": 538.152, | |
| "eval_steps_per_second": 11.284, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0769, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9029032258064517, | |
| "eval_loss": 0.04506096988916397, | |
| "eval_runtime": 5.8645, | |
| "eval_samples_per_second": 528.601, | |
| "eval_steps_per_second": 11.084, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9074193548387097, | |
| "eval_loss": 0.04114566370844841, | |
| "eval_runtime": 5.7197, | |
| "eval_samples_per_second": 541.986, | |
| "eval_steps_per_second": 11.364, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0632, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 519271419317532.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.28786009970392035, | |
| "num_train_epochs": 7, | |
| "temperature": 9 | |
| } | |
| } | |