| { | |
| "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": 0.7293548387096774, | |
| "eval_loss": 3.176466703414917, | |
| "eval_runtime": 2.364, | |
| "eval_samples_per_second": 1311.337, | |
| "eval_steps_per_second": 27.496, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 3.7036, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8590322580645161, | |
| "eval_loss": 1.5857902765274048, | |
| "eval_runtime": 2.433, | |
| "eval_samples_per_second": 1274.147, | |
| "eval_steps_per_second": 26.716, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9109677419354839, | |
| "eval_loss": 0.7998820543289185, | |
| "eval_runtime": 2.397, | |
| "eval_samples_per_second": 1293.285, | |
| "eval_steps_per_second": 27.117, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 1.3611, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9338709677419355, | |
| "eval_loss": 0.4834734797477722, | |
| "eval_runtime": 2.438, | |
| "eval_samples_per_second": 1271.534, | |
| "eval_steps_per_second": 26.661, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.4625, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9416129032258065, | |
| "eval_loss": 0.35008904337882996, | |
| "eval_runtime": 2.443, | |
| "eval_samples_per_second": 1268.932, | |
| "eval_steps_per_second": 26.607, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9461290322580646, | |
| "eval_loss": 0.2929627001285553, | |
| "eval_runtime": 2.448, | |
| "eval_samples_per_second": 1266.34, | |
| "eval_steps_per_second": 26.552, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.1991, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 519927215063004.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.3626187503283364, | |
| "num_train_epochs": 9, | |
| "temperature": 5 | |
| } | |
| } | |