| { | |
| "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": 0.5864516129032258, | |
| "eval_loss": 0.1933201551437378, | |
| "eval_runtime": 5.533, | |
| "eval_samples_per_second": 560.278, | |
| "eval_steps_per_second": 11.748, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.3095, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.09366016089916229, | |
| "eval_runtime": 5.65, | |
| "eval_samples_per_second": 548.672, | |
| "eval_steps_per_second": 11.504, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8912903225806451, | |
| "eval_loss": 0.06257741898298264, | |
| "eval_runtime": 5.4549, | |
| "eval_samples_per_second": 568.294, | |
| "eval_steps_per_second": 11.916, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1078, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9025806451612903, | |
| "eval_loss": 0.04725448787212372, | |
| "eval_runtime": 5.4608, | |
| "eval_samples_per_second": 567.683, | |
| "eval_steps_per_second": 11.903, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0683, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9151612903225806, | |
| "eval_loss": 0.039166562259197235, | |
| "eval_runtime": 5.5371, | |
| "eval_samples_per_second": 559.86, | |
| "eval_steps_per_second": 11.739, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9167741935483871, | |
| "eval_loss": 0.03421633318066597, | |
| "eval_runtime": 5.4396, | |
| "eval_samples_per_second": 569.898, | |
| "eval_steps_per_second": 11.949, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0536, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 638004691747332.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.9205924553180014, | |
| "num_train_epochs": 10, | |
| "temperature": 18 | |
| } | |
| } | |