| { | |
| "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": 0.5861290322580646, | |
| "eval_loss": 0.1950901746749878, | |
| "eval_runtime": 1.47, | |
| "eval_samples_per_second": 2108.896, | |
| "eval_steps_per_second": 44.219, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.315, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8309677419354838, | |
| "eval_loss": 0.09493588656187057, | |
| "eval_runtime": 1.4721, | |
| "eval_samples_per_second": 2105.842, | |
| "eval_steps_per_second": 44.155, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8793548387096775, | |
| "eval_loss": 0.0637393370270729, | |
| "eval_runtime": 1.4665, | |
| "eval_samples_per_second": 2113.9, | |
| "eval_steps_per_second": 44.324, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1093, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.896774193548387, | |
| "eval_loss": 0.04817795380949974, | |
| "eval_runtime": 1.468, | |
| "eval_samples_per_second": 2111.77, | |
| "eval_steps_per_second": 44.279, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0692, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9119354838709678, | |
| "eval_loss": 0.039378490298986435, | |
| "eval_runtime": 1.4633, | |
| "eval_samples_per_second": 2118.489, | |
| "eval_steps_per_second": 44.42, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9180645161290323, | |
| "eval_loss": 0.03448617830872536, | |
| "eval_runtime": 1.4734, | |
| "eval_samples_per_second": 2103.981, | |
| "eval_steps_per_second": 44.116, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0542, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 662375493608136.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.17426119812070984, | |
| "num_train_epochs": 10, | |
| "temperature": 13 | |
| } | |
| } | |