| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.997305550565835, | |
| "global_step": 6955, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7353, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6787, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.703482985496521, | |
| "eval_runtime": 198.0791, | |
| "eval_samples_per_second": 16.221, | |
| "eval_steps_per_second": 1.015, | |
| "step": 1391 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6239, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5056, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5142, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.634272575378418, | |
| "eval_runtime": 197.806, | |
| "eval_samples_per_second": 16.243, | |
| "eval_steps_per_second": 1.016, | |
| "step": 2783 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4472, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3735, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3854, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6563898324966431, | |
| "eval_runtime": 197.807, | |
| "eval_samples_per_second": 16.243, | |
| "eval_steps_per_second": 1.016, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3094, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2745, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2816, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6947186589241028, | |
| "eval_runtime": 197.5706, | |
| "eval_samples_per_second": 16.263, | |
| "eval_steps_per_second": 1.017, | |
| "step": 5567 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2027, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2017, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.7372682094573975, | |
| "eval_runtime": 197.1547, | |
| "eval_samples_per_second": 16.297, | |
| "eval_steps_per_second": 1.02, | |
| "step": 6955 | |
| } | |
| ], | |
| "max_steps": 6955, | |
| "num_train_epochs": 5, | |
| "total_flos": 29845171273728.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |