| { | |
| "epoch": 1.0, | |
| "eval_dpo_losses": 0.6394718289375305, | |
| "eval_logits/chosen": -2.6543400287628174, | |
| "eval_logits/rejected": -2.6202285289764404, | |
| "eval_logps/chosen": -261.381103515625, | |
| "eval_logps/rejected": -247.60108947753906, | |
| "eval_loss": 0.6789140701293945, | |
| "eval_positive_losses": 0.2623763382434845, | |
| "eval_rewards/accuracies": 0.7179999947547913, | |
| "eval_rewards/chosen": 0.2321232557296753, | |
| "eval_rewards/margins": 0.1223447397351265, | |
| "eval_rewards/margins_max": 0.4485361576080322, | |
| "eval_rewards/margins_min": -0.15482433140277863, | |
| "eval_rewards/margins_std": 0.20253273844718933, | |
| "eval_rewards/rejected": 0.1097785010933876, | |
| "eval_runtime": 428.1146, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 4.672, | |
| "eval_steps_per_second": 0.292, | |
| "train_loss": 0.6795008113577053, | |
| "train_runtime": 46141.1523, | |
| "train_samples": 61134, | |
| "train_samples_per_second": 1.325, | |
| "train_steps_per_second": 0.083 | |
| } |