| { | |
| "epoch": 3.0, | |
| "eval_log_odds_chosen": 0.44554343819618225, | |
| "eval_log_odds_ratio": -0.6395100355148315, | |
| "eval_logits/chosen": 319.3428649902344, | |
| "eval_logits/rejected": 262.5517883300781, | |
| "eval_logps/chosen": -1.3908888101577759, | |
| "eval_logps/rejected": -1.6692452430725098, | |
| "eval_loss": 2.1094672679901123, | |
| "eval_nll_loss": 1.7835990190505981, | |
| "eval_rewards/accuracies": 0.5571428537368774, | |
| "eval_rewards/chosen": -0.6954444050788879, | |
| "eval_rewards/margins": 0.13917820155620575, | |
| "eval_rewards/rejected": -0.8346226215362549, | |
| "eval_runtime": 201.4763, | |
| "eval_samples": 553, | |
| "eval_samples_per_second": 2.745, | |
| "eval_steps_per_second": 0.347 | |
| } |