| { | |
| "epoch": 0.9987943737441393, | |
| "eval_logits/chosen": -3.4056549072265625, | |
| "eval_logits/rejected": -3.3998796939849854, | |
| "eval_logps/chosen": -0.5611971616744995, | |
| "eval_logps/rejected": -0.6183008551597595, | |
| "eval_loss": 1.5195728540420532, | |
| "eval_rewards/accuracies": 0.5691489577293396, | |
| "eval_rewards/chosen": -0.5611971616744995, | |
| "eval_rewards/margins": 0.057103704661130905, | |
| "eval_rewards/rejected": -0.6183008551597595, | |
| "eval_runtime": 434.9992, | |
| "eval_samples": 2994, | |
| "eval_samples_per_second": 6.883, | |
| "eval_steps_per_second": 0.432, | |
| "total_flos": 0.0, | |
| "train_loss": 1.5151263257976253, | |
| "train_runtime": 19305.9847, | |
| "train_samples": 59720, | |
| "train_samples_per_second": 3.093, | |
| "train_steps_per_second": 0.024 | |
| } |