| { | |
| "epoch": 1.9960291197882198, | |
| "eval_logits/chosen": -2.7966158390045166, | |
| "eval_logits/rejected": -2.8938040733337402, | |
| "eval_logps/chosen": -747.97314453125, | |
| "eval_logps/rejected": -1448.79833984375, | |
| "eval_loss": 0.043147582560777664, | |
| "eval_rewards/accuracies": 0.9375, | |
| "eval_rewards/chosen": -3.0406525135040283, | |
| "eval_rewards/margins": 6.358960151672363, | |
| "eval_rewards/rejected": -9.399612426757812, | |
| "eval_runtime": 1.7064, | |
| "eval_samples": 100, | |
| "eval_samples_per_second": 58.602, | |
| "eval_steps_per_second": 1.172, | |
| "total_flos": 0.0, | |
| "train_loss": 0.26563876598520053, | |
| "train_runtime": 5267.2293, | |
| "train_samples": 96652, | |
| "train_samples_per_second": 36.699, | |
| "train_steps_per_second": 0.143 | |
| } |