| { | |
| "epoch": 1.0, | |
| "eval_logits/chosen": 13.463111877441406, | |
| "eval_logits/rejected": 13.527791023254395, | |
| "eval_logps/chosen": -15128.9541015625, | |
| "eval_logps/rejected": -14704.759765625, | |
| "eval_loss": 0.3621964752674103, | |
| "eval_pred_label": 8140.01611328125, | |
| "eval_rewards/accuracies": 0.2420634925365448, | |
| "eval_rewards/chosen": -150.60052490234375, | |
| "eval_rewards/margins": -4.306910991668701, | |
| "eval_rewards/rejected": -146.29360961914062, | |
| "eval_runtime": 245.5331, | |
| "eval_samples": 2000, | |
| "eval_samples_per_second": 8.146, | |
| "eval_steps_per_second": 0.257, | |
| "eval_use_label": 11931.984375 | |
| } |