{ "epoch": 1.0, "eval_logits/chosen": 0.8123594522476196, "eval_logits/rejected": 0.8536277413368225, "eval_logps/chosen": -430.860595703125, "eval_logps/rejected": -484.1859436035156, "eval_loss": 0.5485682487487793, "eval_rewards/accuracies": 0.7089999914169312, "eval_rewards/chosen": -1.4556782245635986, "eval_rewards/margins": 0.7474843859672546, "eval_rewards/rejected": -2.203162670135498, "eval_runtime": 1172.6703, "eval_samples": 2000, "eval_samples_per_second": 1.706, "eval_steps_per_second": 0.853 }