{ "epoch": 1.0, "eval_logits/chosen": -2.9068422317504883, "eval_logits/rejected": -2.864485025405884, "eval_logps/chosen": -189.702392578125, "eval_logps/rejected": -173.6967315673828, "eval_loss": 0.11297444999217987, "eval_rewards/accuracies": 0.8643617033958435, "eval_rewards/chosen": -3.773209571838379, "eval_rewards/margins": 9.667583465576172, "eval_rewards/rejected": -13.440792083740234, "eval_runtime": 242.5683, "eval_samples": 1495, "eval_samples_per_second": 6.163, "eval_steps_per_second": 0.194 }