{ "epoch": 1.0, "eval_logits/chosen": 98.27266693115234, "eval_logits/rejected": 98.26683044433594, "eval_logps/chosen": -32.204383850097656, "eval_logps/rejected": -35.79214859008789, "eval_loss": 0.49918678402900696, "eval_rewards/accuracies": 0.489202618598938, "eval_rewards/chosen": 0.023878853768110275, "eval_rewards/margins": 0.00644117034971714, "eval_rewards/rejected": 0.017437683418393135, "eval_runtime": 103.7538, "eval_samples": 343, "eval_samples_per_second": 3.306, "eval_steps_per_second": 0.414 }