{ "epoch": 0.9989701338825953, "eval_logits/chosen": 0.44437986612319946, "eval_logits/rejected": 0.559954047203064, "eval_logps/chosen": -337.15203857421875, "eval_logps/rejected": -430.6039733886719, "eval_loss": 0.5154678821563721, "eval_rewards/accuracies": 0.75, "eval_rewards/chosen": -0.8051658868789673, "eval_rewards/margins": 0.8549233078956604, "eval_rewards/rejected": -1.660089373588562, "eval_runtime": 82.8218, "eval_samples_per_second": 12.074, "eval_steps_per_second": 0.193 }