{ "epoch": 1.0, "eval_logits/chosen": -1.9167659282684326, "eval_logits/rejected": -1.693016767501831, "eval_logps/chosen": -129.38612365722656, "eval_logps/rejected": -152.45419311523438, "eval_loss": 0.6502523422241211, "eval_rewards/accuracies": 0.7272727489471436, "eval_rewards/chosen": 0.15327082574367523, "eval_rewards/margins": 0.10365187376737595, "eval_rewards/rejected": 0.04961895942687988, "eval_runtime": 64.2131, "eval_samples": 674, "eval_samples_per_second": 10.496, "eval_steps_per_second": 0.343 }