{ "epoch": 1.0, "eval_logits/chosen": -1.34141206741333, "eval_logits/rejected": -2.0515904426574707, "eval_logps/chosen": -241.21115112304688, "eval_logps/rejected": -1564.379150390625, "eval_loss": 0.0018388006137683988, "eval_rewards/accuracies": 1.0, "eval_rewards/chosen": -0.4458274245262146, "eval_rewards/margins": 9.664073944091797, "eval_rewards/rejected": -10.109901428222656, "eval_runtime": 6.9758, "eval_samples": 64, "eval_samples_per_second": 9.175, "eval_steps_per_second": 0.287 }