{ "epoch": 2.9865871833084947, "eval_log_odds_chosen": 0.27211081981658936, "eval_log_odds_ratio": -0.7039158344268799, "eval_logits/chosen": 296.6807556152344, "eval_logits/rejected": 267.234619140625, "eval_logps/chosen": -1.0933129787445068, "eval_logps/rejected": -1.2508149147033691, "eval_loss": 1.5379581451416016, "eval_nll_loss": 1.4702765941619873, "eval_rewards/accuracies": 0.5467625856399536, "eval_rewards/chosen": -0.054665643721818924, "eval_rewards/margins": 0.007875093258917332, "eval_rewards/rejected": -0.06254073977470398, "eval_runtime": 112.3528, "eval_samples": 553, "eval_samples_per_second": 4.922, "eval_steps_per_second": 1.237, "total_flos": 0.0, "train_loss": 1.4583633707431025, "train_runtime": 13840.413, "train_samples": 5364, "train_samples_per_second": 1.163, "train_steps_per_second": 0.036 }