all_results.json

{
    "epoch": 0.998691442030882,
    "eval_dpo_lambda": 0.9499999284744263,
    "eval_logits/chosen": -0.9673975706100464,
    "eval_logits/rejected": -1.0031267404556274,
    "eval_logps/chosen": -338.55499267578125,
    "eval_logps/rejected": -360.69610595703125,
    "eval_loss": 0.5627262592315674,
    "eval_rewards/accuracies": 0.7329999804496765,
    "eval_rewards/chosen": -0.3790811598300934,
    "eval_rewards/margins": 0.5175721645355225,
    "eval_rewards/rejected": -0.8966532945632935,
    "eval_runtime": 561.3254,
    "eval_samples": 2000,
    "eval_samples_per_second": 3.563,
    "eval_steps_per_second": 0.891,
    "total_flos": 0.0,
    "train_loss": 0.5879578035582537,
    "train_runtime": 40532.5341,
    "train_samples": 61134,
    "train_samples_per_second": 1.508,
    "train_steps_per_second": 0.012
}