all_results.json

{
    "epoch": 0.9221902017291066,
    "eval_logits/chosen": -0.03889964893460274,
    "eval_logits/rejected": 0.12152793258428574,
    "eval_logps/chosen": -125.9035415649414,
    "eval_logps/ref_chosen": -125.96058654785156,
    "eval_logps/ref_rejected": -162.4300994873047,
    "eval_logps/rejected": -162.40286254882812,
    "eval_loss": 0.5148054361343384,
    "eval_rewards/accuracies": 0.5277777910232544,
    "eval_rewards/chosen": 0.0005705787916667759,
    "eval_rewards/margins": 0.00029826530953869224,
    "eval_rewards/rejected": 0.0002723135403357446,
    "eval_runtime": 145.6449,
    "eval_samples": 1080,
    "eval_samples_per_second": 7.415,
    "eval_sft_loss": 0.42570608854293823,
    "eval_steps_per_second": 1.854,
    "total_flos": 0.0,
    "train_loss": 0.5319805860519409,
    "train_runtime": 668.7287,
    "train_samples": 1387,
    "train_samples_per_second": 2.074,
    "train_steps_per_second": 0.015
}