all_results.json

{
    "epoch": 2.9876543209876543,
    "eval_logits/chosen": -1.9506410360336304,
    "eval_logits/rejected": -1.891175389289856,
    "eval_logps/chosen": -250.91085815429688,
    "eval_logps/rejected": -234.10423278808594,
    "eval_loss": 0.5232799053192139,
    "eval_rewards/accuracies": 0.8217592835426331,
    "eval_rewards/chosen": 0.7968389987945557,
    "eval_rewards/margins": 2.6544954776763916,
    "eval_rewards/rejected": -1.8576565980911255,
    "eval_runtime": 227.7341,
    "eval_samples_per_second": 15.176,
    "eval_steps_per_second": 0.237,
    "total_flos": 4280357159436288.0,
    "train_loss": 0.30565077164941584,
    "train_runtime": 13036.7565,
    "train_samples_per_second": 7.158,
    "train_steps_per_second": 0.028
}