all_results.json

{
    "epoch": 0.9982631930527722,
    "eval_logits/chosen": 0.016398662701249123,
    "eval_logits/rejected": 0.006773484870791435,
    "eval_logps/chosen": -0.3364824056625366,
    "eval_logps/rejected": -0.40113401412963867,
    "eval_loss": 1.2413759231567383,
    "eval_rewards/accuracies": 0.5853658318519592,
    "eval_rewards/chosen": -0.3364824056625366,
    "eval_rewards/margins": 0.06465159356594086,
    "eval_rewards/rejected": -0.40113401412963867,
    "eval_runtime": 428.2667,
    "eval_samples": 1961,
    "eval_samples_per_second": 4.579,
    "eval_steps_per_second": 0.287,
    "total_flos": 0.0,
    "train_loss": 1.2480301234145237,
    "train_runtime": 21322.7535,
    "train_samples": 59876,
    "train_samples_per_second": 2.808,
    "train_steps_per_second": 0.022
}