all_results.json

{
    "epoch": 0.998691442030882,
    "eval_logits/chosen": -0.8399364352226257,
    "eval_logits/rejected": -0.8775278329849243,
    "eval_logps/chosen": -316.6497802734375,
    "eval_logps/rejected": -344.29437255859375,
    "eval_loss": 0.5597857236862183,
    "eval_rewards/accuracies": 0.7239999771118164,
    "eval_rewards/chosen": -0.25838080048561096,
    "eval_rewards/margins": 0.5088632106781006,
    "eval_rewards/rejected": -0.7672439813613892,
    "eval_runtime": 355.4568,
    "eval_samples": 2000,
    "eval_samples_per_second": 5.627,
    "eval_steps_per_second": 1.407,
    "total_flos": 0.0,
    "train_loss": 0.5919944611485399,
    "train_runtime": 28847.4829,
    "train_samples": 61134,
    "train_samples_per_second": 2.119,
    "train_steps_per_second": 0.017
}