eval_results.json

{
    "epoch": 0.8502598016060463,
    "eval_dpo_loss": 0.6905742287635803,
    "eval_logits": -1.6666735410690308,
    "eval_logps": -92.34442901611328,
    "eval_loss": 0.41158440709114075,
    "eval_objective": 0.4097130298614502,
    "eval_ranking_idealized": 0.8731883764266968,
    "eval_ranking_idealized_expo": 0.5320910811424255,
    "eval_ranking_simple": 0.5289855003356934,
    "eval_regularize": 0.4097130298614502,
    "eval_runtime": 307.9574,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.801,
    "eval_steps_per_second": 1.568,
    "eval_wo_beta": 17.63062286376953
}