eval_results.json

{
    "epoch": 0.8502598016060463,
    "eval_dpo_loss": 0.6902037858963013,
    "eval_logits": -1.7611440420150757,
    "eval_logps": -99.38214111328125,
    "eval_loss": 0.6893583536148071,
    "eval_objective": 0.6902037858963013,
    "eval_ranking_idealized": 0.8731883764266968,
    "eval_ranking_idealized_expo": 0.5320910811424255,
    "eval_ranking_simple": 0.5305383205413818,
    "eval_regularize": 0.6902037858963013,
    "eval_runtime": 308.1577,
    "eval_samples": 5790,
    "eval_samples_per_second": 18.789,
    "eval_steps_per_second": 1.567,
    "eval_wo_beta": 9.357504844665527
}