all_results.json

{
    "epoch": 4.988190836088805,
    "eval_dpo_loss": 1141.8062744140625,
    "eval_logits": -0.6597133278846741,
    "eval_logps": -82.09796142578125,
    "eval_loss": 2234.166259765625,
    "eval_objective": 2265.87939453125,
    "eval_ranking_idealized": 0.5092975497245789,
    "eval_ranking_idealized_expo": 0.5092975497245789,
    "eval_ranking_simple": 0.5123966932296753,
    "eval_regularize": 2265.87939453125,
    "eval_runtime": 260.9946,
    "eval_samples": 5790,
    "eval_samples_per_second": 22.184,
    "eval_steps_per_second": 0.927,
    "total_flos": 0.0,
    "train_loss": 743.9546574996276,
    "train_runtime": 35228.1764,
    "train_samples": 50802,
    "train_samples_per_second": 7.21,
    "train_steps_per_second": 0.025
}