all_results.json

{
    "epoch": 0.99,
    "eval_logits/chosen": 0.8022828698158264,
    "eval_logits/rejected": 0.8200842142105103,
    "eval_logps/chosen": -91.75594329833984,
    "eval_logps/rejected": -94.22090911865234,
    "eval_loss": 0.6932105422019958,
    "eval_rewards/accuracies": 0.24799999594688416,
    "eval_rewards/chosen": -0.0002857264771591872,
    "eval_rewards/margins": -4.674374213209376e-05,
    "eval_rewards/rejected": -0.00023898274230305105,
    "eval_runtime": 273.9391,
    "eval_samples": 2000,
    "eval_samples_per_second": 7.301,
    "eval_steps_per_second": 0.456,
    "train_loss": 0.6931231824975265,
    "train_runtime": 1147.8486,
    "train_samples": 6113,
    "train_samples_per_second": 5.326,
    "train_steps_per_second": 0.083
}