all_results.json

{
    "epoch": 0.9981298423724285,
    "eval_abs_diff": 3.19630765914917,
    "eval_all_logps_1": -8025.169921875,
    "eval_all_logps_1_values": -8025.169921875,
    "eval_all_logps_2": 414.86090087890625,
    "eval_all_logps_2_values": 414.86090087890625,
    "eval_logits/chosen": 1.7087222337722778,
    "eval_logits/rejected": 1.8237613439559937,
    "eval_logps/chosen": -19.160812377929688,
    "eval_logps/rejected": -21.144468307495117,
    "eval_loss": 0.7512779235839844,
    "eval_original_losses": 2.0757226943969727,
    "eval_rewards/accuracies": 0.7056451439857483,
    "eval_rewards/chosen": -47.90203094482422,
    "eval_rewards/margins": 4.959139823913574,
    "eval_rewards/rejected": -52.861167907714844,
    "eval_runtime": 70.4078,
    "eval_samples": 1961,
    "eval_samples_per_second": 27.852,
    "eval_steps_per_second": 0.881,
    "eval_weight": 0.36962273716926575,
    "total_flos": 0.0,
    "train_loss": 0.9884350126254227,
    "train_runtime": 7236.0008,
    "train_samples": 59876,
    "train_samples_per_second": 8.275,
    "train_steps_per_second": 0.065
}