qwen3-8b-base-new-dpo-ultrafeedback-4xh200-batch-128-q_t-0.43-s_star-0.4-20260429-230725/train_results.json

{
    "epoch": 0.9989528795811519,
    "total_flos": 0.0,
    "train_loss": 2.690188397901363,
    "train_runtime": 4372.8359,
    "train_samples": 61135,
    "train_samples_per_second": 13.981,
    "train_steps_per_second": 0.109
}