zephyr-7b-dpo-full/all_results.json

{
    "epoch": 0.9990319457889641,
    "total_flos": 0.0,
    "train_loss": 1.2622329578843228,
    "train_runtime": 8815.3772,
    "train_samples": 66084,
    "train_samples_per_second": 7.496,
    "train_steps_per_second": 0.059
}