Luminia-8B-RP-DPO/train_results.json

{
    "epoch": 1.0,
    "num_input_tokens_seen": 75616,
    "total_flos": 3443013082939392.0,
    "train_loss": 1.2913588443434383,
    "train_runtime": 65.9669,
    "train_samples_per_second": 1.258,
    "train_steps_per_second": 1.258
}