huatuo-o1-Med-REFL-LoraAdapter/train_results.json

{
    "epoch": 0.9997605937275557,
    "num_input_tokens_seen": 93801152,
    "total_flos": 4.318246812239528e+18,
    "train_loss": 0.2454147177009747,
    "train_runtime": 22841.5151,
    "train_samples_per_second": 1.463,
    "train_steps_per_second": 0.011
}