Qwen2.5-7B-Open-R1-GRPO/train_results.json

{
    "total_flos": 0.0,
    "train_loss": 0.0,
    "train_runtime": 1.0788,
    "train_samples": 5501,
    "train_samples_per_second": 25495.021,
    "train_steps_per_second": 908.385
}