Qwen2.5-1.5B-Open-R1-Code-GRPO/train_results.json

{
    "total_flos": 0.0,
    "train_loss": 0.0025880077524270744,
    "train_runtime": 10060.4286,
    "train_samples": 131188,
    "train_samples_per_second": 3.976,
    "train_steps_per_second": 0.05
}