Qwen2.5-3B-Open-R1-Code-GRPO / all_results.json
Yukang's picture
Model save
643d511 verified
{
"total_flos": 0.0,
"train_loss": 9.785836571644918e-07,
"train_runtime": 72.7583,
"train_samples": 35735,
"train_samples_per_second": 3518.501,
"train_steps_per_second": 6.872
}