Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
zztheaven's picture
Model save
fc17fe7 verified
{
"total_flos": 0.0,
"train_loss": 138538.8512365978,
"train_runtime": 59124.4888,
"train_samples": 13120,
"train_samples_per_second": 0.222,
"train_steps_per_second": 0.055
}