Qwen2.5-7B-Open-R1-Distill / all_results.json
howardzhou's picture
Model save
c35df41 verified
{
"epoch": 0.9995169859925938,
"total_flos": 1627220267237376.0,
"train_loss": 0.5273892322612792,
"train_runtime": 41682.4586,
"train_samples": 112817,
"train_samples_per_second": 4.768,
"train_steps_per_second": 0.037
}