Qwen2.5-7B-Open-R1-Distill / train_results.json
Lansechen's picture
Model save
4e4e32c verified
{
"total_flos": 353216432701440.0,
"train_loss": 0.5622668473232394,
"train_runtime": 6715.662,
"train_samples": 16610,
"train_samples_per_second": 6.44,
"train_steps_per_second": 0.1
}