Qwen2.5-0.5B-Open-R1-Distill / train_results.json
allendou's picture
Model save
6240c63 verified
{
"total_flos": 34729623552.0,
"train_loss": 1.2833720207214356,
"train_runtime": 27.0374,
"train_samples": 16610,
"train_samples_per_second": 2.959,
"train_steps_per_second": 0.37
}