Qwen2.5-1.5B-Open-R1-Distill / train_results.json
atmatechai's picture
Model save
1f0ab9f verified
raw
history blame
213 Bytes
{
"total_flos": 77344140165120.0,
"train_loss": 0.7063807840505357,
"train_runtime": 17902.8015,
"train_samples": 16610,
"train_samples_per_second": 0.303,
"train_steps_per_second": 0.051
}