Qwen2.5-1.5B-Open-R1-Distill / train_results.json
susumuota's picture
Model save
0207fc5 verified
raw
history blame
212 Bytes
{
"total_flos": 76973799899136.0,
"train_loss": 0.8025016911636443,
"train_runtime": 750.6152,
"train_samples": 16610,
"train_samples_per_second": 28.807,
"train_steps_per_second": 0.225
}