Qwen2.5-0.5B-Open-R1-Distill / train_results.json
czczup's picture
Model save
6b16b3b verified
{
"total_flos": 1.1596255879128678e+17,
"train_loss": 0.571030998982272,
"train_runtime": 553.8515,
"train_samples": 72441,
"train_samples_per_second": 95.222,
"train_steps_per_second": 0.744
}