Qwen2.5-1.5B-Open-R1-Distill / train_results.json
lixiangtian's picture
Model save
c251766 verified
{
"epoch": 1.0,
"total_flos": 76916824473600.0,
"train_loss": 0.8026753050335765,
"train_runtime": 1039.2266,
"train_samples": 16610,
"train_samples_per_second": 20.794,
"train_steps_per_second": 0.163
}