Qwen2.5-1.5B-Open-R1-Distill / train_results.json
wxnfifth's picture
Model save
ada336f verified
raw
history blame
244 Bytes
{
"epoch": 0.997779422649889,
"total_flos": 76745898196992.0,
"train_loss": 0.7870961399389658,
"train_runtime": 4760.8197,
"train_samples": 16610,
"train_samples_per_second": 4.539,
"train_steps_per_second": 0.071
}