Qwen2.5-1.5B-Open-R1-Distill / train_results.json
AlanWuxm's picture
Model save
82df75b verified
{
"epoch": 0.9999069680900549,
"total_flos": 8.247315480402985e+17,
"train_loss": 0.638783668202896,
"train_runtime": 14784.0018,
"train_samples": 16610,
"train_samples_per_second": 1.454,
"train_steps_per_second": 0.182
}