Qwen2.5-1.5B-Open-R1-Distill / train_results.json
jeffff's picture
Model save
e6170b6 verified
{
"total_flos": 76774385909760.0,
"train_loss": 0.7816258675974271,
"train_runtime": 9729.0212,
"train_samples": 16610,
"train_samples_per_second": 2.223,
"train_steps_per_second": 0.035
}