Qwen2.5-1.5B-Open-R1-Distill / train_results.json
Thomas-Chou's picture
Model save
4609027 verified
raw
history blame
213 Bytes
{
"total_flos": 76973799899136.0,
"train_loss": 0.8022714451219909,
"train_runtime": 1175.8601,
"train_samples": 16610,
"train_samples_per_second": 18.389,
"train_steps_per_second": 0.144
}