Qwen2.5-1.5B-Open-R1-Distill / train_results.json
aimarbenzemamessi's picture
Model save
c85dbab verified
{
"total_flos": 76973799899136.0,
"train_loss": 0.8025265755738027,
"train_runtime": 1304.1525,
"train_samples": 16610,
"train_samples_per_second": 16.58,
"train_steps_per_second": 0.13
}