Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
Blancy's picture
Model save
287364c verified
{
"total_flos": 0.0,
"train_loss": 0.011807907519011864,
"train_runtime": 16456.6289,
"train_samples": 1000,
"train_samples_per_second": 0.061,
"train_steps_per_second": 0.002
}