Qwen2.5-0.5B-Open-R1-Code-GRPO / train_results.json
js2025's picture
Model save
fbf8a09 verified
{
"total_flos": 0.0,
"train_loss": 0.00930938980280871,
"train_runtime": 450.9542,
"train_samples": 16,
"train_samples_per_second": 0.035,
"train_steps_per_second": 0.009
}