Qwen3-1.7B-Open-R1-Code-GRPO / train_results.json
Blancy's picture
Model save
f71573d verified
{
"total_flos": 0.0,
"train_loss": 0.04123398642398541,
"train_runtime": 18693.2835,
"train_samples": 1086,
"train_samples_per_second": 0.058,
"train_steps_per_second": 0.002
}