Qwen3-0.6B-Open-R1-GRPO / train_results.json
Blancy's picture
Model save
1dcb3d0 verified
{
"total_flos": 0.0,
"train_loss": 0.014487237919349916,
"train_runtime": 13938.4252,
"train_samples": 1000,
"train_samples_per_second": 0.072,
"train_steps_per_second": 0.003
}