Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
shuheikurita's picture
Model save
6a84fbf verified
{
"total_flos": 0.0,
"train_loss": 0.017702588886127465,
"train_runtime": 38308.166,
"train_samples": 72441,
"train_samples_per_second": 1.891,
"train_steps_per_second": 0.007
}