Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
byteXWJ's picture
Model save
6805b54 verified
{
"total_flos": 0.0,
"train_loss": 0.3506326697535499,
"train_runtime": 41982.506,
"train_samples": 72441,
"train_samples_per_second": 1.726,
"train_steps_per_second": 0.015
}