Qwen2.5-0.5B-Open-R1-GRPO / train_results.json
jdqqjr's picture
Model save
69d8ac2 verified
raw
history blame contribute delete
197 Bytes
{
"total_flos": 0.0,
"train_loss": 0.003593232001742867,
"train_runtime": 3630.59,
"train_samples": 72441,
"train_samples_per_second": 0.2,
"train_steps_per_second": 0.017
}