Qwen2.5-0.5B-Open-R1-GRPO / train_results.json
zzzch's picture
Model save
40836d4 verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": 0.006115545869233169,
"train_runtime": 3082.8277,
"train_samples": 72441,
"train_samples_per_second": 0.324,
"train_steps_per_second": 0.02
}