Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
lihuigu's picture
Model save
ac9ef5e verified
raw
history blame contribute delete
198 Bytes
{
"total_flos": 0.0,
"train_loss": 0.03553462658770556,
"train_runtime": 3215.6389,
"train_samples": 175,
"train_samples_per_second": 0.054,
"train_steps_per_second": 0.013
}