Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
caijanfeng's picture
Model save
c47a239 verified
raw
history blame contribute delete
202 Bytes
{
"total_flos": 0.0,
"train_loss": 0.027384539475202127,
"train_runtime": 37232.0457,
"train_samples": 72441,
"train_samples_per_second": 1.946,
"train_steps_per_second": 0.008
}