Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
Emo3331's picture
Model save
4f70696 verified
raw
history blame contribute delete
199 Bytes
{
"total_flos": 0.0,
"train_loss": 27.60521244561175,
"train_runtime": 16907.8148,
"train_samples": 72441,
"train_samples_per_second": 4.284,
"train_steps_per_second": 0.038
}