Qwen2.5-R1-Distill-GRPO-h / train_results.json
samitizerxu's picture
Model save
4c7f5a8 verified
raw
history blame contribute delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": 4.045824278607414e-05,
"train_runtime": 33605.9584,
"train_samples": 127,
"train_samples_per_second": 0.004,
"train_steps_per_second": 0.001
}