Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
riddickz's picture
Model save
e2e5b66 verified
raw
history blame contribute delete
195 Bytes
{
"total_flos": 0.0,
"train_loss": -7.761021455128988e-10,
"train_runtime": 74.9189,
"train_samples": 10,
"train_samples_per_second": 0.4,
"train_steps_per_second": 0.04
}