Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
dadadar's picture
Model save
d994b6f verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": 2591852.7904117624,
"train_runtime": 66018.0639,
"train_samples": 20000,
"train_samples_per_second": 0.303,
"train_steps_per_second": 0.038
}