Qwen2.5-3B-Open-R1-GRPO / train_results.json
wlzhou's picture
Model save
514375f verified
raw
history blame contribute delete
196 Bytes
{
"total_flos": 0.0,
"train_loss": 1.0999555667603271,
"train_runtime": 38247.6092,
"train_samples": 20,
"train_samples_per_second": 0.209,
"train_steps_per_second": 0.01
}