Qwen2.5-0.5b-GRPO-math / all_results.json
chinmaydk99's picture
Model save
545f519 verified
raw
history blame contribute delete
202 Bytes
{
"total_flos": 0.0,
"train_loss": 0.00017194549279035223,
"train_runtime": 4065.9132,
"train_samples": 45000,
"train_samples_per_second": 0.59,
"train_steps_per_second": 0.037
}