Qwen2.5-1.5B-Open-R1-GRPO / train_results.json
AaronHuangWei's picture
Model save
0ab1069 verified
raw
history blame contribute delete
200 Bytes
{
"total_flos": 0.0,
"train_loss": 1.286716509427883,
"train_runtime": 229250.8975,
"train_samples": 72441,
"train_samples_per_second": 0.316,
"train_steps_per_second": 0.003
}