Qwen2.5-3B-Open-R1-GRPO / train_results.json
LlameUser's picture
Model save
0e1ca88 verified
{
"total_flos": 0.0,
"train_loss": 12868.4468021695,
"train_runtime": 382532.2881,
"train_samples": 40893,
"train_samples_per_second": 0.107,
"train_steps_per_second": 0.007
}