Qwen-2.5-7B-Simple-RL / train_results.json
Ligeng-Zhu's picture
Model save
692c299 verified
{
"total_flos": 0.0,
"train_loss": -0.00014772719968559928,
"train_runtime": 12261.9552,
"train_samples": 7500,
"train_samples_per_second": 0.612,
"train_steps_per_second": 0.005
}