Qwen-2.5-7B-base-Simple-RL / train_results.json
kkish's picture
Model save
434c1cb verified
{
"total_flos": 0.0,
"train_loss": 0.0013394545973730192,
"train_runtime": 6585.1769,
"train_samples": 7500,
"train_samples_per_second": 1.139,
"train_steps_per_second": 0.009
}