Qwen-2.5-7B-Simple-RL / train_results.json
zijianh's picture
Model save
80941cc verified
{
"total_flos": 0.0,
"train_loss": 0.04329907550508606,
"train_runtime": 7330.3865,
"train_samples": 7500,
"train_samples_per_second": 1.023,
"train_steps_per_second": 0.008
}