Qwen-2.5-7B-Simple-RL / all_results.json
zhimeng's picture
Model save
6303069 verified
{
"total_flos": 0.0,
"train_loss": 0.034317180164094664,
"train_runtime": 31905.6637,
"train_samples": 7500,
"train_samples_per_second": 0.235,
"train_steps_per_second": 0.004
}