Qwen-2.5-7B-Simple-RL / train_results.json
coolcui's picture
Model save
839c63e verified
{
"total_flos": 0.0,
"train_loss": 0.048550591330397234,
"train_runtime": 12478.5367,
"train_samples": 7500,
"train_samples_per_second": 0.601,
"train_steps_per_second": 0.005
}