dag_qwen_sft_v0 / all_results.json
aaabiao
add model
1c46159
{
"epoch": 1.9937952430196484,
"total_flos": 3.6089290785072087e+18,
"train_loss": 0.3119487636316861,
"train_runtime": 2571.753,
"train_samples_per_second": 24.051,
"train_steps_per_second": 0.187
}