Qwen2.5-7B-SFT / all_results.json
YYT-t's picture
Model save
0ad28ad verified
{
"total_flos": 1.846643377789993e+17,
"train_loss": 0.30200490220900506,
"train_runtime": 1464.9032,
"train_samples_per_second": 6.741,
"train_steps_per_second": 0.106
}