When I loaded the smallest GPT-2 model weights from Hugging Face and performed inference using both flash attention and a manually implemented attention under the same seed setting, I obtained consistent results within each method individually. However, the results between the two methods were not consistent, and the manually implemented attention seemed to produce more reasonable outputs. Is this normal?
When I loaded the smallest GPT-2 model weights from Hugging Face and performed inference using both flash attention and a manually implemented attention under the same seed setting, I obtained consistent results within each method individually. However, the results between the two methods were not consistent, and the manually implemented attention seemed to produce more reasonable outputs. Is this normal?