Fred Turkington

icon indicating a website link http://z3ugma.github.io

icon location Madison, WI icon location Digital health, Flask web development, binary file format reverse engineering, HL7

Results 22 comments of Fred Turkington

Encountered flash_attn_2_cuda error while running finetune_lora.sh

your error is > flash_attn_2_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZN3c104cuda9SetDeviceEi Check out https://github.com/Dao-AILab/flash-attention/issues/451

Experiment with implementing AWQ for BERT models

Would love this for image captioning with quantized speedup. The `kosmos-2` model from Microsoft would be another good candidate