Michael O'Mahony
michaelomahony
AI & ML interests
None yet
Organizations
None yet
michaelomahony's activity
Performance reduction from using 8bit or 4bit quantized model
#58 opened 12 months ago
by
michaelomahony
what is the prompt used for instruction tuning, and why the model is pre-trained on refineweb but also instruction-tuned with it?
3
#30 opened about 1 year ago
by
zerolyn
Slow inference
9
#33 opened about 1 year ago
by
BigArt
How did you manage to quantize the model?
7
#3 opened 12 months ago
by
SaffalPoosh
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1652562749455-noauth.png)
Output formatting not enforceable
1
#43 opened about 1 year ago
by
Rick458
![](https://cdn-avatars.huggingface.co/v1/production/uploads/1644391889370-620292ea9dab2e6e083d031f.jpeg)
4th inference in a row does not work for Falcon7B in 8 or 4 bit
2
#31 opened about 1 year ago
by
max0uu