Michael O'Mahony

michaelomahony's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes

New activity in tiiuae/falcon-7b-instruct about 1 year ago

Performance reduction from using 8bit or 4bit quantized model

#58 opened about 1 year 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 over 1 year ago by zerolyn

Slow inference

9

#33 opened over 1 year ago by BigArt

New activity in ichitaka/falcon-40b-instruct-8bit about 1 year ago

How did you manage to quantize the model?

7

#3 opened about 1 year ago by SaffalPoosh

New activity in tiiuae/falcon-7b-instruct over 1 year ago

Output formatting not enforceable

1

#43 opened over 1 year ago by Rick458

4th inference in a row does not work for Falcon7B in 8 or 4 bit

2

#31 opened over 1 year ago by max0uu