
hehehehe
@luck_not_shit
Followers
77
Following
213
Media
10
Statuses
107
RT @vpj: Wrote an annotated Triton implementation of Flash Attention 2. (Links in reply). This is based on the flash attention implementati….
0
5
0
RT @janetacarr: docker is supposed to solve the "works on my machine" problem. but often I find that it just adds another layer to the "wor….
0
104
0
I need to let a LLM "talk" to swift core data. Need a language both the DB and the LLM talks so the obvious solution is SQL. SQL won't work on a key value store though. I wonder how hard would it be to write a SQL like driver for core data. #DoingdumbShitTillImNotDumbAnymore.
1
0
0
Pip package: Docs: Github:
github.com
LLM Analytics. Contribute to labmlai/inspectus development by creating an account on GitHub.
We’ve open-sourced our LLM attention visualization library. It generates interactive visualizations of attention matrices with just a few lines of Python code in notebooks. @luck_not_shit cleaned up and polished the existing code to make it open source.
0
0
0
Now Our visualization library Inspectus can visualize values related to tokens in LLM outputs. This demo shows some outputs from using entropyx (by @_xjdr) on Llama 3. Had fun making this. (jk I didn’t). 🔗👇
1
1
6