·   · 731 posts
  •  · 0 friends
  •  · 0 followers

merry christmas and happy new year eve

 

 

https://www.tiktok.com/@huggyattack/video/7367979440548187435

jlos shooting star was an offspring of a kpop group.

since the script is doing well and the both tesla m40s are heating up too much, i've purchaised two p40s. to get executable scripts isn't that easy and there are no free llms which can make code refraction, see https://aider.chat/docs/leaderboards/refactor.html

"The refactoring benchmark requires a large context window to work with large source files. Therefore, results are available for fewer models."

there are also smaller coding LLM with large context window size. the DeepSeek Coder V2 0724 (deprecated) q4 version needs 144gb vram.

#z-stammtisch

powered by the-stammtisch.com

0 0 0 0 0 0
  • 15
Comments (0)
    Info
    Category:
    Created:
    Updated:
    Featured Posts
    the first proof of concept semi-liquid neural network (advanced LLM) run positive (SELI NEORAD for short)
    https://github.com/DuskoPre/AutoCoder/wiki i'm not just testing LLMs but also creating a semi-liquid neural network (advanced LLM) with chromadb: https://github.com/DuskoPre/liquid-neural-network-with-chromadb-cache it seems there is a mayor upgrade for LLMs possible trough classifications (managing-system) of inputs to the LLM. the model is still static but trough the implementations of the classifier it can be made semi-liquid.   #dichipcoin
    Free Website Counter
    Free Website Counter
    SARS-CoV-2 web counter web counter