All the winds of change are blowing these days in the AI/ML space. Seriously, things are changing very rapidly. I’m contemplating running some Auto-GPT on my desktop with the latest Llama 2 model maybe. Right now it would call out to OpenAI via API. Right now I’m starting to dig into this GitHub that is relevant to that aim. 

https://github.com/rhohndorf/Auto-Llama-cpp

That project could one day be very interesting, but right now it’s way more POC than delivered. 

Things for me shifted a bit and I ended up looking at this Google Colab notebooks:

https://github.com/camenduru/text-generation-webui-colab/blob/main/README.md

https://colab.research.google.com/github/camenduru/text-generation-webui-colab/blob/main/llama-2-7b-chat.ipynb (this took about 5 minutes to run)

At this point, I’m wondering what other things people do with Gradio that are interesting. 

Apparently, you can go out to https://huggingface.co/spaces and see a ton of spaces that showcase some of the ML apps made by the community at large.

It looks like people are making Spaces over on Hugging Face that allow people to run and execute Auto-GPT. A lot of these were built out during April of 2023.

Need to spend time this week digging into localGPT and privateGPT installations for my main workstation.

Content consumed this week:

“The Impact of chatGPT talks (2023) – Keynote address by Prof. Yann LeCun (NYU/Meta)” https://www.youtube.com/watch?v=vyqXLJsmsrk&list=PLKemzYMx2_Ot1MZ_er2vFiINdJEgDO8Hg
“Llama 2: Open Foundation and Fine-Tuned Chat Models” https://arxiv.org/pdf/2307.09288.pdf

“Stanford CS229 Machine Learning I Naive Bayes, Laplace Smoothing I 2022 I Lecture 6” https://www.youtube.com/watch?v=ADj95edZc0w