This repository is used to manage the presentations given at Huntsville AI meetups. It provides a collection of Issues, Cards, and Files to plan and create the content needed for a presentation.
This week we'll be talking about the release of a LLM that is licensed under Apache 2.0 and available through HuggingFace. For those on the Discord channel (see below for how to join), there's been a good discussion of when (not if) the open source community or another third party would be able to match the models provided by Google, Meta, & OpenAI in a way that could be used commercially.
As you can see, even this latest news is a bit dated by current AI standards. Over the last few days, there have been 2 major models that we've been tracking:
The MPT-7B is comparable with LLaMA-7B from Meta and available under the Apache 2.0 license - making it the first I've seen that is available to repackage and run locally in your own stack. There is also a chat based model available from MosiacLM under a more restrictive CC-By-NC-SA-4.0 (non-commercial use only) license, but at least we're headed in the right direction!
This week we will attempt to load and use the MPT-7B model and see what we can do with it.
Complete the following items to get a presentation ready for Huntsville AI
Add the file to present (prefer Jupyter Notebooks or Markdown formated files) to the folder structure. For multiple files, create a directory following the naming convention and add the files to it.
Naming convention
We use a convention of starting the filenames with a date (year/month/day) so that the files are still sorted by date even when in alphabetical format.
jperiodlangley
commented
1 year ago 
Description:
This week we'll be talking about the release of a LLM that is licensed under Apache 2.0 and available through HuggingFace. For those on the Discord channel (see below for how to join), there's been a good discussion of when (not if) the open source community or another third party would be able to match the models provided by Google, Meta, & OpenAI in a way that could be used commercially.
Based on a leaked Google document, the timeline of progress in this direction is below. The article is worth a few minutes to read: https://simonwillison.net/2023/May/4/no-moat/
As you can see, even this latest news is a bit dated by current AI standards. Over the last few days, there have been 2 major models that we've been tracking:
https://huggingface.co/bigcode/starcoder https://huggingface.co/mosaicml/mpt-7b The StarCoder model is similar to the one used in GitHub CoPilot, AWS CodeWhisperer, etc. It uses the BigCode Open RAIL-M v1 License Agreement.
The MPT-7B is comparable with LLaMA-7B from Meta and available under the Apache 2.0 license - making it the first I've seen that is available to repackage and run locally in your own stack. There is also a chat based model available from MosiacLM under a more restrictive CC-By-NC-SA-4.0 (non-commercial use only) license, but at least we're headed in the right direction! This week we will attempt to load and use the MPT-7B model and see what we can do with it.
Complete the following items to get a presentation ready for Huntsville AI
Adding material to the presentations repository
Add the file to present (prefer Jupyter Notebooks or Markdown formated files) to the folder structure. For multiple files, create a directory following the naming convention and add the files to it.
Naming convention
We use a convention of starting the filenames with a date (year/month/day) so that the files are still sorted by date even when in alphabetical format.
YYMMDD_Session_Description.extension