sweep VS StableLM

What's the difference with https://sweep.dev ?

- https://sweep.dev have a bot where you create a GitHub comment and it writes the PR to fix it. They have between 30% and 70% success rate. This is pretty bad but they're one of the best today

6️⃣ Sweep AI 🧹

https://github.com/sweepai/sweep arguably already does the work of an "average" programmer

The GitHub repo I researched and found was a repo that utilizes UI to create code changes based on user bug reports https://github.com/sweepai/sweep. This feature works with small errors and is aimed to eliminate developers from opening tickets that involve small bug fixes or refractors. I know that learning a new language often has many set backs and learning curves. After finding this project I believe I can learn some basics of Python while also utilizing UI.

I wanted to share how we got Sweep, an AI-powered junior developer, to use the latest pytorch documentation. We wrote a simple webcrawler using Playwright, which bypasses a lot of issues while scraping. Also, because we only need to crawl documentation, we can filter out common tags in markdown. This lets us accurately crawl and index any docs site. Try it now at https://github.com/sweepai/sweep! You can add docs by going to your sweep.yaml and adding both the keyword and the url.

We have that! You can directly append to our system prompt here to customize Sweep https://github.com/sweepai/sweep/blob/main/sweep.yaml

https://github.com/Stability-AI/StableLM?tab=readme-ov-file#...

https://mistral.ai/news/announcing-mistral-7b/

looking at the 3b results (here https://github.com/Stability-AI/StableLM#stablelm-alpha-v2 ?), it looks like Mistral (which outperforms Llama-2 13b) is far more powerful

Does this mean Stability gave up on StableLM?

I notice that the repo hasn’t been updated since April, and a question asking for an update has been ignored for at least a month: https://github.com/Stability-AI/StableLM/issues/83

I'm not "ignoring" StableLM, if anything it's the impetus for my post. The alpha models were so bad and unusable that it seems they may have simply abandoned the project. It's clear they basically didn't know what they were doing, which is silly for a company of their size and specialization.

1) StableLM released a checkpoint at 800B for their 3B and 7B at 800B tokens with 4096 context size, but perform very poorly on different benchmarks and finetuning is discouraged with such a weak base model

It is the best open-source model currently available. Falcon-40B outperforms LLaMA, StableLM, RedPajama, MPT, etc. See the OpenLLM Leaderboard.

Someone shared this link on another recent post

Github: https://github.com/Stability-AI/StableLM