James Snodgrass
James Snodgrass
Check your account with your LLM provider. They have hard limits on number of requests, context window, token usage, etc. Sometimes their error message in the API is cryptic like...
It looks like you have 2 options. You can use an LLM through an API or you can run an LLM locally. If you don't have a GPU, I recommend...
Yes, I just watched someone use it to make a full stack website.
> Hey @Whytehorse2022, would you like to share the link? https://www.youtube.com/watch?v=iwLe6UWyaS4
It does have limitations, bugs, and costs. I like to break it down into boiler plate code generation. I might have it make me a login page with profiles. It...
I'm also interested in a description of the training data.
The documentation says it is not optimized for any particular task. It does say it has 8 bit quantization. It's unaligned. So the answer is no, it isn't crippled.
> @Whytehorse2022 if this is an issue with using my fork, please open an issue in that repo, I don't want to pollute this thread with changes that I'm making...
Or here's a github project that allows you to hover over math and get mathML rendering. Pretty slick. https://github.com/yousefvand/mathover