rgs224

was able to get around waitlist: copilot-workspace.githubnext.com /<owner>/<repo>?task=<description>

this is very cool, reminds me of the tldraw feature

I kept having to debug prompt issues with open models

So I built OpenAI's Tokenizer page for all tokenizers on HuggingFace: Llama, Mistral, GPT2, MPT, Persimmon, T5 etc

http://tokenwiz.rahul.gs

One of the contributors here: here's a live demo: https://la-inst.caldera.app/

Check out the console for DOM updates streamed from the server. React is running on Node.