Oh boy I could really use an LLM fine-tuned on the NIPS docs right now.
Discussion
* RAG’d, not fine-tuned. Although that might work too
🤔 what I do is use a custom GPT that I just add the link to the NIPs GitHub repo as part of the system prompt. Works well for me. Just gotta say “please search the Nostr NIPs repo before answering”
Downloading and fine tuning may be overkill