having bad LLMs can allow us to find truth faster. reinforcement algorithm could be: "take what a proper model says and negate what a bad LLM says". then the convergence will be faster with two wings!
Are the nostr data sets you are using to train llms public?
Please Login to reply.
datasets are not. but notes are public..
Why? 👀
not uploading datasets could result in more diverse LLMs based on Nostr. which i prefer at this point.
