Peter Kowalski
Peter Kowalski
> But there is a problem that it doesn't seem to stop by itself, it will continue to generate the next line of `### Human` and continuing another response. I...
Yes, I'm talking about rev1, so we need to change llama_token_eos() or what? Alternative solution: https://github.com/ggerganov/llama.cpp/commit/9fd062fd2e7e9f2f14d66c3d64dce3f967604103 [UNVERIFIED] I've recompiled main.cpp with this patch, and it works for me well enough...
> This is just the same as the `--antiprompt` option. The patch above has `\n\n` in the end but I think without them it would be better. You can also...
> I've repeated your modifications, but nothing changes - it still shows "### Human:" each time... Who knows how to make clean answers? ( So that only '>' instead of...
I asked ALPACA: "Why did you choose this name?" Answer: John Smith was a popular choice for spies during World War II as it could be easily changed to fit...
You can start your prompts with something like "Your name is Adam." etc.