AlexPerkin

Results 5 comments of AlexPerkin

Maybe this will help `https://github.com/zylon-ai/private-gpt/issues/1584`

It may be necessary to change the tokenizer according to [Proper tokenizer.model is absent ](https://huggingface.co/meta-llama/Meta-Llama-3-8B/discussions/12)

Try for example `node_parser = SentenceWindowNodeParser.from_defaults(window_size=20)`

Could you please provide an example chatdocs.yml for a custom LangChain prompt. Thank you in advance

For my part, I would like to offer a prompt designed for Llama 2, which significantly improves the quality of answers: ```shell retriever: custom_prompt: "[INST] Use the following pieces of...