Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Factual AI Q&A – Answers based on Huberman Lab transcripts

huberman.rile.yt

Post reply on HN
Print view View original

41–45 of 45 posts

Re: Show HN: Factual AI Q&A – Answers based on Huberman Lab transcripts

#41

Post by rohit89 »

jonathan-adly wrote: post #22

I did the same for the FDA drug label database and 100% believe that this the future for search. Semantic search layer for context then the large language layer for human answers. Tip - you don’t actually need GPT-3 level embedding for a decent semantic search. Sentence transformers paired with one of their models is good enough. I like this: https://huggingface.co/sentence-transformers/multi-qa-MiniLM... - since it’…

Did you use the sentence-transformers model as-is or did you need to fine-tune it for medical data?

Re: Show HN: Factual AI Q&A – Answers based on Huberman Lab transcripts

#42

Post by rileyt »

jonathan-adly wrote: post #22

I did the same for the FDA drug label database and 100% believe that this the future for search. Semantic search layer for context then the large language layer for human answers. Tip - you don’t actually need GPT-3 level embedding for a decent semantic search. Sentence transformers paired with one of their models is good enough. I like this: https://huggingface.co/sentence-transformers/multi-qa-MiniLM... - since it’…

It was less than $2 to embed all 100+ episodes with the new OpenAI embeddings and was as easy as just making a bunch of API calls. Pretty hard to beat that experience.

Re: Show HN: Factual AI Q&A – Answers based on Huberman Lab transcripts

#43

Post by krashidov »

rileyt wrote: post #14
krashidov wrote: post #8

This is a really amazing application of GPT. Did you fine tune a gpt3 model? If so, how did you implement its ability to say “I don’t know?”

It's not fine tuned. You literally just add something like "if the answer to the question isn't in the context, say 'I don't know'" It's wild.

So do you have the entire Huberman podcast transcript in the context of the prompt?

Re: Show HN: Factual AI Q&A – Answers based on Huberman Lab transcripts

#44

Post by jonathan-adly »

rohit89 wrote: post #41
jonathan-adly wrote: post #22

I did the same for the FDA drug label database and 100% believe that this the future for search. Semantic search layer for context then the large language layer for human answers. Tip - you don’t actually need GPT-3 level embedding for a decent semantic search. Sentence transformers paired with one of their models is good enough. I like this: https://huggingface.co/sentence-transformers/multi-qa-MiniLM... - since it’…

Did you use the sentence-transformers model as-is or did you need to fine-tune it for medical data?

As is.. all it’s doing is pulling relevant context to the question. GPT-3 is doing all the heavy natural language lifting.
Post reply on HN