Your website's content -> Q&A bot / chatbot

#1

Post by mpaepper » Tue, Mar 21, 2023, 8:42 PM UTC

Your website's content -> Q&A bot / chatbot
github.com

Re: Your website's content -> Q&A bot / chatbot

#2

Post by layoric » Tue, Mar 21, 2023, 10:43 PM UTC

Nice to have tools like this to wrap up features, definitely makes these types of solutions more accessible, thanks!

It would be nice to know from your experience if there is a kind of rule of thumb for calculating cost of fine tuning and running a solution like this against a docs site?

Re: Your website's content -> Q&A bot / chatbot

#3

Post by wedn3sday » Tue, Mar 21, 2023, 11:01 PM UTC

I would absolutely love to take our internal Wiki and use this against it.

Re: Your website's content -> Q&A bot / chatbot

#4

Post by nico » Tue, Mar 21, 2023, 11:37 PM UTC

Awesome!

Are you planning on adding agent/tools support?

It would be cool to use this with internal data, then allow clients to chat with a bot fine-tunes on their data, but that can also run queries, or get reports for specific dates, or charts, all via tools.

Re: Your website's content -> Q&A bot / chatbot

#5

Post by mdotk » Wed, Mar 22, 2023, 12:18 AM UTC

Make this a Wordpress plugin and I'd pay for it

Re: Your website's content -> Q&A bot / chatbot

#6

Post by mdmglr » Wed, Mar 22, 2023, 12:21 AM UTC

Re: Your website's content -> Q&A bot / chatbot

#7

Post by mdmglr » Wed, Mar 22, 2023, 12:47 AM UTC

mdotk wrote: post #5

Make this a Wordpress plugin and I'd pay for it

Along with paying the openAI API fees? Perhaps a cache layer for common qa.

Re: Your website's content -> Q&A bot / chatbot

#8

Post by itake » Wed, Mar 22, 2023, 1:10 AM UTC

mdotk wrote: post #5

Make this a Wordpress plugin and I'd pay for it

Have you tested the recall of embedding search? I'm not sure about the latest results, but 1-2 years ago, it had 50-70% recall :-/

Re: Your website's content -> Q&A bot / chatbot

#9

Post by grogenaut » Wed, Mar 22, 2023, 1:43 AM UTC

Reading the readme makes me think it's only searching the top 4 most likely docs via the embeddings, not the wiki at any time? or am I misunderstanding how this works? With embeddings being close to just term vector matching via dot(?) product?

So basically get all the sub-prhases/sounds -> vector -> check vector db for closest matching documents -> send to gpt for summarization and answering the quetsion.

If that's ture wouldn't that have severe limitations with scattered information? I guess it would help you get answers and walk the data better than the "I don't even know the term" problem with google?

Re: Your website's content -> Q&A bot / chatbot

#10

Post by wmf » Wed, Mar 22, 2023, 2:16 AM UTC

grogenaut wrote: post #9

Reading the readme makes me think it's only searching the top 4 most likely docs via the embeddings, not the wiki at any time? or am I misunderstanding how this works? With embeddings being close to just term vector matching via dot(?) product? So basically get all the sub-prhases/sounds -> vector -> check vector db for closest matching documents -> send to gpt for summarization and answering the quetsion. If that's…

They probably took this approach because it's the only thing you can do with the OpenAI APIs (for now). Training your own corpus will be the way to go once it's possible.