Hacker News Reader

Comment by pqdbr | Hacker News Reader

Page is down, but here is a description of the service by the founder himself:

https://www.producthunt.com/posts/customgpt

Does anyone knowledgeable in AI know how feasible the claim of training your custom ChatGPT really is? Because this would be the holy grail we’ve all been waiting for, no?

View on HN

Training a custom chatbot is easy. Having it reliably retrieve the data you want is hard. For any given question, you can get it to look at a maximum of 4k tokens (16k characters, or 3k or so words). So when you want your chatbot to have access to a LOT of data, you usually end up having to work around that - typically by:

1. chunking up your documents into ~2k word chunks, 2. and then getting the query, doing a search using that query, and getting the top N results back, 3. running the prompt on each result, and then 4. gradually combining the results of each run of the prompt.

Each step is a little lossy, and there are quite a few steps there, so the bigger the corpus gets, the lossier/less useful your bot is.

Now that being said, there are ways to improve that, but for now at least, its unlikely that a single slackbot is going to solve all your knowledge management problems for you.

My guess is that they are simply finetuning the model on your provided data: https://platform.openai.com/docs/guides/fine-tuning

I wonder if it's actually "trained" or is it rather "imagine you've generated all that fancy text content, which follows now: ${content}".

I also though of this (prompt engineering), but I think that would hit the token limit pretty fast.

Reply on news.ycombinator.com