Live data from Hacker News

Show HN: WhyBot, making GPT-4 question itself

whybot-khaki.vercel.app

1–10 of 34 posts

Show HN: WhyBot, making GPT-4 question itself

#1
Hi HN — we’re John and Vish! We built WhyBot, a tool to help you deeply explore a question or topic. You ask a question, and WhyBot responds by building an ever-expanding knowledge graph. It does this by recursively generating answers and follow-up questions. You can change its persona to change the flavor of the generations (try toddler mode!).

We originally built this for the AngelList Agent Hackathon (https://twitter.com/AqeelMeetsWorld/status/16502799744050421...) and got a lot of interest from folks asking to play around with it. So we thought it’d be fun to brush it up and release it as a web app! It’s a work in progress and we plan on adding more features, such as saving, sharing, focusing on one branch and potentially executing code.

We hope you enjoy playing around with it and would love to hear any of your feedback or thoughts.

Show HN: WhyBot, making GPT-4 question itself
whybot-khaki.vercel.app

Re: Show HN: WhyBot, making GPT-4 question itself

#5
This is incredibly!

I've been benchmarking several LLMs to identify a possible bug in a snippet of code given no context on what is expected from the code. Neither chatgpt, vacuna, open assistant or mpt where able to identify the bug. When I fed the same prompt into this app, the initial reply said there were no bugs, but the first branch immenditly questioned itself with the expected line of thought to identify the bug!

I might subscribe to chatgpt just to use this app, great work!

I'd love it if this got open sourced.

Here's the prompt if you want to test it out:

Can you find any bugs in the following snippet?

```ts

const queuePositions = [

1,

2,

3

];

const pairs: number[][] = [];

for (let i = 0; i pairs.push(queuePositions.slice(i, i + 2));

```

Re: Show HN: WhyBot, making GPT-4 question itself

#7
I have not seen a better demonstration on visualizing unlimited/ open ended reasoning like this.

I really love this.

I'm very curious (and I thought about this before which shows the Zeitgeist) if this can change or enhance the general reasoning skills of people.

Our generation already google things constantly, so much that an older couple told me once that their kids don't ask them stuff but Google it.

But the reasoning aspect was missing.

Now you can't just Google facts but also ask reasoning questions without worries.

Ignoring potential miss guidence in these models and hopping for an emerging reasoning truth through wisdom of the masses.

Self fulfilling prophecy perhaps?

Post reply on HN