sharegpt VS ChatGPT

he also created cool projects like https://oneword.domains/, https://sharegpt.com/, https://novel.sh/ and https://extrapolate.app/

Vicuna is another instruction-focused LLM rooted in LLaMA, developed by researchers from UC Berkeley, Carnegie Mellon University, Stanford, and UC San Diego. They adapted Alpaca’s training code and incorporated 70,000 examples from ShareGPT, a platform for sharing ChatGPT interactions.

We can say that a 13B model per language is reasonable. Then it means we need to create a democratic way for teaching coding by examples and solutions and algorithms, that we create, curate and use open-source. Much like sharegpt.com but for coding tasks, solutions ways of thinking. We should be wary of 'enforcing' principles rather showing different approaches, as all approaches can have advantages and disadvantages.

You can see the url in the comment, https://sharegpt.com and if you go there it gives you the option for installing the chrome extension, after that it shouldn’t be hard to use it

ShareGPT - Share your prompts and your entire conversations

Or used https://sharegpt.com

you can gets lots of ideas here > https://sharegpt.com/ (180,000+ prompts)

I would love to read the whole conversation: Check out this cool little GPT sharing extension: https://sharegpt.com - that way the code snippets can be copied easily

> Cool, that’s really the only point I’m making.

To be clear, I'm saying that I don't know if they are, not that we know that it's not the same.

It's not at all clear that humans do much more than "that basic token sequence prediction" for our reasoning itself. There are glaringly obvious auxiliary differences, such as memory, but we just don't know how human reasoning works, so writing off a predictive mechanism like this is just as unjustified as assuming it's the same. It's highly likely there are differences, but whether they are significant remains to be seen.

> Not necessarily scaling limitations fundamental to the architecture as such, but limitations in our ability to develop sufficiently well developed training texts and strategies across so many problem domains.

I think there are several big issues with that thinking. One is that this constraint is an issue now in large part because GPT doesn't have "memory" or an ability to continue learning. Those two need to be overcome to let it truly scale, but once they are, the game fundamentally changes.

The second is that we're already at a stage where using LLMs to generate and validate training data works well for a whole lot of domains, and that will accelerate, especially when coupled with "plugins" and the ability to capture interactions with real-life users [1]

E.g. a large part of human ability to do maths with any kind of efficiency comes down to rote repetition and generating large sets of simple quizzes for such areas is near trivial if you combine an LLM at tools for it to validate its answers. And unlike with humans where we have to do this effort for billions of humans, once you have an ability to let these models continue learning you make this investment in training once (or once per major LLM effort).

A third is that GPT hasn't even scratched the surface in what is available in digital collections alone. E.g. GPT3 was trained on "only" about 200 million Norwegian words (I don't have data for GPT4). Norwegian is a tiny language - this was 0.1% of GPT3's total corpus. But the Norwegian National Library has 8.5m items, which includes something like 10-20 billion words in books alone, and many tens of billions more in newspapers, magazines and other data. That's one tiny language. We're many generations of LLM's away from even approaching exhausting the already available digital collections alone, and that's before we look at having the models trained on that data generate and judge training data.

[1] https://sharegpt.com/

Be careful. Back when ChatGPT was in beta, a similar project was resulting in mass account bans. They really don't like people using unofficial APIs.

I use chatgpt, and I mainly use a tool on github called ChatGPT , which is basically a reverse engineered one that you could use on your terminal with your ChatGPT API key. So i thought on running the tool and ask it, this was the first response I got

Have a look at this project. Also worth checking out his other repos, such as this one (not sure what exactly is different) or this one for the bing chat.

Or https://github.com/acheong08/ChatGPT of course.

Sure, first thing is to get this in V1 mode running such that you can chat via the command line. This is currently not working for me though, could be because OpenAl changed something on their chat site or because I'm messing something up. One that is running I can show you how to setup a small project locally that looks like the OpenAl API but then relays queries to the reverse engineered chat API using the ChatGPT package

What does this repo https://github.com/acheong08/ChatGPT actually do? I thought in the end it gives me an API key that I can use for apps that work with the GPT-API, but that is stated nowhere on the repo.

I asked chatGPT to code a 2 speaker debate function using the unofficial API from revChatGPT ( https://github.com/acheong08/ChatGPT )

About chatgpt-wrapper, I think you can migrate to https://github.com/acheong08/ChatGPT. It is lighter with a query limit, but I think it is still good.