In my comment I've been referencing https://arxiv.org/pdf/2005.14165.pdf, specifically section 3.9.1 where they summarize results of the arithmetic tasks.

Check out this one: https://thegradient.pub/othello/

In it, researchers built a custom LLM trained to play a board game just by predicting the next move in a series of moves, with no input at all about the game state. They found evidence of an internal representation of the current game state, although the model had never been told what that game state looks like.

GPT3 is pretty bad at it compared to alternatives (although it's hard to compete with calculators on that field), but if it was just repeating after the training dataset it would be way worse. From the study I've linked in my other comment (https://arxiv.org/pdf/2005.14165.pdf):

On addition and subtraction, GPT-3 displays strong proficiency when the number of digits is small, achieving 100% accuracy on 2 digit addition, 98.9% at 2 digit subtraction, 80.2% at 3 digit addition, and 94.2% at 3-digit subtraction. Performance decreases as the number of digits increases, but GPT-3 still achieves 25-26% accuracy on four digit operations and 9-10% accuracy on five digit operations, suggesting at least some capacity to generalize to larger numbers of digits.

To spot-check whether the model is simply memorizing specific arithmetic problems, we took the 3-digit arithmetic problems in our test set and searched for them in our training data in both the forms " + =" and " plus ". Out of 2,000 addition problems we found only 17 matches (0.8%) and out of 2,000 subtraction problems we found only 2 matches (0.1%), suggesting that only a trivial fraction of the correct answers could have been memorized. In addition, inspection of incorrect answers reveals that the model often makes mistakes such as not carrying a “1”, suggesting it is actually attempting to perform the relevant computation rather than memorizing a table.

LLMs do replicate a small subset of human cognition, but not the full scope. This can result in human-like behavior, but it’s important to be aware of the limitations.

The biggest limitation is the misalignment in goals. LLMs won’t perform a very deep analysis of their input because they don’t need to. Their goal isn’t honest discussion, a pursuit for truth, or even having a coherent set of beliefs about the world. Their only goal is to sound plausible. And, as it turns out, it’s not too hard to just bullshit your way through the Turing test.

That ignores all the papers on emergent features of LLMs and the fact they are basically black boxes. Yes, we “trained” them to write what we want to hear. But we don’t really understand what happens inside of it. We can’t categorically claim things like “they are only regurgitating what they heard”. Because that is not a scientific or even philosophical statement.

If you think about it for a second, it’s also applicable to human beings…

I've actually developed quite a bit with gpt4 and have beta access and have developed quite some fancy prompts if I do say so myself.

Telling me 'isn't it obvious' doesn't make it more obvious to me.

Hey, that comment's a bit off the mark. Transformers don't just memorize chunks of text, they're way more sophisticated than that. They use attention mechanisms to figure out what parts of the text are important and how they relate to each other. It's not about memorizing, it's about understanding patterns and relationships. The paper you linked doesn't say anything about these models just regurgitating information.