A friend once told me that virtue is like going to the gym. You practice daily, start with smaller weights (virtuous acts), and review how well you did on a regular basis. You ask "am I getting better at this?" rather than "am I morally perfect?"
If you aren't on the level of the moral greats, you start small and try to build up, the same way you'd start by running a 5k before running an ultramarathon.
I hope others out there find this viewpoint as helpful as I have.
Are these companies developing InfiniBand-class interconnects to pair with their custom chips? Without equivalent fabric, they can’t replace NVIDIA GPUs for large-scale training.
STEM often overlooks the fundamental work that was done in philosophy that led to breakthroughs within STEM. For example, Claude Shannon's undergraduate philosophy course is what taught him boolean algebra, which ultimately led him to design digital circuits.
https://bentley.umich.edu/news-events/magazine/the-elegant-p...
Hmm, yea, I agree with you on several points. For one, we don't fully understand the internal mechanisms of LLMs. I'm also with you on Markov chains and autocomplete tools not having an understanding of the underlying concepts. They merely use statistical patterns in the data.
Based on what you've said, it sounds like your take is that unless we can specify the exact mechanism by which LLMs understand, we have no business saying that they understand. In a lot of cases, this is a reasonable approach. In many areas, if someone tells you X, and you ask for a mechanism of action, and they can't produce one, you have solid grounds for thinking they're bullshitting.
But this case isn't quite the same. We know that LLMs learn to represent their inputs in a high-dimensional vector space (embeddings) and learn the relationships between those vectors. We also see them effectively solve problems in a variety of domains using this representation. I think these two ingredients: having a semantic representation and being able to effectively solve problems amount to something like "understanding." The lack of both properties is why I'd say Markov chains and autocomplete tools don't "understand" -- they haven't learned an effective representation of the underlying phenomena. (I'd also argue this is similar to us as humans. We don't have a good understanding of the human brain or precise mechanisms of action underlying thought. All we know is we as humans have semantic representations and can effectively solve problems.)
bigger note on your chess point: this example illustrates that LLMs are "semi-decidable." We thought they were bad at chess, but we just hadn't discovered the right way to prompt. More generally, we can confirm when an LLM is good at X because we feed it a prompt that produces performance in X, but given the size of the input space we're dealing with here, we can't confirm that LLMs are bad at X just because we haven't seen them do well at it. Maybe we just haven't discovered the right prompt. (These input spaces are massive, by the way. ChatGPT-3.5, for example, has a context window of 4,096 tokens, so if we were considering only the English alphabet, we're looking at more than 26^{4,096} possible inputs.)
What would an LLM have to do to convince you it was good at math? Check out this recent post by OpenAI where one of their models is solving 60%+ of problems from a high school math competition dataset: https://openai.com/research/improving-mathematical-reasoning...
Author here. First off, thank you for reading and for your thoughts.
I provided examples that I thought would be intuitive for humans to help folks understand that an understanding of the underlying phenomena is useful for next token prediction (I've added this as a note).
Could you share what part of the article came across as suggesting that LLMs "magically" acquire whatever ability helps them to predict? I'd like to make that section clearer, so that doesn't come across.
Re: "LLMs are not particularly good at arithmetic". There are published results that show that LLMs using certain techniques reach close to 100% accuracy on 8-digit addition: https://arxiv.org/pdf/2206.07682.pdf.
There are also recent results from OpenAI where their model obtained solid results on high school math competition problems, which are harder than arithmetic: https://openai.com/research/improving-mathematical-reasoning...
I haven't looked into counting syllables or recognizing haikus but I bet that this is a result of tokenization and not an inability of the model to create a representation of the underlying phenomena.
If you aren't on the level of the moral greats, you start small and try to build up, the same way you'd start by running a 5k before running an ultramarathon.
I hope others out there find this viewpoint as helpful as I have.