The science might be legit here, but I'm getting really, really tired of the way every single piece of writing to come out of Anthropic is written in some kind of self-aggrandising, wooey wonderous 'our model has developed a genetic mutation that makes it have feelings' bs style. Regardless of what they're trying to communicate, those undertones are always there. It's annoying and disingenuous. Homeopathy 'this-water…
> written in some kind of self-aggrandising, wooey wonderous
They are drunk on their own kool-aid. To the rest of us it is very annoying, and makes me want to say: it is just a freaking weights machine, stop.
Pretty sure i've also seen research on this spanning models. i.e. similar thinking shapes emerge regardless of which providers model it is, including US vs Chinese which hints at some sort of universality
Recall isn't naturally bidirectional, even for humans. If you are learning vocabulary in a new language, it's common advice to practice both target > source and source > target. Doing only one-way often makes you much better recalling that single direction than both.
I would need further convincing that humans do not naturally tend towards bidirectional recall. Perhaps I’m just on alert anytime I see an LLM-ism that’s met with a claim that the same or similar phenomena holds true in humans as well.
If you can touch type try this:
Go through they keys once alphabetical 'abcd...' recalling the movement from the letter. Then do it the other way round and move your fingers in spatial sequence to the keys and try to recall the letter.
You will find the former much easier if you did not by chance also memorize the keyboard layout for some reason.
IMO consciousness is different from working memory, at least to a certain degree. The inner mind (working memory) is different from consciousness from sensory stimuli. You can stand outside and take in the environment and have a relatively quiet inner mind. When you think about math or philosophy a different type of conscious experience arises than sensory stimuli experience. That doesn't mean consciousness is completely isolated from working memory but there is some distinction there I can't describe fully.
Edit: I also think as someone else said, we already know the intermediate layers can contain a lot of adjacent words related to the topic without explicitly outputting those words. These could just be related embedding intermediate vectors that activate but aren't outputted.
This reminded me of some weird quirk/experiment I found with LLMs that I found while messing around, maybe someone can explain it or something. Open any AI chatbot that isn't cheating by connecting to the Internet (so disable web search). Claude, DeepSeek, Kimi, whatever. Ask them this question: "What was that weird band from michigan from the 2000s that wore coloured ties" You will probably get a wrong answer, or if…
"The reversal curse", it rarely shows up in practice but you found a case when it did. The "knowledge landscape" an LLM uses is "directional". It's easy to reach "a quirky music band from Michigan known for colored ties" when you stand at "Tally Hall". But if you stand at "a quirky music band from Michigan known for colored ties", it's harder to reach "Tally Hall" from there. For the "latent knowledge graph" an LLM u…
"The reversal curse", it rarely shows up in practice but you found a case when it did. The "knowledge landscape" an LLM uses is "directional". It's easy to reach "a quirky music band from Michigan known for colored ties" when you stand at "Tally Hall". But if you stand at "a quirky music band from Michigan known for colored ties", it's harder to reach "Tally Hall" from there. For the "latent knowledge graph" an LLM u…
If this kind of thing holds true for humans we now may understand synesthesia, perhaps we find ourselves with a large enough study pool to map out clever tricks to influence each other.
It certainly holds true for humans. The brain stores relational information in a sequential pattern that is not automatically reversible. One of the best examples is the alphabet. Everyone learns it in school, so the pattern A->B->C->... is trivial to recite for most people. Now, if I gave you a random letter to start with and asked you to to recite the remaining letters until Z, you'll probably find it is still pretty easy. But if I asked you to cite the letters backwards to A, most people would suddenly struggle with this task because they never learned or used the alphabet that way in school. You need to train specifically to link this kind of information backwards in your brain.
https://claude.ai/share/2b0f85a2-e7b8-4f62-91a0-eca61bdeabec Fable 5 on low gets the answer with web search turned off, one-shot!
What does it say to the second question? I've found Claude is one of the worst models with regards to pop culture knowledge like this, even compared to the Chinese open ones. Just curious, not really relevant to the initial post but I don't pay for it so I only have access to Sonnet. https://claude.ai/share/5e7e09b2-a75a-4024-b261-9a1a4e063a8b this is mostly hilariously wrong. wrong tie colors, they did not replace t…
"I've found Claude is one of the worst models with regards to pop culture knowledge like this, "
Is that bad?
I want to use models for coding and reasoning capabilities, not pop trivia knowledge they can get with web search.