I somewhat agree, but I think that the language example is not a good one. As Anthropic have demonstrated[0], LLMs do have "conceptual neurons" that generalise an abstract concept which can later be translated to other languages.
The issue is that those concepts are encoded in intermediate layers during training, absorbing biases present in training data. It may produce a world model good enough to know that "green" and "verde" are different names for the same thing, but not robust enough to discard ordering bias or wording bias. Humans suffer from that too, albeit arguably less.
The issue is that those concepts are encoded in intermediate layers during training, absorbing biases present in training data. It may produce a world model good enough to know that "green" and "verde" are different names for the same thing, but not robust enough to discard ordering bias or wording bias. Humans suffer from that too, albeit arguably less.
[0] https://transformer-circuits.pub/2025/attribution-graphs/bio...