Parts-of-Speech as Emergent Categories in SAE Latent Space
- 类型：arxiv
- 标识：2609.29362
- 链接：https://arxiv.org/abs/2609.29362
- 主分类：multimodal
- 形态：method
- TLDR：Sparse AutoEncoders (SAEs) offer a promising way to inspect language model representations, but it is still unclear what kind of linguistic structure their latents expose. We use part-of-speech (PoS) categories as a controlled test case to study whether morpho-syntactic information is encoded by individual latents or by structured groups of features. We find that PoS distinctions are highly recoverable from SAE activations, but do not align with one-to-one latent / category mappings. This recoverability is not reducible to lexical memorisation, and Open and Closed PoS classes differ substantia
- 待LLM分类：否
- 来源文件：
- /inbox/tom/_candidates/2026-09-25-agent-rag-longcontext-candidates.json