Reasoning Beyond Chain-of-Thought: A Latent Computational Mode in Large Language Models
This work addresses the fundamental understanding of reasoning mechanisms in LLMs for AI researchers, revealing a latent computational mode beyond CoT.
The study investigated whether Chain-of-Thought prompting is the unique mechanism for triggering reasoning in large language models, finding that steering a single latent feature can improve accuracy without explicit prompting, achieving performance comparable to standard CoT prompting in large models.
Chain-of-Thought (CoT) prompting has improved the reasoning performance of large language models (LLMs), but it remains unclear why it works and whether it is the unique mechanism for triggering reasoning in large language models. In this work, we study this question by directly analyzing and intervening on the internal representations of LLMs with Sparse Autoencoders (SAEs), identifying a small set of latent features that are causally associated with LLM reasoning behavior. Across multiple model families and reasoning benchmarks, we find that steering a single reasoning-related latent feature can substantially improve accuracy without explicit CoT prompting. For large models, latent steering achieves performance comparable to standard CoT prompting while producing more efficient outputs. We further observe that this reasoning-oriented internal state is triggered early in generation and can override prompt-level instructions that discourage explicit reasoning. Overall, our results suggest that multi-step reasoning in LLMs is supported by latent internal activations that can be externally activated, while CoT prompting is one effective, but not unique, way of activating this mechanism rather than its necessary cause.