How LLMs Learn: Tracing Internal Representations with Sparse Autoencoders

• URL: http://arxiv.org/abs/2503.06394v1
• Date: Sun, 09 Mar 2025 02:13:44 GMT
• Title: How LLMs Learn: Tracing Internal Representations with Sparse Autoencoders
• Authors: Tatsuro Inaba, Kentaro Inui, Yusuke Miyao, Yohei Oseki, Benjamin Heinzerling, Yu Takagi,
• Abstract summary: Large Language Models (LLMs) demonstrate remarkable multilingual capabilities and broad knowledge.<n>We analyze how the information encoded in LLMs' internal representations evolves during the training process.
• Score: 30.36521888592164
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Models (LLMs) demonstrate remarkable multilingual capabilities and broad knowledge. However, the internal mechanisms underlying the development of these capabilities remain poorly understood. To investigate this, we analyze how the information encoded in LLMs' internal representations evolves during the training process. Specifically, we train sparse autoencoders at multiple checkpoints of the model and systematically compare the interpretative results across these stages. Our findings suggest that LLMs initially acquire language-specific knowledge independently, followed by cross-linguistic correspondences. Moreover, we observe that after mastering token-level knowledge, the model transitions to learning higher-level, abstract concepts, indicating the development of more conceptual understanding.

Related papers

• How do Large Language Models Understand Relevance? A Mechanistic Interpretability Perspective [64.00022624183781]
  Large language models (LLMs) can assess relevance and support information retrieval (IR) tasks. We investigate how different LLM modules contribute to relevance judgment through the lens of mechanistic interpretability.
  arXiv  Detail & Related papers  (2025-04-10T16:14:55Z)
• Explicit Learning and the LLM in Machine Translation [20.630120942837564]
  This study explores the capacity of large language models (LLMs) for explicit learning. Using constructed languages generated by means as controlled test environments, we designed experiments to assess an LLM's ability to explicitly learn and apply grammar rules. Supervised fine-tuning on chains of thought significantly enhances LLM performance but struggles to generalize to typologically novel or more complex linguistic features.
  arXiv  Detail & Related papers  (2025-03-12T14:57:08Z)
• A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models [40.67240575271987]
  Large Language Models (LLMs) have revolutionized natural language processing, yet their internal mechanisms remain largely opaque.<n> mechanistic interpretability has attracted significant attention from the research community as a means to understand the inner workings of LLMs.<n>Sparse Autoencoders (SAEs) have emerged as a promising method due to their ability to disentangle the complex, superimposed features within LLMs into more interpretable components.
  arXiv  Detail & Related papers  (2025-03-07T17:38:00Z)
• The Rise and Down of Babel Tower: Investigating the Evolution Process of Multilingual Code Large Language Model [59.357993924917]
  We study the evolution of multilingual capabilities in large language models (LLMs) during the pre-training process.<n>We propose the Babel Tower Hypothesis, which describes the entire process of LLMs acquiring new language capabilities.<n>We propose a novel method to construct an optimized pre-training corpus for multilingual code LLMs.
  arXiv  Detail & Related papers  (2024-12-10T08:28:57Z)
• How Do Multilingual Language Models Remember Facts? [50.13632788453612]
  We show that previously identified recall mechanisms in English largely apply to multilingual contexts.<n>We localize the role of language during recall, finding that subject enrichment is language-independent.<n>In decoder-only LLMs, FVs compose these two pieces of information in two separate stages.
  arXiv  Detail & Related papers  (2024-10-18T11:39:34Z)
• Unveiling LLMs: The Evolution of Latent Representations in a Dynamic Knowledge Graph [15.129079475322637]
  This work unveils the factual information an Large Language Models represents internally for sentence-level claim verification. We propose an end-to-end framework to decode factual knowledge embedded in token representations from a vector space to a set of ground predicates. Our framework employs activation patching, a vector-level technique that alters a token representation during inference, to extract encoded knowledge.
  arXiv  Detail & Related papers  (2024-04-04T17:45:59Z)
• Identifying Semantic Induction Heads to Understand In-Context Learning [103.00463655766066]
  We investigate whether attention heads encode two types of relationships between tokens present in natural languages. We find that certain attention heads exhibit a pattern where, when attending to head tokens, they recall tail tokens and increase the output logits of those tail tokens.
  arXiv  Detail & Related papers  (2024-02-20T14:43:39Z)
• How Proficient Are Large Language Models in Formal Languages? An In-Depth Insight for Knowledge Base Question Answering [52.86931192259096]
  Knowledge Base Question Answering (KBQA) aims to answer natural language questions based on facts in knowledge bases. Recent works leverage the capabilities of large language models (LLMs) for logical form generation to improve performance.
  arXiv  Detail & Related papers  (2024-01-11T09:27:50Z)
• Supervised Knowledge Makes Large Language Models Better In-context Learners [94.89301696512776]
  Large Language Models (LLMs) exhibit emerging in-context learning abilities through prompt engineering. The challenge of improving the generalizability and factuality of LLMs in natural language understanding and question answering remains under-explored. We propose a framework that enhances the reliability of LLMs as it: 1) generalizes out-of-distribution data, 2) elucidates how LLMs benefit from discriminative models, and 3) minimizes hallucinations in generative tasks.
  arXiv  Detail & Related papers  (2023-12-26T07:24:46Z)
• Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention [53.896974148579346]
  Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains. The enigmatic black-box'' nature of LLMs remains a significant challenge for interpretability, hampering transparent and accountable applications. We propose a novel methodology anchored in sparsity-guided techniques, aiming to provide a holistic interpretation of LLMs.
  arXiv  Detail & Related papers  (2023-12-22T19:55:58Z)
• Probing Large Language Models from A Human Behavioral Perspective [24.109080140701188]
  Large Language Models (LLMs) have emerged as dominant foundational models in modern NLP. The understanding of their prediction processes and internal mechanisms, such as feed-forward networks (FFN) and multi-head self-attention (MHSA) remains largely unexplored.
  arXiv  Detail & Related papers  (2023-10-08T16:16:21Z)
• IERL: Interpretable Ensemble Representation Learning -- Combining CrowdSourced Knowledge and Distributed Semantic Representations [11.008412414253662]
  Large Language Models (LLMs) encode meanings of words in the form of distributed semantics. Recent studies have shown that LLMs tend to generate unintended, inconsistent, or wrong texts as outputs. We propose a novel ensemble learning method, Interpretable Ensemble Representation Learning (IERL), that systematically combines LLM and crowdsourced knowledge representations.
  arXiv  Detail & Related papers  (2023-06-24T05:02:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.