LatentQA: Teaching LLMs to Decode Activations Into Natural Language

• URL: http://arxiv.org/abs/2412.08686v1
• Date: Wed, 11 Dec 2024 18:59:33 GMT
• Title: LatentQA: Teaching LLMs to Decode Activations Into Natural Language
• Authors: Alexander Pan, Lijie Chen, Jacob Steinhardt,
• Abstract summary: We introduce LatentQA, the task of answering open-ended questions about model activations in natural language. We propose Latent Interpretation Tuning (LIT), which finetunes a decoder LLM on a dataset of activations and associated question-answer pairs. Our decoder also specifies a differentiable loss that we use to control models, such as debiasing models on stereotyped sentences and controlling the sentiment of generations.
• Score: 72.87064562349742
• License:
• Abstract: Interpretability methods seek to understand language model representations, yet the outputs of most such methods -- circuits, vectors, scalars -- are not immediately human-interpretable. In response, we introduce LatentQA, the task of answering open-ended questions about model activations in natural language. Towards solving LatentQA, we propose Latent Interpretation Tuning (LIT), which finetunes a decoder LLM on a dataset of activations and associated question-answer pairs, similar to how visual instruction tuning trains on question-answer pairs associated with images. We use the decoder for diverse reading applications, such as extracting relational knowledge from representations or uncovering system prompts governing model behavior. Our decoder also specifies a differentiable loss that we use to control models, such as debiasing models on stereotyped sentences and controlling the sentiment of generations. Finally, we extend LatentQA to reveal harmful model capabilities, such as generating recipes for bioweapons and code for hacking.

Related papers

• Tokens, the oft-overlooked appetizer: Large language models, the distributional hypothesis, and meaning [31.632816425798108]
  Tokenization is a necessary component within the current architecture of many language models. We show that tokenization pretraining can be a backdoor for bias and other unwanted content. We relay evidence that the tokenization algorithm's objective function impacts the large language model's cognition.
  arXiv  Detail & Related papers  (2024-12-14T18:18:52Z)
• Interpret the Internal States of Recommendation Model with Sparse Autoencoder [26.021277330699963]
  RecSAE is an automatic, generalizable probing method for interpreting the internal states of Recommendation models. We train an autoencoder with sparsity constraints to reconstruct internal activations of recommendation models. We automated the construction of concept dictionaries based on the relationship between latent activations and input item sequences.
  arXiv  Detail & Related papers  (2024-11-09T08:22:31Z)
• LLMs can learn self-restraint through iterative self-reflection [57.26854891567574]
  Large Language Models (LLMs) must be capable of dynamically adapting their behavior based on their level of knowledge and uncertainty associated with specific topics. This adaptive behavior, which we refer to as self-restraint, is non-trivial to teach. We devise a utility function that can encourage the model to produce responses only when it is confident in them.
  arXiv  Detail & Related papers  (2024-05-15T13:35:43Z)
• Let Models Speak Ciphers: Multiagent Debate through Embeddings [84.20336971784495]
  We introduce CIPHER (Communicative Inter-Model Protocol Through Embedding Representation) to address this issue. By deviating from natural language, CIPHER offers an advantage of encoding a broader spectrum of information without any modification to the model weights. This showcases the superiority and robustness of embeddings as an alternative "language" for communication among LLMs.
  arXiv  Detail & Related papers  (2023-10-10T03:06:38Z)
• On Conditional and Compositional Language Model Differentiable Prompting [75.76546041094436]
  Prompts have been shown to be an effective method to adapt a frozen Pretrained Language Model (PLM) to perform well on downstream tasks. We propose a new model, Prompt Production System (PRopS), which learns to transform task instructions or input metadata, into continuous prompts.
  arXiv  Detail & Related papers  (2023-07-04T02:47:42Z)
• Augmented Language Models: a Survey [55.965967655575454]
  This survey reviews works in which language models (LMs) are augmented with reasoning skills and the ability to use tools. We refer to them as Augmented Language Models (ALMs) The missing token objective allows ALMs to learn to reason, use tools, and even act, while still performing standard natural language tasks.
  arXiv  Detail & Related papers  (2023-02-15T18:25:52Z)
• elBERto: Self-supervised Commonsense Learning for Question Answering [131.51059870970616]
  We propose a Self-supervised Bidirectional Representation Learning of Commonsense framework, which is compatible with off-the-shelf QA model architectures. The framework comprises five self-supervised tasks to force the model to fully exploit the additional training signals from contexts containing rich commonsense. elBERto achieves substantial improvements on out-of-paragraph and no-effect questions where simple lexical similarity comparison does not help.
  arXiv  Detail & Related papers  (2022-03-17T16:23:45Z)
• Disentangling Generative Factors in Natural Language with Discrete Variational Autoencoders [0.0]
  We argue that continuous variables may not be ideal to model features of textual data, due to the fact that most generative factors in text are discrete. We propose a Variational Autoencoder based method which models language features as discrete variables and encourages independence between variables for learning disentangled representations.
  arXiv  Detail & Related papers  (2021-09-15T09:10:05Z)
• Interactively Generating Explanations for Transformer Language Models [14.306470205426526]
  Transformer language models are state-of-the-art in a multitude of NLP tasks. Recent methods aim to provide interpretability and explainability to black-box models. We emphasize using prototype networks directly incorporated into the model architecture.
  arXiv  Detail & Related papers  (2021-09-02T11:34:29Z)
• Exploring Neural Models for Parsing Natural Language into First-Order Logic [10.62143644603835]
  We study the capability of neural models in parsing English sentences to First-Order Logic (FOL) We model FOL parsing as a sequence to sequence mapping task where given a natural language sentence, it is encoded into an intermediate representation using an LSTM followed by a decoder which sequentially generates the predicates in the corresponding FOL formula.
  arXiv  Detail & Related papers  (2020-02-16T09:22:32Z)

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.