paper-with-me

홈 › Papers

Eliciting Latent Predictions from Transformers with the Tuned Lens

2023-03-14 · Nora Belrose, Zach Furman, Logan Smith, Danny Halawi, Igor Ostrovsky, Lev McKinney, Stella Biderman, Jacob Steinhardt

We analyze transformers from the perspective of iterative inference, seeking to understand how model predictions are refined layer by layer. To do so, we train an affine probe for each block in a frozen pretrained model, making it possible to decode every hidden state into a distribution over the vocabulary. Our method, the \emph{tuned lens}, is a refinement of the earlier ``logit lens'' technique, which yielded useful insights but is often brittle. We test our method on various autoregressive language models with up to 20B parameters, showing it to be more predictive, reliable and unbiased than the logit lens. With causal experiments, we show the tuned lens uses similar features to the model itself. We also find the trajectory of latent predictions can be used to detect malicious inputs with high accuracy. All code needed to reproduce our results can be found at https://github.com/AlignmentResearch/tuned-lens.

📄 PDF Abstract BibTeX arXiv:2303.08112

Code (2)

alignmentresearch/tuned-lens 공식 구현 pytorch
neelnanda-io/transformerlens 공식 구현 jax

Tasks

Language Modelling

Methods 이 논문이 사용한 방법론

Test 설명 없음

Similar Papers 제목 키워드 기반

Does Transformer Interpretability Transfer to RNNs?

2024-04-09 · Gonçalo Paulo, Thomas Marshall, Nora Belrose

Recent advances in recurrent neural network architectures, such as Mamba and RWKV, have enabled RNNs to match or exceed the performance of equal-size transformers in terms of language modeling perplexity and downstream e…

Language ModelingLanguage ModellingMamba

SimLens for Early Exit in Large Language Models: Eliciting Accurate Latent Predictions with One More Token

2025-07-23 · Ming Ma, Bowen Zheng, Zhongqiao Lin, Tianming Yang arxiv

Intermediate-layer predictions in large language models (LLMs) are informative but hard to decode accurately, especially at early layers. Existing lens-style methods typically rely on direct linear readout, which is simp…

Indic-TunedLens: Interpreting Multilingual Models in Indian Languages

2026-01-29 · Mihir Panchal, Deeksha Varshney, Mamta, Asif Ekbal arxiv

Multilingual large language models (LLMs) are increasingly deployed in linguistically diverse regions like India, yet most interpretability tools remain tailored to English. Prior work reveals that LLMs often operate in …

QLENS: Towards A Quantum Perspective of Language Transformers

2025-10-13 · Aditya Gupta, Kirandeep Kaur, Vinayak Gupta, Chirag Shah arxiv

In natural language processing, current methods for understanding Transformers are successful at identifying intermediate predictions during a model's inference. However, these approaches function as limited diagnostic c…

Towards eliciting latent knowledge from LLMs with mechanistic interpretability

2025-05-20 · Bartosz Cywiński, Emil Ryd, Senthooran Rajamanoharan, Neel Nanda

As language models become more powerful and sophisticated, it is crucial that they remain trustworthy and reliable. There is concerning preliminary evidence that models may attempt to deceive or keep secrets from their o…