Interpreting the Second-Order Effects of Neurons in CLIP

• URL: http://arxiv.org/abs/2406.04341v2
• Date: Mon, 24 Jun 2024 02:14:18 GMT
• Title: Interpreting the Second-Order Effects of Neurons in CLIP
• Authors: Yossi Gandelsman, Alexei A. Efros, Jacob Steinhardt,
• Abstract summary: We interpret the function of individual neurons in CLIP by automatically describing them using text. We present the "second-order lens", analyzing the effect flowing from a neuron through the later attention heads, directly to the output. Our results indicate that a scalable understanding of neurons can be used for model deception and for introducing new model capabilities.
• Score: 73.54377859089801
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We interpret the function of individual neurons in CLIP by automatically describing them using text. Analyzing the direct effects (i.e. the flow from a neuron through the residual stream to the output) or the indirect effects (overall contribution) fails to capture the neurons' function in CLIP. Therefore, we present the "second-order lens", analyzing the effect flowing from a neuron through the later attention heads, directly to the output. We find that these effects are highly selective: for each neuron, the effect is significant for <2% of the images. Moreover, each effect can be approximated by a single direction in the text-image space of CLIP. We describe neurons by decomposing these directions into sparse sets of text representations. The sets reveal polysemantic behavior - each neuron corresponds to multiple, often unrelated, concepts (e.g. ships and cars). Exploiting this neuron polysemy, we mass-produce "semantic" adversarial examples by generating images with concepts spuriously correlated to the incorrect class. Additionally, we use the second-order effects for zero-shot segmentation and attribute discovery in images. Our results indicate that a scalable understanding of neurons can be used for model deception and for introducing new model capabilities.

Related papers

• Confidence Regulation Neurons in Language Models [91.90337752432075]
  This study investigates the mechanisms by which large language models represent and regulate uncertainty in next-token predictions. Entropy neurons are characterized by an unusually high weight norm and influence the final layer normalization (LayerNorm) scale to effectively scale down the logits. token frequency neurons, which we describe here for the first time, boost or suppress each token's logit proportionally to its log frequency, thereby shifting the output distribution towards or away from the unigram distribution.
  arXiv  Detail & Related papers  (2024-06-24T01:31:03Z)
• Identifying Interpretable Visual Features in Artificial and Biological Neural Systems [3.604033202771937]
  Single neurons in neural networks are often interpretable in that they represent individual, intuitively meaningful features. Many neurons exhibit $textitmixed selectivity$, i.e., they represent multiple unrelated features. We propose an automated method for quantifying visual interpretability and an approach for finding meaningful directions in network activation space.
  arXiv  Detail & Related papers  (2023-10-17T17:41:28Z)
• Multimodal Neurons in Pretrained Text-Only Transformers [52.20828443544296]
  We identify "multimodal neurons" that convert visual representations into corresponding text. We show that multimodal neurons operate on specific visual concepts across inputs, and have a systematic causal effect on image captioning.
  arXiv  Detail & Related papers  (2023-08-03T05:27:12Z)
• Neuron to Graph: Interpreting Language Model Neurons at Scale [8.32093320910416]
  This paper introduces a novel automated approach designed to scale interpretability techniques across a vast array of neurons within Large Language Models. We propose Neuron to Graph (N2G), an innovative tool that automatically extracts a neuron's behaviour from the dataset it was trained on and translates it into an interpretable graph.
  arXiv  Detail & Related papers  (2023-05-31T14:44:33Z)
• Disentangling Neuron Representations with Concept Vectors [0.0]
  The main contribution of this paper is a method to disentangle polysemantic neurons into concept vectors encapsulating distinct features. Our evaluations show that the concept vectors found encode coherent, human-understandable features.
  arXiv  Detail & Related papers  (2023-04-19T14:55:31Z)
• And/or trade-off in artificial neurons: impact on adversarial robustness [91.3755431537592]
  Presence of sufficient number of OR-like neurons in a network can lead to classification brittleness and increased vulnerability to adversarial attacks. We define AND-like neurons and propose measures to increase their proportion in the network. Experimental results on the MNIST dataset suggest that our approach holds promise as a direction for further exploration.
  arXiv  Detail & Related papers  (2021-02-15T08:19:05Z)
• Compositional Explanations of Neurons [52.71742655312625]
  We describe a procedure for explaining neurons in deep representations by identifying compositional logical concepts. We use this procedure to answer several questions on interpretability in models for vision and natural language processing.
  arXiv  Detail & Related papers  (2020-06-24T20:37:05Z)
• Non-linear Neurons with Human-like Apical Dendrite Activations [81.18416067005538]
  We show that a standard neuron followed by our novel apical dendrite activation (ADA) can learn the XOR logical function with 100% accuracy. We conduct experiments on six benchmark data sets from computer vision, signal processing and natural language processing.
  arXiv  Detail & Related papers  (2020-02-02T21:09:39Z)
• Learning Neural Activations [2.842794675894731]
  We explore what happens when the activation function of each neuron in an artificial neural network is learned from data alone. This is achieved by modelling the activation function of each neuron as a small neural network whose weights are shared by all neurons in the original network.
  arXiv  Detail & Related papers  (2019-12-27T15:52:07Z)

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.