# interpretability **Repository Path**: SavenNeer/interpretability ## Basic Information - **Project Name**: interpretability - **Description**: No description available - **Primary Language**: Unknown - **License**: Apache-2.0 - **Default Branch**: master - **Homepage**: None - **GVP Project**: No ## Statistics - **Stars**: 0 - **Forks**: 0 - **Created**: 2024-10-12 - **Last Updated**: 2024-10-12 ## Categories & Tags **Categories**: Uncategorized **Tags**: None ## README # PAIR Interpretability This repo contains code and articles on [PAIR](https://ai.google/pair) interpretability projects. ### Patchscopes: A Unifying Framework for Inspecting Hidden Representations of Language Models The [Patchscopes mini-site](https://pair-code.github.io/interpretability/patchscopes) contains a brief introduction to the [longer paper](https://arxiv.org/abs/2401.06102), and will have the code uploaded once it is a little cleaner also. ### Visualizing and Measuring the Geometry of BERT [bert-tree](https://github.com/PAIR-code/interpretability/tree/master/bert-tree) and [context-atlas](https://github.com/PAIR-code/interpretability/tree/master/context-atlas) are repos for two interactive blogposts/visualizations for the paper [Visualizing and Measuring the Geometry of BERT ](https://arxiv.org/abs/1906.02715): 1. [Language, trees, and geometry in neural networks](https://pair-code.github.io/interpretability/bert-tree/) explores the geometry of syntactic information in BERT ([bert-tree](https://github.com/PAIR-code/interpretability/tree/master/bert-tree)) 2. [Language, Context, and Geometry in Neural Network](https://pair-code.github.io/interpretability/bert-tree/) explores semantics and context in BERT. See the accompanying tool, [Context Atlas](https://storage.googleapis.com/bert-wsd-vis/demo/index.html?#word=lie), for more details ([context-atlas](https://github.com/PAIR-code/interpretability/tree/master/context-atlas)) ### Deep dreaming on text [text-dream](https://github.com/PAIR-code/interpretability/tree/master/text-dream) contains different experiments and tools to work with deep dreaming for text. ### LinguisticLens [data-synth-syntax](https://github.com/PAIR-code/interpretability/tree/master/data-synth-syntax) contains [LinguisticLens](https://storage.googleapis.com/data-synth-trees/demo/index.html), a tool for visualizing generated text data.