{"list":"other","total":90,"rows":[{"id":"1706.03762","title":"Attention Is All You Need","created":"2017-06-12","cited_by_count":60361},{"id":"1705.07874","title":"A Unified Approach to Interpreting Model Predictions","created":"2017-05-22","cited_by_count":6555},{"id":"1602.04938","title":"\"Why Should I Trust You?\": Explaining the Predictions of Any Classifier","created":"2016-02-16","cited_by_count":5070},{"id":"1703.01365","title":"Axiomatic Attribution for Deep Networks","created":"2017-03-04","cited_by_count":3384},{"id":"2202.05262","title":"Locating and Editing Factual Associations in GPT","created":"2022-02-10","cited_by_count":2691},{"id":"1312.6034","title":"Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps","created":"2013-12-20","cited_by_count":2371},{"id":"1311.2901","title":"Visualizing and Understanding Convolutional Networks","created":"2013-11-12","cited_by_count":2290},{"id":"2211.00593","title":"Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small","created":"2022-11-01","cited_by_count":1165},{"id":"1706.03825","title":"SmoothGrad: removing noise by adding noise","created":"2017-06-12","cited_by_count":1109},{"id":"2209.11895","title":"In-context Learning and Induction Heads","created":"2022-09-24","cited_by_count":964},{"id":"1806.07421","title":"RISE: Randomized Input Sampling for Explanation of Black-box Models","created":"2018-06-19","cited_by_count":838},{"id":"2212.03827","title":"Discovering Latent Knowledge in Language Models Without Supervision","created":"2022-12-07","cited_by_count":767},{"id":"2304.14997","title":"Towards Automated Circuit Discovery for Mechanistic Interpretability","created":"2023-04-28","cited_by_count":724},{"id":"2310.06824","title":"The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets","created":"2023-10-10","cited_by_count":712},{"id":"1704.05796","title":"Network Dissection: Quantifying Interpretability of Deep Visual Representations","created":"2017-04-19","cited_by_count":708},{"id":"1708.06519","title":"Learning Efficient Convolutional Networks through Network Slimming","created":"2017-08-22","cited_by_count":676},{"id":"2303.08112","title":"Eliciting Latent Predictions from Transformers with the Tuned Lens","created":"2023-03-14","cited_by_count":623},{"id":"1704.03296","title":"Interpretable Explanations of Black Boxes by Meaningful Perturbation","created":"2017-04-11","cited_by_count":612},{"id":"1712.01312","title":"Learning Sparse Neural Networks through $L_0$ Regularization","created":"2017-12-04","cited_by_count":583},{"id":"2203.14680","title":"Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space","created":"2022-03-28","cited_by_count":532},{"id":"2304.14767","title":"Dissecting Recall of Factual Associations in Auto-Regressive Language Models","created":"2023-04-28","cited_by_count":528},{"id":"1806.10758","title":"A Benchmark for Interpretability Methods in Deep Neural Networks","created":"2018-06-28","cited_by_count":476},{"id":"1911.02508","title":"Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods","created":"2019-11-06","cited_by_count":443},{"id":"2403.19647","title":"Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models","created":"2024-03-28","cited_by_count":435},{"id":"2210.13382","title":"Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task","created":"2022-10-24","cited_by_count":431},{"id":"1801.01489","title":"All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneously","created":"2018-01-04","cited_by_count":373},{"id":"2309.16042","title":"Towards Best Practices of Activation Patching in Language Models: Metrics and Methods","created":"2023-09-27","cited_by_count":359},{"id":"2310.02207","title":"Language Models Represent Space and Time","created":"2023-10-03","cited_by_count":316},{"id":"2310.15916","title":"In-Context Learning Creates Task Vectors","created":"2023-10-24","cited_by_count":312},{"id":"2310.15213","title":"Function Vectors in Large Language Models","created":"2023-10-23","cited_by_count":308},{"id":"2306.03819","title":"LEACE: Perfect linear concept erasure in closed form","created":"2023-06-06","cited_by_count":267},{"id":"1910.08485","title":"Understanding Deep Networks via Extremal Perturbations and Smooth Masks","created":"2019-10-18","cited_by_count":259},{"id":"2009.05041","title":"Understanding the Role of Individual Units in a Deep Neural Network","created":"2020-09-10","cited_by_count":255},{"id":"1512.02479","title":"Explaining NonLinear Classification Decisions with Deep Taylor Decomposition","created":"2015-12-08","cited_by_count":250},{"id":"1802.07814","title":"Learning to Explain: An Information-Theoretic Perspective on Model Interpretation","created":"2018-02-21","cited_by_count":248},{"id":"1705.07857","title":"Real Time Image Saliency for Black Box Classifiers","created":"2017-05-22","cited_by_count":246},{"id":"2301.04213","title":"Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models","created":"2023-01-10","cited_by_count":243},{"id":"1612.08220","title":"Understanding Neural Networks through Representation Erasure","created":"2016-12-24","cited_by_count":224},{"id":"2011.14878","title":"Explaining by Removing: A Unified Framework for Model Explanation","created":"2020-11-21","cited_by_count":218},{"id":"2404.15255","title":"How to use and interpret activation patching","created":"2024-04-23","cited_by_count":214},{"id":"2305.00586","title":"How does GPT-2 compute greater-than?: Interpreting mathematical abilities in a pre-trained language model","created":"2023-04-30","cited_by_count":202},{"id":"2308.09124","title":"Linearity of Relation Decoding in Transformer Language Models","created":"2023-08-17","cited_by_count":200},{"id":"2304.05969","title":"Localizing Model Behavior with Path Patching","created":"2023-04-12","cited_by_count":195},{"id":"1707.01213","title":"Data-Driven Sparse Structure Selection for Deep Neural Networks","created":"2017-07-05","cited_by_count":181},{"id":"1412.6856","title":"Object Detectors Emerge in Deep Scene CNNs","created":"2014-12-22","cited_by_count":179},{"id":"2006.14032","title":"Compositional Explanations of Neurons","created":"2020-06-24","cited_by_count":172},{"id":"1910.13413","title":"Feature relevance quantification in explainable AI: A causal problem","created":"2019-10-29","cited_by_count":167},{"id":"1702.04595","title":"Visualizing Deep Neural Network Decisions: Prediction Difference Analysis","created":"2017-02-15","cited_by_count":166},{"id":"2310.15154","title":"Linear Representations of Sentiment in Large Language Models","created":"2023-10-23","cited_by_count":163},{"id":"2307.15771","title":"The Hydra Effect: Emergent Self-repair in Language Model Computations","created":"2023-07-28","cited_by_count":156},{"id":"2010.00577","title":"Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking","created":"2020-10-01","cited_by_count":153},{"id":"2307.09458","title":"Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla","created":"2023-07-18","cited_by_count":149},{"id":"1807.08024","title":"Explaining Image Classifiers by Counterfactual Generation","created":"2018-07-20","cited_by_count":144},{"id":"2201.11114","title":"Natural Language Descriptions of Deep Visual Features","created":"2022-01-26","cited_by_count":137},{"id":"2209.02535","title":"Analyzing Transformers in Embedding Space","created":"2022-09-06","cited_by_count":132},{"id":"2207.13243","title":"Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks","created":"2022-07-27","cited_by_count":129},{"id":"1802.00124","title":"Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers","created":"2018-02-01","cited_by_count":112},{"id":"2001.00396","title":"Restricting the Flow: Information Bottlenecks for Attribution","created":"2020-01-02","cited_by_count":110},{"id":"2310.08744","title":"Circuit Component Reuse Across Tasks in Transformer Language Models","created":"2023-10-12","cited_by_count":108},{"id":"2401.12181","title":"Universal Neurons in GPT2 Language Models","created":"2024-01-22","cited_by_count":102},{"id":"1910.12336","title":"CXPlain: Causal Explanations for Model Interpretation under Uncertainty","created":"2019-10-27","cited_by_count":98},{"id":"2002.09815","title":"Neuron Shapley: Discovering the Responsible Neurons","created":"2020-02-23","cited_by_count":98},{"id":"2305.16130","title":"Language Models Implement Simple Word2Vec-style Vector Arithmetic","created":"2023-05-25","cited_by_count":98},{"id":"1903.07435","title":"The emergence of number and syntax units in LSTM language models","created":"2019-03-18","cited_by_count":94},{"id":"2303.09435","title":"Jump to Conclusions: Short-Cutting Transformers With Linear Transformations","created":"2023-03-16","cited_by_count":90},{"id":"2305.15054","title":"A Mechanistic Interpretation of Arithmetic Reasoning in Language Models using Causal Mediation Analysis","created":"2023-05-24","cited_by_count":89},{"id":"2404.07129","title":"What needs to go right for an induction head? A mechanistic study of in-context learning circuits and their formation","created":"2024-04-10","cited_by_count":89},{"id":"2106.00786","title":"The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanations","created":"2021-06-01","cited_by_count":87},{"id":"2307.09476","title":"Overthinking the Truth: Understanding how Language Models Process False Demonstrations","created":"2023-07-18","cited_by_count":81},{"id":"2311.17030","title":"Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patching","created":"2023-11-28","cited_by_count":80},{"id":"2310.04625","title":"Copy Suppression: Comprehensively Understanding an Attention Head","created":"2023-10-06","cited_by_count":76},{"id":"1805.12233","title":"How Important Is a Neuron?","created":"2018-05-30","cited_by_count":73},{"id":"2002.12915","title":"The Implicit and Explicit Regularization Effects of Dropout","created":"2020-02-28","cited_by_count":71},{"id":"2104.03514","title":"Low-Complexity Probing via Finding Subnetworks","created":"2021-04-08","cited_by_count":64},{"id":"2004.14992","title":"How do Decisions Emerge across Layers in Neural Models? Interpretation with Differentiable Masking","created":"2020-04-30","cited_by_count":63},{"id":"2006.09481","title":"Efficient nonparametric statistical inference on population feature importance using Shapley values","created":"2020-06-16","cited_by_count":61},{"id":"2406.16778","title":"Finding Transformer Circuits with Edge Pruning","created":"2024-06-24","cited_by_count":60},{"id":"2010.13984","title":"Interpretation of NLP models through input marginalization","created":"2020-10-27","cited_by_count":44},{"id":"2402.15390","title":"Explorations of Self-Repair in Language Models","created":"2024-02-23","cited_by_count":39},{"id":"2305.13417","title":"VISIT: Visualizing and Interpreting the Semantic Information Flow of Transformers","created":"2023-05-22","cited_by_count":32},{"id":"1802.03788","title":"Influence-Directed Explanations for Deep Convolutional Networks","created":"2018-02-11","cited_by_count":31},{"id":"2309.05973","title":"Circuit Breaking: Removing Model Behaviors with Targeted Ablation","created":"2023-09-12","cited_by_count":27},{"id":"2404.11534","title":"Decomposing and Editing Predictions by Modeling Model Computation","created":"2024-04-17","cited_by_count":19},{"id":"2401.12631","title":"A Reply to Makelov et al. (2023)'s \"Interpretability Illusion\" Arguments","created":"2024-01-23","cited_by_count":17},{"id":"2007.01283","title":"Floodgate: inference for model-free variable importance","created":"2020-07-02","cited_by_count":16},{"id":"2311.04131","title":"Towards Interpretable Sequence Continuation: Analyzing Shared Circuits in Large Language Models","created":"2023-11-07","cited_by_count":15},{"id":"1911.03898","title":"Understanding Multi-Head Attention in Abstractive Summarization","created":"2019-11-10","cited_by_count":12},{"id":"2104.04488","title":"Explaining Neural Network Predictions on Sentence Pairs via Learning Word-Group Masks","created":"2021-04-09","cited_by_count":12},{"id":"2310.07325","title":"An Adversarial Example for Direct Logit Attribution: Memory Management in GELU-4L","created":"2023-10-11","cited_by_count":11},{"id":"2205.15750","title":"Variable importance without impossible data","created":"2022-05-31","cited_by_count":5}],"next_cursor":null}