paper-with-me

Papers

Ecco: An Open Source Library for the Explainability of Transformer Language Models

2021-08-01 · ACL 2021 5 · J Alammar

Our understanding of why Transformer-based NLP models have been achieving their recent success lags behind our ability to continue scaling these models. To increase the transparency of Transformer-based language models, we present Ecco {--} an open-source library for the explainability of Transformer-based NLP models. Ecco provides a set of tools to capture, analyze, visualize, and interactively explore the inner mechanics of these models. This includes (1) gradient-based feature attribution for natural language generation (2) hidden states and their evolution between model layers (3) convenient access and examination tools for neuron activations in the under-explored Feed-Forward Neural Network sublayer of Transformer layers. (4) convenient examination of activation vectors via canonical correlation analysis (CCA), non-negative matrix factorization (NMF), and probing classifiers. We find that syntactic information can be retrieved from BERT{'}s FFNN representations in levels comparable to those in hidden state representations. More curiously, we find that the model builds up syntactic information in its hidden states even when intermediate FFNNs indicate diminished levels of syntactic information. Ecco is available at https://www.eccox.io/

📄 PDF Abstract BibTeX

Code (1)

jalammar/ecco 공식 구현 pytorch

Tasks

Text Generation

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
Residual Connection 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Multi-Head Attention 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

$EvoAl^{2048}$

2024-08-15 · Bernhard J. Berger, Christina Plump, Rolf Drechsler

As AI solutions enter safety-critical products, the explainability and interpretability of solutions generated by AI products become increasingly important. In the long term, such explanations are the key to gaining user…

Interpreto: An Explainability Library for Transformers

2025-12-10 · Antonin Poché, Thomas Mullor, Gabriele Sarti, Frédéric Boisnard 외 arxiv

Interpreto is an open-source Python library for interpreting HuggingFace language models, from early BERT variants to LLMs. It provides two complementary families of methods: attribution methods and concept-based explana…

Text Generation

CLAIMED, a visual and scalable component library for Trusted AI

2021-03-04 · Romeo Kienzler, Ivan Nesic

Deep Learning models are getting more and more popular but constraints on explainability, adversarial robustness and fairness are often major concerns for production deployment. Although the open source ecosystem is abun…

Adversarial RobustnessFairness

Utilizing Explainability Techniques for Reinforcement Learning Model Assurance

2023-11-27 · Alexander Tapley, Kyle Gatesman, Luis Robaina, Brett Bissey 외

Explainable Reinforcement Learning (XRL) can provide transparency into the decision-making process of a Deep Reinforcement Learning (DRL) model and increase user trust and adoption in real-world use cases. By utilizing X…

Decision MakingDeep Reinforcement Learningmodelreinforcement-learning+1

Transformers: State-of-the-Art Natural Language Processing

2020-10-01 · EMNLP 2020 11 · Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond 외

Recent progress in natural language processing has been driven by advances in both model architecture and model pretraining. Transformer architectures have facilitated building higher-capacity models and pretraining has …

Image ClassificationObject RecognitionSpeech Recognition