Picture for Javier Ferrando

Javier Ferrando

A Primer on the Inner Workings of Transformer-based Language Models

Add code
May 02, 2024
Viaarxiv icon

LM Transparency Tool: Interactive Tool for Analyzing Transformer Language Models

Add code
Apr 10, 2024
Viaarxiv icon

Information Flow Routes: Automatically Interpreting Language Models at Scale

Add code
Feb 27, 2024
Viaarxiv icon

Neurons in Large Language Models: Dead, N-gram, Positional

Add code
Sep 09, 2023
Viaarxiv icon

Automating Behavioral Testing in Machine Translation

Add code
Sep 07, 2023
Viaarxiv icon

Explaining How Transformers Use Context to Build Predictions

Add code
May 21, 2023
Viaarxiv icon

Toxicity in Multilingual Machine Translation at Scale

Add code
Oct 06, 2022
Figure 1 for Toxicity in Multilingual Machine Translation at Scale
Figure 2 for Toxicity in Multilingual Machine Translation at Scale
Figure 3 for Toxicity in Multilingual Machine Translation at Scale
Figure 4 for Toxicity in Multilingual Machine Translation at Scale
Viaarxiv icon

Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer

Add code
May 23, 2022
Figure 1 for Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
Figure 2 for Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
Figure 3 for Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
Figure 4 for Towards Opening the Black Box of Neural Machine Translation: Source and Target Interpretations of the Transformer
Viaarxiv icon

On the Locality of Attention in Direct Speech Translation

Add code
Apr 19, 2022
Figure 1 for On the Locality of Attention in Direct Speech Translation
Figure 2 for On the Locality of Attention in Direct Speech Translation
Figure 3 for On the Locality of Attention in Direct Speech Translation
Figure 4 for On the Locality of Attention in Direct Speech Translation
Viaarxiv icon

Measuring the Mixing of Contextual Information in the Transformer

Add code
Mar 14, 2022
Figure 1 for Measuring the Mixing of Contextual Information in the Transformer
Figure 2 for Measuring the Mixing of Contextual Information in the Transformer
Figure 3 for Measuring the Mixing of Contextual Information in the Transformer
Figure 4 for Measuring the Mixing of Contextual Information in the Transformer
Viaarxiv icon