Name: Richard Diehl Martinez
Type: User
Bio: CS PhD student at the University of Cambridge. Exploring the intersection of machine learning, neuroscience, and sociology.
Location: Cambridge, UK
Blog: richarddiehlmartinez.com
Richard Diehl Martinez's Projects
Use attention distributions to classify types of linguistic bias.
A demo script for how to run auto-restarting jobs on the Cambridge CSD3 compute cluster.
higher is a pytorch library allowing users to obtain higher order gradients over losses spanning training loops rather than individual training steps.
Experimentation around resizing models to different shapes and the effect on model performance
A library for training multilingual language models using Subset Masked Language Modeling Tasks (SMLMT)
Code and data for the paper, "Automatically Neutralizing Subjective Bias in Text"
Basic setup for interactive nvim
The hub for EleutherAI's work on interpretability and learning dynamics
🔥 A tool for visualizing and tracking your machine learning experiments. This repo contains the CLI and Python API.