#datasets

Datasets Determine the Difficulty of Interpretability

Recently Goodfire has come out swinging with two posts indicating that neurons in transformers are "more interpretable than previously assumed", or simply "already clean". In the "Neurons as Interpretable Basis" post they offer three reasons for this claim: First, neurons are already functionally sparse - pruning 92 % of the weakest MoE neuron activations only increases the loss by 2.9% Second, the role of a neuron can be split into different "computational functions". Effectively, we can name the contexts in which a neuron fires, and assign reasons to those firings. Most neurons only have one

LessWrong · EN

Opening the datasets behind our AI weather models

Foundational precipitation and geostationary satellite datasets to observe the planet clearly.

DeepMind Blog · EN

Show HN: HackerDatasets-Discover remarkable datasets on HN

Karpathy's HackerTube inspired me to try something similar for datasets. I wanted a place to explore data for project idea and read what people on HN had said about it. The English version starts with five resources: Common Crawl, OpenStreetMap, Our World in Data, World Bank Open Data, and NYC Taxi Trips. Each has an introduction, metadata, selected HN comments linked to their orginal threads, and related resources found in those discussions. I started with Chinese version which has a larger and complete catalog. This is a small English edition site to see if you may need or like it. I would c

Show HN · EN