Categories: FAANG

Language models can explain neurons in language models

We use GPT-4 to automatically write explanations for the behavior of neurons in large language models and to score those explanations. We release a dataset of these (imperfect) explanations and scores for every neuron in GPT-2.

How can we tell if AI is lying? New method tests whether AI explanations are truthful

Given the recent explosion of large language models (LLMs) that can make convincingly human-like statements, it makes sense that there's been a deepened focus on developing the models to be able to explain how they make decisions. But how can we be sure that what they're saying is the truth?

June 6, 2025

In "AI/ML News"

ExpertLens: Activation Steering Features Are Highly Interpretable

This paper was accepted at the Workshop on Unifying Representations in Neural Models (UniReps) at NeurIPS 2025. Activation steering methods in large language models (LLMs) have emerged as an effective way to perform targeted updates to enhance generated language without requiring large amounts of adaptation data. We ask whether the…

November 8, 2025

In "FAANG"

Vertex AI Example-based Explanations improve ML via explainability

September 1, 2022

In "FAANG"

AI Generated Robotic Content

Next Is AI electricity or the telephone? »

Previous « Reshaping IT automation with IBM Watson Code Assistant

Published by

AI Generated Robotic Content

Tags: ai/mlfaang

3 years ago

Using depth maps and weight noising to get better character LoRAs

A few weeks ago I introduced a new method for training style LoRAs which has…

4 hours ago

AI/ML Research

The Statistics of Token Selection: Logits, Temperature, and Top-P Walkthrough

When large language models, or LLMs for short, produce outputs, several criteria are at stake,…

4 hours ago

FAANG

Process financial documents using Amazon Bedrock Data Automation

Financial institutions process thousands of documents daily, including tax forms, loan statements, and purchase orders.…

4 hours ago

FAANG

Introducing Google AI Threat Defense to help you outpace the adversary

aside_block <ListValue: [StructValue([('title', 'Summary of today’s news'), ('body', <wagtail.rich_text.RichText object at 0x7f00683723a0>), ('btn_text', ''), ('href',…

4 hours ago