the word 'simple' and 'Santa' in Santa Barbara

What this neuron does

This neuron mostly lights up on the word 'Santa' in 'University of California, Santa Barbara' and on pieces of words meaning 'simple' or 'simplify'. When it was turned up, the model started repeating the word 'simple' again and again.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens a prefilled issue on GitHub).

Fires most on

The text windows where this neuron's activation was highest. Amber marks positive activation; the outlined token is the peak. Hover a token for its strength.

1. #1 peak

   cing into swirling marine vortexes called “gyres”. Here's a recording of the Hangout in case you missed it...⏎In 1997, Sea Captain Charles Moore discovered an area in the mid-Pacific Ocean the size of Texas that

2. #2 peak

   informatics Group and the Computational Bioscience Research Center created the tool, along with artificial intelligence researcher and current KAUST Provost Lawrence Carin and clinical collaborators from Harbin Medical University in China.⏎The

3. #3 peak

   and the ability to flex and stretch Led by materials science Associate Professor Michael Arnold ... - Read More⏎A team of chemistry and materials science experts from University of California Santa Barbara and The Dow

4. #4 peak

   they were looking for—an unexplained bump emerging from the data.⏎“I was far from convinced,” says Incandela, a professor at the University of California, Santa Barbara and the former spokesperson of the CMS experiment at the Large Hadron Collider.⏎

5. #5 peak

   whole of the Pepys’s Ballad collection has been digitised and is available on the ‘English Broadside Ballad Archive’ hosted by the University of California Santa Barbara.⏎Broadside Ballad: The North Country Lovers (‘John and

6. #6 peak

   have not yet been directly compared with simpler antiretroviral drug regimens in terms of their safety, efficacy, feasibility

Steering

Push this neuron up or down while the model writes, and watch what changes. The outputs were generated ahead of time, so nothing runs in your browser.

Most negative on

Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.

1. #1 peak

   be safe and effective, review suggests⏎By Robert Preidt⏎MONDAY, Oct. 15 (HealthDay News) -- A technique called the "mother's kiss" is a safe and effective way to remove foreign objects from the

2. #2 peak

   Step 1 - Addition, but don't forget to simplify⏎Step 2 - Drag the correct answer⏎Step 3 - Addition, but don't forget to simplify⏎Step 4 - Multiple choice⏎Try to answer all the 1

3. #3 peak

   math by knowing what to expect.⏎How to Find Common Factors⏎An Efficient Method to Find Common Factors⏎How to Simplify Expressions⏎A straight forward approach to simplifying expressions⏎Integer Practice Problems - Addition⏎Review and Practice Problems for adding

4. #4 peak

   And then the numerators. 1 + 3 = 4. We then get the answer 345 Can we simplify? No, that's not possible.⏎The answer is 315 + 35 = 3

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

Activation histogram

How often this neuron fires at each strength. Almost always it does close to nothing (the tall bars in the middle); the snippets above are the rare moments it fired hard, amber for positive and blue for negative.

Logit lens

The words this neuron nudges the model toward (amber) or away from (blue) when it fires. A hint, not a label: many neurons work through later layers instead.