'well' in well-dressed, well-known, well-off

What this neuron does

This neuron fires on the word 'well' when it appears in phrases like 'well-dressed', 'well known', 'well off' or 'well-to-do'. Turning it up made the model write comparisons and repeated phrases, which does not clearly match this word.

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

   activity. During the 1840's, this tiny village, once a mile long, was larger than Charlotte and hosted a collection of the wealthy and well-dressed. After 1849, Gold Hill lost a great deal

2. #2 peak

   и) of the Don, Kuban, Terek and Ural rivers and the Ukrainian Cossacks (козаки), respectively. Somewhat less well known are the Polish Cossacks (Kozacy) and the Tatar

3. #3 peak

   Savannah, Georgia after Jacob encountered financial difficulties. It is not known how much formal schooling Phoebe had, but she was clearly well read and worldly. In 1856, she married a non-J

4. #4 peak

   .⏎This is because climate change appears to be altering the weather conditions wide-ranging birds encounter before, during + after their migration. With some species no longer being as well adapted to, or synchronised with, the conditions they encounter during their travels as

5. #5 peak

   the foremost combat weapons.⏎Acknowledging the significance of the availability of good shields for successful warfare, Carolingian rulers took special care to ensure that their soldiers were well-supplied with ample shields.⏎Shields were used by both the cavalrymen and

6. #6 peak

   composition.⏎The composer was born in 1810 in the village of Zelazowa Wola, not far from Warsaw. His family was not well off, but they had a strong appreciation for music. Chopin’s mother could

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

   1 USA. They are now with the Department of Construction, Chaoyang University of Technology, Taichung 41349, Taiwan (corresponding author's e-mail: firstname.lastname@example.org, email@example.com).⏎Ch

2. #2 peak

   the trends of infectious diseases and the mechanisms of Adaptive diseases.⏎Department of Pediatrics, Soroka University Medical Center, Beer-Sheva, Israel⏎- *Corresponding Author:⏎- Yeongha Wook⏎Department of Pediatrics,⏎Sor

3. #3 peak

   Identify leaks⏎- Determine if any water fixtures or appliances are not working properly⏎- Identify irrigation system problems such as damaged or misdirected spray heads, non-functioning rain sensors or soil moisture sensors⏎- Determine how water is used at the

4. #4 peak

   Like a whole cadre of other prosthetic researchers and makers, they are interested in breaking down the process, figuring out how to turn 3-D printing into something self-sustaining

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.