unclear: token just before a k sound

What this neuron does

The neuron is hard to pin down. Its strongest moments come just before a k sound, such as the word before a k in a formula or the start of a word like karmas. Turning it down made the model repeat the letter k over and over.

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

   (9/10)k-2, where k is the number of digits of 2n. The number of digits can

2. #2 peak

   to “zero out” the data blocks. To do so, do one of the following:⏎Use the metainit command with the -k option. The -k option recreates the RAID 5 volume without initializing it, and sets the

3. #3 peak

   V and generally applied up to and including 52 kV . For technical and economic reasons, the service voltage of medium voltage distribution networks rarely exceeds 35 kV.⏎In this chapter, networks which operate at 1000

4. #4 peak

   alternatives are:⏎- Does one karma produce one birth?⏎- Does one karma cause several births?⏎- Do several karmas result in several births?⏎- Do several karmas give rise to only one birth?⏎There are many karmas

5. #5 peak

   How Many Teeth Do Koi Have?⏎The exact number of teeth which a koi carp has depends on its sub-species as well as on the age of the fish. Therefore, it is impossible to answer this

6. #6 peak

   ) where they proved that a(n) is not more than the smallest number k such that the total number of ones in the binary expansion of numbers from 1 to k is at least n − 1. Which means they proved that a(

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

   grandchildren of the ones stayed behind. The interesting story of The First Turks in America will be immortalized at a museum to be established in Turkey.⏎THE FIRST TURKS IN AMERICA⏎1893, the 400th

2. #2 peak

   so loud and aggressive anyway?⏎Why are geese so loud?⏎Most people are familiar with the honk of the Canadian goose. Sometimes, it seems like they honk non-stop. Geese make so much noise for several reasons:⏎● To

3. #3 peak

   bodies buried underneath earthquake rubble, or to land a spacecraft on low-gravity asteroids, or - as the Army suggests - to sneak inside buildings where enemy combatants lurk," notes the SFGate.⏎Stanford robotics engineer Mark Cutkosky

4. #4 peak

   the internet. Either they can’t afford a connection, or none exists where they live. Of all the efforts to bring those people online, Google’s “Project Loon” sounds the most far-fetched.⏎When the search company announced in June

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.