"for" in same-or-true comparisons

What this neuron does

This neuron fires on the word "for" when a sentence says something is the same, true, or matching for another group, as in "the same is true for businesses" or "the corresponding percentage for whites". Turning it up gave repetitive or off-topic text rather than anything clearly related.

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

   Over the same period the percentage of black adults with a four year college degree jumped from 2.2% to 22.8%. The corresponding percentage for whites in 201

2. #2 peak

   object, and draw a line through them. The best points to use are the corners of an object for which it is easier to match shadow and object. Repeat for as many clearly defined shadow and object points as possible. As you do this,

3. #3 peak

   your code units per conceptual relation as opposed to functional relation. For example, instead of putting all your controllers in one folder, do something like:⏎And the same for deeper code units concepts:⏎Consumer/Event/SomeEvent/SomeEventInterface

4. #4 peak

   in plant sciences. I have worked at a number of institutes and the number of male and female group leaders and department heads is about equal, this is also true for the more junior roles.⏎3.-How has the profile of your profession changed since

5. #5 peak

   insurance, and retirement contributions. Whatever is left is your Net Income and it will be considerably less than $1,500.⏎The same idea holds true for businesses. The difference between gross income and net income can be enormous, and the

6. #6 peak

   at the 2019 Dairy Cattle<|endoftext|>General best practices for performance⏎You must plan for every feature in your app, and the same is true for performance. Planning for performance consists of determining what your performance-critical scenarios will be

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

   i.e., "I don't want to see it again").⏎In #2 "saw" is past tense and neutral about the present. (see The<|endoftext|>In the field of meta-ethics, scholars have identified key problems in the

2. #2 peak

   before and after the main meals. As mouth wash they used another salt solution<|endoftext|>About Apollo Missions"That’s one small step for a man, one giant leap for mankind...” nnOn 20 July 1969, Neil

3. #3 peak

   .⏎• In “The American Frugal Housewife,” (the Better Homes and Gardens of its day) the author recommends using ear wax as lip balm.<|endoftext|>Health care employment has been the bright spot in the otherwise lackluster recent jobs

4. #4 peak

   see practical); the noun from the verb within the earlier spelling practise, on the sample of pairs akin to advise, recommendation.⏎Word Origin Middle English (in legal<|endoftext|>Keywords: Sargent Mountain⏎Historical Items Showing 1 of 1 View

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.