"of" after a noun in "X of Y"

What this neuron does

This neuron lights up on the word "of" when it joins a noun to what follows, as in "course of action", "speed of the computer" or "frame of the car". When it was turned up, the model drifted into repeated filler words like "the process of the human life".

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

   been no previous indication that Chloroquine has been harmful to pregnant woman.⏎I think that it is important to emphasize that patient panic does not lead to a run of finding this medication on the internet. This drug should be targeted for prevention to high

2. #2 peak

   of pus⏎- Tiny red spots on the area at the back of the roof of the mouth (soft or hard palate)⏎- Swollen, tender lymph nodes in

3. #3 peak

   positive or negative.<|endoftext|>Think of your body like a car. Where the muscles are the shock systems and the skeleton (joints/bones) is the frame of the car. The job of the shock system is to absorb any force that is

4. #4 peak

   the image, the speed of the computer, the number files, and the size and resolution of the selected window

5. #5 peak

   about virtuous living by thinking over the qualities of each virtue. 2. Meditation can help a child learn to think for themselves, and determine a best course of action by reflecting on possible solutions. 3. If the child is being raised

6. #6 peak

   feel even worse than before as the infection takes hold.⏎Once you know what is causing your symptoms, you can talk to a pharmacist or doctor about the best course of

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

   and philosophical principles, this is a thorough exploration of the concept and practice of⏎liberalism.ABOUT THE SERIES: The Very Short Introductions series from Oxford University Press contains hundreds of titles in almost every subject area. These pocket-sized

2. #2 peak

   grass-eating ox.⏎Has a nation ever exchanged its gods?(But they were not gods!)Yet My people have exchanged their Gloryfor useless idols.⏎“Tarshish was your trading partner because of your great wealth of every kind. They exchanged

3. #3 peak

   “Civil Rights Act” of 1964. The whole world watched in awe and amazement as Neil Armstrong became the first human being to ever set foot on the moon. The

4. #4 peak

   the input-output pair. It is called supervised learning because the process of an algorithm learning from the training dataset can be thought of as a teacher

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.