"to" after "proved" or "proves"

What this neuron does

The neuron fires on the word "to" right after "proved" or "proves", as in "proved to be". It is one of the most consistent patterns here. Turning it up or down did not make the model write anything about proving things, so the steering gave no clear sign.

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

   Wiliam’s falling off with his wealthy maternal uncles and going to his elder brother Edward to become a tradesman like him. Edward, however, proves to be a cynical and tyrant master and Wiliam is soon obliged to leave

2. #2 peak

   hold it by hand all based on what part of the room one would like to listen to and get the acoustic image from. The 40 cm camera proved to be very handy in a measurement situation where it was not given beforehand exactly where the

3. #3 peak

   do things they enjoy,<|endoftext|>86 is conjectured to be the largest power of 2 not containing a zero. This simply stated conjecture has proven itself to be proof-resistant. Let us see why.⏎What is the probability that the nth

4. #4 peak

   low, it was very difficult to detect any real change in sound pressure level from different measurement positions. Also trying to locate the origin by hearing in different positions proved to be futile when trying to determine source position. In addition to hearing the noise in

5. #5 peak

   end of the war.⏎Before World War I⏎Even before World War I⏎, the German military was already looking forward to replacing the heavy machine guns which proved to be such a success in that war. The MG13⏎was one of

6. #6 peak

   “inside” lots were sold as high as $1000.00. Within four months, these men<|endoftext|>Our discourse by now may have proved to the reader that the⏎British Empire is not as mythical as the propaganda would have

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

   2012 June 23⏎Explanation: As seen from Frösön island in northern Sweden the Sun did set a day

2. #2 peak

   m2 = 10-19barn). It is estimated neutrino cross-section for interaction increases linearly with energy of incident neutrino.⏎Reference: Griffiths, David, Introduction to Elementary Particles, Wiley, 19

3. #3 peak

   questions you need to know. I have her whole file and know her well.” As the conversation progresses,

4. #4 peak

   er.2008 May 27⏎Credit: Phoenix, HiRISE, NASA, JPL-Caltech, Univ. Arizona⏎Explanation: The Phoenix lander's footpads are about the size of a dinner plate

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.