first digit '1' of a year

What this neuron does

This neuron lights up on the digit 1 at the start of a year, such as the 1 in 1871 or 1937. Turning it up made the model's text drift into repetitive, unrelated sentences, so the effect on writing is not clear.

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

   and was the House's lead manager in the Johnson impeachment proceedings. As Chairman of the House Committee on Reconstruction, Butler authored the Ku Klux Klan Act of 1871 and coauthored the landmark Civil Rights Act of 18

2. #2 peak

   Tax Act of 1937⏎On August 2, 1937, President Franklin Roosevelt signed the Marihuana Tax Act of 1937 into law. The act required all producers, laboratories, dealers,

3. #3 peak

   Studies and senior fellow in the Institute of Irish Studies, Queen’s University, Belfast. His books on Irish history and government include Ireland and Anglo-American Relations, 1899-1921 (1969) and The

4. #4 peak

   use material, especially from the Internet, without needing the author's permission or paying licensing fees and royalties. Section 107 of the Copyright Act of 1976 specifically provides for “fair use” of copyrighted materials for teaching,

5. #5 peak

   0⏎The French and Indian War, was the North American phase of the Seven Years War in Europe. It was a war fought between Britain and France from 1755 to 1760 in territory that became Canada and the

6. #6 peak

   tal from Bessarion.Converted from Orthodoxy to Roman Catholicism about 1427, Trebizond conducted his own school at Venice (1428-1

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

   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(30) ≤ 16.Sh

2. #2 peak

   given_arr = [1, 2, 3, 4, 5] i = len(given_arr) - 1 while(i >= 0): print(given_arr[i]) i = i - 1⏎In this

3. #3 peak

   3, 42]⏎-- Without library⏎f :: Bool -> [Int]⏎f b = if b then xs else ⏎-- With library⏎g :: Bool -> [Int]⏎g = mwhen xs⏎See? It’s an amazing

4. #4 peak

   to abcteach⏎Free Account (Settings)⏎An 8 page study unit for Ron

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.