the last digit of a decade year

What this neuron does

This neuron lights up on the final zero in decade years like 1950s, 1970s or 1980s, where the year is written with its last digit split off. Turning it up made the model write about unrelated things, so the steering does not show this pattern.

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

   ten. The number 58⏎rounded to the nearest ten would be 60.⏎Look at the number on⏎each shell. Click

2. #2 peak

   11Vegetarian Diets lower blood pressure for Children . 12 An impressive number of studies, dating back to the early 1920s, show that veg-<|endoftext|>If you take a walk through the Salt Lake

3. #3 peak

   lags behind the industry high of 55 million boxes, reached in the mid 1990's. However, according to Ocean Spray's Grapefruit Juice Marketing Manager Dan Th

4. #4 peak

   late 19th century. The site was excavated by the American Foundation for the Study of Man (AFSM) in the early 1950s and by the Italian Mission to Oman (IMTO) since 19

5. #5 peak

   Over the ensuing years the technology was improved and refined until it became more widely used in hospitals across the country, and indeed globally, in the 1970s.⏎Today, regular scans are a standard part of the care for all pregnant

6. #6 peak

   to be admitted to a prestigious university to get the online courses.⏎North Korea appears to be in roughly the same position as China in the mid-1980s, when China created the Chinese Central Television and Radio University, which is now

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

   eaten between 6:30am and 7:30am and often consists of rice or soup. Lunch is generally eaten between 12:00noon and 1:00pm. Many people eat out, grabbing

2. #2 peak

   0101100.00010000.0yyyyyyy.yyyyyyyy => 172.16.0.0/17 SUBNET2 : 1010

3. #3 peak

   0101100.00010001.0yyyyyyy.yyyyyyyy => 172.17.0.0/17⏎- We need a network with <|endoftext|>|Classification

4. #4 peak

   second number is always lower than the first one. No number will begin with zero.⏎For each expression, print two lines with two given numbers, the second number

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.