'because' and 'that' after a reason

What this neuron does

This neuron fires on the word 'because' (or 'that') when it introduces a reason, as in 'the reason is because...'. Turning it up or down did not clearly change the writing in a way that matches this.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

   decision affects most of the outcome in this play. “To turn your households’ rancor to pure love.” Now the feuding families eventually stopped but only because Romeo and Juliet died. This was mainly because of the Friar.⏎The Cap

2. #2 peak

   his own for three hours, and until he got back she was confused about why he didn’t come when she called him.⏎Why would no one be puzzled? Because there’s one he and there’s one she and life is pronominally simple

3. #3 peak

   of a ‘slowing’ of the metabolism is that older people burn fewer calories on average per kilo of bodyweight than do younger people. But that’s because their lean body mass is less.⏎In a book called Biomarkers, authors William

4. #4 peak

   pe climates, usually border desert climates. In both climate groups, the temperature change between day and night is considerable. One reason for the wide swings in temperature is that the skies are clear and the air is dry. Clouds would reflect much of

5. #5 peak

   0 minutes – it’s essentially moving in a straight line. The reason the path is all bendy is because Swift itself is in motion! Swift orbits the Earth, circling us once every

6. #6 peak

   . At the end of the decade, the median American family had 30% more purchasing power than at the beginning. Inflation was minimal, in part because of Eisenhower’s efforts to balance the federal budget. Unemployment remained low, about

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

   (in the United States) 911 immediately. Always seek the advice of your doctor before starting or changing treatment. Medical professionals who provide responses to health-related

2. #2 peak

   The EPA conducted tests and found that architectural coatings were a large contributing factor to this sort of damage, and that it was therefore essential to limit the amount of emissions from this source.⏎VOC regulations also strive to protect the water

3. #3 peak

   quotes above come from a June 2016 piece he penned on LinkedIn, which ended “

4. #4 peak

   which could negate the effects of gravity. However, as sometimes happens in the Toobworld Dynamic, its use in the 2010 TV movie based on that novel was not its first mention in the TV Universe.⏎By the 2

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.