closing bracket after an acronym

What this neuron does

This neuron fires on the closing bracket that ends a short name or abbreviation written in brackets, such as the "(IP)" in "ingress protection (IP)" or "(TNC)" in a company description. Turning it up made the text repetitive and dull.

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

   As it turns out, "cuteness" improves cognitive function, as evidenced by the pattern of popular cat pictures used as visual mnemonics ("mems") on the Memrise platform. If it can help people learn world capitals and the

2. #2 peak

   and military units in the course of the history of Eastern Europe and adjacent territories. The main and most numerous group is the Russian Cossacks (казаки) of the Don, Kuban, Terek and Ural rivers and the Ukrainian

3. #3 peak

   is, distributors are a higher link than dealers. In practice, the distributor can be either a large company, for example, a transnational corporation (TNC) or an IP.⏎Both the dealer and the distributor are engaged in sales, replen

4. #4 peak

   it was acceptable to discharge Marcellus brines into rivers. Therefore, brines were discharged to some PA rivers through publicly operated treatment works (POTW) and industrial

5. #5 peak

   water submersion. Although “waterproof” does not have a technical definition, most companies who use the term have evaluated the product’s ingress protection (IP) rating in a laboratory (“ingress” means “entrance” and is just

6. #6 peak

   computing.⏎Structure: Modeling Real-World Scenarios with Graphs⏎A graph (or network) is a data structure. It is composed of vertices (dots) and edges (lines). Many real-world scenarios can be modeled as a graph.

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

   - 10 Why is Taoism the best religion?⏎- 11 What does Tao mean?⏎- 12 Is Daoism a religion?⏎- 13 What religions are banned in China?⏎- 14 What are two

2. #2 peak

   daily requirement vitamin A⏎- Almost 40 percent of your daily requirement of vitamin C⏎- About 15 percent of your daily requirement of magnesium⏎- Around 18 percent of your daily requirement of potassium⏎- Roughly

3. #3 peak

   that the celery is safe for cats to eat.⏎Cats can definitely eat celery and benefit from it. Cats can eat some veggies in moderate, well, actually, small amounts. Cats can eat celery

4. #4 peak

   not trying.⏎13. Remember that failure is an event, not a person.⏎14. I expect him to get over the shock of his failure.⏎15. The best teacher is experience learned from failures.⏎16. A hatred

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.