articles after building and room nouns

What this neuron does

This neuron fires on small words like 'a' and 'the' that come right after a place or building noun, such as 'school', 'hall' or 'casino'. Its outputs point toward words like 'library' and 'hall'. Turning it up made the model keep talking about libraries, halls and similar buildings.

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

   recording their heart rates and blood pressure during varying exercise intensities. The very nature of this activity meant that we had to leave the standard classroom and venture into the school gymnasium.⏎With this is mind I wanted to ensure that the data that

2. #2 peak

   2nd class stone or brick cottage with a thatched roof. It had three front windows and two rooms occupied. There were four out-offices a stable, a

3. #3 peak

   by fire in October 1863.⏎For the next 30 years the court moved into a number of temporary quarters: a school, a remodeled jail, a two-story building on the west side of the square and a Baptist

4. #4 peak

   Looming over the busy port, the centrepiece of Dover Castle is the Great Tower - the largest keep in Britain - a four storey building housing a basement,

5. #5 peak

   sing together, the happier we’ll be.⏎I loved singing The More We Get Together when I was a little girl. I remember singing it at school, in the car, at

6. #6 peak

   by river Po’s⏎tributaries and by important old routes. The complexes’ panoramic and prominent locations on hills or mountains and their subdivisions⏎into chapels, let them become territorial reference points easy to identify; we don’t have to forget

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

   towards parking or bike lanes? Bus lanes or street trees? Rain gardens or sidewalks with café dining? The answer likely varies in different situations, but how do we make those

2. #2 peak

   ano, the bridge was completed in 1922. Paolo Alcazaren describes, “the piers of the bridge were festooned with statuary of mermen on dolphins. The bridge balustrades were ornately molded

3. #3 peak

   Candles and devices using this oil-burning technique were the main means used to light homes, employed in candelabras, wall sconces, and elegant chandeliers.⏎It wasn't until the Victorian era that oil lamps improved, thanks

4. #4 peak

   . Hang a long bow and a quiver of arrows conspicuously in your hall or library, and you will soon discover that no exquisite painting or bit of statuary will receive from guests more attention than will be accorded to these ancient weapons

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.