the word 'Mount' in place names

What this neuron does

This neuron lights up on the word 'Mount' when it begins a mountain or volcano name, such as Mount Everest or Mount Etna. Turning it up did not produce any clear theme in the written text; the output drifted into unrelated or repetitive wording.

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

   going to do about it?"⏎How about it?<|endoftext|>Published on Feb 20, 2013⏎CBS News RAW: A new eruption on Mount Etna occurred at its South-Eastern crater. This event generated tall lava

2. #2 peak

   the High Plains.⏎The Rocky Mountains extend from Alaska in the North to New Mexico in the South. Mountainous regions surround high plateaux. The highest mountain is Mount Elbert 14,431 feet.

3. #3 peak

   the mountain, failed to hear him. They stuck close to the boundary markers, determined to close in and learn what Moses had seen on his quick trip up Mt. Sinai.⏎Suddenly several huge bolts of lightning cracked down from above. Like gigantic swords

4. #4 peak

   esite stratovolcano in the Islands of Four Mountains, a volcanic group in the Aleutian Arc. Like all stratovolcanoes, Mount Cleveland grew as explosive eru<|endoftext|>EMBL scientists use ‘mini organs’ to clarify

5. #5 peak

   (17,400 ft); it receives an average of 450 mm (18 in) of precipitation a year. The climate of Mount Everest is extreme

6. #6 peak

   same in both cases seems to confirm the constancy of h / c. (Annual Report of the Mount Wilson Observatory, 1935-6.) Back.<|endoftext|>I’m

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

   ugs, crayfish, tadpoles and frogs and fish. They’ll also mix in some plant stuff.⏎They are ground nesters, building platforms, or mounds, of cattails or

2. #2 peak

   . Natural shorelines are often more resilient against erosion and have more beneficial relationships to the surrounding environment and animals compared to armored shorelines, such as bulkheads, seawalls, and rock revetments.⏎Working with partner organizations and local groups,

3. #3 peak

   standing calf raises, chest flyes and pull-down exercises on these machines. Other types of weight-training exercises require you to move your own body weight. These include triceps dips<|endoftext|>A paper clip

4. #4 peak

   of hunters and gatherers.⏎Now, new research in the current issue of the journal Geoarchaeology, offers compelling evidence that one of the massive earthen mounds at Poverty Point was constructed in less than 90 days, and perhaps

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.