colon after a said or explains verb

What this neuron does

The neuron fires on the colon that comes right after a speaker phrase, like "Professor Bill Davies said:" just before a quote. Turning it up made the model produce strings of asterisks and blanks, which have nothing to do with quotes.

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

   device, and warn them when they are at risk of a seizure and how they can modify their behaviour or change their situation to prevent this.”⏎Professor Schelter added: “We live in a world where there is a massive amount of data, and

2. #2 peak

   unique partnership between the BBC, the British Museum and 350 museums and institutions across the country.⏎Director of The Manchester Museum, Dr Nick Merriman said: "It's

3. #3 peak

   how brain circuits are arranged.⏎Andrew Leuchter, director of the Semel Institute’s TMS clinical and research centre (University of California, LA) explains: "By pulsing it with energy repeatedly, we’re changing the way that area

4. #4 peak

   to a global market.⏎One of the problems of existing vaccines against foot and mouth disease is identifying which animals have been vaccinated and which haven't.⏎Dr Charleston says: 'The complete absence of some viral proteins from this new vaccine will also allow companion

5. #5 peak

   used to help reduce the engine's fuel consumption.⏎Commenting on the projects, Project lead and Lecturer in the Department of Mechanical Engineering, Dr Copeland said: "These two research projects are at the cutting edge of research in turbomachinery

6. #6 peak

   Professor Bill Davies has been exploring the challenges likely to be posed by global population growth for the UK’s first free online course in food security.⏎Professor Bill Davies said: “Food security is one of the grand challenges facing the world, there is going

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

   by Sam.⏎Example of Present Perfect: Active: Many tourists have visited that castle. Passive: That castle has been visited by many tourists.⏎Example of Future (Will): Active: Tonight, John will wash the dishes. Passive: Tonight, the

2. #2 peak

   Related research" (further below) or search for a different version of it.⏎Volume (Year): 42 (2011)⏎Issue (Month): 2 (June)⏎|Contact details of provider:|| Web page:

3. #3 peak

   . I'll see her tomorrow.⏎In #1 "going to" implies a previous arrangement or understanding.⏎In #2 "will" does not carry the implication of an arrangement, and can be a spontaneous decision

4. #4 peak

   his father not died? (9) They had returned the home? (10)You had finished this book?⏎Past Perfect Tense Negative⏎Subject + had not + verb 3rd from + object.⏎Past Perfect Interrogative Negative

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.