comma after a 'said' or 'answered' verb

What this neuron does

This neuron fires on the comma or colon right after a speech verb like "answered", "asked" or "yelled", just before a quotation starts. Turning it up did not produce a clear change in the stories.

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

   Upon seeing the many dinosaurs displayed in the museum, the boys asked if evolutionists are right saying the dinosaurs lived tens of millions of years ago.⏎Lisle answered, "They like to believe that dinosaurs are millions of years old and that the earth

2. #2 peak

   Stephanie looks up and nods but does not say anything. Dr. Smith introduces herself to Stephanie’s caseworker, Hannah, then turns back to Stephanie and asks, “Tell me how you have been doing over the past year, Stephanie.”⏎Step

3. #3 peak

   to meet. The women carried signs that read, "Down with the National Conference without women!", "Stop Injustice!", and "Equal Rights!" They also shouted, "Away with the National Conference without Women!"⏎Alio resigned from the Pre

4. #4 peak

   evolved from a banana, does it?"⏎Ryan started to wonder about how the universe began, asking Lisle, "What about the Big Bang?"⏎Lisle answered, "The Big Bang teaches<|endoftext|>Fig.1 above refer to a point moving along

5. #5 peak

   in later.⏎It took fifteen minutes to complete the scene. When it was over, DeMille yelled, “Cut!” and turned to his assistant, all smiles. “That was great

6. #6 peak

   dens, and birds of the sky have nests, but the Son of Man has no place to lay His head.”⏎New American Standard BibleAnd Jesus said to him, "The foxes have holes and the birds of the air [have] nests

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

   if you just ask “Is your person [name]?”.⏎Players are quick to understand that by asking questions that divide the candidates into two groups, such as “is your person female?”, you can reduce the candidate pool more quickly.⏎If

2. #2 peak

   always train new users on how to use these tools.⏎But searching tools alone are usually not enough. For example, a user might search for a term such as "Individual"⏎not knowing that you stored information about human beings under the class "Person

3. #3 peak

   about how it is going to work, he is uncertain about it. When he is pretty sure of how it is going to work, and he tells you, "This is the way it's going to work, I'll bet," he still

4. #4 peak

   link="none" columns="2" size="medium" ids="2568,256

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.