comma or 'to' before a quotation

What this neuron does

This neuron fires on the comma or word just before a direct quotation begins, such as 'wrote,' or 'said to' followed by quoted speech. Turning it up made the text produce fragments that look like quoted speech, though often jumbled.

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

   how the text may add to the child’s knowledge of the topic. For example, if you were reading a book with your child about dogs, you might note, “A Pomeranian, just like Aunt Elaine’s dog! But Princess is

2. #2 peak

   its relative T. rex, which lived 10-12 million years later. Dr. Mark Loewen, the study’s lead author, noted, “The width of the back of the skull of Lythronax allowed it

3. #3 peak

   the Lord appeared to them, and the glory of the Lord shone around them, and they were filled with great fear. 10 And the angel said to them, “Fear not, for behold, I bring you good news of

4. #4 peak

   Piano.⏎Robert Schumann described Chopin’s music as a “cannon buried in flowers,” and in reviewing his musical contemporary in 1831 proclaimed, “Hats off, gentlemen, a genius!” This appreciation has born the test

5. #5 peak

   delivered by Capt. William Peirce on the ship Desire to the Puritan settlement at Salem, north of Boston. Massachusetts Bay Colony Governor John Winthrop wrote, “Mr. Peirce, in the Salem ship, the Desire, returned

6. #6 peak

   , p. 137).⏎Also, in an earlier time, Jean-Francois Lyotard (1984) wrote, “In contemporary society and culture — postindustrial society, postmodern culture —

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

   closely related to northern Nagari which took its identifiable shape as early as the tenth century A.D. The modern Devanagari, which is now used for writing and printing Sanskrit, Hindi, Nepali, Rajasthani and Marathi

2. #2 peak

   well as I will create a user who has full priviliges to access drupal database:⏎mysql> create database database_name;⏎mysql> CREATE USER user_name@localhost IDENTIFIED BY '

3. #3 peak

   3<|endoftext|>: Grade-2 English Games⏎Sight Words is a simple vocabulary and memory exercise for kids. It is designed to introduce kids to some common dolch sight words. Early readers recognize sight words from having memorized them. In this

4. #4 peak

   .⏎In Years 11 and 12, NSW schools offer a wide variety of languages, catering for beginning students to background speakers.⏎Our Faculty Members⏎Mr John Norberry: email@example.com⏎Mrs Laura Wade: firstname.lastname@example.org

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.