Neuron Explorer

Qwen3-0.6B

'of' inside set phrases like 'course of'

A neuron in layer 21 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron lights up on the word 'of' in fixed phrases such as 'course of action', 'one of the largest' and 'Table of Contents'. It goes quiet or flips negative on 'of' in 'of course', and turning it up made the model drift into odd 'of' and 'the of' loops.

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

    by University of Minnesota students. You can see all of the details involved in making it in the video below.<|endoftext|>Create a Graph⏎Library Home || Full Table of Contents || Suggest a Link || Library Help⏎|National Center for Education Statistics

  2. #2 peak

    grammatical error in this sentence. It should be punctuated with a question mark, not a period. Second, the title of fourth-grade social studies’ course of study is North Carolina History. This question does not address

  3. #3 peak

    about virtuous living by thinking over the qualities of each virtue. 2. Meditation can help a child learn to think for themselves, and determine a best course of action by reflecting on possible solutions. 3. If the child is being raised

  4. #4 peak

    of the state of Oaxaca. He was president of Mexico five times. He helped draft the Plan of Ayutla, demanding the deposition of Santa Anna and the implementation the Constitution of

  5. #5 peak

    is applicable. In general, these models are applied to problems with the following characteristics:⏎The problem has a well-defined sole objective.⏎There are alternative courses of action to resolve the problem.⏎The decision variables are continuous and they can accept any

  6. #6 peak

    Achievement Gap⏎Each year more than 10,000 youth in Minnesota disengage from school, and Minnesota’s academic achievement gap has grown to one of the largest in the nation.⏎Have you considered leaving school? Are you currently dis

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.

turned down a lot
Prompt

Here is something I have been thinking about lately: course of course, course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course course

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

    opportunities or even just advice. They can connect⏎directly with employers and recruiters who can help them reach their goals faster than they ever thought possible. And, of course, Guru99 notes that online job boards function as the perfect resource for

  2. #2 peak

    strap on our virtual reality goggles, and hit the weights.⏎Researchers from the University of Wisconsin-Madison are doing something like that (although, of course, the weights don't weigh anything) by creating a virtual reality simulator called

  3. #3 peak

    Airplane” was once quoted as saying “If you can remember anything about the 60’s, then you weren’t really there!!”. He was, of course, referring to the rampant drug use which has become intrinsically associated with

  4. #4 peak

    answered or increasingly misplace items -- forget appointments, forget medications. If these are new changes, then that's a worry sometimes for Alzheimer's disease. And, of course, if this forgetfulness increases, then other people have to help the individual

Connections

Show on map

Which neurons in earlier layers feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.

Fed by earlier layers

  1. L19:715 strongly excites
  2. L20:2800 excites
  3. L20:2860 excites
  4. L17:317 excites 'of' after a noun phrase
  5. L20:1917 excites
  6. L20:1356 excites
  7. L20:2599 suppresses
  8. L19:2507 suppresses
  9. L20:2962 suppresses "of" in "most X of all" superlatives
  10. L19:2128 excites
  11. L19:2622 excites
  12. L19:1503 excites
  13. L16:961 excites
  14. L18:1954 suppresses
  15. L19:3054 excites
  16. L19:650 weakly suppresses
  17. L18:769 weakly excites
  18. L17:1771 weakly excites
  19. L17:2409 weakly suppresses
  20. L20:55 weakly suppresses
  21. L7:656 weakly excites
  22. L20:302 weakly excites
  23. L18:445 weakly excites
  24. L1:1940 weakly excites

Feeds later layers

  1. L26:711 strongly suppresses 'years' after an age number
  2. L23:567 strongly suppresses
  3. L22:142 strongly excites 'of' after a quantity or noun phrase
  4. L27:447 strongly suppresses
  5. L23:258 strongly suppresses
  6. L25:2020 strongly suppresses
  7. L26:115 strongly excites
  8. L26:920 strongly suppresses
  9. L22:288 strongly suppresses
  10. L24:860 suppresses
  11. L26:279 excites
  12. L25:18 excites
  13. L27:658 suppresses
  14. L22:1993 suppresses
  15. L24:2753 excites
  16. L23:816 suppresses
  17. L23:2665 suppresses
  18. L24:1558 suppresses
  19. L26:3056 suppresses
  20. L26:309 excites
  21. L27:1582 weakly excites
  22. L24:359 weakly suppresses
  23. L26:1320 weakly suppresses
  24. L26:54 weakly suppresses

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.

1 values between -112.38 and -109.171 values between -48.24 and -45.031 values between -19.38 and -16.171 values between -16.17 and -12.964 values between -12.96 and -9.7617 values between -9.76 and -6.5537 values between -6.55 and -3.341260 values between -3.34 and -0.142572 values between -0.14 and 3.077 values between 3.07 and 6.283 values between 6.28 and 9.493 values between 9.49 and 12.691 values between 12.69 and 15.90offnegativepositiveoften

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.

Promotes

  1. знания
  2. issant
  3. 불구하고
  4. gend
  5. 什么地方
  6. zac
  7. יסט
  8. 并且
  9. 并与
  10. 并将

Suppresses

  1. of
  2. course
  3. course
  4. _of
  5. →of
  6. Of
  7. of
  8. Of
  9. của
  10. -course