Neuron Explorer

Qwen3-0.6B

opening parenthesis in citations and asides

A neuron in layer 16 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 fires on an opening bracket that follows a citation title, a year, or a copyright line, such as 'Copyright (c)' or 'Volume (Year)'. Its strongest negative cases are brackets that open a side note after a sentence, like '(click anywhere to close)'. Steering produced looping or off-topic text, which does not help explain it.

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

    ter, Taslima and Simonovic, Slobodan P., "A General Overview of Multi-objective Multiple-participant Decision Making for Flood Management" (2002). Water Resources

  2. #2 peak

    of creating a “thermostatic mechanism” that would counter-balance the dominant presence of electronic media in the lives of children.⏎How to Cite⏎Copyright (c) 2017 Robert Albrecht⏎Copyrights and publishing

  3. #3 peak

    water, were the major factors shaping phytoplankton communities.⏎- Abstract views: 1777⏎- PDF: 842⏎Copyright (c) 2009 Silvia TAVERNINI, Daniele N

  4. #4 peak

    outside the team • The tasks the team perform have consequences that affect others inside or outside the organisation Figure 6.1 Characteristics of an effective team 13 (e) " .⏎Members of the team perform their roles and responsibilities with discipline,

  5. #5 peak

    a different version under "Related research" (further below) or search for a different version of it.⏎Volume (Year): 42 (2011)⏎Issue (Month):

  6. #6 peak

    2022 (during Earth Science Week). Geologic Map Day is a special event designed to promote

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: the idea of a “two mind, two mind, two mind, two mind, two mind, two mind, two mind, two mind, two mind, two mind, two, two, two, two, two, two, two

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

    . However, showing you is much cooler than explaining it.⏎Thanks for Subscribing!⏎You'll need to confirm your email, so please check your inbox.⏎(click anywhere to close)<|endoftext|>According to findings of a major HIV prevention trial presented Monday

  2. #2 peak

    solving of problems using Intel Visual Fortran compiler with IMSL Libraries. The use of other software packages (Maple, MathLab, Mathematica) or compilers (?, ?++, Visual Basic) for code is

  3. #3 peak

    ,⏎- R is the electrical resistance,⏎- I is the electric current,⏎- V is the voltage.⏎(Note: this is not true where the impedance is complex)⏎|This page uses Creative

  4. #4 peak

    This helped the environmentalists a lot.⏎(Note: The background section of this page uses most of the same words as Terra Fut

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. L10:66 strongly suppresses opening parenthesis after a citation or term
  2. L6:468 strongly excites
  3. L4:64 strongly excites
  4. L11:50 strongly excites
  5. L9:982 strongly suppresses
  6. L0:11 strongly excites
  7. L1:102 excites
  8. L7:100 suppresses
  9. L14:2454 excites
  10. L10:142 suppresses opening parenthesis mid-sentence
  11. L4:45 suppresses
  12. L15:608 suppresses
  13. L8:770 excites
  14. L1:155 suppresses
  15. L9:2249 suppresses
  16. L3:10 excites
  17. L0:1622 excites
  18. L11:1579 suppresses
  19. L3:57 excites
  20. L5:330 excites opening parenthesis after a noun
  21. L6:409 excites
  22. L14:90 suppresses
  23. L3:1388 excites
  24. L0:86 excites

Feeds later layers

  1. L20:335 strongly suppresses
  2. L21:187 strongly suppresses
  3. L17:238 strongly suppresses
  4. L18:682 strongly suppresses
  5. L19:1099 strongly suppresses
  6. L19:889 suppresses
  7. L18:910 suppresses
  8. L17:304 suppresses opening parenthesis after a claim
  9. L17:1509 suppresses
  10. L24:336 suppresses
  11. L20:104 excites
  12. L22:475 suppresses
  13. L18:1501 excites
  14. L19:453 suppresses
  15. L26:88 suppresses
  16. L25:415 suppresses
  17. L19:671 suppresses
  18. L22:453 suppresses
  19. L26:208 suppresses
  20. L21:1585 suppresses
  21. L22:246 suppresses
  22. L21:210 excites
  23. L18:1420 suppresses
  24. L24:125 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 -9.75 and -9.421 values between -9.08 and -8.751 values between -7.41 and -7.081 values between -6.74 and -6.413 values between -6.41 and -6.071 values between -6.07 and -5.744 values between -5.74 and -5.411 values between -5.41 and -5.071 values between -5.07 and -4.741 values between -4.74 and -4.403 values between -4.07 and -3.731 values between -3.73 and -3.402 values between -3.07 and -2.731 values between -2.73 and -2.402 values between -2.40 and -2.061 values between -2.06 and -1.734 values between -1.39 and -1.069 values between -1.06 and -0.732 values between -0.73 and -0.3916 values between -0.39 and -0.06188 values between -0.06 and 0.281588 values between 0.28 and 0.611580 values between 0.61 and 0.95462 values between 0.95 and 1.2832 values between 1.28 and 1.611 values between 1.61 and 1.951 values between 3.29 and 3.62offnegativepositiveoften

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. (
  2. (
  3. (
  4. raphics
  5. (
  6. 专业
  7. YPE
  8. 波
  9. AGO
  10. frei

Suppresses

  1. )<=
  2. .*?)
  3. )</
  4. +)\
  5. .*)
  6. []):
  7. {})
  8. sts
  9. 妲
  10. казать