Neuron Explorer

Qwen3-0.6B

opening parenthesis after a citation or term

A neuron in layer 10 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 an opening bracket that comes right after a name, citation, or term, where the text is about to add an explanation, a year, or a translation. Turning it up made the model drift into repetitive, off-topic rambling, which doesn't say much about the neuron itself.

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

    ak Dun v. Yukon, 2015 YKCA 18 (the “appeal decision”), the Yukon Court of Appeal unanimously allowed the Yuk

  2. #2 peak

    innovative drugs have been introduced, from Pfizer's (<|endoftext|>Information identified as archived on the Web is for reference, research or recordkeeping

  3. #3 peak

    the use of English and official language in the content classrooms, typically mathematics and science. According to Cummins, bilingual education is defined as “the used of two (or more) language of instruction at some point in a student’s school career”

  4. #4 peak

    muscle group is necessary for optimal muscle growth. For instance, Fonseca et al. (2014) showed that combining different lower-body exercises resulted in muscle growth

  5. #5 peak

    , 168 note) who remarks: — "Um dies (i.e. Varro's description of the extent of Roma quadrata) mit

  6. #6 peak

    “ponte las pilas,” (get to work,) are the speak of Latino Pachucos of the

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: "I know the words for it, and I know the meaning. And I know it's a very simple concept. But how to think about it? I know that it's a very simple concept. It is the most basic level of

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

    send, (and hopefully, receive), a valentine.⏎In the 1840s

  2. #2 peak

    innovative ways to diagnose the disorder earlier (as we reported here, here and here) and on new genes that contribute to it (which we described here).⏎(More on TIME.com: Health Checkup: Kids and Mental Health)⏎For the new

  3. #3 peak

    with opportunities for continued learning and for future employment, both domestically and internationally, in many areas such as commerce,<|endoftext|>Ancient stone circle holds air of mystery (it's in Britain, but it's not Stonehenge)⏎However, don't

  4. #4 peak

    marks the celebration the birth of Jesus Christ. Families usually gather on Christmas Eve or Christmas day to give presents. Households are decorated with lights and Christmas trees (and much more).⏎This brief synopsis illustrates that the major holidays of various traditions share a

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. L4:64 strongly excites
  2. L7:100 strongly suppresses
  3. L9:982 strongly suppresses
  4. L4:45 strongly suppresses
  5. L6:468 strongly excites
  6. L6:409 excites
  7. L0:11 excites
  8. L1:155 suppresses
  9. L5:59 suppresses
  10. L9:886 suppresses
  11. L4:409 excites
  12. L7:732 excites
  13. L5:1061 suppresses
  14. L8:770 excites
  15. L1:102 excites
  16. L9:2249 suppresses
  17. L5:330 excites opening parenthesis after a noun
  18. L2:450 suppresses
  19. L8:354 excites opening parenthesis before a citation year
  20. L3:57 excites
  21. L0:1622 excites
  22. L4:327 weakly suppresses
  23. L1:195 weakly suppresses
  24. L3:1388 weakly excites

Feeds later layers

  1. L16:264 strongly suppresses opening parenthesis in citations and asides
  2. L11:50 strongly suppresses
  3. L16:869 strongly suppresses
  4. L12:1287 strongly suppresses
  5. L18:682 strongly suppresses
  6. L17:1509 strongly suppresses
  7. L12:2933 strongly suppresses
  8. L16:1967 strongly suppresses
  9. L20:335 suppresses
  10. L20:1451 suppresses
  11. L17:304 suppresses opening parenthesis after a claim
  12. L19:889 suppresses
  13. L21:187 suppresses
  14. L20:104 excites
  15. L12:2745 suppresses
  16. L15:1374 suppresses
  17. L11:2191 suppresses unclear: brackets and parentheses in text
  18. L11:907 suppresses
  19. L15:2486 suppresses bracket or parenthesis inside a quote
  20. L25:213 suppresses
  21. L18:1365 suppresses
  22. L19:453 suppresses
  23. L19:2339 suppresses
  24. L15:608 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 -7.24 and -7.031 values between -6.20 and -6.001 values between -6.00 and -5.791 values between -5.79 and -5.581 values between -5.37 and -5.162 values between -4.75 and -4.543 values between -4.54 and -4.331 values between -4.33 and -4.121 values between -4.12 and -3.922 values between -3.92 and -3.711 values between -3.71 and -3.503 values between -3.50 and -3.293 values between -3.29 and -3.085 values between -3.08 and -2.885 values between -2.88 and -2.672 values between -2.67 and -2.462 values between -2.46 and -2.254 values between -2.25 and -2.051 values between -1.84 and -1.633 values between -1.63 and -1.421 values between -1.42 and -1.213 values between -1.21 and -1.015 values between -1.01 and -0.802 values between -0.80 and -0.594 values between -0.59 and -0.384 values between -0.38 and -0.1710 values between -0.17 and 0.03506 values between 0.03 and 0.241551 values between 0.24 and 0.451287 values between 0.45 and 0.66445 values between 0.66 and 0.8747 values between 0.87 and 1.07offnegativepositiveoften

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. LETE
  2. レーション
  3. rene
  4. roperty
  5. Penguin
  6. 增值
  7. 从此
  8. 可靠
  9. isi
  10. contin

Suppresses

  1. 硐
  2. *);⏎
  3. 箴
  4. *)⏎
  5. &)
  6. "")⏎⏎
  7. &);⏎
  8. ]="
  9. ComVisible
  10. *)"