Neuron Explorer

Qwen3-0.6B

comma in 'not X, but Y' contrasts

A neuron in layer 18 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 comma in sentences that set up a 'not this, but that' contrast, such as 'not just a vision of the future, but...'. It seems to be waiting for the second half of the contrast. Turning it up mostly made the writing wander and repeat, so the steering results do not show a clear theme.

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

    1960.”⏎New technologies often adopt the terms and design characteristics of the tech that came before them. So this wasn’t simply the television set of tomorrow, it was the radio set of tomorrow — now with TV!⏎This amazing bit of

  2. #2 peak

    and feeling?⏎Piper urges us to think for the glory of God. He demonstrates from Scripture that glorifying God with our minds and hearts is not either-or, but both-and. Thinking carefully about God fuels passion and affections for God.

  3. #3 peak

    of dieting could trigger a metabolism gone haywire.⏎In this theory, it’s not stubbornness or a mental disorder that keeps anorexics from eating, it’s their own bodies. The theory could explain why it can be so difficult

  4. #4 peak

    .⏎It's an experiment that's been in the works for a couple of years at this point—an attempt to solve the problem of too-dark roads not with lights, but with glowing strips of paint. This week, crews painted the Oss stretch

  5. #5 peak

    it was likely some jobs would be lost but expected it to happen over a long period of time.⏎“It’s not just taking workers out and putting a machine in, it’s going to happen really quite gradually for a long time yet.” Professor At

  6. #6 peak

    phalia and Bavaria continue to reverberate. They have once again raised our awareness of the fact that climate change is not just a vision of the future, but has already long since begun – that we can no longer afford to continue our

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: not only not a thing that is not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not not

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

    is also quite a common reason as to why limit calculation can be difficult for a student. A student who has math anxiety wouldn’t not dislike math but would simply feel fear of failure when they see a difficult question/task

  2. #2 peak

    , has never even been a teacher. He professes to know how teachers should teach in the entire country and what they should teach, and yet, he is not a teacher. Coleman worked for McGraw-Hill (textbook publishing company)

  3. #3 peak

    How many inventions can you think of that were first made in the first decade of the first century AD? No, I can’t think of any either, and not just because it was a long time ago. There

  4. #4 peak

    hold.⏎Goats brows hard, leaving little in their wake of appetite. This is not a system I would recommend to sensitive ecosystems that are fully intact, not without a lot of supervision. Tethering goats takes

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. L16:752 excites
  2. L10:111 suppresses
  3. L13:822 excites
  4. L11:1095 suppresses
  5. L4:151 excites
  6. L0:2047 weakly suppresses
  7. L8:13 weakly suppresses
  8. L17:443 weakly excites
  9. L17:1644 weakly suppresses
  10. L0:2806 weakly excites
  11. L15:2515 weakly excites
  12. L7:64 weakly suppresses
  13. L15:1293 weakly suppresses
  14. L3:1588 weakly suppresses
  15. L8:7 weakly excites
  16. L4:9 weakly excites
  17. L8:1337 weakly excites
  18. L11:169 weakly excites
  19. L16:2137 weakly suppresses
  20. L7:2977 weakly excites
  21. L16:735 weakly suppresses
  22. L10:504 weakly suppresses
  23. L12:2951 weakly suppresses
  24. L8:10 weakly excites

Feeds later layers

  1. L19:1811 strongly excites
  2. L22:635 excites
  3. L24:1779 excites
  4. L19:2544 excites
  5. L21:1717 excites
  6. L25:2002 excites
  7. L22:896 excites
  8. L23:434 suppresses
  9. L20:1873 suppresses
  10. L23:466 excites
  11. L19:2523 excites
  12. L26:406 suppresses
  13. L21:2799 excites
  14. L19:2171 excites
  15. L23:2518 excites sentence-initial 'Not' starting a clause
  16. L21:1974 weakly excites
  17. L21:544 weakly excites
  18. L20:3031 weakly excites
  19. L19:2451 weakly excites
  20. L19:259 weakly suppresses
  21. L23:1127 weakly excites
  22. L21:2310 weakly suppresses
  23. L21:3046 weakly excites 'at' in 'not at all' and 'at all'
  24. L19:3025 weakly excites

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 -6.61 and -5.901 values between -5.90 and -5.182 values between -5.18 and -4.474 values between -3.75 and -3.046 values between -3.04 and -2.324 values between -2.32 and -1.61114 values between -1.61 and -0.891418 values between -0.89 and -0.182309 values between -0.18 and 0.5416 values between 0.54 and 1.253 values between 1.25 and 1.978 values between 1.97 and 2.686 values between 2.68 and 3.405 values between 3.40 and 4.111 values between 4.11 and 4.832 values between 5.54 and 6.263 values between 6.26 and 6.971 values between 6.97 and 7.691 values between 7.69 and 8.401 values between 9.83 and 10.551 values between 11.98 and 12.691 values between 21.27 and 21.98offnegativepositiveoften

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. 而是
  2. sondern
  3. anymore
  4. nor
  5. 那么简单
  6. אלא
  7. nor
  8. necessarily
  9. any
  10. ched

Suppresses

  1. 并不会
  2. 不知
  3. atica
  4. 不明白
  5. haven
  6. 并非
  7. 没有办法
  8. olean
  9. tér
  10. 憬