Neuron Explorer

Qwen3-0.6B

opening parenthesis after a name or title

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 fires on an opening bracket that comes right after a name, title or heading, such as 'Elizabeth Woodville (c. 1439-1492)'. It goes the other way at the end of a sentence just before a new heading. Turning it up made the text repeat phrases like 'the role of the', so the steering does 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

    BMC Musculoskeletal Disorders volume 4, Article number: 4 (2003)⏎The purpose of this study is to present an automated system

  2. #2 peak

    Pretty Balloons (counting/number sentences) (Resource Book Only) eBookKindergarten|Grade

  3. #3 peak

    2. Stephen Ruane, 13. James Grealish, 14. John Grealish, 15. James Grealish (Michl), 16. John Carr (Pat), 17.

  4. #4 peak

    Simon, who owned horses and property, is skeptical about the longevity of the new parks.⏎“How are they going to make a park in the<|endoftext|>Elizabeth Woodville (c. 1439-1492) was the daughter

  5. #5 peak

    = 2 x 1/4 or 1/4 is 1/2 of 1/2<|endoftext|>Messier 52 (also known as M 52 or NGC 7654)

  6. #6 peak

    condition that you copy this document without modifications and in its entirety, including this copyright notice.⏎Book Two, Chapter Twenty-Four, "Man and the Evolution" (Part 2 of 3)⏎If indeed man has evolved in Nature out of

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 have a a . I have a a . I have a a. I have a a. I have a a. I have a a. I have a a. I have a

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

    by beautiful natural surroundings, making it an ideal place for meditation and spiritual relaxation. In this article, we will tell you about the history and architecture of the temple.⏎Dhyunsar Mahadev Temple History⏎The history of the temple is

  2. #2 peak

    is the average of these conditions over extended periods. This distinction is key in understanding current weather patterns, including the cold wave, within the broader context of climate change.⏎El Niño and Jet Stream Patterns: Unpredictable Weather Catalysts⏎Recent

  3. #3 peak

    the way you think in order to become more positive. By following these tips, you can start living a more confident life – no matter what your current situation is.⏎Negative self-talk is harmful and can hold us back from achieving our goals.⏎Negative

  4. #4 peak

    can be worn as a nearly invisible patch on the skin. The small size frees the patient from concerns with holding the pump in place and concealing it under clothing.⏎The insulin Nanopump, developed by Debiotech and industrialized by ST

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. L17:1568 strongly suppresses
  2. L16:190 strongly suppresses end-of-document boundary token
  3. L16:234 strongly excites
  4. L17:1176 suppresses
  5. L15:2070 excites
  6. L14:1132 suppresses
  7. L5:203 excites
  8. L17:2848 excites
  9. L17:1906 excites
  10. L1:1842 suppresses
  11. L17:1024 suppresses
  12. L6:2439 weakly excites
  13. L15:1325 weakly excites
  14. L16:313 weakly excites
  15. L17:1316 weakly suppresses document boundaries and line ends
  16. L16:1411 weakly suppresses
  17. L4:658 weakly excites
  18. L12:1310 weakly excites
  19. L16:2189 weakly excites
  20. L15:987 weakly excites
  21. L0:2540 weakly excites
  22. L1:816 weakly suppresses
  23. L3:70 weakly excites
  24. L12:2961 weakly excites

Feeds later layers

  1. L19:550 excites
  2. L20:1162 excites
  3. L19:1940 excites
  4. L22:597 excites end of document token
  5. L19:1201 excites
  6. L19:975 excites
  7. L19:801 suppresses
  8. L21:186 suppresses
  9. L22:820 excites
  10. L19:909 excites
  11. L20:1487 excites
  12. L19:624 excites
  13. L26:449 suppresses
  14. L21:466 excites
  15. L24:11 excites
  16. L19:1069 excites
  17. L19:2615 suppresses
  18. L19:558 suppresses
  19. L25:397 weakly excites
  20. L21:264 weakly suppresses
  21. L23:374 weakly excites
  22. L19:2591 weakly excites
  23. L21:301 weakly excites
  24. L19:1479 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 -14.14 and -13.641 values between -10.62 and -10.121 values between -10.12 and -9.612 values between -9.61 and -9.111 values between -9.11 and -8.611 values between -8.61 and -8.101 values between -8.10 and -7.601 values between -7.60 and -7.101 values between -7.10 and -6.604 values between -6.60 and -6.094 values between -6.09 and -5.592 values between -5.59 and -5.094 values between -5.09 and -4.586 values between -4.58 and -4.085 values between -3.58 and -3.071 values between -3.07 and -2.573 values between -2.57 and -2.076 values between -2.07 and -1.579 values between -1.57 and -1.06152 values between -1.06 and -0.561955 values between -0.56 and -0.061663 values between -0.06 and 0.4559 values between 0.45 and 0.9512 values between 0.95 and 1.454 values between 1.45 and 1.963 values between 1.96 and 2.463 values between 2.96 and 3.472 values between 4.47 and 4.971 values between 5.48 and 5.98offnegativepositiveoften

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. ().__
  2. ST
  3. ocal
  4. 三方
  5. قة
  6. وث
  7. abeth
  8. dist
  9. miss
  10. 硬

Suppresses

  1. ckeditor
  2. 儆
  3. DisplayStyle
  4. și
  5. Guaranteed
  6. Ariel
  7. ioutil
  8. ism
  9. elly
  10. idian