Neuron Explorer

Qwen3-0.6B

'of' after a noun

A neuron in layer 1 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 the word 'of' when it follows a noun such as 'designs', 'contents' or 'modifications'. When it was turned up, the model's text did not visibly change in a way that matched this pattern, so the link to the output is unclear.

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

    ⏎This compendium comprises articles on legal and technological issues of Internet Governance in sphere of human rights prepared by the group of legal and technical scholars of information studies of the Higher School of Economics. This compound is devoted to the workshop on ‘Human

  2. #2 peak

    especially when the hunting pressure increases.⏎Last season I spent some time examining stomach contents of harvested deer in the pine woods of East Feliciana Parish and in the bottom

  3. #3 peak

    in the early 1930s. But that didn’t stop companies from speculating about TV designs of the future. Like this Marconi-brand “radio set of 196

  4. #4 peak

    ERS.⏎The post 1929 Stock Market Crash and the Great Depression. first appeared on essaypanel.com.<|endoftext|>This tutorial shows how to send modifications of code in the right way: by using patches.⏎

  5. #5 peak

    is identified as a counsel, usually with a flat area, of a complete or component of a region. The job of any map is usually to describe spatial interactions of distinct characteristics that this map aspires to signify. There are several forms of maps

  6. #6 peak

    deepening sense of self and trust in their ability to stand up for themselves, with peers and beyond.⏎This approach to teaching them to argue—through our behaviors of patience, reconsidering, acknowledging, and welcoming their influence in our thinking—takes

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 been thinking about something. I have been thinking about something. I have been thinking about something. I have been thinking about something. I have been thinking about something. I have been thinking about something. I have been thinking about something

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

    ty to Animals (ASPCA), "Animals in Entertainment" refers to any animal(s) used to act, perform, fight and/or kill for the enjoyment of humans. The term encompasses many different forms of entertainment – from circuses to movies

  2. #2 peak

    eyes.<|endoftext|>In January, CLBB hosted a panel discussion, “Empathy: The Development and Disintegration of Human Connection,” meant to explore the remarkable capacity of humans to relate to others,

  3. #3 peak

    America placed first, second or third in 20 events while Greece scored in 46 events and Germany placed in 13 competitions. To the delight of the hometown crowd, Greek runner Spyridon Louis won the marathon. The first

  4. #4 peak

    treaties ‘recall that human rights are universal and shall apply to all individuals, and stressing therefore its commitment to guarantee the equal dignity of all human beings and the enjoyment of rights and freedoms of all individuals without discrimination on any ground’ .However, despite

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. L0:1257 weakly suppresses
  2. L0:1525 weakly excites
  3. L0:1531 weakly excites
  4. L0:1220 weakly suppresses
  5. L0:277 weakly excites
  6. L0:15 weakly excites
  7. L0:1219 weakly suppresses
  8. L0:405 weakly suppresses
  9. L0:2846 weakly excites
  10. L0:178 weakly suppresses
  11. L0:1887 weakly suppresses
  12. L0:1604 weakly excites
  13. L0:2183 weakly excites
  14. L0:2547 weakly excites
  15. L0:2300 weakly excites
  16. L0:1224 weakly suppresses
  17. L0:120 weakly excites
  18. L0:510 weakly suppresses
  19. L0:451 weakly suppresses
  20. L0:2917 weakly suppresses
  21. L0:2528 weakly suppresses
  22. L0:919 weakly suppresses
  23. L0:968 weakly excites "and" joining two nouns
  24. L0:1123 weakly suppresses

Feeds later layers

  1. L2:1837 excites
  2. L7:416 weakly excites
  3. L20:2141 weakly suppresses
  4. L2:1124 weakly suppresses
  5. L16:1089 weakly suppresses
  6. L22:1657 weakly suppresses
  7. L7:243 weakly excites
  8. L5:546 weakly excites
  9. L3:1187 weakly suppresses
  10. L26:37 weakly excites
  11. L17:289 weakly excites
  12. L19:988 weakly suppresses
  13. L18:2883 weakly suppresses
  14. L16:1133 weakly suppresses
  15. L17:2353 weakly suppresses
  16. L16:2997 weakly suppresses
  17. L22:470 weakly suppresses
  18. L15:649 weakly suppresses
  19. L5:630 weakly excites
  20. L7:2291 weakly suppresses
  21. L20:887 weakly suppresses
  22. L9:2921 weakly suppresses
  23. L20:2121 weakly suppresses
  24. L16:1098 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.

4 values between -2.23 and -2.121 values between -2.12 and -2.011 values between -2.01 and -1.903 values between -1.90 and -1.782 values between -1.78 and -1.672 values between -1.67 and -1.562 values between -1.56 and -1.442 values between -1.44 and -1.331 values between -1.33 and -1.222 values between -1.22 and -1.106 values between -1.10 and -0.996 values between -0.99 and -0.882 values between -0.88 and -0.763 values between -0.76 and -0.657 values between -0.65 and -0.542 values between -0.54 and -0.429 values between -0.42 and -0.3171 values between -0.31 and -0.20487 values between -0.20 and -0.082261 values between -0.08 and 0.03752 values between 0.03 and 0.15182 values between 0.15 and 0.2654 values between 0.26 and 0.3712 values between 0.37 and 0.487 values between 0.48 and 0.604 values between 0.60 and 0.714 values between 0.71 and 0.824 values between 0.82 and 0.943 values between 0.94 and 1.055 values between 1.05 and 1.161 values between 1.16 and 1.282 values between 1.28 and 1.391 values between 1.39 and 1.502 values between 1.73 and 1.841 values between 2.18 and 2.30offnegativepositiveoften

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. arges
  2. lore
  3. Fon
  4. ARGE
  5. Rendering
  6. leet
  7. .Fill
  8. conna
  9. وع
  10. ASTE

Suppresses

  1. //{⏎
  2. rowable
  3. 那就是
  4. pire
  5. ishop
  6. серии
  7. 的就是
  8. 因为她
  9. iteit
  10. 但她