Neuron Explorer

Qwen3-0.6B

'of' inside institution names

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 fires on the word 'of' inside the names of schools, institutes and organisations, such as 'Higher School of Economics' or 'Harvard T.H. Chan School of Public Health'. Its negative side is weak and mixed. Turning it up did not produce text that clearly relates to institution names.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens an email to the maker with the neuron already filled in).

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

    980-4369⏎DOMINGUES, Beatriz Helena. The guarani jesuitic missions in the context of Enlightenment. História [online]. 2006, vol.25, n.1

  2. #2 peak

    the Internet: legal frames and technological implications’ organized by the Higher School of Economics on the 7th Internet Governance Forum (Baku, Azerbaijan,

  3. #3 peak

    ium 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 Rights on

  4. #4 peak

    Harvard T.H. Chan School of Public Health, Sept. 2012, https://www.hsph

  5. #5 peak

    hn School of Medicine at Mount Sinai.<|endoftext|>Visualise different economic scenarios, how they might be

  6. #6 peak

    checkout.⏎Feng ZW, Cao HF, Zhou XP (1999) The effects of acid precipitation on the ecological environment and restoration. Press of Chinese Environmental Sciences, Beijing, pp 1–3⏎Lei HC,

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: What is it, and what is it, and what is it, and what is it, and what is it, and what is it, and what is it, and what is it, and what is it, and what is it

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

    power of a successfully communicated thought, from one human mind to another, is one of the greatest forces we know. But like the tango, it takes two to communicate

  2. #2 peak

    ance Inc. illustrates how this can be applied to a major change initiative. The company’s CEO initiated a major organizational change to move from a more traditional, top-down culture to a servant-leadership culture. A 18

  3. #3 peak

    , it also means that pollutants that originate indoors are getting trapped inside.⏎“People normally<|endoftext|>Spring Dawn by Meng Haoran 春眠不觉晓 处处闻啼茑 夜来风雨声 花落知多少⏎Pin

  4. #4 peak

    aching a wider audience effectively! : Because an instructor can handle a small mob at a time only. When every other student bombards him with questions and raises hand in every micro second, the teacher sits down with head in his hands.⏎· It

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. L4:42 suppresses
  2. L9:207 suppresses
  3. L4:92 suppresses
  4. L9:789 excites
  5. L9:46 suppresses
  6. L3:130 excites
  7. L3:23 suppresses
  8. L9:2581 excites
  9. L7:92 excites
  10. L3:672 excites
  11. L4:267 excites
  12. L3:249 weakly excites
  13. L8:277 weakly excites
  14. L7:2621 weakly suppresses
  15. L3:2650 weakly excites
  16. L9:1139 weakly suppresses
  17. L3:615 weakly suppresses
  18. L2:678 weakly suppresses
  19. L4:76 weakly excites
  20. L9:991 weakly suppresses
  21. L2:974 weakly excites
  22. L2:1592 weakly suppresses
  23. L9:1470 weakly suppresses
  24. L9:429 weakly excites

Feeds later layers

  1. L11:2472 weakly excites
  2. L25:2562 weakly suppresses
  3. L17:2512 weakly suppresses
  4. L23:542 weakly suppresses
  5. L27:131 weakly suppresses
  6. L24:1133 weakly suppresses
  7. L17:1404 weakly suppresses
  8. L15:1226 weakly excites
  9. L20:1726 weakly suppresses
  10. L25:574 weakly excites
  11. L19:1549 weakly suppresses
  12. L13:2397 weakly excites
  13. L25:170 weakly excites
  14. L15:2225 weakly excites
  15. L22:1403 weakly suppresses
  16. L25:2149 weakly suppresses
  17. L11:2147 weakly suppresses
  18. L24:777 weakly suppresses
  19. L23:1806 weakly suppresses
  20. L15:2795 weakly suppresses
  21. L13:825 weakly excites
  22. L16:2715 weakly suppresses
  23. L18:1833 weakly excites
  24. L17:2758 weakly 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 -1.05 and -0.992 values between -0.68 and -0.627 values between -0.62 and -0.565 values between -0.56 and -0.5022 values between -0.50 and -0.4441 values between -0.44 and -0.3859 values between -0.38 and -0.31115 values between -0.31 and -0.25150 values between -0.25 and -0.19278 values between -0.19 and -0.13462 values between -0.13 and -0.07647 values between -0.07 and -0.01886 values between -0.01 and 0.05549 values between 0.05 and 0.11313 values between 0.11 and 0.17188 values between 0.17 and 0.2498 values between 0.24 and 0.3044 values between 0.30 and 0.3621 values between 0.36 and 0.426 values between 0.42 and 0.484 values between 0.48 and 0.541 values between 0.54 and 0.603 values between 0.60 and 0.661 values between 0.66 and 0.722 values between 0.78 and 0.851 values between 0.85 and 0.912 values between 1.33 and 1.39offnegativepositiveoften

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. 第一批
  3. ян
  4. gänge
  5. 责任人
  6. 每个
  7. 第一次
  8. 法定代表
  9. огда
  10. 第三个

Suppresses

  1. rij
  2. udo
  3. orial
  4. irement
  5. Signup
  6. Signup
  7. كم
  8. abra
  9. rement
  10. exas