Neuron Explorer

Qwen3-0.6B

unclear: 'your' in contact boilerplate

A neuron in layer 10 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on words in website and notice boilerplate, most often 'your' in phrases like 'contact your local office' or 'contact your librarian'. The trigger is not very consistent, and the steering outputs did not show a clear theme.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    list:⏎Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.⏎Contact your local office<|endoftext|>The point in the Moon's orbit where it lies diametrically

  2. #2 peak

    list:⏎Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.⏎Contact your local office<|endoftext|>It’s PrAACtically Martin Luther King Jr Day⏎Looking

  3. #3 peak

    reasons why your holiday shipments may not arrive on time. COVID-19 disruptions, congested ports, semiconductor shortages, stretched trucking capacity — the list goes on and on.⏎

  4. #4 peak

    range of therapies includes the following diseases of the head: acoustic neuromas / vestibular schwannomas, meningiomas, angiomas - arteriovenous malformations (AVM), metastases, trigeminal neuralgia and

  5. #5 peak

    can however freely search the site and view the abstracts and keywords for each book and chapter.⏎If you think you should have access to this title, please contact your librarian.⏎To troubleshoot, please check our FAQs, and if you can't

  6. #6 peak

    abstracts and keywords for each book and chapter.⏎If you think you should have access to this title, please contact your librarian.<|endoftext|>Information on creating school gardens for outdoor learning, compiled by the National

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: the design of a similar system. I have been considering the process of decision-making and how it might be applied to different scenarios. 1. In the following: "A" has been designed to be a control that has been tested in 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

    extrasolar planets, half of all that have been discovered. One recent discovery was an Earth-like planet orbiting in the habitability zone of its star where liquid water

  2. #2 peak

    collision that took place 1.3 billion years ago. When it occurred, about three times the mass of the sun was converted to energy in a fraction of a second.⏎The discovery has been accepted for publication in

  3. #3 peak

    as a cold wind blows⏎through a dark, dark cave.⏎Those words are split across multiple pages. The pattern of rhyming two lines and ending with a dark, dark cave continues. This sets a

  4. #4 peak

    the annexation of Austria by Nazi Germany, the rebellious actions of Benito Mussolini’s Italy and the ongoing boycott of Uruguay.⏎When France was chosen to host the World Cup two years before the tournament, South American giants Uruguay and Argentina announced

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. L8:1988 weakly excites
  2. L8:2969 weakly suppresses
  3. L9:2777 weakly excites
  4. L7:1735 weakly suppresses
  5. L9:2438 weakly excites
  6. L0:2356 weakly suppresses
  7. L8:1857 weakly excites
  8. L6:2982 weakly excites
  9. L8:2282 weakly suppresses
  10. L7:211 weakly suppresses
  11. L6:168 weakly excites
  12. L6:1381 weakly suppresses
  13. L9:2514 weakly excites
  14. L4:1962 weakly excites
  15. L7:479 weakly suppresses
  16. L8:2727 weakly suppresses
  17. L5:3045 weakly suppresses
  18. L0:951 weakly excites
  19. L9:1800 weakly suppresses 'hands' after 'wash your'
  20. L8:2393 weakly excites
  21. L9:855 weakly suppresses
  22. L5:655 weakly excites
  23. L6:605 weakly excites
  24. L2:2669 weakly suppresses

Feeds later layers

  1. L12:359 excites
  2. L12:2283 weakly suppresses
  3. L13:2735 weakly suppresses
  4. L14:545 weakly excites
  5. L11:1946 weakly excites
  6. L13:2210 weakly suppresses
  7. L16:2949 weakly excites
  8. L11:436 weakly suppresses
  9. L16:1510 weakly suppresses
  10. L12:1436 weakly suppresses
  11. L11:428 weakly suppresses
  12. L14:1716 weakly excites
  13. L21:2696 weakly suppresses
  14. L21:2890 weakly suppresses
  15. L12:888 weakly suppresses
  16. L12:2010 weakly suppresses
  17. L22:1015 weakly suppresses
  18. L13:2422 weakly excites
  19. L18:987 weakly suppresses
  20. L22:2111 weakly suppresses
  21. L13:1919 weakly excites
  22. L13:1528 weakly suppresses
  23. L12:2013 weakly excites
  24. L15:128 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 -1.67 and -1.592 values between -1.59 and -1.522 values between -1.52 and -1.441 values between -1.36 and -1.291 values between -1.21 and -1.142 values between -1.14 and -1.063 values between -1.06 and -0.986 values between -0.98 and -0.914 values between -0.91 and -0.837 values between -0.83 and -0.765 values between -0.76 and -0.6811 values between -0.68 and -0.6026 values between -0.60 and -0.5327 values between -0.53 and -0.4541 values between -0.45 and -0.3859 values between -0.38 and -0.3098 values between -0.30 and -0.22142 values between -0.22 and -0.15243 values between -0.15 and -0.07606 values between -0.07 and 0.01698 values between 0.01 and 0.08650 values between 0.08 and 0.16521 values between 0.16 and 0.23329 values between 0.23 and 0.31215 values between 0.31 and 0.39109 values between 0.39 and 0.4661 values between 0.46 and 0.5418 values between 0.54 and 0.6112 values between 0.61 and 0.695 values between 0.69 and 0.771 values between 0.77 and 0.841 values between 0.99 and 1.071 values between 1.30 and 1.37offnegativepositiveoften

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. edException
  2. admit
  3. 雳
  4. elta
  5. lie
  6. enie
  7. 녕
  8. #print
  9. entifier
  10. gie

Suppresses

  1. Meetings
  2. iox
  3. caret
  4. harmless
  5. terminates
  6. Ter
  7. .clearRect
  8. castle
  9. eras
  10. rad