Neuron Explorer

Qwen3-0.6B

sentence-initial 'Not' starting a clause

A neuron in layer 23 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 word 'Not' when it begins a new sentence or clause, as in 'Not surprisingly' or 'Not too long ago'. When it was turned up, the model started repeating the word 'sure' over and over.

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

    20th / 21st-century philosophy|⏎|Main interests||Linguistics · Psychology⏎Philosophy of language⏎Politics · Ethics⏎|Notable ideas||Generative grammar|⏎Avram Noam Chomsky (born

  2. #2 peak

    of limiting individual impact on climate change.⏎The event hosted three food sustainability specialists Julia Davies (Nottingham Trent University), who gave a global food security perspective, Hannah Ford (University

  3. #3 peak

    try and eat more produce. Of course, fruits and vegetables supply vitamins, minerals, and fiber, which we need to achieve and maintain overall health, too.⏎Not surprisingly, a steady intake of fruits

  4. #4 peak

    associated with investments in renewable energy. And in countries with a low GDP per capita, nuclear electricity production clearly associates with CO2 emissions that tend to be higher.”⏎Not surprisingly, the study attracted criticism. Some summarily dismissed it as pseudos

  5. #5 peak

    States and Mexico; The opposite occurs with Cubans and Puerto Ricans living in the United States.⏎This is reflected in a survey carried out between Axios and Noticias Telemundo. Cubans have settled very well in Florida since the times

  6. #6 peak

    2011, the UN Special Representative to Kosovo called for an independent investigation into Kosovan organ trafficking allegations to be spearheaded by the UN Security Council.⏎Noting the changing forms of slavery, including the rise in global sex trafficking, UN

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 can still see some former students in the former school, and I also can feel some feelings of stillness and separation between former and new, or maybe the former and latter parts of the former school. I have both feelings and feelings of

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

    to protect the earth’s climate. The global climate crisis is non-negotiable. #Move the Date: We need to do everything we can

  2. #2 peak

    the damage.⏎Emergency preparation is now an important part of a city's defense. We can no longer depend on historical weather patterns to be a guide. Sea-level rise and climate

  3. #3 peak

    : Sometimes treats⏎There are certain foods that your bunny should only have in moderation such as foods that<|endoftext|>Moving the boundaries (on Shoftim)⏎"You shall not move the boundary of your neighbour, which the first ones made his boundary when you

  4. #4 peak

    soliloquy in Act 1 Scene 2 where he recognizes that "self-slaughter" is against God's commandment (i.e. Thou shall not kill). Here, Hamlet is being over dramatic: youthful hyperbole. His

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. L20:1414 excites
  2. L22:635 suppresses
  3. L4:151 excites
  4. L15:786 excites
  5. L17:443 excites
  6. L4:9 excites
  7. L11:1095 suppresses
  8. L16:752 excites
  9. L10:111 suppresses
  10. L10:504 suppresses
  11. L10:386 excites
  12. L17:1644 suppresses
  13. L1:509 excites 'to' after 'unable'
  14. L19:259 excites
  15. L18:2220 suppresses
  16. L9:54 excites
  17. L18:1763 excites comma in 'not X, but Y' contrasts
  18. L19:2184 suppresses
  19. L14:2441 excites
  20. L20:212 weakly excites
  21. L21:2799 weakly suppresses
  22. L7:32 weakly excites
  23. L0:2681 weakly excites
  24. L15:1293 weakly suppresses

Feeds later layers

  1. L25:1756 weakly excites
  2. L25:2845 weakly excites
  3. L25:1264 weakly suppresses
  4. L25:1985 weakly excites
  5. L27:1117 weakly excites
  6. L27:618 weakly excites
  7. L27:2268 weakly excites
  8. L26:670 weakly excites
  9. L25:1094 weakly excites
  10. L26:2916 weakly excites
  11. L25:1096 weakly suppresses
  12. L24:590 weakly suppresses
  13. L27:2044 weakly excites
  14. L26:677 weakly excites
  15. L27:938 weakly excites
  16. L24:960 weakly excites
  17. L27:2827 weakly excites
  18. L26:1395 weakly suppresses
  19. L24:2690 weakly excites
  20. L26:3012 weakly excites
  21. L24:1964 weakly excites
  22. L25:1674 weakly excites
  23. L26:1975 weakly excites
  24. L27:398 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.

3 values between -8.27 and -6.037 values between -6.03 and -3.7914 values between -3.79 and -1.553798 values between -1.55 and 0.7052 values between 0.70 and 2.9412 values between 2.94 and 5.187 values between 5.18 and 7.435 values between 7.43 and 9.673 values between 9.67 and 11.911 values between 14.15 and 16.403 values between 18.64 and 20.881 values between 27.61 and 29.851 values between 32.10 and 34.341 values between 79.20 and 81.44offnegativepositiveoften

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. sure
  2. sure
  3. Sure
  4. 太
  5. Sure
  6. ched
  7. ches
  8. 那么
  9. CHED
  10. 太高

Suppresses

  1. 前者
  2. ◄
  3. べき
  4. offee
  5. ainer
  6. 后者
  7. plier
  8. rowsing
  9. ⊘
  10. former