Neuron Explorer

Qwen3-0.6B

"of" in "most X of all" superlatives

A neuron in layer 20 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 responds to the word "of" in phrases like "worst of all", "strangest of all" or "the most ... of any state", where a superlative is about to be completed. It turns slightly negative on "very" after "the". Turning it up made the model stutter "of all of all of all", which is the same phrase it fires on.

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

    alligator wasn’t feeling⏎His feet felt draggy.⏎His skin felt baggy.⏎His tail wouldn’t swish this way and that.⏎And, worst of all, his big jaw wouldn’t SNAP.⏎Only two lines actually rhymed

  2. #2 peak

    and the telescope) can see. Those objects are getting farther away all the time because the universe is expanding, and moreover the expansion is accelerating. Strangest of all, the luminous matter of the galaxies—everything we see shining in the

  3. #3 peak

    to find brief and simple answers that are appropriate to their questions; answers that they can understand and that do not overwhelm them with too many words.⏎Perhaps most difficult of all, it involves examining our own feelings and beliefs so that we can talk to

  4. #4 peak

    severe cases<|endoftext|>Considering the fact that half of you reading this right now are probably wearing one, It’s only right to start this week off with the most commonplace of all acronymic items– the OCBD (oxford cloth button down.) While

  5. #5 peak

    invented in the 19th century. While it is possible to get food-borne illnesses from many different foods, raw milk is one of the riskiest of all, according to the Centers for Disease Control and Prevention.⏎

  6. #6 peak

    iamond⏎DIAMOND, a mineral universally recognized as chief among precious stones; it is the hardest, the most imperishable, and also the most brilliant of minerals. These qualities alone have made it supreme as a

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 the most efficient way to model a system where a person could say a message in a room, say, A most efficient model would be a language, which is a representation, which can be used to model both the message and the

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

    and philosophical principles, this is a thorough exploration of the concept and practice of⏎liberalism.ABOUT THE SERIES: The Very Short Introductions series from Oxford University Press contains hundreds of titles in almost every

  2. #2 peak

    12 percent of the nation’s women and girls. That steady hum of progress in Dhaka--despite its unscrupulousness--has helped the very poorest break free of poverty.⏎Aside from poverty alleviation, many also claim that

  3. #3 peak

    system, and multithreading.⏎Also included is coverage of some of Java's most powerful features,⏎such as generics, lambda expressions, and the very important⏎modules feature. A number of recent additions to the Java language, such

  4. #4 peak

    two complement each other. In simple terms, these two creatures do not have the same nutritional requirements which means, they cannot be fed with the same food. The very small breed rabbits may be okay to

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. L17:1771 excites
  2. L19:2622 excites
  3. L13:1659 weakly suppresses
  4. L16:961 weakly excites
  5. L18:2347 weakly suppresses
  6. L18:3048 weakly suppresses
  7. L19:2036 weakly suppresses
  8. L17:2241 weakly excites the word 'very'
  9. L17:152 weakly suppresses mid-word fragments, often 'intric'
  10. L19:3054 weakly excites
  11. L18:2241 weakly excites
  12. L18:1631 weakly excites
  13. L15:1196 weakly excites
  14. L5:729 weakly suppresses
  15. L8:1074 weakly excites
  16. L13:1249 weakly excites
  17. L17:459 weakly suppresses
  18. L8:506 weakly suppresses
  19. L18:1619 weakly suppresses
  20. L11:223 weakly suppresses
  21. L19:715 weakly excites
  22. L6:2430 weakly suppresses
  23. L7:2266 weakly excites
  24. L18:1174 weakly suppresses

Feeds later layers

  1. L23:567 strongly excites
  2. L26:711 strongly excites 'years' after an age number
  3. L27:447 strongly excites
  4. L25:2020 strongly excites
  5. L21:1718 excites
  6. L22:142 suppresses 'of' after a quantity or noun phrase
  7. L26:279 suppresses
  8. L23:258 excites
  9. L26:920 excites
  10. L26:115 suppresses
  11. L22:1993 excites
  12. L22:288 excites
  13. L27:769 excites
  14. L23:2665 excites
  15. L22:477 excites
  16. L21:2802 suppresses 'of' inside set phrases like 'course of'
  17. L23:1712 excites
  18. L24:2753 suppresses
  19. L21:2646 excites
  20. L24:860 excites
  21. L26:1052 weakly excites
  22. L25:1241 weakly excites
  23. L27:658 weakly excites
  24. L21:1306 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.

2 values between -7.00 and -5.822 values between -4.63 and -3.443 values between -3.44 and -2.2616 values between -2.26 and -1.072680 values between -1.07 and 0.121159 values between 0.12 and 1.3017 values between 1.30 and 2.494 values between 2.49 and 3.6811 values between 3.68 and 4.865 values between 4.86 and 6.052 values between 6.05 and 7.242 values between 9.61 and 10.801 values between 11.99 and 13.171 values between 14.36 and 15.551 values between 16.73 and 17.921 values between 19.11 and 20.291 values between 39.28 and 40.47offnegativepositiveoften

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. of
  2. _of
  3. Of
  4. Of
  5. of
  6. OfFile
  7. →of
  8. _Of
  9. OfString
  10. _OF

Suppresses

  1. ever
  2. umont
  3. greatest
  4. 一个是
  5. iclass
  6. 首家
  7. -Saharan
  8. EVER
  9. highest
  10. 쉽