Neuron Explorer

Qwen3-0.6B

"most" meaning "very" or "the most"

A neuron in layer 19 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 "most" when it means "the most" or "very," as in "at the most basic level" or "made it most useful." On the other side it fires on "so" in phrases like "changed so much." Turning it down made the model repeat "a lot of," which fits a theme of amounts and degrees.

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

    Now how does this hypothesis compare to that of Homer? Having lived almost four-centuries prior to Plato and nearly ten-centuries prior to Plotinus, Homer was most likely completely unacquainted with the ‘mind-body dualism’ theory held

  2. #2 peak

    and this was scarcely developed before the year 1746. The consummate hardness of the diamond, in spite of its high price, has made it most useful for purposes of grinding, polishing and drilling. Numerous attempts have been made to

  3. #3 peak

    a way of defining and solving problems by combining logic with insight, and by finding patterns and structure. At the most basic level, we use the abstract concept of “number” to understand what we

  4. #4 peak

    and Their Effectiveness⏎The recent approval of the commercial growing of Golden Rice (GR) based on guidelines set by a 5-interdepartmental circular is a most welcome, most awaited development.⏎Vitamin A deficiency (VAD) continues

  5. #5 peak

    too dim; however, it counts as sixth-closest to the Sun at 8.3 light years.⏎The seventh-closest is a star most denizens of Earth would recognise: Sirius (alpha Canis Majoris), with

  6. #6 peak

    , maintaining verb tense cohesiveness is important. However, understanding the different types of tense can make you, well, tense.⏎At the most fundamental level, there are three tenses: past tense, present tense and future

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: a lot of a lot a lot a lot a lot a lot a lot a lot a lot a a a a a a a a a a a a a a a a a a a a a a a a a a a a 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

    contrast, an `io soaring over the Panaewa farm lots bears no resemblance to another hawk. It must have arrived a long time ago to have changed so much. But how long ago was that?⏎The answers have come recently by a

  2. #2 peak

    An example of such an occurrence is an earthquake. Serious interventions ought to be applied in a case such a tragedy occurs. Japan is one of the countries where earthquakes occur. According to Reid (2019), earthquakes in Japan happen around one

  3. #3 peak

    keeps your nitric oxide levels high. Which means closing tanning beds and beaches is actually horrible for heart health.⏎One of the reasons many think that viruses spread so rapidly during the winter and slow during the summer is the amount of UV radiation we

  4. #4 peak

    order to use natural uranium to fuel a reactor, it has to be extremely stingy with neutrons. The coolant moderator combination had to be one that did not absorb neutrons to any great extent. To make natural uranium “burn” every available

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. L18:623 strongly suppresses
  2. L15:2176 excites
  3. L18:2077 excites
  4. L17:1688 excites
  5. L16:1577 suppresses
  6. L16:1789 excites
  7. L18:940 excites
  8. L13:2109 excites
  9. L16:1273 suppresses
  10. L18:1309 excites
  11. L18:1870 excites
  12. L14:1103 excites
  13. L17:1318 excites
  14. L18:2287 excites
  15. L14:1875 excites
  16. L11:922 excites
  17. L17:2292 suppresses
  18. L16:2362 suppresses
  19. L17:1921 suppresses
  20. L9:292 excites unclear: weak scattered tokens
  21. L16:2647 excites
  22. L18:2158 weakly suppresses
  23. L16:1720 weakly suppresses
  24. L9:1603 weakly suppresses

Feeds later layers

  1. L23:530 strongly suppresses
  2. L27:545 strongly suppresses
  3. L23:538 suppresses
  4. L21:542 suppresses
  5. L26:670 suppresses
  6. L20:2030 suppresses
  7. L27:620 suppresses
  8. L24:915 suppresses
  9. L21:699 suppresses
  10. L22:867 suppresses
  11. L20:1773 suppresses
  12. L20:1733 suppresses
  13. L20:1919 suppresses
  14. L27:755 suppresses
  15. L23:283 suppresses
  16. L24:1052 suppresses
  17. L23:2373 suppresses
  18. L24:1197 suppresses
  19. L20:914 suppresses
  20. L26:2494 suppresses
  21. L20:733 suppresses
  22. L20:851 suppresses
  23. L20:2341 suppresses
  24. L23:143 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.

1 values between -9.63 and -9.041 values between -6.71 and -6.131 values between -6.13 and -5.551 values between -5.55 and -4.975 values between -4.97 and -4.394 values between -4.39 and -3.8012 values between -3.80 and -3.2215 values between -3.22 and -2.6423 values between -2.64 and -2.0631 values between -2.06 and -1.4762 values between -1.47 and -0.89401 values between -0.89 and -0.312663 values between -0.31 and 0.27443 values between 0.27 and 0.8697 values between 0.86 and 1.4456 values between 1.44 and 2.0234 values between 2.02 and 2.6017 values between 2.60 and 3.1812 values between 3.18 and 3.778 values between 3.77 and 4.355 values between 4.35 and 4.933 values between 4.93 and 5.513 values between 5.51 and 6.102 values between 6.10 and 6.681 values between 6.68 and 7.263 values between 7.26 and 7.841 values between 9.59 and 10.171 values between 11.92 and 12.502 values between 13.08 and 13.66offnegativepositiveoften

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. ために
  2. _|
  3. |_
  4. ~
  5. '_
  6. theano
  7. 냥
  8. ~~
  9. voices
  10. //{⏎

Suppresses

  1. 很深
  2. 非常
  3. very
  4. 很大
  5. 较多
  6. Very
  7. .large
  8. 很
  9. 大量
  10. 很长