Neuron Explorer

Qwen3-0.6B

"any" in negated sentences

A neuron in layer 3 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 "any" when it follows a negation, such as "don't contain any", "isn't showing any" or "hasn't received any". Its negative side is weak and not clearly meaningful. Turning it up did not give a clear, on-topic effect in the steered text.

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

    or can be manufactured under circumstances that lack quality control.⏎It is known that more than 30 percent of the counterfeit drugs that are available today don’t contain any active ingredients whatsoever. “Counterfeit drugs include preparations without active ingredients, toxic preparations

  2. #2 peak

    retaliated.⏎- It is the heaviest fighting between the two bitter enemies since 2014 and isn’t showing any signs of slowing.⏎- An emergency meeting of the UNSC has been conducted for

  3. #3 peak

    . Let’s just say we wouldn’t go making any plans for distant space travel or a personal vegetable garden on the moon just yet.⏎

  4. #4 peak

    that oil would wash up on that island," he said. "I’m not aware that any significant oil did occur on that island.”⏎Newstead hasn’t received any reports, of areas crucial to bird nesting, that were damaged.⏎He attributes that

  5. #5 peak

    figure 182. Without exerting any pressure, carefully detach the lose threads form the signatures, and keep them aside.⏎

  6. #6 peak

    72 and they realized it could read the tapes and process the data. Allen had graduated and was enrolled at Washington State University. Since neither Gates nor Allen had any hardware design experience they were initially stumped. The computer community in Seattle at that

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 important for you? I have been thinking about lately: what is the most important for you? I have been thinking about lately: what is the most important for you? I have been thinking about lately: what is 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

    de Ribereños y⏎Armadores, an association of fishermen and producers in Mexico.⏎"Fishermen are desperate."⏎Gilly blames the most recent El Niño in the Pacific Ocean for⏎forcing some Humboldt squid to migrate away. Others

  2. #2 peak

    of strange attractors (J.C. Sprott). Attractors with dueling symmetry (C.A. Reiter). A new feature in Hénon's map (M. Michelitsch, O.E. Rössler

  3. #3 peak

    for the effects of age, sex, race, and alcohol consumption, depression was twice as frequent in those who had intermediate cardiovascular health and two and a half times more frequent in those who had poor cardiovascular health when compared to those who

  4. #4 peak

    essence that forms the basis of her aspect of manifestation; they are the essential content by which all worldly forms are shaped.⏎In Filianic Scripture, the Janyati are compared to the splitting of white light through a prism, creating a plethora

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. L2:2253 excites
  2. L2:1891 weakly excites
  3. L2:1215 weakly excites
  4. L2:111 weakly suppresses
  5. L0:178 weakly suppresses
  6. L0:1301 weakly suppresses
  7. L0:2681 weakly excites
  8. L2:3003 weakly suppresses
  9. L2:2328 weakly suppresses
  10. L0:1799 weakly suppresses
  11. L0:1224 weakly suppresses
  12. L2:1416 weakly suppresses
  13. L2:2806 weakly suppresses
  14. L1:2378 weakly excites
  15. L2:2696 weakly suppresses
  16. L1:146 weakly excites
  17. L0:2380 weakly excites
  18. L1:2683 weakly excites
  19. L2:510 weakly suppresses
  20. L1:1029 weakly excites
  21. L0:564 weakly suppresses
  22. L2:269 weakly excites
  23. L1:301 weakly excites
  24. L2:2173 weakly suppresses

Feeds later layers

  1. L4:731 excites
  2. L16:1009 excites
  3. L9:59 excites "a" meaning per (a day, a year)
  4. L23:2339 excites
  5. L17:1516 excites
  6. L16:1309 excites
  7. L17:2233 excites 'any' after negatives or 'more'
  8. L23:2403 excites unclear: 'no matter' and 'several' before reasons
  9. L17:2902 excites
  10. L4:137 weakly excites
  11. L5:9 weakly excites
  12. L25:283 weakly excites
  13. L16:2055 weakly excites
  14. L17:1524 weakly excites
  15. L17:2348 weakly excites
  16. L15:519 weakly excites
  17. L14:2546 weakly excites
  18. L4:1938 weakly suppresses
  19. L22:146 weakly excites
  20. L26:10 weakly excites
  21. L4:283 weakly excites
  22. L16:1910 weakly excites
  23. L14:1463 weakly suppresses
  24. L7:224 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.

5 values between -0.50 and -0.3840 values between -0.38 and -0.26260 values between -0.26 and -0.141349 values between -0.14 and -0.021854 values between -0.02 and 0.09335 values between 0.09 and 0.2143 values between 0.21 and 0.338 values between 0.33 and 0.451 values between 0.45 and 0.572 values between 0.57 and 0.691 values between 1.75 and 1.871 values between 2.46 and 2.582 values between 2.58 and 2.701 values between 2.70 and 2.821 values between 3.05 and 3.171 values between 3.29 and 3.411 values between 3.88 and 4.001 values between 4.00 and 4.122 values between 4.12 and 4.24offnegativepositiveoften

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. nisi
  2. ")[
  3. 例外
  4. except
  5. […]
  6. 次要
  7. […]
  8. кроме
  9. amine
  10. [mid

Suppresses

  1. ycled
  2. ovie
  3. Din
  4. jeta
  5. glich
  6. ICLES
  7. edom
  8. filme
  9. :⏎⏎⏎⏎
  10. (||