Neuron Explorer

Qwen3-0.6B

unclear: noun heads in expository text

A neuron in layer 12 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 the main noun of a phrase in informative writing, such as 'holes' in 'black holes' or 'chains' in 'food chains'. Its meaning is hard to pin down, and turning it up made the model's text drift into random lists and counting rather than a clear topic.

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

    supermassive black hole in the new galaxy develops an insatiable appetite. However, this ferocious appetite is unsustainable.⏎The growth of supermassive black holes, which are found in the centers of all normal galaxies (including our Milky Way

  2. #2 peak

    er will harpoon itself into place to keep it from ricocheting back off into space.⏎It may hold secrets on the origins of life⏎Comets are balls of ice and dirt left over after the dawn of the solar system, and some

  3. #3 peak

    number using prime factorization.⏎Algebra.com [offsite link]⏎Algebra.com has a variety of sample problems plus online tutors and more.⏎Converting Decimals -

  4. #4 peak

    slide. There is also a child-sized theater with a stage curtain for dramatic presentations. Science content will include life cycles, invasive species that threaten our lakes, food chains, water quality, and identification of fish species.⏎Mounts of game birds and

  5. #5 peak

    educate our patients on the signs and symptoms of gum disease and how to treat, and ultimately prevent it.⏎Did you know that most people do not even know that they have periodontal disease? This is because a lot of the time, pain

  6. #6 peak

    - Adding Mixed Numbers⏎Learn how to Add mixed numbers with like and unlike denominators.⏎Fractions - Adding Unlike Denominators - LCM⏎Least common multiples are very useful in finding the least common denominator when adding and subtracting fractions.

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 want to add to the problem. I want to add to the problem. So that's what I want to add to the the problem. But I want to add to the the problem. So that's the problem. But I want

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

    if you fall into one of the high risk categories listed below. This will keep your doctor on the

  2. #2 peak

    gone from technology being a shiny object–a TV–to being an indispensable part of one’s everyday life. Technology is changing as rapidly as you change the channels on the TV set. The challenge and opportunity in education is to understand what’s needed

  3. #3 peak

    of a tile.⏎So what can I do now?⏎Once the glaze on tiles has worn, it can't be rejuvenated in any way.⏎Likewise,

  4. #4 peak

    3 seconds I had to have a tripod and a cable release, also with an exposure of 33 seconds I would need to use the bulb mode as my cameras set exposures stops at 30 seconds.⏎Second get the exposure right; this

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. L11:2210 suppresses
  2. L9:374 excites
  3. L7:651 excites
  4. L11:937 excites
  5. L7:34 suppresses
  6. L9:55 excites
  7. L11:119 weakly suppresses
  8. L5:705 weakly suppresses
  9. L11:860 weakly suppresses
  10. L11:172 weakly excites
  11. L11:1294 weakly suppresses
  12. L10:661 weakly suppresses
  13. L10:624 weakly excites
  14. L8:275 weakly suppresses
  15. L11:114 weakly excites
  16. L11:1247 weakly suppresses
  17. L10:2421 weakly excites
  18. L11:223 weakly suppresses
  19. L7:327 weakly excites
  20. L10:967 weakly excites
  21. L9:1726 weakly suppresses
  22. L10:1644 weakly suppresses
  23. L7:103 weakly suppresses
  24. L9:1553 weakly suppresses

Feeds later layers

  1. L17:1706 strongly excites
  2. L17:1974 suppresses
  3. L17:1848 excites
  4. L18:1238 excites
  5. L16:2507 excites
  6. L20:807 excites
  7. L13:1647 excites
  8. L13:2063 suppresses
  9. L19:654 suppresses
  10. L21:913 excites
  11. L19:626 excites
  12. L19:592 suppresses
  13. L25:990 excites
  14. L24:488 excites
  15. L16:1317 excites
  16. L25:374 excites
  17. L15:1920 excites
  18. L20:1857 excites
  19. L21:392 excites
  20. L17:340 excites
  21. L27:1052 excites
  22. L13:1912 excites
  23. L20:240 suppresses
  24. L25:423 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 -1.36 and -1.271 values between -0.97 and -0.872 values between -0.87 and -0.771 values between -0.77 and -0.675 values between -0.67 and -0.5732 values between -0.57 and -0.47123 values between -0.47 and -0.38285 values between -0.38 and -0.28471 values between -0.28 and -0.18808 values between -0.18 and -0.081149 values between -0.08 and 0.02605 values between 0.02 and 0.12220 values between 0.12 and 0.2288 values between 0.22 and 0.3238 values between 0.32 and 0.4223 values between 0.42 and 0.5214 values between 0.52 and 0.619 values between 0.61 and 0.716 values between 0.71 and 0.819 values between 0.81 and 0.914 values between 0.91 and 1.014 values between 1.01 and 1.112 values between 1.11 and 1.211 values between 1.31 and 1.411 values between 1.41 and 1.501 values between 1.50 and 1.601 values between 1.60 and 1.701 values between 1.70 and 1.801 values between 1.90 and 2.001 values between 2.10 and 2.201 values between 2.49 and 2.59offnegativepositiveoften

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. adoras
  2. ших
  3. URN
  4. tures
  5. _dicts
  6. semblies
  7. arsers
  8. iefs
  9. ящих
  10. шие

Suppresses

  1. a
  2. 一份
  3. 一个
  4. 活动
  5. 一颗
  6. a
  7. 恭喜
  8. 把我
  9. 因为你
  10. 因为我