Neuron Explorer

Qwen3-0.6B

number word before a list of reasons

A neuron in layer 17 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on words like 'two', 'several' and 'a few' when the text is about to list a set number of reasons, ingredients or ways. Turning it up did not produce a clear, related change in the output.

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

    We have a centuries-old accumulation of armaments that shows no sign of evaporating.⏎But there are two things that are needed for a gun to work: the gun and the ammunition.

  2. #2 peak

    city of⏎Gundisapur in 'Ylam at the eastern bank of today's Shatt-il-Arab.⏎These remarks indicate two ingredients in the formation of early Mesopotamian⏎Christianity, that have

  3. #3 peak

    and that of human civilization go hand in hand.⏎In fact, historians believe that yogurt played a pivotal role in the development of human societies. This is due to two main reasons. First, by allowing milk to be stored for longer, more people

  4. #4 peak

    ⏎- Door. …⏎- Lower Door Skin.⏎Why are car bumpers made of plastic?⏎Plastic is the preferred material for modern car bumpers for several reasons. … First and foremost, plastic is much lighter and more aerodynamic than

  5. #5 peak

    blog post we will use a Python script to work out how many lego bricks would be needed to build planet Earth!⏎To solve our problem we will use the following data:⏎We will work out the number of lego bricks needed in several step

  6. #6 peak

    , given up to an animal rescue group, or put to sleep.⏎Compared to other states’ cruelty laws, Texas animal cruelty laws are unique in a couple of ways. First, Texas criminal cruelty laws only apply to domesticated animals, such

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't get over the fact that the (time) (time) (time) (time) (time) (time) (time) (time) (time) (time) (time) (time) (time)

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

    has a lot of items that can’t be moved, then a portable play yard is a good idea–a place where he can handle anything he finds. But again he should have new things to discover every day.⏎The word “gentle”

  2. #2 peak

    in Joules⏎c) The amount of heat added to the gas in Joules⏎d) The change in energy of the gas in Joules⏎I've

  3. #3 peak

    honors courses?⏎You must have come across this term while searching through college websites. Basically honor courses are a “distinctive

  4. #4 peak

    , and pigment granules. Understanding these components will not only deepen your knowledge of hair but also help you identify them and comprehend their significance. So, let's dive in!⏎The cuticle is the outermost layer of the hair shaft, serving

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. L16:655 excites 'see' cross-references and web boilerplate
  2. L16:1961 suppresses
  3. L15:427 excites
  4. L15:2777 suppresses
  5. L14:1738 excites
  6. L16:2539 suppresses
  7. L16:888 excites
  8. L15:717 suppresses
  9. L15:2158 suppresses
  10. L16:1127 suppresses opening bracket after an imperial measurement
  11. L15:2258 excites
  12. L13:3004 suppresses
  13. L15:1119 suppresses
  14. L14:667 excites
  15. L15:695 suppresses
  16. L16:737 excites
  17. L15:2145 weakly suppresses
  18. L15:1441 weakly excites
  19. L15:2871 weakly excites
  20. L5:1039 weakly suppresses
  21. L16:2665 weakly suppresses
  22. L16:1680 weakly suppresses
  23. L15:2610 weakly suppresses
  24. L11:513 weakly excites words inside idioms and fixed phrases

Feeds later layers

  1. L19:2149 excites
  2. L18:548 excites
  3. L18:1483 excites
  4. L19:2239 excites
  5. L18:2261 suppresses
  6. L19:2727 excites
  7. L21:2585 excites
  8. L18:2843 suppresses
  9. L18:2572 suppresses
  10. L18:2359 excites
  11. L18:468 weakly excites
  12. L20:2268 weakly excites
  13. L20:3049 weakly excites
  14. L21:2353 weakly excites
  15. L19:2843 weakly excites
  16. L18:132 weakly excites
  17. L26:2817 weakly excites
  18. L20:2127 weakly suppresses
  19. L18:3046 weakly suppresses
  20. L18:959 weakly suppresses
  21. L21:2273 weakly excites
  22. L23:1409 weakly excites
  23. L21:2256 weakly excites
  24. L18:731 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 -1.88 and -1.684 values between -1.68 and -1.482 values between -1.48 and -1.284 values between -1.28 and -1.075 values between -1.07 and -0.8712 values between -0.87 and -0.6739 values between -0.67 and -0.46186 values between -0.46 and -0.261029 values between -0.26 and -0.061990 values between -0.06 and 0.14490 values between 0.14 and 0.35101 values between 0.35 and 0.5524 values between 0.55 and 0.757 values between 0.75 and 0.953 values between 0.95 and 1.161 values between 1.16 and 1.361 values between 1.36 and 1.561 values between 1.56 and 1.773 values between 1.97 and 2.172 values between 2.17 and 2.371 values between 2.37 and 2.581 values between 6.02 and 6.22offnegativepositiveoften

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. 几种
  2. 事情
  3. oth
  4. var
  5. 有一些
  6. 几个
  7. ric
  8. 有所不同
  9. 有所
  10. 两种

Suppresses

  1. TemplateName
  2. qrstuvwxyz
  3. ”⏎⏎
  4. agger
  5. ”⏎⏎
  6. _Default
  7. Template
  8. CPP
  9. .assertNot
  10. ({⏎