Neuron Explorer

Qwen3-0.6B

'are' after 'here' introducing a list

A neuron in layer 22 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 'are' when it follows a phrase like 'Here are' or 'what are', right before a list of steps, tips or reasons is about to start. Turning it up made the model drift into loose, rambling text rather than list-making.

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

    this method will help you achieve a bouncing ball animation in probably 5 minutes or less.⏎So without wasting much time, let’s get into it!⏎Here are the steps:⏎- Delete everything by first selecting everything (pressing A) and

  2. #2 peak

    in the IT sector, you've probably heard some of the hype surrounding Terraform by HashiCorp. But what exactly are Terraform's capabilities, and why should your business consider using it?⏎Put simply

  3. #3 peak

    coastlines. It flourishes in rock formations or deep soft sand along the shores of Cape Town.⏎Medusa’s Head Euphorbia Plant Care⏎Here are some factors Medusa’s head plant owners must consider and take care of to ensure

  4. #4 peak

    challenging to know what to do to get your middle school child ready for all of the activities, requirements and events that will go on during the school year. Here are a few tips that you can follow that may come in handy as you ready your

  5. #5 peak

    your own beautiful garden at home. Here are some simple tips that you can make your garden look more beautiful.⏎According to expert

  6. #6 peak

    the most frustrating behaviors a dog owner can experience. It not only disturbs your sleep but also gives you the sense that the dog is not comfortable. So what are the reasons why dogs growl?⏎Here’s The Sort Answer To Why Your Dog

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: some of the set of all possible set of all possible set of all possible set of all possible set of all possible set of all set of all set of all set of all set of all set of all set of all set of all set

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

    panels. Sunpower panels are made up of photovoltaic cells that convert sunlight into electrical energy. While Sunpower solar panels have many advantages, they also have a few disadvantages.⏎Sunpower solar panels are a premium product and as such, come

  2. #2 peak

    time. Additionally, smart work can lead to innovation and creativity, as individuals are encouraged to think outside the box and find new solutions to old problems.⏎However, there are also disadvantages to smart work. First, it can lead to a lack of

  3. #3 peak

    .”⏎The process of electrolysis, where hydrogen is removed from water isn’t new, but Constantine says existing technology typically relies on pure water, which presents its own challenges.⏎“That’s really

  4. #4 peak

    as constant innovation and change can lead to burnout and exhaustion.⏎In conclusion, hard work and smart work are two distinct approaches to achieving a goal, each with its own advantages and disadvantages. Hard work can lead to a deep sense of accomplishment and

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. L20:2117 strongly excites
  2. L20:1735 strongly excites
  3. L19:2379 suppresses
  4. L21:913 suppresses
  5. L17:1510 excites
  6. L16:1127 suppresses opening bracket after an imperial measurement
  7. L19:1695 excites
  8. L19:2122 suppresses
  9. L19:2149 excites
  10. L18:1806 suppresses
  11. L18:1565 excites
  12. L12:1590 suppresses
  13. L17:772 suppresses
  14. L18:1256 suppresses
  15. L19:1694 suppresses
  16. L21:1580 excites
  17. L10:910 excites
  18. L19:1424 excites
  19. L15:1770 suppresses
  20. L14:2813 suppresses
  21. L17:1462 excites
  22. L13:1321 suppresses
  23. L17:2031 excites
  24. L19:1331 suppresses

Feeds later layers

  1. L27:551 strongly excites
  2. L24:434 suppresses
  3. L24:489 excites
  4. L25:320 excites
  5. L25:399 excites
  6. L27:227 suppresses
  7. L27:588 excites
  8. L25:423 suppresses
  9. L24:488 suppresses
  10. L27:480 suppresses
  11. L26:111 suppresses
  12. L24:703 suppresses
  13. L25:464 excites
  14. L25:377 suppresses
  15. L27:60 excites
  16. L27:592 excites
  17. L25:628 excites
  18. L25:1586 excites
  19. L26:430 suppresses name fragments inside split words
  20. L23:1123 suppresses
  21. L24:109 excites
  22. L27:1035 suppresses
  23. L27:1181 suppresses
  24. L27:797 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 -6.03 and -5.642 values between -5.64 and -5.243 values between -5.24 and -4.844 values between -4.84 and -4.457 values between -4.45 and -4.055 values between -4.05 and -3.6512 values between -3.65 and -3.2614 values between -3.26 and -2.8613 values between -2.86 and -2.4632 values between -2.46 and -2.0741 values between -2.07 and -1.6777 values between -1.67 and -1.27126 values between -1.27 and -0.88212 values between -0.88 and -0.48727 values between -0.48 and -0.081703 values between -0.08 and 0.31465 values between 0.31 and 0.71200 values between 0.71 and 1.11101 values between 1.11 and 1.5050 values between 1.50 and 1.9034 values between 1.90 and 2.3026 values between 2.30 and 2.6914 values between 2.69 and 3.0910 values between 3.09 and 3.487 values between 3.48 and 3.883 values between 3.88 and 4.288 values between 4.28 and 4.672 values between 4.67 and 5.071 values between 5.07 and 5.471 values between 5.47 and 5.861 values between 5.86 and 6.262 values between 6.26 and 6.661 values between 7.45 and 7.851 values between 8.24 and 8.641 values between 8.64 and 9.041 values between 9.43 and 9.83offnegativepositiveoften

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. option
  2. requirement
  3. feature
  4. argument
  5. constraint
  6. rule
  7. ssue
  8. 也是一个
  9. property
  10. property

Suppresses

  1. 哪些
  2. 的各项
  3. inputs
  4. fields
  5. 些
  6. principles
  7. signs
  8. norms
  9. terms
  10. 的一些