Neuron Explorer

Qwen3-0.6B

'important' and degree words after 'how'

A neuron in layer 10 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 lights up on words that judge how much something matters or how strong it is, especially 'important' right after 'how' (as in 'how important is...'). When it was turned up, the model's replies were unrelated and a bit scrambled, so the steering tells us little about its role.

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

    something new: If you baby-sit for your grandchildren, how about walking to the park instead of playing video games?⏎- Rethink your priorities: How important is an entire afternoon of TV? How about a walk after lunch instead?⏎-

  2. #2 peak

    – coal, shale gas – affect climate change. Thinking long-term is not about pin-pointing what will happen in a given year. It is about understanding how important trends work over time.⏎The same is true of the budget. A longer window

  3. #3 peak

    on governments and businesses - and even individuals - to carefully study what kind of hazards could trigger a disaster, to identify what is exposed to these threats, how sensitive an area is to these hazards and what would be the consequences of a storm like Sandy

  4. #4 peak

    such as diet and exercise can have a deep positive impact on our health and lower high cholesterol levels, it turns out that genetics also plays a crucial part in how high our cholesterol may end up

  5. #5 peak

    due to lockdown and less travel. If we still use our cars less after the pandemic, this could continue to make a huge difference.⏎But, how much of a difference are we talking?⏎Well,

  6. #6 peak

    a tool for assessing the democratic health of European countries across many different dimensions. That includes not just formal dimensions of democracy but also more everyday features of democracy – how important democratic principles and practices are to the cultures of workplaces, to people’s community life

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 do do I should do do I can have a child I have will be have a child I have will be have a baby I will be have a child I have will be a child is also a place is a person is a child

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

    , now you will find below the Solutions of Class 10 Science Chapter 4 Environmental management questions answers. From the subject English of Maharashtra State Board.⏎Below you can see

  2. #2 peak

    information about children's rights available in one place is to build a clearer picture of some of the repeated violations of children's rights in a given country. The eventual goal will be to try to match these violations with possible avenues of redress.⏎We

  3. #3 peak

    1, 2009⏎Computer science focuses on aspects of the computer field that have a well-defined set of abstract concepts and principles. The objective of a concentration in computer science is to demonstrate strong comprehension of those abstract concepts and principles

  4. #4 peak

    how true is that! There are more job opportunities in Analytics and Optimization than there were last year and many IT professionals are prepared to invest time and money for the

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. L9:2773 strongly suppresses
  2. L2:89 strongly suppresses
  3. L7:403 excites
  4. L8:511 excites
  5. L2:180 suppresses 'is' after focus or goal nouns
  6. L5:1038 suppresses
  7. L4:670 excites
  8. L6:217 excites
  9. L9:756 excites
  10. L5:1938 suppresses
  11. L5:224 suppresses
  12. L8:1149 excites
  13. L5:1717 excites
  14. L5:99 excites
  15. L8:174 excites
  16. L9:1258 excites
  17. L7:84 suppresses
  18. L5:2147 excites
  19. L2:638 excites
  20. L2:578 excites
  21. L9:2303 suppresses
  22. L5:395 suppresses
  23. L9:2357 suppresses
  24. L8:1931 excites

Feeds later layers

  1. L15:304 strongly excites
  2. L13:1129 strongly suppresses
  3. L11:578 strongly suppresses
  4. L16:968 strongly excites
  5. L13:1107 strongly suppresses
  6. L15:681 strongly suppresses
  7. L18:857 strongly suppresses
  8. L15:2971 strongly suppresses
  9. L11:782 excites
  10. L13:403 excites
  11. L13:2672 suppresses
  12. L16:1223 suppresses
  13. L15:824 suppresses
  14. L23:220 suppresses
  15. L14:1326 suppresses
  16. L14:850 suppresses
  17. L25:9 excites
  18. L21:262 excites
  19. L12:869 suppresses
  20. L20:586 suppresses
  21. L14:1487 suppresses
  22. L16:1808 suppresses
  23. L18:907 suppresses
  24. L16:881 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 -3.32 and -3.131 values between -2.58 and -2.392 values between -2.20 and -2.024 values between -1.83 and -1.656 values between -1.65 and -1.469 values between -1.46 and -1.2814 values between -1.28 and -1.0929 values between -1.09 and -0.9031 values between -0.90 and -0.7261 values between -0.72 and -0.53140 values between -0.53 and -0.35232 values between -0.35 and -0.161058 values between -0.16 and 0.031087 values between 0.03 and 0.21760 values between 0.21 and 0.40348 values between 0.40 and 0.5858 values between 0.58 and 0.7717 values between 0.77 and 0.9613 values between 0.96 and 1.1412 values between 1.14 and 1.337 values between 1.33 and 1.513 values between 1.51 and 1.703 values between 1.70 and 1.884 values between 1.88 and 2.072 values between 2.07 and 2.262 values between 2.44 and 2.632 values between 2.63 and 2.811 values between 3.00 and 3.181 values between 3.93 and 4.11offnegativepositiveoften

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. ích
  2. bidden
  3. ーター
  4. ople
  5. 些什么
  6. 足够的
  7. ooth
  8. cht
  9. aster
  10. ewhere

Suppresses

  1. 是
  2. 是我
  3. 是他
  4. 是我的
  5. 是由
  6. 是没有
  7. 是不会
  8. adalah
  9. 留下了
  10. 能得到