Neuron Explorer

Qwen3-0.6B

'and' in 'more and more' pairs

A neuron in layer 5 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 'and' in phrases that repeat a comparison, such as 'more and more' or 'stronger and more united'. Turning it up did not produce that phrase and gave unrelated 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

    in 2010 on the promise of a “tory miracle”. He promised a “country that leaves the European Union” will emerge “a stronger and more united nation”. His party is led by Cameron and his government has pushed through

  2. #2 peak

    5-years-of-age, more and more opt not to drive at night. In addition to senior drivers, a significant

  3. #3 peak

    writing career, Vladimir Nabokov was frustrated by both the Freudian and behavorialist schools of psychology that reigned in his day, seeking a richer and more complex view of the mind than either provided. In a long, incis

  4. #4 peak

    1 Julkaistu artikkeli, soviteltu|<|endoftext|>Though it straddles the Arizona and Utah borders, Monument Valley is a symbol bigger and wider-ranging than either of the two states. Watch The Searchers and it’s

  5. #5 peak

    the largest in the kingdom. In the late fourteenth century<|endoftext|>With the ever growing popularity of Linux as a desktop operating system, more and more people are beginning to see it as a usable and valid alternative to proprietary software

  6. #6 peak

    “artificial intelligence.” And while it may sound like something out of a science fiction novel, the truth is that AI is very real and it is becoming more and more commonplace in our everyday lives.⏎Creating A Story With AI⏎So, what

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: the answer to the question. If you can't get it, it will be included in the answer. If you can't get it, it will be excluded from the answer. (1) This is a true reflection of the actual problem

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

    a high likelihood of developing one or more of these common precancers if they spend time in the sun and live long enough. Location makes a difference: The closer to the equator you live, the more likely you are

  2. #2 peak

    cotangent as⏎Both of these values can be decimals. The range of cotangent also has no restrictions; the ratios are any and all numbers,⏎The closer you get to the values where⏎the smaller the number on the bottom of your fraction

  3. #3 peak

    ice. A new model, based on experiments in the laboratory that simulate salty seas, shows that the ocean and ice may be stacked up in multiple layers, more like a club sandwich.⏎Ice comes in different forms depending on pressures. "Ice I

  4. #4 peak

    the course of all food regardless of the time in which need to consume, so you can say desserts as such do not exist in Chinese food, the more close to these in Chinese food are the sweet dishes that are served.<|endoftext|>PREVIOUS :

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. L4:656 excites
  2. L3:346 weakly excites
  3. L4:1584 weakly suppresses
  4. L0:1325 weakly suppresses
  5. L0:560 weakly suppresses
  6. L4:686 weakly suppresses
  7. L3:456 weakly suppresses
  8. L4:2751 weakly suppresses
  9. L2:2895 weakly excites
  10. L0:1924 weakly excites
  11. L1:1144 weakly excites
  12. L4:2264 weakly excites
  13. L1:2906 weakly excites
  14. L3:59 weakly excites
  15. L1:2970 weakly excites
  16. L0:529 weakly excites
  17. L2:220 weakly excites
  18. L0:165 weakly excites
  19. L1:1700 weakly excites
  20. L0:1671 weakly suppresses
  21. L2:3015 weakly suppresses
  22. L4:2117 weakly suppresses
  23. L3:177 weakly suppresses
  24. L0:3050 weakly suppresses

Feeds later layers

  1. L7:2157 strongly excites 'the' after 'like' or 'stay'
  2. L17:1962 suppresses
  3. L19:1716 suppresses
  4. L27:3027 excites
  5. L6:960 weakly excites
  6. L19:1065 weakly suppresses
  7. L15:1333 weakly excites
  8. L13:227 weakly excites
  9. L18:1175 weakly suppresses
  10. L16:3037 weakly suppresses
  11. L18:1309 weakly suppresses
  12. L8:2741 weakly suppresses
  13. L11:338 weakly excites
  14. L12:801 weakly excites
  15. L17:698 weakly suppresses
  16. L7:534 weakly suppresses 'much' or 'many' after 'not' or 'little'
  17. L18:3065 weakly suppresses
  18. L8:312 weakly suppresses
  19. L15:1301 weakly suppresses
  20. L10:1871 weakly excites
  21. L8:1074 weakly suppresses
  22. L11:1537 weakly suppresses
  23. L11:922 weakly suppresses
  24. L8:2694 weakly suppresses

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 -6.92 and -6.711 values between -4.26 and -4.061 values between -3.44 and -3.241 values between -2.83 and -2.631 values between -2.42 and -2.221 values between -2.22 and -2.011 values between -2.01 and -1.813 values between -1.81 and -1.602 values between -1.60 and -1.405 values between -1.40 and -1.193 values between -1.19 and -0.994 values between -0.99 and -0.799 values between -0.79 and -0.5827 values between -0.58 and -0.38227 values between -0.38 and -0.172188 values between -0.17 and 0.031307 values between 0.03 and 0.24109 values between 0.24 and 0.448 values between 0.44 and 0.654 values between 0.65 and 0.853 values between 0.85 and 1.062 values between 1.06 and 1.26offnegativepositiveoften

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. .Builder
  2. 之
  3. feit
  4. ania
  5. ActiveForm
  6. um
  7. ancies
  8. double
  9. .getDouble
  10. uman

Suppresses

  1. -than
  2. vib
  3. 倾向
  4. 亲
  5. ieg
  6. hog
  7. .Lib
  8. ARED
  9. elsinki
  10. Than