Neuron Explorer

Qwen3-0.6B

"of" in "some kind of" or "some type of"

A neuron in layer 1 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 reacts to the word "of" in phrases like "some kind of" or "some type of". It is quiet, or slightly negative, on words like "or" in "either-or" choices. Turning it up or down gave text that wandered off in different directions, so it does not clearly change the topic.

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

    You may need to invest in some type of pesticide;there are many DIY chemical-free recipes available nowadays too.⏎As the summer

  2. #2 peak

    to do some type of exercise every day. It is best to do some kind of aerobic activity without stopping for at least 20 to 30 minutes each

  3. #3 peak

    to yourself. For example, thinking that everything people do or say is some kind of reaction to you. You also compare yourself to others, trying to determine who’s

  4. #4 peak

    embassy. They are often set up between nations that do not have official diplomatic relations. They offer a way for two nations that have tense relations to continue some kind of working contact. The U.S. has operated “interest sections” in places like

  5. #5 peak

    and also points to separate research which estimates that up to 22% of crashes could be caused by some kind of distraction; and drivers who perform a secondary task at the wheel, like using a

  6. #6 peak

    a few days, the CNT and the Socialist General Confederation of Workers (UGT), which since December 1933 had been discussing some kind of working-class alliance, planned a national general strike. It began on October 4

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 am thinking. So I am thinking. I am thinking. So I am thinking. So I am thinking. So I am thinking. So I am thinking. So I am thinking. So I am thinking. So I am thinking.

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

    of motor oil, there are three pieces of information: the API service rating, the viscosity grade and the “Energy Conserving” indicator (the oil either is or it isn’t).⏎API service rating⏎This

  2. #2 peak

    Piper urges us to think for the glory of God. He demonstrates from Scripture that glorifying God with our minds and hearts is not either-or, but both-and. Thinking carefully about God fuels passion and affections for God. Likewise, Christ

  3. #3 peak

    another word this?⏎- Which is short form?⏎- Which is or that is?⏎- What is the means of 7?⏎- What Why means

  4. #4 peak

    programming language.⏎Python does not have a static checking phase,⏎The (C)Python *interpreter* doesn't. Other Python implementations (existing or hypothetical) may or may not have a static checking phase.⏎But static tools need specifications beyond (ie

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. L0:2651 weakly suppresses
  2. L0:1812 weakly suppresses
  3. L0:2793 weakly suppresses
  4. L0:2537 weakly suppresses
  5. L0:1945 weakly suppresses
  6. L0:146 weakly suppresses
  7. L0:395 weakly excites
  8. L0:1433 weakly suppresses
  9. L0:373 weakly excites
  10. L0:1100 weakly excites
  11. L0:2616 weakly suppresses
  12. L0:376 weakly excites
  13. L0:1394 weakly suppresses
  14. L0:983 weakly suppresses "similar" after "are"
  15. L0:1475 weakly suppresses
  16. L0:78 weakly excites the 'S' in 'U.S.'
  17. L0:1720 weakly excites
  18. L0:942 weakly excites
  19. L0:895 weakly excites
  20. L0:5 weakly excites
  21. L0:189 weakly suppresses
  22. L0:427 weakly suppresses
  23. L0:260 weakly excites
  24. L0:2817 weakly suppresses

Feeds later layers

  1. L7:2170 weakly suppresses
  2. L14:2276 weakly excites opening bracket after a term or number
  3. L10:2235 weakly excites
  4. L7:712 weakly excites
  5. L19:1835 weakly suppresses
  6. L6:876 weakly excites
  7. L16:51 weakly suppresses
  8. L12:1823 weakly excites
  9. L3:1114 weakly suppresses
  10. L17:986 weakly excites
  11. L12:2662 weakly suppresses
  12. L18:1975 weakly suppresses
  13. L16:184 weakly suppresses
  14. L3:710 weakly excites
  15. L11:1667 weakly excites
  16. L14:346 weakly excites
  17. L5:1691 weakly suppresses
  18. L13:1592 weakly excites
  19. L21:1993 weakly excites
  20. L16:47 weakly suppresses
  21. L16:1402 weakly suppresses
  22. L14:1296 weakly excites
  23. L11:2954 weakly suppresses
  24. L16:454 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 -2.62 and -2.521 values between -1.77 and -1.681 values between -1.59 and -1.492 values between -1.40 and -1.311 values between -1.21 and -1.123 values between -1.02 and -0.932 values between -0.93 and -0.845 values between -0.84 and -0.744 values between -0.74 and -0.654 values between -0.65 and -0.563 values between -0.56 and -0.467 values between -0.46 and -0.3727 values between -0.37 and -0.2791 values between -0.27 and -0.18402 values between -0.18 and -0.091517 values between -0.09 and 0.011327 values between 0.01 and 0.10316 values between 0.10 and 0.2083 values between 0.20 and 0.2943 values between 0.29 and 0.3834 values between 0.38 and 0.4816 values between 0.48 and 0.577 values between 0.57 and 0.667 values between 0.66 and 0.761 values between 0.85 and 0.943 values between 1.04 and 1.13offnegativepositiveoften

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. 老师的
  2. мент
  3. lication
  4. 師
  5. 师
  6. ulses
  7. case
  8. ilies
  9. isted
  10. 생

Suppresses

  1. 成熟
  2. ja
  3. eno
  4. Frem
  5. frem
  6. buiten
  7. décou
  8. 宰
  9. compte
  10. Mature