Neuron Explorer

Qwen3-0.6B

"on" in "went on to"

A neuron in layer 3 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 "on" in phrases like "went on to" or "going on", where a verb is followed by "on" and another verb. Turning it up did not produce a clear change in what the model wrote.

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

    last week when climate advisors in the U.K. warned that the younger generation will be the ones paying the price of climate change. The Committee on Climate Change went on to say they’re “acutely concerned” over the U.K.’s lack

  2. #2 peak

    were elected, or 39%. The success rate for plurality winners in U.S. Senate races is much worse, as only four of 35 went on to win (11%). Candidates who went through a primary runoff tended to fare

  3. #3 peak

    frying pan?⏎How Is the Process of Coffee Beans Roasting Going On?⏎The process of roasting is required to bring coffee beans to the condition suitable

  4. #4 peak

    to Alzheimer’s in roughly 10% to 15% of cases each year, according to the study.⏎Forty-five percent of the participants went on to develop serious dementia consistent with Alzheimer’s disease. Those who had high cholesterol,

  5. #5 peak

    all American citizens.⏎Houston, tragically, would die before his strategy came to fruition in the Brown suit, but Marshall would argue the case victoriously and go on to become the first African-American Supreme Court justice—always crediting his mentor for

  6. #6 peak

    ’ should be adhered to. Yet, the current amendment goes on to give colossal discretion to the forest bureaucracy in the name of curbing colonial power

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 now a student of the University of the Science of the Future. So, let me know. I am also a student of the University of the Science of the Future. I am also a student of the University of the Science of

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

    o correction factor, allow mathematical calculation of the affects of air density on the wide-open-throttle horsepower and torque. The dyno correction factor is simply the mathematical reciprocal of the relative horsepower value.⏎Originally, all of the major US auto manufacturers were

  2. #2 peak

    perpendicular lines are the lines that intersect at 90 degrees.⏎Answer: Lines with the same slope are parallel and if the slope of one line is the negative reciprocal of the second line, then they are perpendicular.⏎Let's see how to identify

  3. #3 peak

    measure of how easily electricity flows along a certain path through an electrical element. The SI derived unit of conductance is the siemens (formerly referred to as the reciprocal ohm or mho). Oliver Heaviside coined the term in September

  4. #4 peak

    such as polymer melt.⏎Research in the Polymers and Industrial Mathematics group focuses on the mechanics of polymers and other complex fluids, free-surface flows and inverse problems. We are also concerned with the development and implementation of novel numerical and computational

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. L0:2553 weakly excites
  2. L1:2882 weakly suppresses
  3. L2:2974 weakly suppresses
  4. L2:2913 weakly excites
  5. L2:111 weakly suppresses
  6. L1:1246 weakly suppresses "word" after a language origin phrase
  7. L0:178 weakly suppresses
  8. L1:146 weakly excites
  9. L0:1445 weakly excites
  10. L2:1424 weakly excites
  11. L2:1710 weakly suppresses
  12. L2:2426 weakly excites
  13. L1:951 weakly excites
  14. L0:1224 weakly suppresses
  15. L2:66 weakly excites
  16. L2:2125 weakly excites
  17. L2:1494 weakly suppresses
  18. L1:2653 weakly excites
  19. L1:1464 weakly excites
  20. L1:2040 weakly excites
  21. L0:1337 weakly excites
  22. L2:2658 weakly excites
  23. L2:2326 weakly suppresses gravity word pieces (gravit-)
  24. L2:366 weakly suppresses

Feeds later layers

  1. L10:940 suppresses
  2. L9:946 suppresses
  3. L15:423 excites
  4. L4:303 weakly excites
  5. L24:2008 weakly suppresses
  6. L10:2453 weakly suppresses 'of' after publisher or division nouns
  7. L9:1789 weakly suppresses
  8. L4:2864 weakly suppresses
  9. L11:1724 weakly suppresses
  10. L5:2582 weakly suppresses
  11. L13:949 weakly suppresses
  12. L21:1974 weakly excites
  13. L12:552 weakly suppresses
  14. L10:2020 weakly suppresses
  15. L6:1262 weakly excites
  16. L15:808 weakly excites
  17. L19:954 weakly excites
  18. L13:311 weakly excites
  19. L10:1522 weakly excites
  20. L6:1692 weakly excites
  21. L15:2238 weakly excites
  22. L9:2389 weakly suppresses
  23. L13:525 weakly suppresses second half of an opposite pair
  24. L17:1951 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 -6.21 and -6.011 values between -1.42 and -1.221 values between -1.22 and -1.021 values between -0.82 and -0.622 values between -0.62 and -0.42110 values between -0.42 and -0.221666 values between -0.22 and -0.022014 values between -0.02 and 0.18104 values between 0.18 and 0.386 values between 0.38 and 0.581 values between 0.78 and 0.981 values between 1.58 and 1.78offnegativepositiveoften

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. ourn
  2. NR
  3. iol
  4. .IContainer
  5. iom
  6. 设计师
  7. NavController
  8. nr
  9. esign
  10. nghiệm

Suppresses

  1. dır
  2. alık
  3. itez
  4. .bat
  5. backs
  6. сь
  7. .backward
  8. ếu
  9. inspace
  10. oğlu