Neuron Explorer

Qwen3-0.6B

coercion, abuse and harassment

A neuron in layer 11 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 fires on words tied to force, abuse and victimisation, such as 'coercion' in a definition of human trafficking or 'harassment' in a bullying story. It is weakly negative on words about business deals and loans. Turning it up gave text that drifted toward distress, but the link is loose.

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

    growing criminal enterprises in the world.⏎If you’re not familiar with it, U.S. law defines human trafficking as “the use of force, fraud, or coercion to compel a person into commercial sex acts or labor against their will,” and it

  2. #2 peak

    - In search of their lost ones.<|endoftext|>The Verbal Reasoning section of the Graduate Record Examination (GRE) challenges a student’s reading comprehension, vocabulary, and sentence completion skills

  3. #3 peak

    all that is required to ensure each child receives the best education, knowing success in learning cannot be achieved in the same way for every child. They support teachers by providing a map of progression from Graduate through Proficient and Highly Accomplished to Lead

  4. #4 peak

    not drink so much wine that you get drunk, and do not let drinking become a habit.⏎Luke 6:31 (GNBDK)⏎Do for others just what you want them to do for you.⏎To answer the original question,

  5. #5 peak

    location can result from an accident or from forceful muscle contractions. A shoulder dislocation can also happen when an individual tries to pull themselves up by holding onto something with one hand while lifting their other arm.⏎The most common type of injury that

  6. #6 peak

    show-up. He or she can be required to speak, stand in a particular pose, and wear appropriate clothing.⏎Suspects can invoke the Sixth Amendment right to counsel during lineups and showups held after they have been arraigned or indicted.

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 have a 2000 car, and a 2000 car, and a 2000 car, and a 2000 car, and a 2000 car, and a

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

    also home to a species of ant that live in the plants' hollow thorns. In a classic example of mutualism, the ants protect the plant in return for shelter and food, said Meehan, who conducted the research while at Vill

  2. #2 peak

    Unit in Wapato kicked into high gear in 2005 when ARS entered into a research agreement with the California Cherry Advisory Board (CCAB). The collaboration makes sense since Washington

  3. #3 peak

    believe it or not, the global bond market is more than double the stock market.⏎When you purchase a bond from the bond issuer, you are essentially making a loan to the bond issuer. As the bond price is the amount of money investors pay

  4. #4 peak

    tasks. Redefinition is "when students connect to a classroom across the world where they would each write a narrative of the same historical event using the chat and comment section to discuss the differences, and they use the voice comments to discuss 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. L10:1624 suppresses
  2. L10:1995 suppresses formal nouns after verbs of answering or committing
  3. L10:204 excites
  4. L10:508 suppresses
  5. L6:2131 excites
  6. L10:2004 excites
  7. L9:1809 excites
  8. L10:1928 excites
  9. L10:1347 excites
  10. L9:2636 excites unclear: camping and money words
  11. L6:412 suppresses
  12. L10:1430 suppresses
  13. L7:1959 excites
  14. L9:2922 excites
  15. L10:2726 excites
  16. L9:2987 suppresses
  17. L9:782 weakly excites
  18. L10:2788 weakly excites
  19. L8:1862 weakly suppresses
  20. L10:3050 weakly excites
  21. L7:2351 weakly suppresses
  22. L10:2785 weakly suppresses
  23. L4:2185 weakly excites
  24. L6:1632 weakly suppresses

Feeds later layers

  1. L17:3040 suppresses
  2. L12:931 suppresses
  3. L12:2994 suppresses
  4. L20:2536 suppresses words for friend or colleague
  5. L14:2188 suppresses
  6. L17:1162 suppresses
  7. L12:1907 suppresses
  8. L12:2634 suppresses
  9. L20:2940 suppresses
  10. L19:1161 suppresses
  11. L14:1654 suppresses
  12. L15:1649 suppresses
  13. L12:597 excites
  14. L13:802 excites
  15. L13:2397 excites
  16. L22:1822 suppresses
  17. L12:536 suppresses
  18. L20:1794 suppresses
  19. L12:1630 suppresses
  20. L16:2222 suppresses
  21. L12:2739 suppresses
  22. L22:1839 suppresses
  23. L13:2116 suppresses
  24. L17:1986 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 -2.26 and -2.132 values between -2.13 and -2.011 values between -1.50 and -1.381 values between -1.38 and -1.253 values between -1.25 and -1.131 values between -1.13 and -1.007 values between -1.00 and -0.886 values between -0.88 and -0.7518 values between -0.75 and -0.6327 values between -0.63 and -0.5039 values between -0.50 and -0.3885 values between -0.38 and -0.25222 values between -0.25 and -0.131189 values between -0.13 and 0.001476 values between 0.00 and 0.13492 values between 0.13 and 0.25202 values between 0.25 and 0.3872 values between 0.38 and 0.5037 values between 0.50 and 0.637 values between 0.63 and 0.757 values between 0.75 and 0.883 values between 0.88 and 1.001 values between 1.00 and 1.131 values between 1.13 and 1.254 values between 1.25 and 1.381 values between 1.38 and 1.501 values between 1.63 and 1.761 values between 2.13 and 2.261 values between 2.63 and 2.76offnegativepositiveoften

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. parsers
  2. 开辟
  3. 觉悟
  4. annis
  5. ift
  6. efforts
  7. -parse
  8. PROFILE
  9. osph
  10. ifest

Suppresses

  1. 给了我
  2. rous
  3. 对方
  4. 关系
  5. ROUND
  6. hest
  7. hôn
  8. rematch
  9. ymb
  10. iciel