Neuron Explorer

Qwen3-0.6B

'who' after a person noun

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 fires on the word 'who' when it follows a word for a person, like 'a person who', 'someone who' or 'a woman who'. Turning it up did not make the model's writing focus on people, so the output effect is not clear.

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

    <|endoftext|>Acute paranoia is a clinically recognized personality disorder characterized by extreme distrust, hostility and suspicion of others for unfounded and often delusional reasons. A person who suffers from paranoia is limited regarding healthy social interactions including

  2. #2 peak

    .⏎Emotional problems, such as stress, anxiety, or depression, can make a person more forgetful and can be mistaken for dementia. For instance, someone who has recently

  3. #3 peak

    under copyright without permission, the owner may sue for the value of the violation. Most such cases are handled by civil law. In more serious cases, a person who copies a work that is protected under copyright could be arrested, fined or even go

  4. #4 peak

    the calorie count of any food;⏎spends more than 2 hours in the gym AND could be underweight? Health-conscious, overzealous, or someone who could have an eating disorder?⏎If you opted for the last one, you are

  5. #5 peak

    Towns⏎|A religious settlement.⏎|A person who teaches his or her religion to others.⏎|A cowhand; a person who

  6. #6 peak

    type of shoe that limits pronation or inward rolling of the foot), and running style, which can be influenced by strengthening the hip and core muscles.⏎A runner who increases mileage or intensity too quickly, such as by doing too many speed workouts or

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 been thinking about lately, and I have been thinking about lately. I think that is the time I have been thinking about lately. I think that is the time I have been thinking about lately. I think that is the time I

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

    itensis and B. abortus disease since 1958 (4).⏎In Bulgaria, until 1998 serologic screening was mandatory for all cattle, sheep, and goats >12 months of age. Afterward,

  2. #2 peak

    smart devices. From small household appliances to large industrial tools, these objects all have unique identifiers and can communicate with each other. They also have …What is Internet of Things? Read More<|endoftext|>Physicist who slowed the speed of light to speak

  3. #3 peak

    stories in fact tell us about Franciscan fathers who had been⏎fundamental for all complexes’ building. Some had been the inventors, like Bernardino Caimi

  4. #4 peak

    . All these caveats make for a murky takeaway.⏎But a lot of other animals eat meat. The article doesn’t talk about our cooking of meat, which I think is the key to how we’ve evolved.<|endoftext|>Is the rising popularity of

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. L0:1224 suppresses
  2. L0:178 suppresses
  3. L0:1301 suppresses
  4. L0:1389 weakly excites
  5. L0:1817 weakly excites
  6. L0:2000 weakly suppresses
  7. L0:1526 weakly excites
  8. L0:2345 weakly suppresses
  9. L0:489 weakly excites
  10. L0:2250 weakly excites
  11. L0:1167 weakly excites
  12. L0:2178 weakly suppresses
  13. L0:1766 weakly excites
  14. L0:31 weakly suppresses
  15. L0:2350 weakly excites
  16. L0:703 weakly excites
  17. L0:1118 weakly suppresses
  18. L0:2549 weakly suppresses
  19. L0:2481 weakly suppresses
  20. L0:230 weakly suppresses
  21. L0:733 weakly excites
  22. L0:2652 weakly suppresses
  23. L0:972 weakly excites
  24. L0:1561 weakly excites

Feeds later layers

  1. L2:105 suppresses
  2. L3:402 excites
  3. L9:865 excites
  4. L2:302 suppresses
  5. L7:685 excites
  6. L6:958 weakly excites
  7. L14:1798 weakly excites
  8. L5:2773 weakly excites
  9. L3:94 weakly suppresses
  10. L7:2390 weakly excites
  11. L3:1593 weakly suppresses
  12. L7:1382 weakly suppresses
  13. L7:1484 weakly excites
  14. L10:1124 weakly excites
  15. L7:173 weakly suppresses
  16. L21:2185 weakly suppresses
  17. L2:2116 weakly suppresses
  18. L9:2442 weakly excites
  19. L2:560 weakly suppresses
  20. L14:298 weakly excites
  21. L2:1984 weakly suppresses
  22. L15:317 weakly suppresses
  23. L15:840 weakly excites
  24. L15:2728 weakly 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 -2.68 and -2.451 values between -2.01 and -1.781 values between -1.56 and -1.334 values between -1.11 and -0.884 values between -0.88 and -0.6623 values between -0.66 and -0.44207 values between -0.44 and -0.212046 values between -0.21 and 0.011502 values between 0.01 and 0.2474 values between 0.24 and 0.4610 values between 0.46 and 0.696 values between 0.69 and 0.914 values between 0.91 and 1.143 values between 1.14 and 1.368 values between 1.36 and 1.582 values between 1.58 and 1.811 values between 1.81 and 2.033 values between 2.03 and 2.261 values between 2.26 and 2.482 values between 2.48 and 2.711 values between 3.38 and 3.602 values between 3.83 and 4.051 values between 4.50 and 4.721 values between 6.07 and 6.29offnegativepositiveoften

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. hor
  2. iors
  3. ewire
  4. ensored
  5. "go
  6. nels
  7. ród
  8. irectional
  9. 弘
  10. horizontal

Suppresses

  1. imum
  2. vince
  3. ział
  4. INESS
  5. rút
  6. ใจ
  7. ւ
  8. ;)
  9. ươ
  10. ibal