Neuron Explorer

Qwen3-0.6B

unclear: list items and everyday prepositions

A neuron in layer 14 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 that finish an everyday phrase, like 'stay close', 'rest your head on', or the 'nuts' in a list of foods. The link is loose, and turning it up didn't produce a clear theme.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    clothes with someone you trust.⏎DURING A VIOLENT INCIDENT⏎• Call 911 or local police.⏎• If an argument starts, stay close to an exit.⏎• Stay away from the bathroom, kitchen, or any other

  2. #2 peak

    /1<|endoftext|>Some may argue that sleeping is one of the best parts of the day. You get into bed, snuggle under the covers, rest your head on a comfortable cushion, and switch off your thoughts for the

  3. #3 peak

    <|endoftext|>Remember snuggling with your child, book between you, the scent of baby shampoo under your nose? Most of us think such moments are gone after our snuggler learns to read. But they don’t have to be! Save the ritual

  4. #4 peak

    solve their problems.⏎Read also, 10 lines short stories for kids.⏎2. Hungry dogs⏎Three dogs lived in a city. They were very close friends. One day, all three dogs were very hungry.⏎They started searching for

  5. #5 peak

    are potato chips, French fries, pop, red meat and processed meat.⏎Foods that do not contribute to weight gain are fruits and vegetables, whole grains, nuts and yogurt.⏎This is based on following 120,000

  6. #6 peak

    for the day. If you have time constraints, consider options such as protein smoothies, healthy muffins, hardboiled eggs or granola with fruit, nuts and seeds.⏎Eliminate or Min<|endoftext|>Venezuela’s human rights record was

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 a a contradiction in my own life. I have been a person who did not exist, but I am now... I have been a person, I have been a person, I have been a person. I have been 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

    1883 we were walking along past a large field of⏎potatoes in North Oxford, where we beheld the strange spectacle of a⏎pair of bipeds drilling their offspring to march up one potato row and⏎down the other

  2. #2 peak

    have to illustrate the sentence on a large piece of paper in front of the class, while the rest of the class guesses what sentence it is. The first one to correctly say the sentence is the next one to choose a sentence and do the illustration

  3. #3 peak

    fire and air.⏎He was a strong proponent of the sustainable use and cultivation of land and water, proper treatment of animals, plants and birds, and the equal rights of users. In this context the modernity of the Prophet sall Alla

  4. #4 peak

    taught personal responsibility through leading by both word and deed. Yet lacking an in-home role model or role models -- my father grew up as an out-of-wedlock only child -- does not necessarily determine one's fate or destiny. Sometimes one must

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. L10:1270 weakly excites
  2. L12:2148 weakly excites
  3. L3:2994 weakly suppresses
  4. L10:1968 weakly excites
  5. L7:2317 weakly suppresses
  6. L10:2243 weakly excites
  7. L13:397 weakly excites
  8. L12:1615 weakly suppresses
  9. L12:1435 weakly excites
  10. L7:2785 weakly excites
  11. L8:2299 weakly suppresses
  12. L10:2038 weakly excites
  13. L7:2814 weakly excites
  14. L9:760 weakly suppresses
  15. L2:1264 weakly suppresses
  16. L12:2779 weakly suppresses
  17. L10:2637 weakly suppresses evaluative words: good, bad, positive
  18. L8:997 weakly excites
  19. L8:1497 weakly suppresses
  20. L5:724 weakly suppresses
  21. L6:1961 weakly suppresses
  22. L8:1449 weakly suppresses
  23. L8:899 weakly suppresses
  24. L8:229 weakly excites

Feeds later layers

  1. L22:983 weakly excites
  2. L19:2312 weakly suppresses
  3. L25:2313 weakly suppresses
  4. L18:1175 weakly excites
  5. L24:1394 weakly excites
  6. L22:1947 weakly suppresses
  7. L20:2454 weakly excites
  8. L20:1653 weakly suppresses
  9. L20:1625 weakly excites
  10. L27:1961 weakly excites
  11. L15:2809 weakly excites
  12. L21:1325 weakly suppresses
  13. L23:2084 weakly excites
  14. L20:2626 weakly suppresses angle and right-angle phrases
  15. L19:2817 weakly excites
  16. L24:1713 weakly excites
  17. L20:1035 weakly excites
  18. L23:1367 weakly suppresses
  19. L17:1074 weakly excites
  20. L21:234 weakly suppresses
  21. L21:1159 weakly suppresses
  22. L20:1509 weakly suppresses
  23. L19:1097 weakly suppresses
  24. L22:1325 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 -2.10 and -1.981 values between -1.75 and -1.631 values between -1.52 and -1.401 values between -1.40 and -1.281 values between -1.28 and -1.174 values between -1.17 and -1.051 values between -1.05 and -0.945 values between -0.94 and -0.827 values between -0.82 and -0.7112 values between -0.71 and -0.5919 values between -0.59 and -0.4736 values between -0.47 and -0.3684 values between -0.36 and -0.24256 values between -0.24 and -0.131134 values between -0.13 and -0.011697 values between -0.01 and 0.11353 values between 0.11 and 0.22124 values between 0.22 and 0.3464 values between 0.34 and 0.4541 values between 0.45 and 0.5721 values between 0.57 and 0.6914 values between 0.69 and 0.807 values between 0.80 and 0.922 values between 0.92 and 1.037 values between 1.03 and 1.153 values between 1.15 and 1.273 values between 1.27 and 1.384 values between 1.38 and 1.501 values between 1.50 and 1.611 values between 1.61 and 1.732 values between 1.96 and 2.081 values between 2.43 and 2.54offnegativepositiveoften

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. 为期
  2. igte
  3. 捣
  4. -response
  5. ัส
  6. Inspectable
  7. _food
  8. weight
  9. -track
  10. 饼干

Suppresses

  1. 开放
  2. vampires
  3. 败
  4. 敗
  5. .disconnect
  6. .twitch
  7. inema
  8. cuts
  9. célib
  10. yle