Neuron Explorer

Qwen3-0.6B

'only' in 'not only ... but'

A neuron in layer 15 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 'only' in the phrase 'not only ... but also', which sets up a second point. Turning it up made the model write in a repetitive, list-like way, though not obviously about this phrase. The negative side reacts to commas inside lists of names or places.

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

    , and smaller ears than the Black Rat. The former was introduced into England - probably from the East Indies - early in the eighteenth century, and has not only almost entirely displaced the Black Rat in this country, but

  2. #2 peak

    that promotes self-interest, if not an unchecked selfishness.1 Since the 1970s, neoliberalism or free-market fundamentalism has become not only a much-vaunted ideology that now shapes all aspects of life in the United

  3. #3 peak

    are not properly maintained, your home will not only feel very uncomfortable to live in, but you and your family may also become susceptible

  4. #4 peak

    team was interested in a protein (G-protein-coupled receptor) which occurs in various cell types. Had the researchers administered a drug, they would not only have deactivated the receptor in the Purkinje cells, but in all cell types

  5. #5 peak

    bodies, and pigment granules. Understanding these components will not only deepen your knowledge of hair but also help you identify them and comprehend their significance.

  6. #6 peak

    1 m/s. In order to reduce the⏎pressure changes<|endoftext|>Tūtira Restoration Options Presented⏎Scientific modelling of Lake Tūtira has not only given Hawke’s Bay Regional Council a better understanding of how the lake works,

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 think I have the ... (I think I have the ... ?) ... I'm ... I am ... ? ... ? ... ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?

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

    wine, he commanded the seven eunuchs who served him – Methuan, Biztha, Harbon, Bigtha, Anagtha, Zethat and Karkas – to bring before him

  2. #2 peak

    In eleven states, about a third or more of black women were poor (IA, IN, LA, ME, MI, MN, MS, SD, WI, WV, and, VT). Roughly a third of Hispanic women were poor in

  3. #3 peak

    and only one is accepted which rules out the other three. The alternatives are:⏎- Does one karma produce one birth?⏎- Does one karma cause several births?⏎- Do several karmas result in several births?⏎- Do several karmas

  4. #4 peak

    : (1) cervical, lying along the esophagus; (2) clavicular, frequently merging into a single interclavicular sac; (3) anterior thoracic, on the abdominal wall of the thorax; and (

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. L13:2423 strongly excites commas in lists of names or items
  2. L14:2942 strongly excites
  3. L14:810 excites
  4. L14:1975 excites
  5. L12:1647 excites
  6. L14:2412 suppresses
  7. L13:2942 excites
  8. L14:1215 suppresses
  9. L13:2022 suppresses
  10. L13:106 excites
  11. L13:1112 suppresses
  12. L14:2125 excites
  13. L14:2766 excites
  14. L14:318 excites
  15. L13:1830 suppresses
  16. L14:2732 excites
  17. L14:287 excites
  18. L11:2112 excites
  19. L13:912 suppresses
  20. L14:2034 excites
  21. L12:58 excites
  22. L13:2262 excites
  23. L14:1307 excites
  24. L11:1258 excites

Feeds later layers

  1. L16:3064 excites
  2. L16:2880 suppresses
  3. L17:919 suppresses
  4. L17:1029 suppresses
  5. L16:1364 suppresses
  6. L16:2069 excites
  7. L16:1084 suppresses
  8. L16:2200 excites
  9. L16:2902 suppresses
  10. L16:1307 suppresses
  11. L16:1809 suppresses comma inside a list of names or acronyms
  12. L19:867 suppresses
  13. L20:610 suppresses
  14. L16:533 excites
  15. L16:397 suppresses
  16. L16:2922 excites
  17. L19:1605 suppresses
  18. L17:518 excites
  19. L17:997 suppresses
  20. L20:882 suppresses
  21. L16:1141 suppresses
  22. L19:492 excites
  23. L16:2329 suppresses
  24. L16:2360 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.

2 values between -1.87 and -1.611 values between -1.61 and -1.361 values between -1.36 and -1.101 values between -1.10 and -0.842 values between -0.84 and -0.5833 values between -0.58 and -0.321169 values between -0.32 and -0.062385 values between -0.06 and 0.19264 values between 0.19 and 0.4536 values between 0.45 and 0.713 values between 0.71 and 0.973 values between 0.97 and 1.232 values between 1.48 and 1.741 values between 2.00 and 2.261 values between 2.52 and 2.782 values between 3.29 and 3.551 values between 4.32 and 4.581 values between 8.20 and 8.45offnegativepositiveoften

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. patibility
  2. оборот
  3. ableView
  4. ..."⏎
  5. ".⏎⏎
  6. ..."⏎⏎
  7. "⏎⏎⏎⏎
  8. Girlfriend
  9. `.⏎⏎
  10. ...'⏎

Suppresses

  1. itur
  2. urge
  3. &,
  4. 建设和
  5. itos
  6. いけ
  7. 多层次
  8. urg
  9. itas
  10. ino