Neuron Explorer

Qwen3-0.6B

preposition after a noun phrase

A neuron in layer 14 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 lights up on the small connecting word (usually 'of' or 'in') that comes right after a noun and introduces a name or detail, such as 'Battle of St Albans' or 'cells in the central nervous system'. When it was turned up, the model's replies did not show a clear theme.

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

    official language in the content classrooms, typically mathematics and science. According to Cummins, bilingual education is defined as “the used of two (or more) language of instruction at some point in a student’s school career” (cited in Cre

  2. #2 peak

    infect the nervous system. It reactivates later in life in people who have suppressed immune systems. The virus damages oligodendrocytes, the cells in the central nervous system (CNS) which produce myelin, the material that

  3. #3 peak

    ⏎Reproduced from https://style.mla.org/whats-new/⏎Author. When a source has three or more authors, only the first one shown in the source is normally given

  4. #4 peak

    properly. This is the motivation that put the Basel Convention into play.⏎In Cuernavaca, a plant was built and paid for by the US Filter Corporation of Palm Desert, California. The filter was the first granted to an American company making

  5. #5 peak

    and a minor gentleman, Sir Richard Woodville. Elizabeth’s first husband, was Sir John Grey, and they had two sons before he died at the Second Battle of St Albans. In September 14

  6. #6 peak

    was 72.⏎A sister, Casey Camp-Horinek, said her brother had cancer.⏎Mr. Camp, a member of the Ponca Tribe of Oklahoma, was a longtime member of the American Indian Movement, organizing more than

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: of, say, in any of the, I have been, perhaps, encountered in some of, I have been, perhaps I, or, now, in, or, perhaps, I, or, now, I, or, 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

    and educational opportunities. The women’s declaration appropriated the language and form of the Declaration of Independence. Considered shocking in its time, many of the aims and goals contained in the Declaration of Sentiments apply to

  2. #2 peak

    lers’ farm animals destroyed the shoots of new trees and trampled the land. The compacted soil was less fertile and less absorbent. The environmental pressure exerted by the settlers was not only a result of numbers, but also of their subs

  3. #3 peak

    was required and it was introduced as pitta. Vata, pitta and kapha came to be called the Tridosha.⏎Meanwhile, the theory witnessed expansion in another interesting line. Once the idea of the two-fold causes of diseases

  4. #4 peak

    the end they don't look at the writer by what he writes, but rather evaluate him by where he lives, or by the location of "the room" from which he writes, as the Peruvian writer Vargas Llosa noted in

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. L10:1124 suppresses
  2. L7:1451 excites
  3. L12:1823 suppresses
  4. L6:958 suppresses
  5. L9:865 suppresses
  6. L11:775 excites
  7. L8:1703 excites
  8. L13:2988 excites
  9. L11:1381 excites
  10. L10:155 suppresses
  11. L7:2503 suppresses
  12. L12:1531 suppresses
  13. L5:1428 suppresses
  14. L10:669 excites
  15. L13:44 excites inside web addresses and citations
  16. L13:423 excites
  17. L8:982 suppresses
  18. L11:223 suppresses
  19. L13:1108 suppresses
  20. L9:2312 weakly excites 'to' after a noun (a role to play)
  21. L12:958 weakly excites
  22. L12:1222 weakly suppresses
  23. L7:685 weakly excites
  24. L9:392 weakly suppresses "and" inside organisation names

Feeds later layers

  1. L16:2746 suppresses
  2. L19:2621 suppresses
  3. L15:632 suppresses
  4. L15:1464 suppresses
  5. L20:55 excites
  6. L15:2728 suppresses
  7. L16:1739 suppresses
  8. L18:68 suppresses
  9. L17:1167 suppresses
  10. L15:1851 suppresses 'that' after a noun, 'being' in passives
  11. L18:1672 suppresses
  12. L15:817 excites
  13. L16:1441 excites
  14. L15:942 excites
  15. L15:2893 suppresses
  16. L17:2067 suppresses
  17. L16:1322 suppresses
  18. L15:1247 suppresses
  19. L15:1328 suppresses
  20. L22:132 suppresses
  21. L17:2267 suppresses
  22. L17:812 excites
  23. L15:396 excites
  24. L20:1481 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 -6.02 and -5.731 values between -4.28 and -3.991 values between -3.70 and -3.414 values between -2.55 and -2.261 values between -2.26 and -1.977 values between -1.97 and -1.6813 values between -1.68 and -1.3915 values between -1.39 and -1.1025 values between -1.10 and -0.8144 values between -0.81 and -0.52235 values between -0.52 and -0.231957 values between -0.23 and 0.061276 values between 0.06 and 0.35271 values between 0.35 and 0.6424 values between 0.64 and 0.9315 values between 0.93 and 1.227 values between 1.22 and 1.516 values between 1.51 and 1.801 values between 1.80 and 2.092 values between 2.09 and 2.382 values between 5.27 and 5.56offnegativepositiveoften

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. 能否
  2. 所以他
  3. 能不能
  4. ledge
  5. xn
  6. 甚至
  7. fk
  8. ему
  9. 矩
  10. 粉色

Suppresses

  1. @",
  2. ญ
  3. onnen
  4. ์
  5. above
  6. ashtra
  7. :",⏎
  8. downloaded
  9. carved
  10. specified