Neuron Explorer

Qwen3-0.6B

"which" after a noun of "the X of which"

A neuron in layer 3 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 "which" when it comes after a phrase like "the origin of" or "the building of", where it starts a relative clause. Turning it up made the text wander into repeated, unclear phrasing.

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

    as Āmū, Chāch, Xīwa, Samarqand, Chaghāniyān and some others, the origin of which till now remains unknown. Name of Buxārā, which is

  2. #2 peak

    viruses cannot, in general, spread via Web site access.⏎Of course, the above remarks do not always apply. By relaxing the security settings (the details of which depend on both operating system and browser), it is possible to create an environment in

  3. #3 peak

    . So, if you are relying on government, you are in fact relying on your fellow countrymen who work hard each day to earn an income the portion of which would be taken ‘without prior consent’ by the government which would in turn use

  4. #4 peak

    their satellites.⏎In the face of these multiplying instances of treaty-breaking, of violence, and of open warfare, the United States followed a policy the successive stages of which are summed up here.⏎During the years preceding the outbreak of war in Europe in

  5. #5 peak

    the long-awaited and often-troubled advent of the railroad in the state, the building of which shifted from private to public efforts and back again, amid tumultuous social, business

  6. #6 peak

    supported by Tannen’s research and my personal observation and experience, is that there is a difference in communication style and objective between the genders – the understanding of which can aid our effective partnership in the kingdom. This difference in style is a reflection

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 am a person who is a good and honest person, who is always ready for the future, who is always ready for the future, who is a good and honest person, who is always ready for the future, who is 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

    in clinical settings (note, there are now phase II clinica trials of a small molecule activator of cardiac myosin for congestive heart failure;Malik et al., Cardiac Myosin Activation: A Potential Therapeutic Approach for

  2. #2 peak

    66 million years ago - best-known for the extinction of the dinosaurs, the Cretaceous-Paleogene extinction event killed off three-quarters of all species. It's

  3. #3 peak

    historic mammal hair discovered in Cretaceous amberJune 16th, 2010 - 1

  4. #4 peak

    the journal Palaeontology.⏎The South Pyrenean area is rich in dinosaur egg sites, most of which communicate to sauropod eggs from the upper Cretaceous, dating back more than 70 million years ago.⏎During that period

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 weakly suppresses
  2. L2:2692 weakly excites
  3. L0:1301 weakly suppresses
  4. L0:178 weakly suppresses
  5. L1:2305 weakly excites
  6. L2:70 weakly excites
  7. L0:1 weakly suppresses
  8. L1:1464 weakly excites
  9. L2:49 weakly suppresses
  10. L2:2328 weakly suppresses
  11. L1:1321 weakly excites
  12. L1:2910 weakly suppresses
  13. L0:1589 weakly suppresses
  14. L1:1630 weakly suppresses
  15. L1:2452 weakly suppresses
  16. L1:2683 weakly excites
  17. L2:1036 weakly excites
  18. L2:1829 weakly suppresses
  19. L2:2583 weakly excites
  20. L1:168 weakly suppresses
  21. L1:635 weakly excites 'by' after a past participle
  22. L1:1841 weakly suppresses
  23. L1:2444 weakly suppresses
  24. L2:92 weakly excites

Feeds later layers

  1. L4:2517 excites
  2. L7:1103 weakly suppresses 'as' in 'not as ... as' comparisons
  3. L4:117 weakly excites
  4. L5:2103 weakly excites
  5. L5:2417 weakly excites
  6. L12:2103 weakly suppresses
  7. L16:2390 weakly excites
  8. L16:2213 weakly excites
  9. L6:994 weakly excites
  10. L16:302 weakly suppresses
  11. L22:1962 weakly excites
  12. L15:1319 weakly suppresses
  13. L15:2309 weakly suppresses
  14. L23:2050 weakly excites
  15. L6:1751 weakly suppresses
  16. L11:65 weakly suppresses
  17. L7:1129 weakly suppresses
  18. L9:2402 weakly excites
  19. L25:1815 weakly excites
  20. L13:159 weakly suppresses
  21. L18:1803 weakly suppresses
  22. L15:721 weakly excites
  23. L4:1523 weakly excites
  24. L6:1240 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 -0.64 and -0.614 values between -0.43 and -0.405 values between -0.40 and -0.374 values between -0.37 and -0.334 values between -0.33 and -0.3016 values between -0.30 and -0.2623 values between -0.26 and -0.2347 values between -0.23 and -0.2050 values between -0.20 and -0.16110 values between -0.16 and -0.13196 values between -0.13 and -0.09324 values between -0.09 and -0.06588 values between -0.06 and -0.021027 values between -0.02 and 0.01674 values between 0.01 and 0.04388 values between 0.04 and 0.08187 values between 0.08 and 0.11112 values between 0.11 and 0.1546 values between 0.15 and 0.1839 values between 0.18 and 0.2218 values between 0.22 and 0.2523 values between 0.25 and 0.287 values between 0.28 and 0.328 values between 0.32 and 0.353 values between 0.35 and 0.392 values between 0.39 and 0.421 values between 0.49 and 0.521 values between 0.69 and 0.73offnegativepositiveoften

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. 在生活中
  2. sometimes
  3. urch
  4. иров
  5. erm
  6. Lambert
  7. osoph
  8. iscrim
  9. positories
  10. happen

Suppresses

  1. чная
  2. vens
  3. 深交
  4. Kun
  5. 珠江
  6. ocht
  7. 下一
  8. Hv
  9. rrha
  10. plans