Neuron Explorer

Qwen3-0.6B

'in' meaning 'in this case' or 'in between'

A neuron in layer 16 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 word 'in' when it means something like 'in this case', 'in between' or 'in terms of', as in 'anything in between' or 'in that'. When it was turned down, the model got stuck repeating 'in in in'. Turned up, the writing drifted to unrelated topics.

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

    spirits can happen outside. It just requires that we embrace a broader vision of what “outside” means: an urban playground, a remote national park, and everywhere in between. We need to Rethink Outside and center equity when doing so.

  2. #2 peak

    ards have been going in and out of style for years. What is the reason for this changing style of hair and no hair? We’re going to step back in time to the beginning and examine the history of the beard.⏎In prehistoric times

  3. #3 peak

    Engage in Reading Adventures⏎There is no better time to read than during the dog days of summer. Whether it’s sci-fi, historical fiction, or anything in between, students can be encouraged to find things that they are excited to explore during

  4. #4 peak

    capital of the South,” a Jet Age airport promised to put what had once been a minor railroad town within a day’s reach of London and Paris – and, in the minds of some Atlantans, on the same plane of international significance as

  5. #5 peak

    different types of treatment<|endoftext|>The concept of a graph has been around since the dawn of mechanical computing and for many decades prior in the domain of pure mathematics. Due in large part to this golden age of databases, graphs are becoming increasingly popular in software

  6. #6 peak

    ,” the phenomenon captured by the telescope is what’s known as a reflection nebula – a piece of debris left behind by the formation of a newborn star, or in this case, a small, multiple star system known like V380 Orion

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: in I have, on I have, I have, I have on I have on on I have, I have on on again, on I have, I have, I have, I have on I have, on I have, 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

    Comparative Analysis⏎- 人文学院-已发表论文 ⏎“助动词”在汉语中具有较大的作用,使用率高,但是泰国学生经常在学习和使用汉语助动词中产生各种的偏误。本文通过

  2. #2 peak

    and Huntington’s, are characterized by protein aggregation (clumping) in the nerves’ cells, impairing their function.⏎Normally, when a protein is made in the body, it is folded into the 3D shape it needs to perform its

  3. #3 peak

    kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.⏎Translations are not retained in our system. Your use of this feature and the translations is subject

  4. #4 peak

    to be an accessible topic for middle school students.19<|endoftext|>毋忘在莒(Pinyin: Wu Wang Zai Ju) is a famous Chinese idiom

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. L4:289 strongly excites
  2. L3:493 strongly suppresses
  3. L0:781 suppresses
  4. L1:1179 suppresses 'in' after 'result' (results in)
  5. L5:490 excites
  6. L9:1025 excites
  7. L8:351 suppresses 'in' after 'high' or 'dressed'
  8. L10:262 suppresses
  9. L14:1625 excites the word 'in' after a noun
  10. L0:1280 excites
  11. L5:729 suppresses
  12. L4:1966 suppresses
  13. L8:3068 suppresses
  14. L2:2307 excites
  15. L15:871 suppresses
  16. L6:846 suppresses
  17. L15:1328 weakly suppresses
  18. L15:78 weakly excites
  19. L2:1463 weakly suppresses
  20. L14:1389 weakly suppresses
  21. L11:2856 weakly suppresses
  22. L1:1922 weakly suppresses
  23. L15:115 weakly excites
  24. L14:1122 weakly suppresses

Feeds later layers

  1. L17:706 excites
  2. L24:893 suppresses
  3. L21:1935 excites
  4. L19:704 excites
  5. L17:2619 excites
  6. L23:1911 excites
  7. L22:535 excites
  8. L18:869 excites
  9. L20:1574 excites
  10. L24:2163 excites
  11. L25:2032 excites
  12. L19:1345 excites
  13. L27:168 excites
  14. L25:1865 excites 'in' in 'keep in mind'
  15. L23:1366 excites
  16. L17:2821 excites
  17. L26:825 suppresses
  18. L24:2051 excites
  19. L25:2957 excites 'in' starting a word fragment
  20. L27:405 suppresses
  21. L24:2134 excites
  22. L18:2615 excites
  23. L18:698 excites
  24. L18:2832 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.

680 values between -1.19 and -0.592993 values between -0.59 and 0.0193 values between 0.01 and 0.6223 values between 0.62 and 1.2218 values between 1.22 and 1.8210 values between 1.82 and 2.436 values between 2.43 and 3.038 values between 3.03 and 3.635 values between 3.63 and 4.236 values between 4.23 and 4.841 values between 4.84 and 5.445 values between 5.44 and 6.042 values between 6.04 and 6.652 values between 6.65 and 7.253 values between 7.25 and 7.853 values between 7.85 and 8.453 values between 8.45 and 9.063 values between 9.06 and 9.661 values between 9.66 and 10.265 values between 10.26 and 10.876 values between 10.87 and 11.472 values between 11.47 and 12.074 values between 12.07 and 12.674 values between 12.67 and 13.282 values between 13.28 and 13.883 values between 13.88 and 14.483 values between 14.48 and 15.091 values between 15.09 and 15.692 values between 15.69 and 16.291 values between 16.29 and 16.892 values between 16.89 and 17.501 values between 18.10 and 18.701 values between 18.70 and 19.311 values between 19.31 and 19.912 values between 20.51 and 21.111 values between 21.11 and 21.721 values between 21.72 and 22.321 values between 22.32 and 22.92offnegativepositiveoften

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. variants
  2. form
  3. var
  4. asm
  5. emple
  6. /out
  7. StateException
  8. variant
  9. terms
  10. finity

Suppresses

  1. 在外面
  2. сайте
  3. itur
  4. ально
  5. 在学校
  6. 在路上
  7. 在国内
  8. 在国外
  9. 在生活中
  10. 在深圳