Neuron Explorer

Qwen3-0.6B

engineering and measurement wording

A neuron in layer 22 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 tends to fire in technical writing about machines, threads, sizes and materials, such as 'main circuit', 'major diameter' and 'plate thickness'. It is not clear what turning it up or down does, since the generated text shifted 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

    the AC asynchronous motor, improve the operation precision, change the power factor, and Realize overcurrent, overvoltage, overload protection and other functions. The main circuit of the frequency converter can be roughly divided into two categories: one is the

  2. #2 peak

    , generous and clean.⏎Cleaner uniforms must be equipped with appropriate personal protective equipment, such as gloves, water shoes, dust masks, etc., which can not only improve work efficiency, but also reduce or avoid occupational hazards for cleaners.⏎It can

  3. #3 peak

    forced) and a NO-GO gauge is not answered, the threaded part is deemed to be accepted in totality.⏎But in some specific cases where the thread major diameter (for screws) or minor diameter (for nuts) is critical, separate

  4. #4 peak

    For example, you may find a threaded rod measuring M8-1.25 x 1 Meter. The M8 (8mm) is the rod's diameter,

  5. #5 peak

    gauging of threaded parts.⏎Before that, a word about thread gauging practice:⏎In a threaded part, several features are involved – the thread’s major diameter, minor diameter, pitch diameter, pitch (in case of metric threads)

  6. #6 peak

    .22–95 filler wire of size 1.2 mm diameter and the base metal used was IS:2062 structural steel plate of 20 mm thickness. An

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 want to the same point of a point of a point of a point... like a human perspective, so that for every input, output, same point, same point, same point, same point, same point... i.e., the

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

    technology for COVID-19 diagnosis, vaccination, or therapy.⏎Xuexin Duan, School of Precision Instrument and Opto-electronics Engineering, Tianjin University, China⏎Weihua Guan, School of<|endoftext|>The flu is

  2. #2 peak

    1,* , Jingbiao Xu 2⏎Nanjing Xiaozhuang University Affiliated Primary School, Nanjing, Jiangsu 211171, China⏎2 Nanjing Xiaozhuang University, Nan

  3. #3 peak

    many parts of Tibet, with shops as well, where they were not seen previously.⏎Tibet's Overall Population⏎U'Tsang (Tibet Autonomous Region)⏎In the capital Lhasa, Tibetans offficially outnumber

  4. #4 peak

    the Department of Construction, Chaoyang University of Technology, Taichung 41349, Taiwan (

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. L18:233 suppresses
  2. L21:525 excites
  3. L19:615 excites
  4. L18:1050 excites
  5. L20:395 suppresses
  6. L9:207 suppresses
  7. L18:607 excites
  8. L10:873 excites
  9. L3:130 excites
  10. L13:2796 suppresses
  11. L9:232 suppresses
  12. L21:158 excites
  13. L20:859 excites
  14. L7:92 weakly excites
  15. L4:42 weakly suppresses
  16. L4:177 weakly excites
  17. L21:1405 weakly suppresses
  18. L9:281 weakly excites
  19. L9:46 weakly suppresses
  20. L18:768 weakly excites
  21. L10:155 weakly excites
  22. L4:267 weakly excites
  23. L18:1721 weakly excites
  24. L5:46 weakly excites

Feeds later layers

  1. L27:2103 weakly excites
  2. L25:2829 weakly excites
  3. L25:1491 weakly excites
  4. L24:1932 weakly excites
  5. L25:2597 weakly excites
  6. L23:1869 weakly excites
  7. L23:1397 weakly excites
  8. L25:1537 weakly excites
  9. L27:1029 weakly excites
  10. L27:1439 weakly excites
  11. L24:2610 weakly suppresses
  12. L24:889 weakly excites
  13. L27:2749 weakly excites
  14. L27:2656 weakly excites
  15. L24:1567 weakly excites
  16. L23:1879 weakly suppresses
  17. L24:2461 weakly excites
  18. L26:365 weakly suppresses
  19. L24:3010 weakly suppresses
  20. L24:2565 weakly suppresses
  21. L24:214 weakly suppresses
  22. L23:241 weakly suppresses
  23. L27:2544 weakly suppresses
  24. L23:2792 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 -8.71 and -8.281 values between -6.57 and -6.151 values between -4.44 and -4.011 values between -3.58 and -3.152 values between -2.30 and -1.872 values between -1.87 and -1.449 values between -1.44 and -1.0295 values between -1.02 and -0.59810 values between -0.59 and -0.162491 values between -0.16 and 0.27428 values between 0.27 and 0.7053 values between 0.70 and 1.127 values between 1.12 and 1.552 values between 1.55 and 1.981 values between 1.98 and 2.412 values between 2.83 and 3.261 values between 4.97 and 5.401 values between 7.96 and 8.39offnegativepositiveoften

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. etter
  2. |`⏎
  3. 国度
  4. official
  5. bras
  6. OMX
  7. picture
  8. Official
  9. pictured
  10. racially

Suppresses

  1. utan
  2. rends
  3. �建
  4. entar
  5. alike
  6. ervation
  7. uto
  8. eterminate
  9. rightness
  10. 遞