Neuron Explorer

Qwen3-0.6B

capital M inside abbreviations

A neuron in layer 27 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 a capital M that starts a short name or abbreviation, such as the M in MWh, MIF or MSE, especially right after an opening bracket. When it was turned up, the model's writing turned into loops of repeated words that had nothing to do with the topic.

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

    s, flush residues after fossil fuels are burned and cool power plants.⏎Coal uses up to 3.2 cubic metres of water per megawatt hour (MWh) of electricity produced, gas uses up to 1.7 m3

  2. #2 peak

    lab is equipped with nuclear-powered capabilities to vaporize rocks, ingest soil, and determine whether Mars was ever inhabited by smaller forms of life. For updates follow @M

  3. #3 peak

    : livelihoods and health, education, respectable employment and financial stability, and financial security. This is accomplished through the use of the Multi-dimensional Inequality Framework (MIF). This is one of the first initiatives in Vietnam to comprehend the

  4. #4 peak

    least two minutes each time. Brush your teeth in circular strokes, and clean all surfaces, especially the backs, and chewing surfaces.⏎2. Use mouthwash:⏎Mouthwash can help destroy bacteria that cause bad breath and gum disease in your mouth

  5. #5 peak

    10 log10(MAX^2/MSE) = 10 log10(MAX/(MSE)^(1/2))^2 = 20 log10(MAX

  6. #6 peak

    resistance towards oxidation and corrosion. Besides that, volatility loss and deposit of carbon should be kept low for industrial and machinery lubrication.⏎* Malaysian Palm Oil Board (MPOB),⏎6 Persiaran Institusi, Bandar Baru Bangi

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: Likeathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathathath

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

    before the Senate Agriculture Committee, which is considering the carbon-farming legislation. Introduced by senators Mike Braun (R-Indiana) and Debbie Stabenow (D-Michigan), the bill sets up the US Department of Agriculture to certify independent consultants

  2. #2 peak

    recognizing what meaning the word contributes to the passage in which it appears.”⏎For more information, The report card can be found online: www.nationsreportcard.gov.MORE IN Vermont News⏎- Most Popular⏎- Most Emailed⏎-

  3. #3 peak

    have grown, ____10___ have usually come to represent more than a single neighborhood.⏎3. Прочитайте раздел Грамматического справочника Модальные глаголы и их эквиваленты и выполн

  4. #4 peak

    of Pharmacy terminated.⏎1905-Theological Seminary organized in Waco.⏎1907-Separated from Baylor University.⏎1910-Moved to Fort Worth.⏎1918-College of Dentistry organized 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. L22:1138 suppresses
  2. L17:230 excites
  3. L24:1190 excites
  4. L10:311 suppresses
  5. L0:73 excites
  6. L21:1679 excites
  7. L23:270 excites
  8. L20:234 suppresses
  9. L18:553 excites
  10. L2:1563 excites
  11. L19:334 suppresses
  12. L7:291 excites
  13. L18:133 excites
  14. L1:16 excites
  15. L1:35 suppresses
  16. L19:623 excites
  17. L21:505 suppresses
  18. L20:2203 weakly suppresses
  19. L16:428 weakly suppresses unclear: mid-sentence commas, 'most' before a verb
  20. L26:520 weakly excites
  21. L4:393 weakly suppresses
  22. L0:844 weakly excites
  23. L0:1585 weakly suppresses
  24. L9:1040 weakly suppresses

Feeds later layers

None: this is the last layer.

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 -90.44 and -87.621 values between -31.32 and -28.501 values between -22.87 and -20.062 values between -11.61 and -8.803 values between -5.98 and -3.1792 values between -3.17 and -0.353798 values between -0.35 and 2.474 values between 2.47 and 5.281 values between 8.10 and 10.913 values between 10.91 and 13.732 values between 19.36 and 22.17offnegativepositiveoften

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. 干部队伍
  2. หมายเลข
  3. 要做
  4. 也只是
  5. delegated
  6. 不间断
  7. tection
  8. 担负
  9. �
  10. 无敌

Suppresses

  1. arine
  2. ath
  3. ac
  4. ass
  5. ovie
  6. MMMM
  7. anus
  8. AP
  9. ASTER
  10. bps