Neuron Explorer

Qwen3-0.6B

unclear: opening quotes and closing brackets

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 is hard to pin down. It fires most on quotation marks that open a title or quote right after a line break, and on a few closing brackets. When it was turned down, the model repeated closing brackets over and over.

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

    /MSE) = 10 log10(MAX/(MSE)^(1/2))^2 = 20 log10(MAX/(MSE)^(1/2)). Therefore, PSNR = <|endoftext|>In my current position

  2. #2 peak

    (of at most 32K), and a number of libraries (each of at most 32K).⏎GLib (pronounced ``gee-lib'') shared libraries are a way to implement shared libraries on the Pilot that differs from

  3. #3 peak

    beverage products.⏎Let’s raise a fork to the food we love and the people who produce it!<|endoftext|>Deor's Lament: a Modern English Translation⏎"Deor's Lament" is one of the truly great poems of English antiqu

  4. #4 peak

    letto Ticino in Orta, or the preachers⏎like Giovan Battista Aguggiari in Vare<|endoftext|>Photograph by Vanessa Allen⏎"Arteria" by Handiedan (Netherlands) highlights beluga whales

  5. #5 peak

    Coast — Did the Coast Change?" and calling on citizen volunteers to study aerial photographs taken before and after extreme storms to determine if or how the landscape changed.⏎"Computers cannot yet automatically identify damages and geomorphic changes to the coast from the

  6. #6 peak

    eruptions and strips the ocean of its oxygen, according to a new study.⏎And no one knows what's causing the "intermittently catastrophic" changes.⏎"Many geologists believe that geological events are random over time," said geologist Michael

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:))))))))))))))))))))))))))))))))))))))))))))))))

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

    ebook on an iPhone, iPad or Android mobile device you will need the Adobe Digital Editions app, or BlueFire Reader or Txtr app. These are free, too. For more details see this article.⏎|Size: ||359

  2. #2 peak

    involves multiple systemic symptoms or appears to be rapidly worsening should be immediately referred for medical evaluation. Possible DISRs include Stevens-Johnson syndrome and toxic epidermal necrolysis.⏎4. Identify the drug, classify the type of reaction,

  3. #3 peak

    to correct its authors, title, abstract, bibliographic or download information, contact: (Michael McNulty)⏎If references are entirely missing, you can add them using this form.<|endoftext|>Farmers fight drought with move to digital age⏎21

  4. #4 peak

    keep eating fiber to maximize its many health benefits.⏎Dr. Frost is a board-certified colon and rectal surgeon and a fellow of the American Society of Colon and Rectal Surgeons. His practice offers colonoscopy, colon operations and evaluation

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. L25:250 suppresses
  2. L26:712 excites
  3. L26:381 suppresses
  4. L25:36 suppresses
  5. L26:505 excites unclear: opening quote marks after a line break
  6. L16:2 excites
  7. L26:1602 suppresses
  8. L24:260 suppresses
  9. L23:819 excites
  10. L26:345 excites
  11. L21:2490 excites comma inside dates
  12. L20:1904 excites
  13. L19:501 excites
  14. L23:777 excites
  15. L19:223 suppresses
  16. L19:1875 excites
  17. L23:184 suppresses
  18. L20:79 suppresses
  19. L24:1539 excites
  20. L25:112 excites
  21. L18:2126 suppresses
  22. L19:1777 suppresses
  23. L21:147 suppresses
  24. L16:78 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 -12.30 and -10.262 values between -10.26 and -8.222 values between -8.22 and -6.183 values between -6.18 and -4.1419 values between -4.14 and -2.102146 values between -2.10 and -0.071240 values between -0.07 and 1.97235 values between 1.97 and 4.01103 values between 4.01 and 6.0579 values between 6.05 and 8.0929 values between 8.09 and 10.1315 values between 10.13 and 12.1715 values between 12.17 and 14.219 values between 14.21 and 16.251 values between 16.25 and 18.282 values between 18.28 and 20.322 values between 20.32 and 22.361 values between 24.40 and 26.441 values between 28.48 and 30.521 values between 32.55 and 34.591 values between 36.63 and 38.671 values between 67.21 and 69.25offnegativepositiveoften

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. "" ⏎
  2. 什么呢
  3. '' ⏎
  4. ''⏎
  5. (){
  6. loquent
  7. toi
  8. же
  9. емся
  10. الة

Suppresses

  1. )
  2. ),
  3. )⏎
  4. ).
  5. ),⏎
  6. )]
  7. );⏎
  8. ).⏎⏎
  9. )⏎⏎
  10. ):⏎⏎