Neuron Explorer

Qwen3-0.6B

hyphen joining word parts

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

The neuron fires on a hyphen that joins pieces of a long name or compound word, such as 'Ha-Devarim' or 'Qal'a-ye'. When turned up, the model mostly produced runs of blank lines, and turned down it repeated the word 'writing'.

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

    o Arama'ah: Tabulae in grammaticen linguae Chaldaeae (Paris, 1560); Aseret ha-Devarim: Decalogus, with the commentary of Abraham Ibn Ezra,

  2. #2 peak

    possible site for Alexander’s fort, is known today as the Citadel of Herat or Qal’a-ye Ikhtiyar al-Din. The name Ikhtiyar al

  3. #3 peak

    the one armed Tyr was the icon for a judge.⏎<|endoftext|>Sign up for our newsletter⏎Gardening in the Rainy Zone.⏎Pronounced: AL-lee-um ka-ra-tah-vee-EN-see⏎Sun

  4. #4 peak

    . In case you need a good example here is one from Edutopia.http://www.edutopia.org/blog/flip-stem-classroom-ainissa-ramirez and all flipped classrooms do not look alike.⏎To see how technology transforms

  5. #5 peak

    required prospective teachers to pass these tests. Research, however, does not show a relationship between test scores and teacher effectiveness.⏎In one [research study], Marc Claude-Charles Colitti of Michigan State University examined data going back to 196

  6. #6 peak

    of this article, click here: http://www.abc.net.au/news/rural/2017-05-23/the-future-of-automation-in-the-mining-industry/8550636<|endoftext|>

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:writingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwritingwriting

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

    a lot of wood-frame construction, which is flexible and tends to ride out a quake fairly well, said Heidi Faison, outreach director at the Pacific Earthquake Engineering

  2. #2 peak

    tension, difficulty sleeping and difficulty concentrating.⏎Obsessive Compulsive Disorder (OCD)⏎Obs<|endoftext|>A mobile phone user outside a mobile payment facility in Port-au-Prince, Haiti.⏎our goal: to alleviate poverty

  3. #3 peak

    ard (1652-1728), who was assisted by Marshal Ferdinand de Marsin (1656-1706) and the Elect

  4. #4 peak

    About Apollo Missions"That’s one small step for a man, one giant leap for mankind...” nnOn 20 July 1969, Neil Armstrong made the historic first steps on the Moon. In celebration of the 5

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. L26:2003 strongly excites hyphen joining two word parts
  2. L25:1399 strongly suppresses
  3. L20:61 excites
  4. L22:271 excites
  5. L25:36 excites
  6. L26:153 excites
  7. L26:381 excites
  8. L26:1374 suppresses
  9. L23:1351 excites
  10. L19:804 excites
  11. L26:254 excites
  12. L22:1533 excites
  13. L16:27 excites
  14. L26:725 excites
  15. L16:39 suppresses
  16. L0:141 excites the word 'of' in 'middle of'
  17. L26:1018 excites
  18. L25:40 excites
  19. L21:50 suppresses comma after 'So' or intro phrase
  20. L26:230 excites
  21. L21:2309 excites
  22. L26:319 excites
  23. L3:459 excites
  24. L20:1886 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.

8 values between -5.57 and -1.993894 values between -1.99 and 1.603 values between 1.60 and 5.181 values between 5.18 and 8.771 values between 102.01 and 105.601 values between 134.29 and 137.88offnegativepositiveoften

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. ⏎ ⏎
  3. ⏎⏎⏎⏎
  4. �
  5. ,⏎⏎⏎
  6. "⏎⏎⏎
  7. ellaneous
  8. وء
  9. iction
  10. ויות

Suppresses

  1. writing
  2. hot
  3. dam
  4. bar
  5. per
  6. sm
  7. css
  8. cards
  9. cut
  10. fast