Neuron Explorer

Qwen3-0.6B

hyphen inside a compound word

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 the hyphen in words joined by a dash, like Self-Reflection or Coal-to-liquid. Turning it down made the model produce run-on strings of words without spaces, and turning it up made it repeat small words like 'the'.

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

    roversial Qeiyafa Comes to U.S. Museum<|endoftext|>An Argumentative Essay about Mary Oliver’s Poems: Exploring the Power of Nature and Self-Reflection

  2. #2 peak

    were all based on Earth and planetary cycles, most had a spring goddess. The spring goddess in Scandinavian was "Ostra," for Anglo-Saxons it was "Eostre," and the current region of Germany she was called "Eastre

  3. #3 peak

    be even easier to use. Do you live in or near a Cool City? Check the site now to find out.⏎The Worst Alternative Fuel Ever?⏎Coal-to-liquid fuel has

  4. #4 peak

    the documents are surrounded by an inert gas. A cross-section of these cases can be seen in the “Public Vaults,” along with a special documentary about the re-encasement process.⏎This brochure is made possible by the National Archives Foundation with

  5. #5 peak

    05, 2012, from http://www.studymode.com/essays/The-Struggle-For-Power-In-The-994323.html⏎"The Struggle for Power in the U

  6. #6 peak

    /essays/The-Struggle-For-Power-In-The-994323.html>.⏎"The Struggle for Power in the U

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:EXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENTEXPECTMENT

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

    in the Trans-Pecos. Current evidence indicates that mountain lions (Felis concolor), bobcats (Lynx rufus), and coyotes (Canis latrans) are the primary predators involved, although little information is available

  2. #2 peak

    :⏎Fifty-eight mammal species have been reported in the park to date such as:⏎- Tiger (Panthera Tigris),⏎- Clouded Leopard (Neofelis nebulosa),⏎- Swamp Deer (Cervus du

  3. #3 peak

    how many exist.⏎These tentacled titans have now ensnared marine biologists with⏎another riddle: they have left their normal gathering grounds in⏎the Sea of Cortez in Baja California, Mexico. Fishermen worry that

  4. #4 peak

    when using the bathroom, particularly shigella (a foodborne bacteria that causes diarrhea, fever, and stomach cramps) and norovirus (often called "stomach flu" which also causes nausea, vomiting, stomach cramps,

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. L19:1174 strongly excites
  2. L19:124 strongly suppresses
  3. L22:65 strongly suppresses
  4. L16:1421 suppresses
  5. L20:563 excites
  6. L14:140 suppresses
  7. L18:76 suppresses blank spot before a number
  8. L20:59 suppresses
  9. L21:29 excites
  10. L22:14 excites
  11. L18:48 suppresses
  12. L19:963 excites
  13. L25:17 excites
  14. L10:84 suppresses
  15. L25:239 excites
  16. L23:262 suppresses
  17. L12:2582 suppresses
  18. L23:80 excites blank slot before a year or number
  19. L25:222 suppresses
  20. L25:109 excites
  21. L19:1005 excites
  22. L14:348 suppresses
  23. L20:193 excites
  24. L16:723 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 -26.69 and -25.562 values between -16.52 and -15.401 values between -15.40 and -14.271 values between -14.27 and -13.144 values between -13.14 and -12.011 values between -12.01 and -10.883 values between -10.88 and -9.752 values between -9.75 and -8.624 values between -8.62 and -7.493 values between -7.49 and -6.3612 values between -6.36 and -5.238 values between -5.23 and -4.109 values between -4.10 and -2.9727 values between -2.97 and -1.8447 values between -1.84 and -0.712251 values between -0.71 and 0.421445 values between 0.42 and 1.5542 values between 1.55 and 2.6710 values between 2.67 and 3.809 values between 3.80 and 4.936 values between 4.93 and 6.061 values between 6.06 and 7.197 values between 7.19 and 8.324 values between 8.32 and 9.453 values between 9.45 and 10.582 values between 10.58 and 11.711 values between 11.71 and 12.841 values between 13.97 and 15.101 values between 17.36 and 18.48offnegativepositiveoften

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. is
  2. that
  3. .
  4. @
  5. :
  6. =
  7. i
  8. of
  9. and
  10. #

Suppresses

  1. artz
  2. ennent
  3. →se
  4. →en
  5. atem
  6. apor
  7. ourt
  8. TIM
  9. BEGIN
  10. ores