Neuron Explorer

Qwen3-0.6B

unclear: opening quote marks after a line break

A neuron in layer 26 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. Its strongest reactions come on an opening quotation mark that starts a new paragraph in news-style text, but the steered outputs do not share a clear theme.

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

    apweed varieties also are major problems in the Four Corners area. Russian knapweed is the most common, overtaking fields and empty lots like wildfire.⏎"Russian knapweed spreads by black,

  2. #2 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

  3. #3 peak

    been directly observed and quantified in the thicker areas of the planet's shell, Calais said.⏎"Such dyking events had been included in theories, but researchers had never before been

  4. #4 peak

    to play outside," she says.⏎However, she stresses, indoor space is limited, and these homes had on average four people per room living in the homes.⏎"Outdoor play is

  5. #5 peak

    ., who led the research. Her team is presenting the work this week at a scientific meeting, Purines 2010, in Barcelona, Spain.⏎"Acupuncture has been a mainstay of medical treatment in certain parts of the world

  6. #6 peak

    al collections manager at the Field Museum, Chicago, US, said hearing the news that the monkey belonged to a new genus "sent shivers down my spine".⏎"Simply put, the genetics said that it was closely related to baboons, but

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

    ) законодательная власть⏎4) Associate Justice d) беспартийная система⏎5

  2. #2 peak

    ,000 people every year.⏎Also known as vestibular schwannoma, acoustic neurinoma, or acoustic neurilemoma, an acoustic neuroma is a

  3. #3 peak

    ledge SJ, Niebuhr DW, Scher AI, Munger KL and Ascherio A⏎Multiple sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system of unknown etiology

  4. #4 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

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. L19:43 excites
  2. L16:255 excites
  3. L2:1620 excites
  4. L16:3 suppresses
  5. L19:129 excites
  6. L23:1408 suppresses
  7. L5:97 suppresses
  8. L16:17 suppresses
  9. L3:395 suppresses
  10. L19:118 excites
  11. L19:89 excites
  12. L19:233 suppresses
  13. L4:345 excites
  14. L16:137 excites
  15. L20:344 excites hyphen inside compound words
  16. L16:7 suppresses
  17. L22:866 excites
  18. L22:271 suppresses
  19. L20:441 excites
  20. L2:1869 suppresses
  21. L24:248 excites
  22. L2:761 suppresses
  23. L2:404 excites
  24. L0:132 suppresses

Feeds later layers

  1. L27:248 suppresses
  2. L27:128 excites unclear: opening quotes and closing brackets
  3. L27:627 suppresses
  4. L27:1425 suppresses
  5. L27:172 suppresses
  6. L27:1007 suppresses
  7. L27:740 weakly suppresses
  8. L27:1520 weakly excites
  9. L27:2125 weakly excites
  10. L27:1019 weakly excites
  11. L27:2439 weakly suppresses slash inside web addresses
  12. L27:29 weakly suppresses
  13. L27:873 weakly suppresses
  14. L27:2229 weakly suppresses the letter h starting a word fragment
  15. L27:2889 weakly suppresses
  16. L27:657 weakly suppresses
  17. L27:1946 weakly suppresses the letter z inside words
  18. L27:981 weakly suppresses
  19. L27:929 weakly suppresses
  20. L27:541 weakly suppresses
  21. L27:2451 weakly excites
  22. L27:681 weakly suppresses
  23. L27:2518 weakly suppresses the letter f starting a word
  24. L27:1459 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.

20 values between -5.23 and -1.903880 values between -1.90 and 1.422 values between 1.42 and 4.751 values between 4.75 and 8.071 values between 8.07 and 11.401 values between 21.38 and 24.701 values between 54.64 and 57.972 values between 124.49 and 127.81offnegativepositiveoften

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. vant
  2. .terminate
  3. ��
  4. Macron
  5. vably
  6. �
  7. [%
  8. mint
  9. seller
  10. 瘗

Suppresses

  1. ></
  2. ><
  3. > ⏎
  4. >{{$
  5. In
  6. It
  7. 知道
  8. >{
  9. The
  10. Well