Neuron Explorer

Qwen3-0.6B

opening parenthesis after a sentence

A neuron in layer 17 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 reacts to an opening bracket that starts a side note right after a finished sentence, such as a source credit or an explanation. When it is pushed negative, it reacts to the word 'And' that starts a new sentence instead. Steering changed the story in ways that do not clearly follow this pattern.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens an email to the maker with the neuron already filled in).

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

    state’s current coastal Master Plan, which incorporated worst-case scenarios for relative sea-level rise calculated two years ago— which the new figures now make out-of-date. (The state’s estimates of sea-level rise and subsidence are listed on page

  2. #2 peak

    (I say “apparently” because there are differing accounts of what exactly happened). Coleman, riding in the rear seat, was not wearing a seat belt. (She may have been unable to see over the cockpit when strapped in.) Her mechanic

  3. #3 peak

    from traditional South African folk songs, and protest songs from famous musicians from Miriam Makeba and Hugh Masekela to U2 and Stevie Wonder. (Source: PBS)⏎The conclusion of "Have You Heard From Johannesburg?, Free at

  4. #4 peak

    a big one with what looks like a fuzzy beard of brown leaves underneath its green fronds. It’s naturally found around desert oases in the Colorado Desert. (The Colorado Desert is not in Colorado, but is named for the river. Joshua

  5. #5 peak

    , Dr. Patterson has maintained a database now listing 370 genetic disorders in dogs, with 5 to 10 new ones added each year. (By comparison, there are some 5

  6. #6 peak

    Another case is where we need to authenticate some data, i.e. produce a kind of integrity check that only a party with a given private key could produce. (In this case, the general solution is to combine a hash code with encryption.)⏎

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: if, for, say, if, for, if, for, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if, if,

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

    health (not to mention a whole host of other health issues).⏎Read on to learn more about some of the drinks that may contribute to aging your brain faster. And for more, check out The Worst Foods Increasing Your Risk of Dementia.⏎S

  2. #2 peak

    real impact on their future. They will look around at their environment to determine if they want to stay where they are or move somewhere else when they get older. And they may get their first job at 16 or 17, which

  3. #3 peak

    a reader and a textbook. First, it provides extensive readings from the central writings in ethics along with writings from literary figures who offer concrete examples of lived ethics. And second

  4. #4 peak

    and, later, to ensure a multiplicity of routes such that the system could survive a nuclear attack. It was not to keep the system’s data safe. And as ARPANET gave way to the purely academic world of the early internet,

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. L16:489 suppresses
  2. L16:432 suppresses period after month or initial
  3. L14:155 excites sentence-ending periods in prose
  4. L14:128 suppresses
  5. L16:240 suppresses comma after 'sometimes' or 'the truth'
  6. L9:2447 excites
  7. L10:478 suppresses
  8. L14:762 suppresses
  9. L13:53 excites period ending a claim before elaboration
  10. L4:102 excites
  11. L15:313 excites decimal point inside numbers
  12. L11:1194 excites
  13. L5:168 suppresses
  14. L4:7 excites
  15. L16:482 excites
  16. L14:422 excites periods after initials and abbreviations
  17. L13:813 suppresses
  18. L16:203 weakly suppresses
  19. L9:2302 weakly excites
  20. L10:612 weakly excites full stop ending boilerplate or a title
  21. L16:869 weakly excites
  22. L10:1004 weakly excites
  23. L16:285 weakly excites
  24. L3:854 weakly suppresses

Feeds later layers

  1. L19:979 excites
  2. L18:1841 excites
  3. L18:2045 excites
  4. L26:22 excites the 'res' fragment inside words
  5. L18:2217 excites
  6. L18:1403 excites
  7. L18:1791 suppresses
  8. L26:236 weakly suppresses
  9. L24:216 weakly suppresses
  10. L19:2493 weakly excites
  11. L25:1113 weakly suppresses
  12. L24:411 weakly suppresses
  13. L19:1262 weakly suppresses
  14. L27:286 weakly suppresses
  15. L18:621 weakly suppresses
  16. L25:373 weakly suppresses
  17. L18:892 weakly suppresses
  18. L18:1749 weakly suppresses
  19. L18:1216 weakly excites
  20. L24:1853 weakly excites
  21. L20:68 weakly excites
  22. L18:1671 weakly excites
  23. L22:368 weakly suppresses
  24. L19:2416 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.

1 values between -12.38 and -11.831 values between -9.63 and -9.081 values between -9.08 and -8.532 values between -6.88 and -6.332 values between -5.78 and -5.242 values between -5.24 and -4.696 values between -4.69 and -4.144 values between -4.14 and -3.592 values between -3.59 and -3.043 values between -3.04 and -2.4914 values between -2.49 and -1.9411 values between -1.94 and -1.3936 values between -1.39 and -0.8478 values between -0.84 and -0.292358 values between -0.29 and 0.261048 values between 0.26 and 0.81111 values between 0.81 and 1.3652 values between 1.36 and 1.9135 values between 1.91 and 2.4537 values between 2.45 and 3.0018 values between 3.00 and 3.5520 values between 3.55 and 4.1014 values between 4.10 and 4.6512 values between 4.65 and 5.2010 values between 5.20 and 5.7513 values between 5.75 and 6.306 values between 6.30 and 6.854 values between 6.85 and 7.402 values between 7.40 and 7.953 values between 7.95 and 8.501 values between 8.50 and 9.051 values between 9.05 and 9.59offnegativepositiveoften

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. cased
  2. pled
  3. ponent
  4. оказ
  5. cle
  6. unction
  7. LOAT
  8. uum
  9. zá
  10. uits

Suppresses

  1. ,json
  2. ,r
  3. ,options
  4. rão
  5. ?,?,
  6. ,np
  7. ,R
  8. ,’
  9. ertools
  10. though