Neuron Explorer

Qwen3-0.6B

pieces of software and library names

A neuron in layer 11 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on the later pieces of technical names, such as 'lib' in 'GLib', 'Bool' in a Haskell module, '.curve' in a scipy function or '2' in 'urllib2'. It is mostly quiet or slightly negative on names in ordinary text, like 'Journal' or 'Base Ten'. Turning it up or down did not give a clear story.

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

    since it is a function instead of syntax, it can be partially applied and passed around to higher order functions, like so:⏎ghci> :m + Data.Bool.Extras⏎ghci> let yesOrNo = bool

  2. #2 peak

    as resource databases, with a database type of GLib, and a library-specific creator ID. The GLib library will usually contain three

  3. #3 peak

    :|| Web page: http://www.tandfonline.com/VECE20|⏎|Order Information:||Web: http://www.tandfonline.com/pricing/journal/VECE20|⏎When requesting a correction,

  4. #4 peak

    ography. arXiv - Computer Science > Cryptography and Security (21 October 2016) | arXiv: 1610.06918 | PDF⏎We ask whether neural networks can learn

  5. #5 peak

    up your code into a main portion (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

  6. #6 peak

    The widget uses scipy.curve_fit to find the optimal values of the parameters.⏎The widget works only on regression

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: I will be a scientist? I will be a doctor? I will be a teacher? I will be a nurse? I will be a writer? I will be a pilot? I will be a doctor? I will be a researcher?

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

    write to show their mathematical thinking.⏎Contents 4-8 Counting and Cardinality Task Cards (CC) 9-13 Numbers and Operations in Base Ten Task Cards (NBT) 14-18 Operations and Algebraic

  2. #2 peak

    Gall, head of the French Space Agency CNES.⏎Courtesy ESA/Rosetta/MPS for OSIRIS Team MPS/UPD/LAM/IAA/SSO/INTA/UPM/DASP/IDAThe spacecraft will spend

  3. #3 peak

    the brown dwarfs, identified as CFBDSIR 1458+10B, in a paper in the current issue of The Astrophysical Journal.⏎"At such temperatures we expect the brown dwarf to have properties that are different from

  4. #4 peak

    iztli Calmecac performs Aztec dance from Pennsylvania, another in the 2009 Homegrown Concert Series sponsored by the American Folklife Center.⏎Calmecac, Ollin Yoliztli -

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. L8:509 strongly excites
  2. L9:1164 strongly excites
  3. L7:1681 suppresses
  4. L8:1031 excites
  5. L10:2154 excites
  6. L9:2303 suppresses
  7. L6:673 suppresses "to" after adjectives like obvious or meaning
  8. L10:2787 suppresses unclear: document boundaries and digits
  9. L6:2192 suppresses
  10. L10:1642 suppresses
  11. L4:1183 suppresses
  12. L4:1797 suppresses
  13. L9:1366 weakly excites
  14. L3:2128 weakly suppresses
  15. L7:1755 weakly excites
  16. L4:2121 weakly excites
  17. L7:900 weakly excites
  18. L9:814 weakly suppresses
  19. L3:2733 weakly suppresses the word 'form' in 'formative'
  20. L4:1765 weakly excites
  21. L3:1561 weakly excites
  22. L6:728 weakly excites
  23. L9:461 weakly excites
  24. L4:905 weakly suppresses

Feeds later layers

  1. L14:1248 excites
  2. L27:362 excites
  3. L26:174 excites
  4. L16:1446 excites
  5. L24:228 excites
  6. L13:1354 excites
  7. L19:1596 excites
  8. L20:536 excites
  9. L25:387 excites
  10. L17:1674 excites
  11. L19:913 excites
  12. L26:593 excites
  13. L13:1576 excites
  14. L13:1845 excites
  15. L27:875 excites
  16. L25:294 excites
  17. L27:370 excites
  18. L16:2651 excites
  19. L21:389 excites
  20. L23:372 excites
  21. L25:1135 excites
  22. L12:498 suppresses
  23. L21:1842 excites
  24. L19:954 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 -0.73 and -0.691 values between -0.59 and -0.561 values between -0.53 and -0.502 values between -0.50 and -0.462 values between -0.46 and -0.435 values between -0.43 and -0.4011 values between -0.40 and -0.3722 values between -0.37 and -0.3325 values between -0.33 and -0.3043 values between -0.30 and -0.2771 values between -0.27 and -0.23107 values between -0.23 and -0.20118 values between -0.20 and -0.17171 values between -0.17 and -0.14261 values between -0.14 and -0.10354 values between -0.10 and -0.07471 values between -0.07 and -0.04486 values between -0.04 and -0.01535 values between -0.01 and 0.03344 values between 0.03 and 0.06292 values between 0.06 and 0.09206 values between 0.09 and 0.13130 values between 0.13 and 0.1681 values between 0.16 and 0.1968 values between 0.19 and 0.2241 values between 0.22 and 0.2626 values between 0.26 and 0.2915 values between 0.29 and 0.324 values between 0.32 and 0.362 values between 0.36 and 0.394 values between 0.39 and 0.423 values between 0.42 and 0.452 values between 0.45 and 0.491 values between 0.49 and 0.522 values between 0.55 and 0.58offnegativepositiveoften

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. iswa
  2. wipe
  3. ู้
  4. ',(
  5. ",(
  6. ascus
  7. olves
  8. Incorporated
  9. ::_
  10. życ

Suppresses

  1. ?
  2. ??
  3. ?
  4. ???
  5. -
  6. "
  7. 5
  8. "?
  9. 義
  10. 义