Neuron Explorer

Qwen3-0.6B

table pipes and 'non' before a hyphen

A neuron in layer 18 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 mostly lights up on the vertical bar character used to draw table rows in plain text. Its strongest negative cases are the word 'non' before a hyphen, as in 'non-climatic'. The steering results are scattered and do not show 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

    object identifiers) is encouraged.⏎Citing the date when an online work was consulted is now optional.⏎Selected examples of MLA 7 vs. MLA 8⏎|MLA 7||MLA 8|⏎Article from a library⏎|

  2. #2 peak

    look at some examples of deferred revenue and how to determine if something is deferred revenue<|endoftext|>Your location is outside the National Library of Australia. Resources you can access:⏎|Show by title|⏎ACT Heritage Library: The A.C.T. electoral rolls for

  3. #3 peak

    ), and freshwater fish (2,000 species) than any other continent.⏎What continent has the largest animals?⏎List of largest land carnivorans⏎|Rank||Common name||Native range by continent|⏎|1||Polar bear

  4. #4 peak

    PPL) and cross polarised (XPL) . As an example, in a Cornubian granite, the following minerals have the following basic properties:⏎|Quartz||Dull!

  5. #5 peak

    and Schedule⏎|4 years through 8 years||One or two dosesa, 0.

  6. #6 peak

    ease of use and because a simple application tends to grow in complexity with time.⏎The global pattern to follow is MVC and the layering:⏎(View & Controller)⏎|View||HTML templates usually generated using a template engine|⏎|Input Controller||Filter

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: the- 0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0-0

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

    to adapt over the long term. People have considerable knowledge and skills to adapt, often using innovative new solutions, but their ability to respond depends on a host of non-climatic factors that increases people’s vulnerability, such as conflicts that restrict mobility

  2. #2 peak

    as the violence and scope of protests decline. As the protests begin to lessen or become less intense, non-protesters will feel safer leaving their homes.⏎"There are other possible explanations for

  3. #3 peak

    job with just one subunit.⏎The authors stress that they are not arguing against natural selection as a process; they say rather that it can be aided by “non-ad<|endoftext|>Poet and writer Langston Hughes, famous for his elucidations of

  4. #4 peak

    akers, whether in our personal or in our political lives. Yet, few have the skills or ideas to create the conditions in which peace can result. Courses in non-violence are rarely taught in schools, and non-violence is rarely used by

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. L6:1751 excites
  2. L9:3065 excites
  3. L2:668 weakly suppresses
  4. L16:38 weakly suppresses
  5. L17:157 weakly suppresses
  6. L1:516 weakly excites
  7. L0:626 weakly excites
  8. L1:339 weakly suppresses
  9. L16:282 weakly excites
  10. L15:129 weakly suppresses
  11. L16:79 weakly suppresses
  12. L16:1869 weakly excites
  13. L15:122 weakly suppresses
  14. L15:115 weakly excites
  15. L1:2629 weakly suppresses
  16. L4:2819 weakly suppresses
  17. L17:998 weakly excites
  18. L6:1063 weakly excites
  19. L15:89 weakly excites
  20. L13:1229 weakly excites
  21. L4:681 weakly suppresses
  22. L16:118 weakly suppresses
  23. L4:1326 weakly excites
  24. L7:78 weakly excites

Feeds later layers

  1. L19:2429 strongly suppresses
  2. L23:864 suppresses
  3. L19:1053 suppresses
  4. L23:1175 suppresses
  5. L20:1064 suppresses
  6. L19:792 excites
  7. L27:752 suppresses
  8. L22:494 suppresses
  9. L26:459 suppresses opening bracket before an acronym
  10. L25:149 suppresses
  11. L21:378 suppresses
  12. L26:526 suppresses
  13. L27:2159 weakly suppresses
  14. L21:118 weakly suppresses
  15. L19:1305 weakly excites unclear: verbs after a noun or 'that'
  16. L26:596 weakly suppresses
  17. L23:676 weakly suppresses the 'm' in a.m. and p.m
  18. L27:1459 weakly suppresses
  19. L21:2279 weakly excites
  20. L24:266 weakly suppresses
  21. L25:1030 weakly suppresses
  22. L27:807 weakly suppresses
  23. L27:83 weakly suppresses
  24. L26:771 weakly suppresses

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 -41.53 and -40.411 values between -33.71 and -32.591 values between -30.35 and -29.231 values between -28.12 and -27.001 values between -27.00 and -25.881 values between -24.76 and -23.651 values between -21.41 and -20.292 values between -20.29 and -19.172 values between -19.17 and -18.061 values between -18.06 and -16.941 values between -13.58 and -12.471 values between -12.47 and -11.352 values between -11.35 and -10.231 values between -9.11 and -7.993 values between -7.99 and -6.882 values between -6.88 and -5.763 values between -5.76 and -4.644 values between -4.64 and -3.528 values between -3.52 and -2.4013 values between -2.40 and -1.29136 values between -1.29 and -0.173640 values between -0.17 and 0.9581 values between 0.95 and 2.071 values between 2.07 and 3.19offnegativepositiveoften

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. iment
  2. ACHE
  3. ọng
  4. ATIONAL
  5. fingert
  6. NCY
  7. าย
  8. edly
  9. finder
  10. ỡ

Suppresses

  1. -
  2. pat
  3. -C
  4. disc
  5. ť
  6. -M
  7. metric
  8. -c
  9. −
  10. atics