Neuron Explorer

Qwen3-0.6B

"by" after a passive verb

A neuron in layer 13 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 responds to the word "by" when it names who made or did something, as in "created by", "written by" or "betrayed by". Turning it up gave wandering text that did not clearly show the pattern, so the effect is hard to read.

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

    Megan Smith0⏎A Hummingbird’s Amazing Journey⏎by Robert Burleigh⏎- Language: english⏎- Format: hardcover,

  2. #2 peak

    in italics) are related in the sentences to follow. A<|endoftext|>southeast asia 2⏎Home > Preview⏎The flashcards below were created by user⏎on FreezingBlue Flashcards.⏎by early 1900

  3. #3 peak

    and several regiments of infantry, while the heights on both sides were crowned by Russian artillery and infantry. Lord Raglan, perceiving that it was the intention of the Russians to carry off the guns they had captured from the Turks, ordered up

  4. #4 peak

    |⏎Koothus in Silapathikaram⏎by Subashini Pathmanathan⏎Ancient Tamils called their dances by the

  5. #5 peak

    just that. After discussion with another friend, Wes Prichard, Prichard suggested to Wennberg<|endoftext|>Home > Preview⏎The flashcards below were created by user⏎on FreezingBlue Flashcards.⏎What factors could affect feet and leg

  6. #6 peak

    and found a hiding place for them for the remainder of the war.⏎Waterman Pinkhof continued her work sleeping at a different address each night until she was betrayed by someone claiming to be able to assist in Joop Westerweel’s release

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: a question from a guest of a cat. all, a cat. there were three, a white fluffy white, fluffy, white, white., the white fluffy white, white, white, white, white, white, three

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

    , and others reinforcing existing knowledge and challenges. But all have important implications for designing appropriate programme interventions to fill information and communication gaps. They include:⏎- The dominant role of radio, particularly community radio, as an information medium and potentially a

  2. #2 peak

    may have taken, few of them have endured, which helps explain why they've been so easily and so largely forgotten.⏎On earlier occasions, too, the part played by Greece has been pivotal - sometimes positive but sometimes negative. And history has recently

  3. #3 peak

    ,” said Tom Crouch, a senior curator at the museum and the man in charge of coordinating the event.⏎“It is a chance to remind visitors of the role that the Smithsonian has played in the history of flight,” Crouch said, “from

  4. #4 peak

    the historical conception of the county as the place of vicinage in federal capital trials; (2) a Batson-type three-step process for rooting out the influence of race on the decision to prosecute federally; and/or (3) voluntary measures

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. L12:1286 strongly suppresses
  2. L9:1562 excites 'by' before a byline or agent
  3. L12:2810 excites
  4. L12:1529 excites
  5. L12:321 suppresses
  6. L11:2517 excites
  7. L6:2880 suppresses
  8. L12:337 suppresses
  9. L10:2138 excites
  10. L12:2016 suppresses
  11. L12:1040 suppresses
  12. L12:1531 excites
  13. L12:2266 suppresses
  14. L11:2387 suppresses
  15. L10:1054 suppresses 'from' after movement or transfer verbs
  16. L11:1450 suppresses
  17. L12:1066 suppresses
  18. L12:1304 suppresses
  19. L9:1465 excites
  20. L9:355 suppresses
  21. L6:1637 excites
  22. L12:2369 weakly suppresses
  23. L3:2203 weakly suppresses
  24. L11:2497 weakly excites

Feeds later layers

  1. L14:2374 suppresses
  2. L20:1357 suppresses
  3. L14:1711 suppresses
  4. L18:2195 suppresses
  5. L16:630 suppresses
  6. L18:2321 suppresses
  7. L15:3006 suppresses
  8. L19:1832 suppresses
  9. L14:2918 suppresses
  10. L17:1033 suppresses
  11. L16:2530 suppresses
  12. L16:1706 weakly suppresses
  13. L14:1796 weakly suppresses
  14. L19:1503 weakly suppresses
  15. L17:1370 weakly excites
  16. L17:1221 weakly suppresses
  17. L15:2017 weakly suppresses
  18. L14:2007 weakly suppresses
  19. L14:1953 weakly suppresses
  20. L16:1986 weakly excites
  21. L19:2617 weakly suppresses
  22. L17:577 weakly excites
  23. L17:1058 weakly suppresses
  24. L18:2770 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 -2.97 and -2.861 values between -2.76 and -2.651 values between -2.12 and -2.021 values between -1.91 and -1.802 values between -1.49 and -1.383 values between -1.28 and -1.174 values between -1.17 and -1.062 values between -1.06 and -0.963 values between -0.96 and -0.858 values between -0.85 and -0.7511 values between -0.75 and -0.6418 values between -0.64 and -0.5422 values between -0.54 and -0.4348 values between -0.43 and -0.33139 values between -0.33 and -0.22350 values between -0.22 and -0.111111 values between -0.11 and -0.011389 values between -0.01 and 0.10481 values between 0.10 and 0.20164 values between 0.20 and 0.3163 values between 0.31 and 0.4246 values between 0.42 and 0.5221 values between 0.52 and 0.638 values between 0.63 and 0.736 values between 0.73 and 0.843 values between 0.94 and 1.052 values between 1.16 and 1.26offnegativepositiveoften

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. 渥
  2. станавлива
  3. ynthesis
  4. >()->
  5. networking
  6. 后台
  7. Networking
  8. 在网络
  9. mix
  10. 在京

Suppresses

  1. emple
  2. 这名
  3. ôm
  4. donor
  5. unicorn
  6. spacer
  7. Pence
  8. 瓒
  9. óm
  10. Contributors