Neuron Explorer

Qwen3-0.6B

lowercase 'm' inside split words

A neuron in layer 27 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 lights up on the letter m when it appears inside a word that has been broken into pieces, like the m in mastiche or mystery. Turning it up or down gave repeated filler words such as one or none, so its effect on the text is unclear.

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

    time to complete the ultimate objective which is the discovery of the cures.⏎Pandemic board game, Z-MAN Games, Retrieved from https://www.zmangames.com/en/games/pandemic/⏎A game

  2. #2 peak

    , we have to start with ancient Greece, where citizens routinely chewed on the resin (or sap) from the Mastic tree. This substance was dubbed “mastiche

  3. #3 peak

    century that America finally implememented the promise of its ideals to all its citizens and in the process saved

  4. #4 peak

    23:34:592021-10-20 23:34:59assignment 2⏎mynursingpaper.com is a unique service that provides guidance with different types of

  5. #5 peak

    with a bleach solution and to take down your feeder for a week if you notice sick or dead birds.⏎Officials still haven't figured out the cause of the 'mystery' illness that has killed thousands of songbirds. However, the state says

  6. #6 peak

    berg says.⏎Think fat just hangs around and does nothing? It doesn’t⏎By Nanci Hellmich, USA TODAY⏎PHOENIX — Most people think of

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:onononononononononononononononononononononononononononononononononononononononononononononononon

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

    otes (sweet potatoes), jicama (a tuber also known as sengkwang, yam bean, singkamas, Mexican turnip), amaranth (also called pigweeds) and mucuma (a tropical leg

  2. #2 peak

    notice him leaving the party early?► see thesaurus at seeNotice is not used in the progressive. You say: I notice you’re wearing glasses. ✗Don’t say: I’m noticing you’re wearing glasses.2 → be

  3. #3 peak

    the environmental and social risks of increased oil and gas production. Despite the heated debate, neither side has a monopoly on the facts. In this book, Daniel Raimi gives a balanced and accessible view of oil and gas development, clearly and thoroughly

  4. #4 peak

    .⏎Free homework help from online tutors⏎Every Day 1-10 p.m.<|endoftext|>Contact: Kara Villamil, or Mona S. Rowe⏎FOR IMMEDIATE RELEASE⏎DECEMBER 3, 1997⏎Images

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. L23:270 strongly excites
  2. L19:334 suppresses
  3. L20:2203 suppresses
  4. L24:1190 excites
  5. L21:1679 excites
  6. L22:1138 suppresses
  7. L17:230 excites
  8. L18:553 excites
  9. L1:16 excites
  10. L2:1563 excites
  11. L16:73 excites
  12. L0:73 excites
  13. L1:35 suppresses
  14. L26:550 suppresses
  15. L26:2511 suppresses
  16. L8:250 weakly suppresses
  17. L8:14 weakly excites
  18. L4:1372 weakly suppresses
  19. L0:741 weakly excites
  20. L16:66 weakly suppresses
  21. L14:95 weakly excites
  22. L7:922 weakly suppresses
  23. L7:291 weakly excites
  24. L10:311 weakly suppresses

Feeds later layers

None: this is the last layer.

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 -16.03 and -13.842 values between -11.65 and -9.451 values between -9.45 and -7.262 values between -5.07 and -2.8732 values between -2.87 and -0.683852 values between -0.68 and 1.518 values between 1.51 and 3.712 values between 5.90 and 8.092 values between 8.09 and 10.282 values between 10.28 and 12.481 values between 16.86 and 19.061 values between 19.06 and 21.252 values between 69.50 and 71.69offnegativepositiveoften

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. ��
  2. ever
  3. ee
  4. (one
  5. 一头
  6. Matter
  7. eh
  8. 一团
  9. matter
  10. est

Suppresses

  1. inis
  2. inus
  3. inem
  4. ent
  5. on
  6. iss
  7. in
  8. anos
  9. an
  10. utation