Neuron Explorer

Qwen3-0.6B

'ch' starting a word fragment

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 fires on the letters 'ch' when they begin the second half of a longer word, such as chandeliers, chivalrous or chieftain. When it was turned up, the model kept writing the word 'story', and when turned down it kept writing 'say'.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    Candles and devices using this oil-burning technique were the main means used to light homes, employed in candelabras, wall sconces, and elegant chandeliers.⏎It wasn't until the Victorian era that oil lamps improved, thanks

  2. #2 peak

    377. Edward, who loved knightly pursuits such as war, jousting, tournaments and hunting, surrounded himself with warriors, magnates, and chivalrous knights. The Hundred Years' War broke out as a result of a

  3. #3 peak

    !⏎2. Stop buying disposable plastic products⏎Single-use disposable plastic bottles are a big issue – whilst some people do reuse them for a limited time, more get chucked in the bin and head straight for our landfill.⏎There are lots of products

  4. #4 peak

    , Dutch and French.⏎New England: King Phillip’s War is the first supporting idea. * The New England Confederation put to the test during war with Indian chieftain King Phillip (Metacom)—Wampanoag Chief son

  5. #5 peak

    or "vain pledge," or thought to mean "having profitable pledges." The O'Boyles were chieftains in Donegal, ruling west Ulster with the O'Donn

  6. #6 peak

    silver plate, including crosses, chalices, and other spoils of the Church might be as much more.

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: say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say say

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

    you would like to include in your essay. Next, group related ideas together and make sure they are ordered effectively. Finally, label your ideas using main headings and sub

  2. #2 peak

    Testaments, by Robert Jamieson, A.R. Fausset and David Brown at sacred-texts.com⏎neh 3:1THE NAMES AND ORDER OF THEM THAT BUILDED THE WALL OF JERUSALEM. (

  3. #3 peak

    also include idioms & quotations that either use this word or its translations in them or use any of the related words in English or Urdu translations. These idioms or quotations can also be taken as

  4. #4 peak

    well, tense.⏎At the most fundamental level, there are three tenses: past tense, present tense and future tense. Each of these tenses has four sub-categories (or aspects) which better describe the tense used within the writing. These

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. L20:240 suppresses
  2. L18:1433 excites
  3. L19:1427 suppresses
  4. L24:109 suppresses
  5. L20:664 excites
  6. L19:2956 suppresses
  7. L22:939 suppresses
  8. L21:338 suppresses
  9. L19:2480 suppresses
  10. L22:260 excites
  11. L20:807 excites
  12. L18:2391 excites
  13. L22:204 excites
  14. L24:488 suppresses
  15. L18:1556 suppresses
  16. L20:2997 suppresses
  17. L19:2382 suppresses
  18. L21:2537 suppresses
  19. L23:119 excites
  20. L23:218 suppresses
  21. L23:1112 suppresses
  22. L19:2522 excites
  23. L19:592 suppresses
  24. L22:2089 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 -58.50 and -56.442 values between -37.94 and -35.881 values between -29.71 and -27.662 values between -23.54 and -21.497 values between -21.49 and -19.436 values between -19.43 and -17.385 values between -17.38 and -15.3210 values between -15.32 and -13.2610 values between -13.26 and -11.2115 values between -11.21 and -9.1533 values between -9.15 and -7.0949 values between -7.09 and -5.0484 values between -5.04 and -2.98271 values between -2.98 and -0.933136 values between -0.93 and 1.13158 values between 1.13 and 3.1967 values between 3.19 and 5.2421 values between 5.24 and 7.3011 values between 7.30 and 9.363 values between 9.36 and 11.414 values between 11.41 and 13.474 values between 13.47 and 15.532 values between 15.53 and 17.582 values between 17.58 and 19.641 values between 19.64 and 21.693 values between 21.69 and 23.75offnegativepositiveoften

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. story
  2. 忘
  3. aths
  4. story
  5. 文章
  6. 迹
  7. _story
  8. vive
  9. 空间
  10. TemplateName

Suppresses

  1. leg
  2. ям
  3. ~~
  4. say
  5. lü
  6. íc
  7. >Contact
  8. fare
  9. undry
  10. aux