Neuron Explorer

Qwen3-0.6B

ampersand inside names and abbreviations

A neuron in layer 23 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 "&" sign when it sits inside a name or abbreviation, like P&G, F&V or D&RGW. Turning it down pushed the model toward writing blank fill-in-the-line underscores.

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

    . It is a free software. We also strongly recommend that you sign up for an AdobeID at the Adobe website. For more details please see FAQ 1&2. To view this ebook on an iPhone, iPad or Android mobile device you

  2. #2 peak

    . It is a free software. We also strongly recommend that you sign up for an AdobeID at the Adobe website. For more details please see FAQ 1&2. To view this ebook

  3. #3 peak

    victor. How true, the losers are not around to defend their position.⏎In my early days at P&G, we had a saying, if there were three people walking up a hill

  4. #4 peak

    cost of repairing and upgrading the water supply infrastructure is estimated to be approximately R400-million (US$52.8 million).⏎World Cruise on P&O⏎Pass

  5. #5 peak

    70s. These writings are from publications by people outside of industry, mostly college professors. Their perspective is quite different than mine. I started working at P&G in 1951, inside industry. At that time, there

  6. #6 peak

    ) as an environmental indicator. These studies principally showed that meat and dairy are among the largest contributors to GHGE; whereas high intake of fruits, vegetables (F&V), and legumes/pulses/nuts consumption is associated with the lowest GH

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: how to view_ a___________________________________________

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

    identifying organisms that uses two words to name the species is.⏎Their habitat and physical differences⏎In Aristoles system of classification, animals were classified on the basis of ____________________.⏎The greater the number of taxa two organisms have in common, the __________________

  2. #2 peak

    . What have you found?⏎2. 1/4 turn clockwise, forward 2, 1/4 turn anti-clockwise and then forward 3 _______.⏎3. Forward 2, 1/4 turn clockwise, forward

  3. #3 peak

    things? No! Your flexible brain is so good at making sense of patterns that it can see two images in one image.⏎The image at left is both a __________________ and __________________.⏎Your brain also solves mysteries in language. Imagine you find the crypt

  4. #4 peak

    0 in another(),so this prints 20. Thus final answer is 10 0 20. I hope i have been clear enough. ------------------ Come on in !! Drinks are on the house in the Big Moose Saloon

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. L19:1654 excites
  2. L8:7 excites
  3. L21:195 excites
  4. L18:267 excites
  5. L6:7 suppresses
  6. L8:14 excites
  7. L10:33 suppresses
  8. L21:173 excites
  9. L1:8 excites
  10. L20:603 excites
  11. L17:38 weakly suppresses
  12. L20:1664 weakly excites
  13. L6:16 weakly excites
  14. L5:110 weakly suppresses
  15. L21:189 weakly suppresses document break token mid-sentence
  16. L8:10 weakly excites
  17. L16:170 weakly suppresses
  18. L6:13 weakly excites
  19. L8:2 weakly excites
  20. L3:187 weakly excites
  21. L0:2776 weakly suppresses
  22. L18:48 weakly suppresses
  23. L9:7 weakly excites
  24. L18:887 weakly excites

Feeds later layers

  1. L25:292 suppresses
  2. L27:432 excites
  3. L27:635 suppresses
  4. L26:98 suppresses
  5. L26:396 weakly suppresses
  6. L27:2159 weakly suppresses
  7. L24:206 weakly excites
  8. L24:138 weakly excites
  9. L26:575 weakly suppresses
  10. L25:1986 weakly suppresses
  11. L27:802 weakly suppresses
  12. L26:524 weakly suppresses
  13. L26:1750 weakly suppresses
  14. L26:2975 weakly excites
  15. L26:313 weakly excites
  16. L24:2740 weakly suppresses
  17. L27:1373 weakly suppresses
  18. L27:1692 weakly suppresses
  19. L24:659 weakly suppresses
  20. L27:2046 weakly suppresses
  21. L24:1662 weakly excites
  22. L25:161 weakly excites
  23. L27:921 weakly excites
  24. L26:1534 weakly 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 -6.61 and -6.022 values between -4.27 and -3.691 values between -3.69 and -3.113 values between -3.11 and -2.521 values between -2.52 and -1.941 values between -1.94 and -1.3615 values between -1.36 and -0.77170 values between -0.77 and -0.193605 values between -0.19 and 0.3999 values between 0.39 and 0.984 values between 0.98 and 1.561 values between 1.56 and 2.141 values between 2.14 and 2.721 values between 2.72 and 3.311 values between 3.31 and 3.891 values between 3.89 and 4.471 values between 16.14 and 16.72offnegativepositiveoften

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. dilig
  2. *,⏎
  3. *"
  4. **⏎
  5. *,
  6. *&
  7. isations
  8. ―
  9. -----------
  10. !");⏎

Suppresses

  1. _
  2. &
  3. #
  4. amp
  5. //
  6. ype
  7. o
  8. _VIEW
  9. hotmail
  10. &amp