Neuron Explorer

Qwen3-0.6B

the 're' in self-reliant and similar

A neuron in layer 26 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 're' piece in words like self-reliant, self-renew, self-repair and recovery, where a hyphen or a word part comes before it. Its opposite side fires on commas in lists just before a final 'and'. Turning it up made the text fall into repeated 'and' and 'or' words.

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

    a, and hydrangea. Rejuvenation pruning is best done in late winter or early spring.<|endoftext|>Neural stem cells (NSCs) can self-renew and differentiate into neurons and glia. Transplanted NS

  2. #2 peak

    be noted that during the process of clearing the body of toxins, the liver is damaged by exposure to harmful substances. The liver therefore possesses the amazing ability to self-repair and regenerate of damaged tissues.⏎So as one can plainly see, proper liver

  3. #3 peak

    treating women and children.⏎Women can accomplish good work and I have observed they are always above men in their class. This is accounted for in that she is self-reliant. Women know they will meet opposition, and prepare for it; they

  4. #4 peak

    to environmentally relevant behaviors. This observation may inspire researchers to⏎more directly test predictions arising from current theoretical models,⏎gather data on actual behavior rather than relying on self-reports,

  5. #5 peak

    , they’re likely to be fully prepared for snack time but missing important papers or sports equipment. What’s the right thing to do?⏎Your desire to raise self-reliant children is fantastic. But there’s no doubt that passing the baton

  6. #6 peak

    is typically warmed simply by the heat of the students inside.⏎Ventilation is sized carefully at the correct low volume exclusively to provide appropriate fresh air. The central heat-recovery heat exchanger that makes the once-through air system efficient is the largest piece

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: @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

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 get things done and to keep the United States functioning properly.⏎The Three Branches⏎The three branches of the U.S. government include the executive, legislative, and judicial branches. Each one has its own set of abilities. The legislative branch

  2. #2 peak

    , men were believed to grow beards for different reasons than men do today. The three main reasons prehistoric men had beards were for warmth, intimidation, and to defend their face from hits. In Egyptian times, men would dye their

  3. #3 peak

    are created and make sure they comply with the Constitution. It’s easy to see how these groups must work to balance each other as they attempt to create, enforce, and interpret the laws

  4. #4 peak

    aid in keeping racism alive. Researches strongly concur that minorities are overrepresented in all 3 components of the Criminal Justice System; the police, the courts, and corrections. However, researchers are divided as to why this disproportional representation

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. L25:6 strongly suppresses
  2. L19:1245 strongly suppresses comma opening a list of parallel items
  3. L19:290 strongly excites
  4. L21:739 strongly suppresses 'both X and Y' pairing words
  5. L25:1270 strongly suppresses
  6. L20:2746 strongly suppresses
  7. L20:852 strongly excites
  8. L24:205 strongly suppresses
  9. L23:2137 strongly suppresses
  10. L18:1856 strongly excites
  11. L19:2416 strongly suppresses
  12. L21:2699 strongly suppresses
  13. L21:64 strongly suppresses
  14. L17:2529 strongly suppresses
  15. L20:2812 strongly excites
  16. L20:1009 strongly excites
  17. L20:560 strongly excites
  18. L19:628 strongly suppresses
  19. L19:1362 strongly excites
  20. L20:523 strongly excites commas before an elaboration
  21. L19:3055 strongly excites
  22. L18:1327 strongly excites
  23. L19:1092 strongly excites
  24. L20:2864 strongly suppresses

Feeds later layers

  1. L27:289 strongly excites
  2. L27:376 excites
  3. L27:265 excites
  4. L27:1902 suppresses
  5. L27:1698 excites
  6. L27:94 suppresses
  7. L27:1937 excites
  8. L27:258 weakly suppresses
  9. L27:2212 weakly excites
  10. L27:191 weakly excites
  11. L27:52 weakly suppresses
  12. L27:338 weakly excites
  13. L27:498 weakly excites
  14. L27:700 weakly suppresses
  15. L27:542 weakly excites
  16. L27:59 weakly suppresses full stops in running text
  17. L27:1290 weakly suppresses
  18. L27:1443 weakly suppresses
  19. L27:690 weakly excites
  20. L27:2349 weakly suppresses
  21. L27:425 weakly suppresses
  22. L27:467 weakly excites
  23. L27:568 weakly excites
  24. L27:255 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 -26.17 and -25.191 values between -20.30 and -19.321 values between -13.44 and -12.461 values between -11.48 and -10.502 values between -10.50 and -9.525 values between -9.52 and -8.542 values between -8.54 and -7.569 values between -7.56 and -6.584 values between -6.58 and -5.607 values between -5.60 and -4.6213 values between -4.62 and -3.6416 values between -3.64 and -2.6633 values between -2.66 and -1.69569 values between -1.69 and -0.711188 values between -0.71 and 0.27735 values between 0.27 and 1.25523 values between 1.25 and 2.23355 values between 2.23 and 3.21202 values between 3.21 and 4.19113 values between 4.19 and 5.1751 values between 5.17 and 6.1531 values between 6.15 and 7.1316 values between 7.13 and 8.1112 values between 8.11 and 9.093 values between 9.09 and 10.077 values between 10.07 and 11.055 values between 11.05 and 12.033 values between 12.03 and 13.01offnegativepositiveoften

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. and
  2. or
  3. 和
  4. and
  5. 或
  6. And
  7. And
  8. và
  9. 和技术
  10. -and

Suppresses

  1. ')">
  2. @return
  3. aille
  4. entirety
  5. %</
  6. emsp
  7. ')</
  8. nbsp
  9. &apos
  10. pomoc