Neuron Explorer

Qwen3-0.6B

comma after 'and' or 'and then' clause

A neuron in layer 18 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 a comma that comes right after a joining word like 'and' or 'and then', where a new clause is about to start. Turning it up made the writing drift into repetitive, loosely connected sentences.

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

    ,” we get the subject form (An Luan), and, by following genitive case rules, we get the preposition form (Dé Lu

  2. #2 peak

    in protecting the courthouse.⏎Thereafter, they get an opportunity to learn about the legal aspects of the court system from a judge who explains how trials work. And, then, when they least expect it, they get to experience a trial themselves –

  3. #3 peak

    is time to stop reading my words, bundle up the kids and hike into the Arctic. Why not today! Children will almost certainly see seals just yards away and, if the timing is right, go nose to nose with a mighty polar bear that

  4. #4 peak

    .⏎- In the case of Col. Shaw, it seems that he chose to live his life a certain way (leading his African American troops into battle) and, now that he's dead, he's off on the sidelines, checking out all

  5. #5 peak

    emitted by light sources attract insects. A study was conducted on moths. It was observed that, only the males of the species got attracted towards the light source, mistaking it

  6. #6 peak

    the early 1990’s, rats were fed GM tomatoes. Actually, they refused to eat them, so they had to be force fed. And, rats aren’t the only animals who’ve declined a snack of GMOs. Smith

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: "I am" is a good one to be considered a "good" item to be considered a "good" item is good" is a good item" is "good" is "good" is "good" is "good"

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

    risk of strep throat infection:⏎- Young age. Strep throat occurs most commonly in children.⏎- Time of year. Although strep throat can occur anytime,<|endoftext|>The dangers of feeding lorikeets⏎Lorikeet feeding habits⏎

  2. #2 peak

    premise had yet to be tested. John enlisted coleopterist Peter Kovarik as a partner in this study. The two decided that in addition to butterflies, scarabaeine scarabs and hister beetles would become part of the

  3. #3 peak

    a pH greater than 5.2 or under drought conditions. The tubers are most susceptible, but stems and roots may occasionally be infected. No above-ground symptoms are apparent on infected

  4. #4 peak

    have any photos of this monument yet. Why don't you be the first to send us one?⏎If Google Street View is available, the image is from the best available vantage point looking, if possible, towards

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. L16:1223 excites
  2. L9:19 excites
  3. L8:511 excites
  4. L17:1028 excites
  5. L17:980 excites
  6. L13:796 weakly suppresses
  7. L14:1993 weakly excites
  8. L16:1886 weakly excites
  9. L17:639 weakly suppresses
  10. L15:403 weakly suppresses period inside an abbreviation
  11. L6:250 weakly excites
  12. L14:1487 weakly excites
  13. L16:1442 weakly suppresses
  14. L4:2942 weakly excites 'yet' and 'haven't' after have
  15. L17:990 weakly excites
  16. L12:1284 weakly suppresses
  17. L17:2684 weakly suppresses
  18. L16:1100 weakly excites
  19. L15:1379 weakly suppresses
  20. L15:2447 weakly suppresses
  21. L12:1344 weakly excites
  22. L16:489 weakly suppresses
  23. L16:714 weakly suppresses
  24. L14:1457 weakly suppresses

Feeds later layers

  1. L19:1590 excites
  2. L19:2098 excites
  3. L19:1463 excites
  4. L22:160 suppresses
  5. L21:140 suppresses
  6. L19:543 excites
  7. L22:132 excites
  8. L21:1594 excites
  9. L26:585 excites
  10. L25:277 excites
  11. L20:1245 excites
  12. L21:818 excites
  13. L19:31 weakly excites
  14. L20:1281 weakly suppresses
  15. L24:116 weakly suppresses
  16. L21:801 weakly excites
  17. L26:259 weakly suppresses
  18. L19:167 weakly excites
  19. L19:2946 weakly excites
  20. L19:398 weakly suppresses
  21. L23:1137 weakly excites
  22. L27:684 weakly excites
  23. L24:784 weakly excites
  24. L19:1765 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.

3 values between -7.82 and -6.963 values between -6.96 and -6.112 values between -6.11 and -5.264 values between -5.26 and -4.414 values between -4.41 and -3.5612 values between -3.56 and -2.7021 values between -2.70 and -1.8527 values between -1.85 and -1.00438 values between -1.00 and -0.153234 values between -0.15 and 0.7083 values between 0.70 and 1.5626 values between 1.56 and 2.4122 values between 2.41 and 3.268 values between 3.26 and 4.115 values between 4.11 and 4.965 values between 4.96 and 5.821 values between 6.67 and 7.521 values between 7.52 and 8.371 values between 8.37 and 9.231 values between 9.23 and 10.082 values between 10.08 and 10.931 values between 10.93 and 11.781 values between 11.78 and 12.631 values between 13.49 and 14.341 values between 14.34 and 15.191 values between 25.41 and 26.27offnegativepositiveoften

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. amid
  2. arend
  3. 辕
  4. Tonight
  5. aboard
  6. amidst
  7. imates
  8. вокруг
  9. usting
  10. Amid

Suppresses

  1. nào
  2. ampa
  3. pass
  4. pass
  5. trans
  6. 資
  7. reserved
  8. col
  9. cape
  10. odos