Neuron Explorer

Qwen3-0.6B

"other" and contrast words

A neuron in layer 16 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires most on the word "other" when one thing is being set against another, as in "the other peaks" or "the other plants". Words like "lower" and "south" also trigger it when they mark a contrasting side. Turning it up did not bring out a clear theme in the continuations.

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

    , there is a rocky giant called Mt. San Gorgonio. It is 11,502 feet tall—a half a head higher than the other peaks flanking it in the San Bernardino range. By early spring, when most

  2. #2 peak

    pered by factors such as genetic differences, upbringing, and lifestyle choices. By studying identical twins, however, the researchers were able to control for genetics and limit the other factors, as the twins grew up in the same households and reported similar lifestyles.⏎

  3. #3 peak

    of the vertebrae. Nerves branch from the spinal cord, pass through openings (foraminae) between the vertebrae, and branch to the lower body.⏎Primary Medical Reviewer William H. Blahd,

  4. #4 peak

    a horticultural term—it’s sort of like “vegetable,” in that you can kind of call anything a vegetable—but palms are not at all like the other plants commonly referred to as trees. They don’t have wood, for one thing

  5. #5 peak

    . Some philosophers who reject this metaphysical view have swung in the other direction and rejected the idea of the self altogether. David Hume said that the

  6. #6 peak

    1.4 million light-years away from Earth, so far away that it floats freely in space, unperturbed by the Milky Way.⏎Unlike the other hobbit galaxies, Leo T includes both old and young stars. It also contains

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 have to be sure that I have a good example, like I have to have a good example, like I have to have a good example, like, for example, maybe, like, like, like, maybe, like, for

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

    s, and no Back-Cover Texts. A copy of the license is included in the page “GNU Free Documentation License”.⏎The copyright and license notices on this page only apply to

  2. #2 peak

    Awards program combines knowledge of STEM concepts with hands on activities.<|endoftext|>The sheer amount of gold in Fort Knox was so great that Auric Goldfinger, the villain of the 1964 James Bond film which carries his name, knew he

  3. #3 peak

    am presenting it here because it has been proven to work on many projects of all shapes and sizes. At the very least, your story should contain all of the elements described in the template. The story template looks like this:⏎<|endoftext|>What is a

  4. #4 peak

    For thousands of years, rubies were considered the gemstones of love, energy, passion, and power. In a class of its own

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. L11:875 strongly suppresses
  2. L10:1377 strongly suppresses
  3. L11:331 strongly excites
  4. L15:1848 strongly suppresses
  5. L4:73 strongly suppresses
  6. L10:1162 strongly excites
  7. L13:1234 strongly suppresses
  8. L9:662 strongly suppresses
  9. L10:120 strongly suppresses
  10. L4:49 strongly suppresses
  11. L11:1168 strongly excites
  12. L15:869 strongly excites
  13. L15:1611 strongly suppresses
  14. L5:225 excites
  15. L2:205 excites
  16. L15:888 excites
  17. L13:824 suppresses
  18. L9:24 excites "the" before a named organisation or title
  19. L7:155 excites
  20. L3:344 excites
  21. L6:426 suppresses
  22. L14:2498 suppresses
  23. L9:59 excites "a" meaning per (a day, a year)
  24. L9:2198 excites

Feeds later layers

  1. L18:1547 excites
  2. L17:1542 suppresses
  3. L18:641 excites
  4. L17:2190 excites 'the' starting an object phrase
  5. L17:2970 excites
  6. L17:1179 suppresses
  7. L17:188 suppresses 'a' before a noun phrase
  8. L17:1652 suppresses
  9. L17:30 excites
  10. L18:1084 suppresses
  11. L22:1522 excites
  12. L18:314 excites
  13. L17:2135 excites 'the' after 'of' and similar words
  14. L19:2227 excites
  15. L18:799 excites
  16. L19:2815 excites
  17. L21:614 excites
  18. L19:3016 excites
  19. L20:901 suppresses
  20. L17:1481 excites 'the' after 'some of' or 'most of'
  21. L21:1741 excites
  22. L19:1543 excites
  23. L19:2115 excites
  24. L18:171 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.

1 values between -5.68 and -5.371 values between -4.74 and -4.431 values between -4.11 and -3.801 values between -3.80 and -3.487 values between -3.48 and -3.174 values between -3.17 and -2.867 values between -2.86 and -2.5410 values between -2.54 and -2.2312 values between -2.23 and -1.9123 values between -1.91 and -1.6029 values between -1.60 and -1.2854 values between -1.28 and -0.97102 values between -0.97 and -0.66284 values between -0.66 and -0.341100 values between -0.34 and -0.031728 values between -0.03 and 0.29212 values between 0.29 and 0.60113 values between 0.60 and 0.9265 values between 0.92 and 1.2339 values between 1.23 and 1.5435 values between 1.54 and 1.8616 values between 1.86 and 2.1718 values between 2.17 and 2.4911 values between 2.49 and 2.8017 values between 2.80 and 3.124 values between 3.12 and 3.433 values between 3.43 and 3.742 values between 3.74 and 4.061 values between 4.06 and 4.371 values between 4.37 and 4.691 values between 4.69 and 5.002 values between 5.32 and 5.632 values between 5.94 and 6.262 values between 6.57 and 6.89offnegativepositiveoften

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. converse
  2. dens
  3. ost
  4. od
  5. oses
  6. adows
  7. ones
  8. oden
  9. 一種
  10. jaws

Suppresses

  1. ún
  2. ()>⏎
  3. pillar
  4. (QtGui
  5. stood
  6. batim
  7. .yang
  8. fighter
  9. ręcz
  10. urate