Neuron Explorer

Qwen3-0.6B

the "un" prefix splitting a word

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 letters "un" when they start a word stem that continues after them, like "unearned" or "unspoken" split off from the rest of the word. Turned up, the model got stuck repeating fragments like "ly" and "out of".

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

    with 100% plagiarism free content. We prepare quality content and notes for Characteristics of Linear Programming topic under Operations Management theory and study material. These are avail for subscribed users and

  2. #2 peak

    anything but an asset. Cash is a current asset, and serves to increase the net worth of whoever is in possession of the cash. But when the cash is unearned, or you haven't performed the service or provided the product that the cash

  3. #3 peak

    this new field are also valid. Some community nervousness is certainly justified since “these technologies offer the potential for great benefit, [but] it is not benefit unalloyed.”12⏎However, banning research is also unlikely to be

  4. #4 peak

    (Mt. 12: 41).<|endoftext|>“There shall never be any bond slaverie, villinage or Captivitie amongst us unles it be lawfull Captives taken in just warres, and such strangers

  5. #5 peak

    as an energy source. Researchers had believed that keeping the donor and acceptor layers as pure as possible was the best way to ensure that the excitons could travel unimpeded, so that solar cells could capture the maximum amount of energy.⏎NC

  6. #6 peak

    Too Much”⏎Some people feel more than others.⏎They are often told—spoken and unspoken—that they are ‘too much,’ ‘too intense,’ ‘too sensitive,’

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: what can be in outfit同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用同用

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

    the food we consume which regularly introduces new elements into this already-unstable milieu.⏎Disruption to gut flora is known to have an adverse effect on

  2. #2 peak

    windows provide natural daylight, and the sun's rays help keep rooms warmer in winter. If you want more light, add insulated windows to the home's south side-unless you live in the desert or an extremely hot

  3. #3 peak

    -only models.⏎But why would any school board adopt such an obviously flawed sex education program?⏎For one, American society has an extensive history of supporting abstinence-only-until-marriage (AOUM) programming. This is the result of

  4. #4 peak

    , breathing exercises, pranayama (breath control), and self-control. When you have almost any queries concerning wherever along

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. L16:3 suppresses
  2. L2:761 suppresses
  3. L20:61 suppresses
  4. L20:212 excites
  5. L4:216 excites
  6. L20:1414 excites
  7. L16:160 excites
  8. L10:431 excites
  9. L4:151 excites
  10. L18:853 suppresses
  11. L19:118 excites
  12. L17:391 excites
  13. L18:1104 suppresses
  14. L1:259 suppresses
  15. L21:625 excites
  16. L0:1168 suppresses
  17. L22:271 suppresses
  18. L5:499 excites
  19. L16:208 suppresses
  20. L3:459 suppresses
  21. L2:238 excites
  22. L16:48 excites
  23. L18:127 excites
  24. L2:598 excites

Feeds later layers

  1. L27:2661 weakly suppresses
  2. L27:185 weakly excites
  3. L27:668 weakly suppresses
  4. L27:2770 weakly suppresses
  5. L27:744 weakly excites
  6. L27:919 weakly suppresses
  7. L27:2589 weakly suppresses
  8. L27:765 weakly excites
  9. L27:1630 weakly suppresses
  10. L27:1815 weakly suppresses
  11. L27:1737 weakly excites
  12. L27:2009 weakly suppresses
  13. L27:2640 weakly suppresses
  14. L27:925 weakly excites
  15. L27:698 weakly suppresses
  16. L27:847 weakly suppresses
  17. L27:1384 weakly suppresses
  18. L27:538 weakly suppresses
  19. L27:850 weakly suppresses
  20. L27:453 weakly suppresses
  21. L27:1525 weakly suppresses
  22. L27:2897 weakly suppresses
  23. L27:638 weakly excites
  24. L27:699 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.

1 values between -34.97 and -31.241 values between -23.79 and -20.061 values between -20.06 and -16.333 values between -16.33 and -12.612 values between -12.61 and -8.883 values between -8.88 and -5.1519 values between -5.15 and -1.423847 values between -1.42 and 2.319 values between 2.31 and 6.035 values between 6.03 and 9.761 values between 9.76 and 13.495 values between 13.49 and 17.213 values between 17.21 and 20.941 values between 20.94 and 24.672 values between 28.40 and 32.121 values between 32.12 and 35.851 values between 58.22 and 61.941 values between 61.94 and 65.671 values between 65.67 and 69.401 values between 110.40 and 114.13offnegativepositiveoften

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. ly
  2. gli
  3. ibling
  4. ughter
  5. ungle
  6. imal
  7. (rawValue
  8. apor
  9. теперь
  10. ive

Suppresses

  1. 可
  2. 会让
  3. 用
  4. 懂
  5. 同
  6. 知
  7. 出
  8. 会对
  9. linear
  10. 会