Neuron Explorer

Qwen3-0.6B

the word 'long' in phrases

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 lights up on the word 'long' in set phrases like 'a long way', 'long ago', 'long before' and 'long-haired'. Turning it up made the model write more about long periods of time, so it seems tied to the idea of duration.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    of the pelvic floor and the muscles that support the bladder neck and vagina and give these organs tone and strength. A strong healthy set of pelvic floor muscles goes a long way to preventing urinary stress incontinence and pelvic floor prolapse.⏎There is

  2. #2 peak

    still farmland back then, a good six miles north of teeming downtown and this was long before public transit. Maps of New York City as late as 184

  3. #3 peak

    this important message!<|endoftext|>Forty-eight years ago, an event occurred that shattered the American psyche and permanently altered society as we know it.⏎The Beatles, a long-haired quartet from Liverpool, England, appeared on the Ed Sullivan show in New

  4. #4 peak

    dwelling houses and field systems. This was in days when there were still wolves in the British Isles, long before the Romans introduced rabbits and horse chestnut trees and in a time when there

  5. #5 peak

    more premature deaths each year than automobile accidents, which was long considered the leading cause. According to the study, particulate air pollution causes at

  6. #6 peak

    necessary context for understanding. But it’s the strange bits of information that can really capture a person’s curiosity and help us relate to the people that lived long ago. They’re also the most fun things to share with our guests and get

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 want to go to a “first step” level, which is the minimum level of “full” “swaption” offering, which is the first “step” of “P0” or “P0” level. But I

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

    order to resolve the dispute without the use of force. The rebels, who hail from the autonomous island province of Sulu in the southwestern Philippines, had been given until midnight on Tuesday

  2. #2 peak

    ), Part 4, Chapter 7, eBook.<|endoftext|>Transcript for A New Horizon, segment 05 of 13⏎The Eastern

  3. #3 peak

    an infected tick. Discover the latest research on babesiosis here.⏎Transcriptional analysis of p30 major outer membrane protein genes of Ehrlichia

  4. #4 peak

    1–2. Sign up to view the full content.⏎This preview has intentionally blurred sections. Sign up to view the full version.View Full Document⏎Unformatted text preview: 60 | | | | CHAPTER 1 FUNCTIONS AND MODELS

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. L10:852 strongly excites
  2. L4:1978 excites
  3. L5:2290 excites
  4. L17:1608 excites
  5. L16:155 excites
  6. L8:918 excites
  7. L17:509 suppresses
  8. L10:273 suppresses
  9. L12:1703 excites
  10. L5:2731 weakly suppresses
  11. L17:378 weakly suppresses
  12. L10:937 weakly suppresses
  13. L1:2130 weakly suppresses
  14. L7:1101 weakly suppresses
  15. L1:27 weakly excites
  16. L2:2166 weakly suppresses 'in' after high or low
  17. L9:515 weakly suppresses
  18. L6:3028 weakly suppresses
  19. L10:1502 weakly excites
  20. L17:1445 weakly excites
  21. L17:725 weakly excites
  22. L4:2251 weakly excites
  23. L9:2572 weakly suppresses
  24. L5:1093 weakly excites

Feeds later layers

  1. L19:1712 strongly excites
  2. L22:2698 excites 'long' in 'in the long run'
  3. L24:712 excites 'long' in 'long-term' and 'long-lasting'
  4. L20:2982 excites
  5. L27:1469 excites
  6. L22:1699 excites
  7. L22:1465 excites
  8. L19:1768 suppresses
  9. L19:2901 excites
  10. L20:2368 weakly suppresses
  11. L20:1736 weakly excites
  12. L25:2988 weakly excites
  13. L19:2850 weakly excites
  14. L23:1490 weakly excites
  15. L22:1915 weakly suppresses
  16. L21:2289 weakly excites
  17. L21:2818 weakly excites
  18. L22:2755 weakly suppresses
  19. L20:1163 weakly suppresses
  20. L22:622 weakly excites
  21. L27:2815 weakly suppresses
  22. L24:1809 weakly excites
  23. L26:2136 weakly excites
  24. L22:1893 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.

2 values between -2.94 and -2.0840 values between -2.08 and -1.231573 values between -1.23 and -0.382209 values between -0.38 and 0.4732 values between 0.47 and 1.3314 values between 1.33 and 2.189 values between 2.18 and 3.039 values between 3.03 and 3.882 values between 3.88 and 4.748 values between 4.74 and 5.593 values between 5.59 and 6.444 values between 6.44 and 7.291 values between 7.29 and 8.141 values between 9.85 and 10.701 values between 30.30 and 31.16offnegativepositiveoften

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. -lasting
  2. dài
  3. 很长
  4. lasting
  5. lasts
  6. lengths
  7. lasted
  8. 最长
  9. length
  10. طول

Suppresses

  1. %\
  2. #{
  3. irates
  4. matter
  5. -degree
  6. ula
  7. rade
  8. gression
  9. itals
  10. >}</