Neuron Explorer

Qwen3-0.6B

'the' after 'like' or 'stay'

A neuron in layer 7 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 tends to fire on the word 'the' when it comes after 'like' in comparisons, such as 'like the grocery store'. Its strongest negative cases involve 'and' between superlatives. Turning it up gave repetitive, wandering text.

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

    hundreds in a puritan society. In his novel,The Scarlet Letter, Nath<|endoftext|>What they learned after only seven years is worth remembering (or learning, as the case may be).⏎The United States has operated under two constitutions. The first

  2. #2 peak

    large telescope or sending a spacecraft to another planet, for example, requires policymakers to embrace a common goal and to stay the course for many years, shielding scientific workers from the shifting whims of politicians.⏎The

  3. #3 peak

    Cooper.⏎Buy the hardcover edition of⏎I Heard the Willow Weep for $6.00. (Retail price: $1

  4. #4 peak

    But now that she is a reader, surround her with print. Get a newspaper delivered. Get her a library card and make the library a regular stop, like the grocery store and the dry cleaner. And get over your view of what "proper

  5. #5 peak

    landed on the “dark side” of the moon! The “dark side” … like the evil side? Nah—it’s actually the South Pole-Aitken Basin area of

  6. #6 peak

    One of the Greatest Forces⏎The power of a successfully communicated thought, from one human mind to another, is one of the greatest forces we know. But like the tango, it takes two to communicate

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: the most commonestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestestest

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

    Amazon.com.<|endoftext|>The world’s seabeds are littered with wrecks and treasures. Despite many decades of increasingly active searches for the best, the rarest and the most valuable of them

  2. #2 peak

    class on code.org, jump into CircuitPython to learn Python and hardware together, TinyGO, or even use the Arduino IDE. Circuit Playground Express is the newest and best Circuit

  3. #3 peak

    the party has not completely abandoned state planning and nationalisation. But less has changed politically. The party - based on the Soviet model - runs the country in much the same way it has always done.⏎What about the future?⏎Some believe the opening

  4. #4 peak

    writing career, Vladimir Nabokov was frustrated by both the Freudian and behavorialist schools of psychology that reigned in his day, seeking a richer and more complex view of the mind than either provided. In a long, incis

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. L5:2787 strongly excites 'and' in 'more and more' pairs
  2. L4:1727 weakly excites
  3. L2:473 weakly excites
  4. L4:682 weakly suppresses
  5. L4:344 weakly excites
  6. L2:2182 weakly excites
  7. L2:70 weakly excites
  8. L5:158 weakly suppresses
  9. L1:2801 weakly excites
  10. L6:16 weakly excites
  11. L5:2844 weakly excites
  12. L4:801 weakly suppresses
  13. L0:2842 weakly suppresses
  14. L0:960 weakly excites
  15. L5:225 weakly excites
  16. L6:1004 weakly suppresses
  17. L2:3065 weakly excites
  18. L5:2345 weakly excites
  19. L3:2071 weakly excites
  20. L1:2032 weakly excites
  21. L4:2270 weakly excites
  22. L1:1809 weakly suppresses
  23. L6:107 weakly suppresses 'organisms' and 'Code' in reference contexts
  24. L6:1542 weakly excites

Feeds later layers

  1. L10:94 strongly suppresses
  2. L19:395 suppresses
  3. L17:446 suppresses
  4. L8:423 suppresses
  5. L16:286 suppresses
  6. L8:71 suppresses
  7. L22:969 suppresses
  8. L16:592 suppresses
  9. L15:2378 suppresses
  10. L24:118 suppresses
  11. L16:1231 suppresses
  12. L16:283 suppresses
  13. L9:929 suppresses 'each' in 'each other'
  14. L21:327 suppresses
  15. L16:415 suppresses
  16. L13:74 suppresses
  17. L13:2580 suppresses
  18. L16:582 suppresses
  19. L14:713 suppresses
  20. L8:1032 suppresses
  21. L15:1470 suppresses
  22. L27:2557 suppresses
  23. L25:2986 suppresses '1' in 21st or 11th, before 'st'
  24. L20:944 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 -1.84 and -1.741 values between -1.74 and -1.641 values between -1.43 and -1.331 values between -1.23 and -1.122 values between -1.12 and -1.022 values between -1.02 and -0.921 values between -0.92 and -0.822 values between -0.82 and -0.716 values between -0.71 and -0.615 values between -0.61 and -0.5116 values between -0.51 and -0.4135 values between -0.41 and -0.30127 values between -0.30 and -0.20500 values between -0.20 and -0.101220 values between -0.10 and 0.011122 values between 0.01 and 0.11543 values between 0.11 and 0.21187 values between 0.21 and 0.3171 values between 0.31 and 0.4230 values between 0.42 and 0.5215 values between 0.52 and 0.627 values between 0.62 and 0.722 values between 0.72 and 0.831 values between 0.83 and 0.932 values between 1.03 and 1.131 values between 1.13 and 1.241 values between 1.24 and 1.341 values between 1.44 and 1.541 values between 1.54 and 1.652 values between 1.85 and 1.951 values between 1.95 and 2.061 values between 2.16 and 2.26offnegativepositiveoften

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. heels
  2. SPDX
  3. voi
  4. però
  5. vu
  6. vero
  7. HostException
  8. võ
  9. nik
  10. уш

Suppresses

  1. est
  2. 最快
  3. er
  4. 了一个
  5. 级以上
  6. 更多
  7. more
  8. ar
  9. ­s
  10. ingly