Neuron Explorer

Qwen3-0.6B

names in website boilerplate

A neuron in layer 17 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on names and product words that appear in website footers, copyright lines and page blurbs, such as 'collections' in a lesson-planning site ad, 'Foundation' in a software licence, or 'Smith' in a company notice. Turned up, the model did not write about these, so its effect on the output is unclear.

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

    Free Trial to Unlock the Full Review⏎Why Lesson Planet?⏎Find quality lesson planning resources, fast!⏎Share & remix collections to collaborate.⏎Organize your curriculum with collections. Easy!⏎Have time to be more creative & energetic with your students!⏎Students

  2. #2 peak

    Free Trial to Unlock the Full Review⏎Why Lesson Planet?⏎Find quality lesson planning resources, fast!⏎Share & remix collections to collaborate.⏎Organize your curriculum with collections. Easy!⏎Have time to be more creative & energetic with your students!⏎Students

  3. #3 peak

    Why Lesson Planet?⏎Find quality lesson planning resources, fast!⏎Share & remix collections to collaborate.⏎Organize your curriculum with collections. Easy!⏎Have time to be

  4. #4 peak

    family from this group with four members, which is the average family size in the<|endoftext|>Why Lesson Planet?⏎Find quality lesson planning resources, fast!⏎Share & remix collections to collaborate.⏎Organize your curriculum with collections. Easy!⏎Have time to be

  5. #5 peak

    copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, no Front-Cover Texts, and no

  6. #6 peak

    . Prices and availability in WHSmith Stores may vary significantly⏎© Copyright 2013 - 2017 WHSmith and its suppliers.⏎WHSmith High Street Limited Greenbridge Road, Swindon, Wiltshire, United

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 a, I have a, I have a, I have a, I have a, I have a, I have a, I have a, I have a, I have a, I have a, I have a,

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

    of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither BioOne nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied

  2. #2 peak

    |Math Topics:||Graphing of Data|⏎© 1994- Drexel University. All rights reserved.⏎The Math Forum is a research and educational enterprise of the Drexel University School of Education.<|endoftext|>Maybe you’ve

  3. #3 peak

    problems, and the authors also include methods that allow for the solving of nonmathematical problems as mathematical problems.⏎To view this DRM protected ebook on your desktop or laptop you will need to have Adobe Digital Editions installed. It is a free

  4. #4 peak

    nothing — not even a conservative culture or fear of the Taliban<|endoftext|>This preview shows pages 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⏎Un

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. L11:29 excites
  2. L15:1317 excites
  3. L16:248 excites
  4. L15:23 suppresses
  5. L10:61 suppresses
  6. L15:1272 suppresses
  7. L7:22 suppresses
  8. L13:129 suppresses
  9. L14:1641 excites
  10. L16:1543 suppresses
  11. L13:2963 suppresses
  12. L13:2188 excites unclear: weak, scattered tokens
  13. L16:2477 excites
  14. L10:812 excites
  15. L10:2957 excites
  16. L10:1578 excites
  17. L14:932 excites
  18. L15:2628 excites
  19. L15:2839 excites
  20. L10:46 suppresses
  21. L15:1795 suppresses
  22. L10:124 suppresses
  23. L14:2164 excites
  24. L15:1295 weakly excites

Feeds later layers

  1. L26:1082 suppresses
  2. L25:3 weakly suppresses
  3. L23:2606 weakly excites
  4. L19:772 weakly excites
  5. L21:2551 weakly excites
  6. L22:1799 weakly excites
  7. L25:509 weakly excites
  8. L18:348 weakly excites
  9. L25:2769 weakly excites
  10. L23:210 weakly excites verb right before its object
  11. L22:1940 weakly excites
  12. L21:1989 weakly excites
  13. L22:157 weakly excites
  14. L24:716 weakly excites
  15. L25:2354 weakly excites
  16. L23:124 weakly suppresses
  17. L22:2639 weakly suppresses
  18. L25:2606 weakly excites
  19. L25:1453 weakly excites
  20. L22:1822 weakly excites
  21. L20:543 weakly excites
  22. L18:2616 weakly excites
  23. L25:697 weakly excites
  24. L24:1121 weakly excites

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 -4.76 and -4.601 values between -4.29 and -4.131 values between -3.97 and -3.812 values between -3.18 and -3.021 values between -2.70 and -2.542 values between -2.23 and -2.071 values between -2.07 and -1.911 values between -1.75 and -1.591 values between -1.59 and -1.432 values between -1.27 and -1.121 values between -1.12 and -0.965 values between -0.96 and -0.8013 values between -0.80 and -0.6446 values between -0.64 and -0.48192 values between -0.48 and -0.32387 values between -0.32 and -0.16658 values between -0.16 and -0.01723 values between -0.01 and 0.15672 values between 0.15 and 0.31550 values between 0.31 and 0.47362 values between 0.47 and 0.63194 values between 0.63 and 0.7958 values between 0.79 and 0.9528 values between 0.95 and 1.104 values between 1.10 and 1.261 values between 1.26 and 1.421 values between 1.42 and 1.58offnegativepositiveoften

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. nqu
  2. lup
  3. уш
  4. rophe
  5. عون
  6. omap
  7. Lup
  8. createQuery
  9. MATRIX
  10. APH

Suppresses

  1. side
  2. ışı
  3. gibi
  4. mé
  5. ın
  6. bin
  7. сбор
  8. mercial
  9. ž
  10. económica