Neuron Explorer

Qwen3-0.6B

word before a polite or emotional phrase

A neuron in layer 9 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 small connecting words that come right before a set phrase, like 'sorry for', 'thanks for', 'proud to' and 'take heart'. It seems to care about the words leading into an apology, thanks or feeling. Turning it up did not make the writing emotional, so the pattern is hard to confirm from the outputs.

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

    code there is easily understandable though...⏎January 10th, 2004 01:32 AM⏎Tim_axe, sorry to be pedantic, but if you look at his code, x would never be zero

  2. #2 peak

    America's working men and women by setting and enforcing standards, and providing training, education and assistance. For more information, visit www.osha.gov.<|endoftext|>End of the year math review for 2nd grade can be FUN! 2nd

  3. #3 peak

    pressure. As the resulting jets of gas shoot backward, the engine and its craft are thrust forward. However, showing you is much cooler than explaining it.⏎Thanks for Subscribing!⏎You'll need to confirm your email, so please check your

  4. #4 peak

    for college and career.⏎The statistics are sobering, but take heart! There are steps you can take to help prepare your teens for college.⏎College

  5. #5 peak

    elacanth specimen was captured by fishermen in 1938. This inflicted a terrible disappointment on evolutionists. James Leonard Brierley Smith, an instructor in the Rhodes

  6. #6 peak

    to their final size on the map. I'm afraid I was unable to find the original author of the tutorial I used to create these planets, so I apologise for not being able to offer credit where it is due.⏎Make a new image with

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 make a point that: (1) the value of the expression is that the number of the expression is that the value of the expression is that the number of the expression is that the value of the expression is that the number

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

    partners, explore the possible health and economic impacts of using pricing strategies to promote consumption of healthy foods and beverages and discourage consumption of unhealthy products.<|endoftext|>ADS BY GOOGLE⏎In grammarFrom a grammatical point of view, superlative is a type

  2. #2 peak

    <|endoftext|>CONSUMER ADVISORY: Motorists Should 'Share the Road' With Motorcyclists⏎Friday, May

  3. #3 peak

    designer, one of the greatest names in French fashion in the 20th century. In 1985, Caroline Rennolds Milbank wrote, "The most consistently celebrated and influential designer of the past twenty-five years, Y

  4. #4 peak

    . Because it’s in your mind. This important tool is a<|endoftext|>Job Description: To Explore the Unknown⏎Curiosity about how life might have evolved on Mars could

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. L5:436 suppresses
  2. L7:250 excites
  3. L8:235 suppresses
  4. L8:308 excites
  5. L8:937 excites
  6. L8:80 excites
  7. L7:362 suppresses
  8. L7:792 excites
  9. L8:2469 excites
  10. L8:731 weakly excites
  11. L6:150 weakly suppresses
  12. L8:494 weakly suppresses
  13. L8:157 weakly suppresses
  14. L7:260 weakly suppresses
  15. L7:4 weakly excites
  16. L4:366 weakly suppresses
  17. L8:322 weakly excites
  18. L8:2343 weakly suppresses
  19. L6:359 weakly excites
  20. L3:767 weakly suppresses
  21. L7:1420 weakly excites
  22. L7:301 weakly excites
  23. L6:63 weakly suppresses
  24. L7:506 weakly suppresses

Feeds later layers

  1. L10:936 suppresses
  2. L14:438 suppresses
  3. L10:1710 excites
  4. L15:3048 excites
  5. L11:1144 excites
  6. L13:1470 weakly suppresses
  7. L14:591 weakly excites
  8. L10:405 weakly suppresses
  9. L11:1197 weakly excites
  10. L14:2286 weakly suppresses
  11. L13:57 weakly suppresses
  12. L12:85 weakly suppresses
  13. L14:1268 weakly excites
  14. L14:2128 weakly suppresses
  15. L10:725 weakly suppresses
  16. L10:2676 weakly suppresses
  17. L21:1234 weakly excites
  18. L14:1743 weakly suppresses
  19. L12:1072 weakly suppresses
  20. L10:1132 weakly excites
  21. L10:2006 weakly suppresses
  22. L15:3051 weakly excites
  23. L10:2418 weakly suppresses
  24. L12:1338 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 -2.27 and -2.172 values between -2.17 and -2.073 values between -1.69 and -1.592 values between -1.59 and -1.504 values between -1.50 and -1.402 values between -1.40 and -1.306 values between -1.30 and -1.214 values between -1.21 and -1.117 values between -1.11 and -1.027 values between -1.02 and -0.9212 values between -0.92 and -0.8211 values between -0.82 and -0.7322 values between -0.73 and -0.6318 values between -0.63 and -0.5437 values between -0.54 and -0.4491 values between -0.44 and -0.34150 values between -0.34 and -0.25367 values between -0.25 and -0.15620 values between -0.15 and -0.061343 values between -0.06 and 0.04602 values between 0.04 and 0.14282 values between 0.14 and 0.23151 values between 0.23 and 0.3367 values between 0.33 and 0.4328 values between 0.43 and 0.5223 values between 0.52 and 0.6212 values between 0.62 and 0.718 values between 0.71 and 0.819 values between 0.81 and 0.916 values between 0.91 and 1.002 values between 1.00 and 1.102 values between 1.10 and 1.202 values between 1.29 and 1.392 values between 1.39 and 1.482 values between 1.48 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. ми
  2. ptron
  3. deo
  4. สเต
  5. EK
  6. 하겠습니다
  7. #${
  8. comes
  9. theless
  10. #%

Suppresses

  1. opper
  2. olución
  3. atches
  4. aque
  5. сыл
  6. çi
  7. atement
  8. Symbols
  9. accomplish
  10. accompl