Neuron Explorer

Qwen3-0.6B

end-of-document token

A neuron in layer 9 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 hidden marker that separates one web page or document from the next. Turning it up mostly produced rambling, unrelated text, so it does not seem to shape any clear topic.

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

    Chapter 6 Exercise 6.5, we have discussed the Pythagoras theorem. Other important topics related to the chapter are covered in the following exercises.<|endoftext|>What you will learn in this module⏎Understand different tasks that have to be

  2. #2 peak

    two different numbers cant have the same cube). Therefore, by Definition 1, is one-to-one....⏎View Full Document⏎- Spring '09<|endoftext|>part of speech: noun⏎definition 1: an upright line or pole with

  3. #3 peak

    allowed us to propose something that hasn´t yet been possible to study in depth and that now should be taken into serious consideration. Is genomic instability the cause of tum<|endoftext|>The role of the built environment in addressing climate change is clear as buildings account for

  4. #4 peak

    (e) William Sholes - Typewriter⏎(f) Louis Daguerre - Photography⏎- (a) Lewis Waterman⏎(b) Lazlo Biro<|endoftext|>We are

  5. #5 peak

    ions iv. Special Limits c. Will be seen later d. If none of the top three apply, try using the Squeeze Theorem....⏎View Full Document<|endoftext|>A Woman of Salt, by Mary Potter Engel⏎Midrash refers to the

  6. #6 peak

    can be claimed, the final question is what, exactly, has been discovered? Scientists can have complete confidence in their results but remain uncertain about how to interpret them<|endoftext|>After

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: a new program, which is a set of rules that you can use to talk about it again and again and again and again and again and again and again and again and again and again and again and again and again and again and again and

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

    at high risk for developing osteoporosis. Osteoporosis is a condition in which thinning bones become so fragile that they are prone to fracture easily. A person with osteopenia may benefit from treatment aimed

  2. #2 peak

    , and sometimes they also tell us about what it was like when it was young. That’s due to gravitational lensing, a peculiar effect of relativity where the overall gravity of the cluster bends the light of an even more distant galaxy behind it

  3. #3 peak

    you may be entitled to a settlement.⏎The most common type of head injury is a concussion or traumatic brain injury (TBI) and can happen when the body is impacted with enough force to shift the brain and it collides with the skull.

  4. #4 peak

    financial literacy education for grades K-12, college, and post-college.⏎To view this DRM protected ebook on your desktop or laptop you will need to have Adobe Digital Editions installed. It is a free software. We also strongly recommend

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. L0:783 excites end-of-document marker token
  2. L6:3068 weakly excites
  3. L2:2888 weakly suppresses
  4. L5:236 weakly suppresses
  5. L6:994 weakly suppresses
  6. L7:2309 weakly suppresses
  7. L4:158 weakly suppresses
  8. L5:294 weakly suppresses
  9. L7:792 weakly excites
  10. L4:790 weakly suppresses
  11. L0:436 weakly suppresses
  12. L7:383 weakly suppresses
  13. L7:2383 weakly suppresses
  14. L3:186 weakly suppresses
  15. L6:67 weakly excites
  16. L7:1333 weakly excites
  17. L8:2757 weakly excites
  18. L2:1141 weakly suppresses
  19. L7:1423 weakly suppresses
  20. L2:222 weakly excites
  21. L7:1355 weakly suppresses
  22. L6:505 weakly suppresses
  23. L4:1963 weakly suppresses
  24. L0:1937 weakly suppresses

Feeds later layers

  1. L17:712 excites
  2. L17:2097 weakly suppresses
  3. L12:1003 weakly suppresses
  4. L10:1461 weakly suppresses
  5. L19:1732 weakly suppresses
  6. L19:1698 weakly suppresses
  7. L22:389 weakly suppresses
  8. L25:811 weakly suppresses
  9. L24:967 weakly suppresses
  10. L18:2322 weakly excites end-of-text token after a finished passage
  11. L16:588 weakly suppresses
  12. L23:388 weakly suppresses
  13. L27:42 weakly excites
  14. L24:1919 weakly suppresses
  15. L11:3059 weakly suppresses
  16. L17:1402 weakly suppresses
  17. L16:2324 weakly suppresses
  18. L12:2078 weakly excites end of document token
  19. L27:825 weakly suppresses
  20. L17:748 weakly suppresses
  21. L11:1123 weakly suppresses
  22. L24:1849 weakly suppresses
  23. L17:1105 weakly suppresses
  24. L25:2795 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 -1.15 and -0.61613 values between -0.61 and -0.073273 values between -0.07 and 0.479 values between 0.47 and 1.011 values between 1.01 and 1.551 values between 1.55 and 2.091 values between 2.09 and 2.642 values between 14.00 and 14.551 values between 15.63 and 16.171 values between 17.25 and 17.791 values between 17.79 and 18.341 values between 18.88 and 19.421 values between 19.42 and 19.961 values between 19.96 and 20.50offnegativepositiveoften

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. iel
  2. et
  3. el
  4. ens
  5. arian
  6. innacle
  7. !
  8. esar
  9. uel
  10. lector

Suppresses

  1. же
  2. _)
  3. 电子邮件
  4. teness
  5. activex
  6. damer
  7. 又要
  8. ***/⏎
  9. 我们要
  10. @return