Neuron Explorer

Qwen3-0.6B

"quite" in hedged negations

A neuron in layer 6 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 "quite" when it follows a negation or doubt, as in "not quite sure" or "can't quite work out." Turning it up made the model's text drift into unrelated topics, so the steering results don't show this pattern clearly.

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

    regiments to be raised from the population of California to help protect mail routes and the communications lines of the West. Neither Lincoln nor the U.S. War Department quite trusted the Mormons of the Utah Territory to remain loyal to the Union, in spite

  2. #2 peak

    I don't know how to review it. We're adding the slope at a single point to the slope over a big area, which is something I can't quite work out visually.<|endoftext|>The world produces more than enough food to feed everyone,

  3. #3 peak

    exit out of the loop, IE it will continue to call it self over and over until the program finally kills the heap/stack or something. I don't quite know what it is...⏎Also, even if it wasn't a static, it

  4. #4 peak

    u-substitution,⏎- we could say that du is equal to 5 cosine 5x.⏎- Now when you look over here, we don't have quite du there. We have just cosine of 5x dx--⏎- sorry,

  5. #5 peak

    ever wondered what a cobot is?⏎You might have come across this term “cobots” while searching for a solution to improve your production processes but weren’t quite sure what the term referred to. As you’ve probably noticed yourself, cobots

  6. #6 peak

    recognize the goodness of American foreign policy and the inherent democratic character of the United States. Bush increasingly appears somewhat like a frustrated Lyndon Baines Johnson who could not quite understand why “all” Americans did not recognize their prosperity during the democratic upheav

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 number of people who have been in the system. The number of people who have been in the system is equal to the number of people who have been in the system. Let me try to figure out the next possible thing to add.

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

    fought for⏎civil rights, child welfare and womens rights.<|endoftext|>How does Ray Bradbury portray technology in the story "There Will Come Soft Rains?"⏎2 Answers | Add Yours⏎The technology in the story is actually presented as the main character

  2. #2 peak

    Carl Ruetz⏎Page last modified July 14, 2009<|endoftext|>what does hamlet think about suicide? with textual support⏎2 Answers | Add Yours⏎Hamlet's thoughts on suicide occur at two places in the

  3. #3 peak

    in 6 bytes, with the last byte equal to 0.⏎October 3rd, 2010, 07:44 AM⏎We are not going to do you homework.⏎If you want our help, you

  4. #4 peak

    ), and A[1:0] for ARM instruction fetches (nOPC=0 and MAS[1:0]=10).⏎Article last edited on: 2008-09-09 15:

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. L3:255 excites
  2. L3:148 weakly suppresses
  3. L2:3004 weakly suppresses
  4. L5:2155 weakly excites
  5. L5:1803 weakly excites
  6. L0:2768 weakly suppresses
  7. L1:1299 weakly suppresses
  8. L3:52 weakly suppresses
  9. L5:1872 weakly suppresses
  10. L4:413 weakly excites
  11. L2:1988 weakly suppresses
  12. L2:2074 weakly excites
  13. L5:1380 weakly excites
  14. L5:866 weakly excites
  15. L5:1920 weakly suppresses
  16. L1:2669 weakly excites
  17. L0:2722 weakly excites
  18. L2:1081 weakly excites
  19. L4:2883 weakly excites
  20. L2:899 weakly suppresses
  21. L3:767 weakly suppresses
  22. L4:618 weakly suppresses
  23. L1:506 weakly excites
  24. L5:3027 weakly suppresses

Feeds later layers

  1. L7:1992 weakly excites
  2. L13:1958 weakly excites
  3. L23:1443 weakly excites
  4. L7:2773 weakly excites
  5. L7:1770 weakly excites
  6. L14:1083 weakly suppresses
  7. L14:812 weakly suppresses
  8. L9:2679 weakly suppresses
  9. L11:2552 weakly excites
  10. L9:1772 weakly suppresses
  11. L10:2908 weakly excites
  12. L17:1201 weakly suppresses
  13. L13:2488 weakly suppresses
  14. L16:2202 weakly excites
  15. L8:1854 weakly suppresses
  16. L17:2288 weakly excites
  17. L23:1452 weakly suppresses
  18. L8:2912 weakly excites
  19. L10:1758 weakly suppresses
  20. L14:431 weakly suppresses
  21. L8:2999 weakly suppresses
  22. L8:2355 weakly excites
  23. L15:505 weakly suppresses
  24. L9:2961 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 -1.20 and -1.132 values between -0.65 and -0.583 values between -0.58 and -0.5115 values between -0.51 and -0.4440 values between -0.44 and -0.3780 values between -0.37 and -0.30158 values between -0.30 and -0.23328 values between -0.23 and -0.16520 values between -0.16 and -0.09716 values between -0.09 and -0.02795 values between -0.02 and 0.05593 values between 0.05 and 0.12321 values between 0.12 and 0.19187 values between 0.19 and 0.2699 values between 0.26 and 0.3329 values between 0.33 and 0.4013 values between 0.40 and 0.472 values between 0.47 and 0.532 values between 0.53 and 0.601 values between 0.67 and 0.741 values between 0.95 and 1.021 values between 1.23 and 1.301 values between 1.51 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. amber
  2. ESH
  3. atches
  4. 뜬
  5. PropTypes
  6. exhaust
  7. *)&
  8. 昆
  9. computing
  10. domestically

Suppresses

  1. olution
  2. tür
  3. 褓
  4. alary
  5. lsx
  6. mdir
  7. 越高
  8. vio
  9. levation
  10. ians