Neuron Explorer

Qwen3-0.6B

'you' after if, when, and the more

A neuron in layer 8 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 responds to the word 'you' (and nearby words like 'to' or 'is') when it comes right after a word like 'if', 'when' or 'the more'. It is about the generic 'you' in a sentence. Turning it up did not produce any clear theme in the written text.

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

    cold winter and the snow was deep in the woods surrounding the logging camp near the Dungarvon River. But there was plenty of game to be found if you were a good hunter, so the men didn't go hungry. The camp cook

  2. #2 peak

    almost entirely Muslim. “Also, because of her Malay, Arabic comes easily.” Eunice adds, “And, English helps with French. The more languages you know the easier it is to learn.” According to Angela, her colleague is indeed

  3. #3 peak

    , according to⏎"If you're an assistant professor proposing to study it, I don't⏎think you'd get tenure," he says. "But one has to exist."⏎laboratory at Stanford University's Hopkins Marine Station⏎in Pacific Grove

  4. #4 peak

    ’s George Leach Centre Wednesday, as the injustices and struggles of Canada’s indigenous people were recounted, one blanket at a time.⏎“It’s very impactful when you relive it, it’s always got that pain,” said Shirley Horn, Alg

  5. #5 peak

    animals) the the fact that it still needs to be mounted on a "neck" to see much of the field.⏎Acctually, it depends on what you are doing. The greatest advancement of science, wealth and welfare in history has come

  6. #6 peak

    't quite get to the⏎> issue though.⏎> The problem I'm concerned about is the need to define y in the first⏎> place. If one is chasing through a data structure and finds a need to⏎> change something buried within

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 current state of the project is the problem. You know, I'm your assistant? Maybe you can ask me questions for me, but I need to make sure I understand what the current situation is. If I think about the problem,

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

    and science for every school year group. We really hope that you can use some, or all, of this content as you see fit, whether that

  2. #2 peak

    End of Year Math Review is not your typical math printables…this is math review with a twist! A digital version has been added!⏎With this resource, your students will be engaged and having fun all while reviewing what they’ve learned throughout the

  3. #3 peak

    the argument found in our assigned readings, and that in our in-class and online discussions we have already hashed out what we think the readings mean.⏎Instead of asking you to do that low-level thinking again, this assignment asks you to focus on applying

  4. #4 peak

    the rapid social changes around us, there exist certain constants like the family, religion, language, national identity, and democracy.⏎But the fact is, as you’ve heard me say many times before, everything changes and evolves, including the family,

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. L7:169 excites
  2. L4:499 excites
  3. L7:194 excites
  4. L5:2771 weakly suppresses
  5. L7:1375 weakly suppresses
  6. L5:222 weakly suppresses
  7. L1:1063 weakly suppresses
  8. L7:500 weakly excites
  9. L3:678 weakly suppresses
  10. L2:2299 weakly suppresses
  11. L6:234 weakly excites
  12. L1:2392 weakly suppresses
  13. L5:158 weakly suppresses
  14. L5:52 weakly excites
  15. L4:2089 weakly suppresses
  16. L4:515 weakly excites
  17. L1:294 weakly excites
  18. L2:2025 weakly excites
  19. L3:838 weakly suppresses
  20. L6:1564 weakly suppresses
  21. L5:1820 weakly suppresses
  22. L2:1566 weakly suppresses
  23. L7:959 weakly suppresses
  24. L3:981 weakly suppresses

Feeds later layers

  1. L10:1309 excites
  2. L15:1204 excites
  3. L10:293 excites
  4. L9:1279 weakly excites
  5. L23:2177 weakly suppresses
  6. L9:576 weakly suppresses
  7. L17:730 weakly suppresses
  8. L20:592 weakly suppresses
  9. L16:3044 weakly suppresses
  10. L12:976 weakly excites
  11. L19:2133 weakly excites
  12. L12:1702 weakly suppresses
  13. L14:2998 weakly suppresses
  14. L12:937 weakly excites
  15. L17:2643 weakly suppresses
  16. L11:937 weakly suppresses
  17. L9:1417 weakly suppresses
  18. L12:1671 weakly suppresses
  19. L17:1118 weakly excites mid-clause words in technical prose
  20. L13:722 weakly excites
  21. L13:2205 weakly suppresses
  22. L15:86 weakly suppresses
  23. L15:1113 weakly suppresses
  24. L9:1236 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.

1 values between -1.04 and -0.872 values between -0.87 and -0.703 values between -0.70 and -0.5344 values between -0.53 and -0.36313 values between -0.36 and -0.191250 values between -0.19 and -0.021651 values between -0.02 and 0.15467 values between 0.15 and 0.32122 values between 0.32 and 0.4927 values between 0.49 and 0.669 values between 0.66 and 0.835 values between 0.83 and 1.001 values between 1.00 and 1.174 values between 1.51 and 1.681 values between 1.68 and 1.851 values between 1.85 and 2.022 values between 2.02 and 2.191 values between 2.36 and 2.531 values between 2.70 and 2.871 values between 3.21 and 3.381 values between 4.22 and 4.391 values between 5.58 and 5.75offnegativepositiveoften

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. theor
  2. anz
  3. subject
  4. lick
  5. etr
  6. theory
  7. subject
  8. nel
  9. know
  10. .setView

Suppresses

  1. 你们
  2. 成年人
  3. 你們
  4. OUS
  5. 小男孩
  6. phans
  7. educators
  8. exampleModalLabel
  9. 您的孩子
  10. .enqueue