Neuron Explorer

Qwen3-0.6B

'yet' ending a negated clause

A neuron in layer 2 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 fires on the word 'yet' at the end of sentences that say something has not happened so far, like 'haven't been tested yet'. Turning it up did not make the model's writing visibly about this.

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

    .⏎- Design & Thinking Official Trailer (02:28)⏎Design & Thinking is a documentary released this year about design thinking. I haven't seen it yet, but I'm loving the approach -- the cacophony of voices

  2. #2 peak

    became one of the first to author the word aight in published text. He’s even glossed it “(Alright)” for everyone who hadn’t caught on yet.⏎We learn from the OED entry that aight represents a colloquial

  3. #3 peak

    you boring and dull, because you have nothing to talk about, because everything you talk about is a memory. A person cannot talk about something that hasn’t happened yet in reality.⏎“Memory depends very much on the perspicuity, regularity

  4. #4 peak

    isn’t even a “hypothesis”.⏎Being a falsifiable hypothesis doesn’t mean the hypothesis is true...not yet. The hypothesis hasn’t been tested yet.⏎The hypothesis is never true or false by default. Only the tests or evid

  5. #5 peak

    gingivitis?⏎Gingivitis is the first and only stage that periodontal disease can be reversed. This is because the bones haven’t been attacked yet. Bleeding gums is usually one of the first symptoms experienced if you have ging

  6. #6 peak

    ifference-between-a-cote-de-and-pres-de⏎I hope this is helpful.⏎Bonne journée !⏎Sign in to submit your answer⏎Don't have an account yet? Join today⏎Test your French to the CEFR standard<|endoftext|>I believe that

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 role of a teacher in the field of Education. I have been thinking about lately: the role of a teacher in the field of Education. I have been thinking about lately: the role of a teacher in the field of Education. I

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

    /restaurant companies who don’t use fair trade beans, for example). Socially responsible investing (SRI) is basically the practice of putting your money to work only in companies that are trying to make a positive impact on society.⏎You can pick

  2. #2 peak

    the item is re-sold at a higher price, then the profit will go down. With what is variable CPV, you don’t pay for the item only once but you get to sell it more times and therefore gain profit each time.⏎

  3. #3 peak

    .<|endoftext|>Our research goal is to understand the maximum short-term capabilities of electrical machines and investigate control methods to achieve them. Electric machine capabilities and limitations cannot be understood only through electrical specifications and nameplate data, because redundancy is included in the design.

  4. #4 peak

    are unique to each server. Otherwise, the servers try to log to the same files.⏎Even when the preceding precautions are observed, this kind of⏎setup works only with⏎MERGE tables, and not with any of the other⏎storage engines

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. L1:2675 weakly suppresses
  2. L0:1011 weakly suppresses
  3. L1:583 weakly excites
  4. L1:1779 weakly excites
  5. L0:1825 weakly suppresses
  6. L0:2171 weakly suppresses
  7. L0:207 weakly suppresses
  8. L1:2445 weakly excites
  9. L1:1645 weakly excites
  10. L1:1356 weakly suppresses
  11. L1:2446 weakly suppresses
  12. L1:359 weakly excites
  13. L0:2634 weakly suppresses
  14. L0:649 weakly suppresses
  15. L0:816 weakly excites
  16. L1:2978 weakly suppresses
  17. L1:174 weakly excites
  18. L1:466 weakly suppresses
  19. L1:2830 weakly excites
  20. L1:316 weakly excites
  21. L1:445 weakly excites document start marker token
  22. L1:966 weakly suppresses
  23. L1:2642 weakly excites
  24. L1:2607 weakly suppresses

Feeds later layers

  1. L3:40 weakly excites
  2. L6:101 weakly suppresses
  3. L3:4 weakly suppresses
  4. L16:1477 weakly suppresses
  5. L27:425 weakly suppresses
  6. L13:2120 weakly excites
  7. L9:69 weakly suppresses
  8. L21:46 weakly suppresses
  9. L15:740 weakly excites
  10. L7:551 weakly excites
  11. L4:3058 weakly excites
  12. L26:2514 weakly suppresses the word 'just' after have or has
  13. L15:2751 weakly excites
  14. L23:234 weakly excites
  15. L3:1489 weakly suppresses
  16. L5:401 weakly suppresses
  17. L14:282 weakly suppresses
  18. L13:1034 weakly excites
  19. L21:1513 weakly excites
  20. L8:3001 weakly suppresses
  21. L13:1372 weakly suppresses number before an age or week count
  22. L4:2469 weakly excites
  23. L9:2622 weakly excites
  24. L8:149 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 -1.78 and -1.691 values between -1.69 and -1.591 values between -0.83 and -0.744 values between -0.64 and -0.552 values between -0.55 and -0.452 values between -0.45 and -0.3619 values between -0.36 and -0.2780 values between -0.27 and -0.17401 values between -0.17 and -0.081773 values between -0.08 and 0.021216 values between 0.02 and 0.11295 values between 0.11 and 0.2161 values between 0.21 and 0.3027 values between 0.30 and 0.408 values between 0.40 and 0.494 values between 0.49 and 0.591 values between 0.59 and 0.685 values between 0.68 and 0.784 values between 0.97 and 1.061 values between 1.15 and 1.251 values between 1.44 and 1.531 values between 1.91 and 2.01offnegativepositiveoften

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. 략
  2. nc
  3. قات
  4. 赀
  5. 盛会
  6. craft
  7. mise
  8. �
  9. 坊
  10. ty

Suppresses

  1. "}>⏎
  2. igham
  3. ighbor
  4. olleyError
  5. <Student
  6. <footer
  7. nEnter
  8. /org
  9. /MIT
  10. ApiController