Neuron Explorer

Qwen3-0.6B

'of' after nouns like anniversary or onset

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 fires on 'of' right after a noun such as 'anniversary', 'onset', 'undersides', 'issue' or 'use'. It is a word that leads into the thing being described. Its low side is about phrases like 'get the most out' and 'during the life'. Turning it up gave repetitive or unrelated 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

    Communist Party⏎- 27 June 2011⏎- From the section Asia-Pacific⏎China is preparing to mark the 90th anniversary of the founding of the ruling Communist Party. The BBC's Michael Bristow in

  2. #2 peak

    -polluting nations such as<|endoftext|>Learning disabilities are neurologically based conditions that interfere with the acquisition, storage, organization, and use of skills and knowledge. They are identified by deficits in academic functioning and in processing other

  3. #3 peak

    may have had trouble finding them this spring.”⏎Disease symptoms include wilting, yellowing and bloom loss. Homeowners may notice a white growth on undersides of leaves, especially during humid conditions, rainy

  4. #4 peak

    midnight. Auroral observations by the Wideband Imaging Camera on the Imager for Magnetopause-to-Aurora Global Exploration spacecraft are used to determine the onset times of substorms. It is shown that the magnetic phenomena at the earthward site of

  5. #5 peak

    being processed by steam, a container is normally selected from one of the first containers going into the bottom layer of the first retort crate. This container is then set aside where it is not subject

  6. #6 peak

    other colonies, territories and states.⏎In late July of 2019, there will be a large celebration commemorating the 400th anniversary of the first legislative assembly held in North America when Governor Sir George Yeardley,

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 Power of the Ocean, the Power of the People of the Sea of Nations (POC) of the Future of Humanity" (POC) in the United States and the South Africa of the South. This is the time to

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

    up in order to get the most out of your battery.⏎Lithium-Ion Battery: They charge faster than other

  2. #2 peak

    selected the Lockheed Martin Corp. to build the Joint Strike Fighter (JSF). The contract had a potential total value of over $200 billion during the life of the program. The U.S. planned to buy nearly 3,0

  3. #3 peak

    . Bruce Weinstein, better known to some as The Ethics Guy, listed five of them in his book Ethical Intelligence:⏎Do No Harm⏎Make Things Better⏎

  4. #4 peak

    healthy eating as defined by community members themselves.⏎This study includes three of the most deprived electoral wards in Stoke-on-Trent. In each of these areas

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. L8:2257 suppresses
  2. L8:502 excites
  3. L6:1735 excites
  4. L8:1003 suppresses "the" or "of" before a noun phrase
  5. L7:23 suppresses
  6. L8:447 excites
  7. L8:25 suppresses
  8. L8:853 suppresses slot right before a quantity
  9. L5:1721 excites
  10. L8:2095 suppresses
  11. L8:706 suppresses
  12. L6:10 excites
  13. L6:823 suppresses
  14. L8:892 weakly excites
  15. L8:9 weakly excites
  16. L8:433 weakly suppresses
  17. L5:2890 weakly excites
  18. L4:1686 weakly excites
  19. L3:1096 weakly suppresses
  20. L8:500 weakly suppresses 'to' after verbs like continue or help
  21. L6:1602 weakly excites unclear: hyphens and size comparisons
  22. L6:2217 weakly suppresses
  23. L8:913 weakly suppresses
  24. L8:879 weakly suppresses

Feeds later layers

  1. L10:584 suppresses
  2. L10:40 suppresses
  3. L20:1788 excites
  4. L13:2065 excites
  5. L13:471 weakly excites
  6. L11:219 weakly excites
  7. L10:67 weakly excites
  8. L15:487 weakly suppresses
  9. L13:376 weakly excites
  10. L21:2342 weakly excites
  11. L12:653 weakly excites
  12. L13:2469 weakly suppresses
  13. L21:1073 weakly excites
  14. L15:1811 weakly excites
  15. L12:74 weakly suppresses
  16. L15:1168 weakly excites
  17. L13:1062 weakly excites
  18. L17:2302 weakly excites 'in' after a noun
  19. L21:1866 weakly suppresses
  20. L12:1741 weakly excites
  21. L19:2926 weakly excites
  22. L12:1685 weakly suppresses
  23. L24:2657 weakly excites
  24. L11:1461 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 -3.67 and -3.461 values between -3.46 and -3.262 values between -3.26 and -3.051 values between -2.85 and -2.643 values between -2.64 and -2.441 values between -2.44 and -2.236 values between -2.23 and -2.036 values between -2.03 and -1.823 values between -1.82 and -1.6212 values between -1.62 and -1.4110 values between -1.41 and -1.2019 values between -1.20 and -1.0033 values between -1.00 and -0.7958 values between -0.79 and -0.5999 values between -0.59 and -0.38299 values between -0.38 and -0.181324 values between -0.18 and 0.031391 values between 0.03 and 0.24448 values between 0.24 and 0.4486 values between 0.44 and 0.6527 values between 0.65 and 0.8524 values between 0.85 and 1.0613 values between 1.06 and 1.2612 values between 1.26 and 1.476 values between 1.47 and 1.674 values between 1.67 and 1.886 values between 1.88 and 2.093 values between 2.09 and 2.293 values between 2.29 and 2.502 values between 2.50 and 2.702 values between 2.70 and 2.911 values between 3.32 and 3.521 values between 4.35 and 4.55offnegativepositiveoften

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. large
  2. rebellion
  3. =com
  4. borough
  5. создан
  6. gibi
  7. себе
  8. lastName
  9. Rebel
  10. собой

Suppresses

  1. -piece
  2. iations
  3. inue
  4. Material
  5. ideos
  6. 跄
  7. 物
  8. AMENT
  9. inqu
  10. unist