Neuron Explorer

Qwen3-0.6B

'the' inside a mid-sentence phrase

A neuron in layer 20 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 'the' when it comes in the middle of a sentence and a noun phrase is about to follow, such as 'as the ship' or 'without the owner'. The pattern is loose, and the steering outputs don't clearly show a clear theme.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    so by including a specific date or a triggering event on which the trust is to terminate. By way of illustration, the Settlor might include a term requiring the trust to terminate upon the marriage of the trust’s only beneficiary or upon the youngest

  2. #2 peak

    from Southhampton, England on April 10, 1912.⏎This is one of the last photographs of Titanic taken from the shore as the ship departed towards New York City.⏎This photo shows some of the workers from the

  3. #3 peak

    and Practice Problems for adding positive and negative numbers⏎<|endoftext|>In physics, a conservation law states that a particular measurable property of an isolated physical system does not change as the system evolves over time. Exact conservation laws include conservation of energy, conservation of linear

  4. #4 peak

    a minimum, they should be cognizant that it is malicious software that collects personal information, usually with the intent of sending that information to third parties, without the owner's knowledge or consent. If anyone would like to learn the historical, technical

  5. #5 peak

    who they claim to be. Typical ways of doing this include asking for:⏎- Knowledge – Information only the user would know. Potentially a personal question about a favorite book or similar⏎

  6. #6 peak

    , to be much more economically efficient.<|endoftext|>Sepsis is a life-threatening illness caused by your body’s response to an infection. It develops when the chemicals the immune system releases into the bloodstream to fight an infection cause inflammation throughout the entire body

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 *this* *meat* the *this* *the* *the* *the* *the* *the* *the* ** ** *the* ** ** ** * * * **

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

    .<|endoftext|>DualWingGenerator is a power generation with the wing-beating principle. Inspired by birds this flapping generator could make modern wind turbines a thing of the past⏎German company Festo has unveiled DualWingGenerator to generate electricity,

  2. #2 peak

    This image was based upon the head of the great statue of Helios, better known as the Colossus of Rhodes, one of the Seven Wonders of the Ancient World. The Colossus memorialized the war with Demetrios Pol

  3. #3 peak

    reveal the 10 objects they have chosen to tell a history of Manchester and its place in the world.⏎The objects are part of the wider A History of the World project, a unique partnership between the BBC, the British Museum and 3

  4. #4 peak

    help find out more about all the servicemen whose images make up the online exhibition but about whom very little else is known.⏎The collection of photographs - Faces of the First World War - is part of the IWM's preparations to mark the cent

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. L18:861 strongly suppresses
  2. L19:2283 strongly suppresses
  3. L19:1023 excites
  4. L19:644 excites
  5. L17:372 suppresses
  6. L19:695 suppresses
  7. L18:1498 excites
  8. L17:30 suppresses
  9. L18:1359 excites
  10. L17:507 suppresses 'of' after a noun
  11. L16:1320 suppresses
  12. L16:419 excites
  13. L12:435 suppresses
  14. L19:2894 excites
  15. L16:792 excites "the" inside a prepositional phrase
  16. L8:636 suppresses
  17. L18:1513 suppresses
  18. L18:1041 suppresses
  19. L16:1851 suppresses
  20. L9:24 excites "the" before a named organisation or title
  21. L19:466 suppresses
  22. L19:1305 suppresses unclear: verbs after a noun or 'that'
  23. L18:1909 suppresses
  24. L17:56 suppresses

Feeds later layers

  1. L25:525 excites
  2. L21:843 excites
  3. L21:713 suppresses
  4. L21:1119 excites
  5. L23:24 suppresses
  6. L21:309 suppresses
  7. L21:696 excites
  8. L21:2279 suppresses
  9. L23:436 excites
  10. L21:920 excites
  11. L25:284 suppresses
  12. L27:335 excites
  13. L27:592 weakly excites
  14. L24:2040 weakly excites
  15. L27:1038 weakly excites
  16. L23:79 weakly suppresses
  17. L21:1227 weakly excites
  18. L27:533 weakly excites
  19. L22:1318 weakly suppresses
  20. L22:303 weakly suppresses
  21. L27:687 weakly excites
  22. L24:2454 weakly excites
  23. L27:449 weakly excites
  24. L24:268 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.

2 values between -5.97 and -5.554 values between -5.55 and -5.141 values between -5.14 and -4.733 values between -4.73 and -4.322 values between -4.32 and -3.904 values between -3.90 and -3.494 values between -3.49 and -3.085 values between -3.08 and -2.674 values between -2.67 and -2.268 values between -2.26 and -1.8410 values between -1.84 and -1.4350 values between -1.43 and -1.02387 values between -1.02 and -0.611410 values between -0.61 and -0.191754 values between -0.19 and 0.22155 values between 0.22 and 0.6325 values between 0.63 and 1.0411 values between 1.04 and 1.469 values between 1.46 and 1.875 values between 1.87 and 2.287 values between 2.28 and 2.699 values between 2.69 and 3.108 values between 3.10 and 3.524 values between 3.52 and 3.937 values between 3.93 and 4.344 values between 4.34 and 4.754 values between 4.75 and 5.173 values between 5.17 and 5.581 values between 5.58 and 5.991 values between 5.99 and 6.401 values between 7.23 and 7.642 values between 8.05 and 8.462 values between 8.46 and 8.882 values between 10.11 and 10.52offnegativepositiveoften

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. server
  2. 墟
  3. udad
  4. website
  5. tern
  6. device
  7. pherd
  8. 公司在
  9. company
  10. .server

Suppresses

  1. тики
  2. features
  3. Messaging
  4. (features
  5. amu
  6. gets
  7. lots
  8. /features
  9. тик
  10. ockets