Neuron Explorer

Qwen3-0.6B

'to' in 'access to'

A neuron in layer 0 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 'to' when it comes after 'access', as in 'access to technology'. The activations are weak compared with the other neurons here. Turning it up gave unrelated, rambling 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

    and understanding the subject thoroughly to provide required feedback at school. What really is the hurdle here?⏎Access to technology: Not every student has internet at his home. Though it is a pretty

  2. #2 peak

    years has been how to apply the time-honored purposes of the Communications Act to our 21st Century communications platform—broadband Internet—access to which is generally provided by the same companies that provide telephone and cable television

  3. #3 peak

    Organised group of people within a larger group.⏎- Asylum - Refugee. ( Government's permission to stay in their country).⏎- Kid-gloves - To treat with extreme tact

  4. #4 peak

    a host of countries.⏎This article cannot deal with every issue on the summit agenda. What follows are a few examples of the challenges and innovations under discussion.⏎Access to Education⏎At the most fundamental level, the world has still not achieved the commitment

  5. #5 peak

    promoting the sport and achieving. American swimmers hold themselves in high standards and take every opportunity to change the otherwise monotonous sport to an exciting sport. Access to facility is the greatest obstacle but the efforts to contacting new facilities at local levels are

  6. #6 peak

    nectar.⏎|Science Recorder Pro|⏎|Free trial. No ads. Exclusive interviews. Access to all articles. Just $4.99/yr. Subscribe|⏎New research

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 a Power" is a well-known, but not the most. But it's a good one. And it's a good one. I am not sure about. I am not sure about. So, I am not

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

    ὔτε γὰπ τὸ χωρίον τοῦτο, ἐν ᾧ τὸ ἰερὸν φυλάττ<|endoftext|>Iron and Stone by the

  2. #2 peak

    . This should not be confused with the anniversary of the Internet, which happened in 1969 as a Defense Department project

  3. #3 peak

    ⏎As the UK Parliament prepares to celebrate the 150th anniversary of its world-famous clock, E&T examines the history of Big Ben.⏎With

  4. #4 peak

    the story of Baldr's death. In Gylfaginning, Snorri Sturluson tells how Loki, fearing the wrath of the Æsir

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

None: this is the first layer.

Feeds later layers

  1. L19:858 weakly excites
  2. L10:90 weakly suppresses
  3. L27:1129 weakly suppresses
  4. L23:2746 weakly excites
  5. L16:1918 weakly excites
  6. L23:326 weakly suppresses
  7. L2:274 weakly excites
  8. L12:1693 weakly excites
  9. L22:1922 weakly suppresses
  10. L22:2878 weakly suppresses
  11. L16:1115 weakly excites
  12. L25:2828 weakly suppresses
  13. L11:2196 weakly suppresses
  14. L21:971 weakly suppresses
  15. L5:3030 weakly excites
  16. L12:1229 weakly suppresses
  17. L26:1331 weakly excites
  18. L12:1563 weakly suppresses
  19. L21:1499 weakly suppresses question marks ending a question
  20. L2:2842 weakly excites
  21. L26:973 weakly excites
  22. L15:1815 weakly suppresses
  23. L21:21 weakly excites
  24. L9:1276 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 -0.38 and -0.353 values between -0.30 and -0.272 values between -0.27 and -0.246 values between -0.24 and -0.216 values between -0.21 and -0.1825 values between -0.18 and -0.1567 values between -0.15 and -0.1280 values between -0.12 and -0.10169 values between -0.10 and -0.07279 values between -0.07 and -0.04780 values between -0.04 and -0.011437 values between -0.01 and 0.02531 values between 0.02 and 0.05231 values between 0.05 and 0.08134 values between 0.08 and 0.1169 values between 0.11 and 0.1437 values between 0.14 and 0.1624 values between 0.16 and 0.1911 values between 0.19 and 0.224 values between 0.22 and 0.251 values between 0.25 and 0.281 values between 0.28 and 0.312 values between 0.31 and 0.342 values between 0.34 and 0.371 values between 0.37 and 0.401 values between 0.40 and 0.421 values between 0.42 and 0.451 values between 0.45 and 0.481 values between 0.57 and 0.601 values between 0.74 and 0.77offnegativepositiveoften

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. Tür
  2. RecognitionException
  3. iếu
  4. związane
  5. WebpackPlugin
  6. enção
  7. ollen
  8. pien
  9. 殂
  10. IMO

Suppresses

  1. ández
  2. ittest
  3. aces
  4. nghi
  5. acet
  6. idable
  7. Originally
  8. conde
  9. clid
  10. aight