Neuron Explorer

Qwen3-0.6B

"of" after nouns like eve, family, modes

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 "of" when it follows a noun that usually takes it, such as "on the eve of", "a family of", "modes of" or "heirs of". Turning it up didn't produce a clear shift in the model's writing.

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

    named, siblings of the parents were named after, especially the ones who had died already.⏎It is

  2. #2 peak

    2006⏎Part I: The Umayyad Caliphate (661 – 750 CE)⏎5 Tibet on the Eve of the Arrival of the First Muslim Teacher⏎When al-Salit bin-Ab

  3. #3 peak

    .⏎The problem is, at least in part, topographical in nature. West Africa’s geography is dangerous, vast, and difficult to subordinate. On the eve of much of West Africa’s independence from France in 1961, R

  4. #4 peak

    heirs of their cultural⏎heritage. Therefore, present-day Chaldeans and Assy

  5. #5 peak

    -Baptiste Barrelon (18th October 1818 – 31st September 1885) was born in a family of locksmith

  6. #6 peak

    of the 328th who felt Sergeant Bernard Early and other members of the patrol deserved credit, too.⏎On the eve of York’s’ return to America in late May of that year, Major General George

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: I have been working on this project, and I have been thinking about it a lot. It's a bit of a problem that I have been trying to fix, but I haven't really fixed it. I have been thinking about it,

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

    1900 the population had risen to 4.5 million. Of course all these people needed to get around the ever expanding city and so more modes of transport started to appear.⏎Level 2 covers these transportations issues during 1

  2. #2 peak

    of dementias on the brain, and the impact on an individual's cognitive function. A dedicated chapter explains how a diagnosis is reached, including the different modes of assessment, from blood tests carried out by a GP, to specialist scans, highlighting

  3. #3 peak

    of 9, Gabon has a rating of 4, and Malawai has a rating of 1

  4. #4 peak

    odic retrospective memory, even without any signs of dementia.⏎Given these findings, clinicians can help even healthy e-4 carriers to

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. L23:1217 weakly excites
  2. L26:1322 weakly suppresses
  3. L25:1841 weakly suppresses
  4. L19:1615 weakly suppresses
  5. L21:1979 weakly excites the number 5 in a numbered list
  6. L20:899 weakly suppresses
  7. L26:1415 weakly suppresses 're' prefix fragment
  8. L19:1696 weakly excites
  9. L25:961 weakly suppresses years and numbers before the digits
  10. L17:1416 weakly excites
  11. L16:967 weakly excites
  12. L26:2865 weakly excites
  13. L21:2406 weakly suppresses
  14. L25:2555 weakly suppresses
  15. L18:402 weakly suppresses
  16. L25:2006 weakly suppresses
  17. L27:242 weakly suppresses
  18. L20:2446 weakly excites
  19. L20:2217 weakly suppresses 'trying' after 'were' or 'are'
  20. L18:666 weakly suppresses
  21. L12:1386 weakly suppresses
  22. L4:1973 weakly excites
  23. L26:308 weakly suppresses
  24. L9:1508 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 -0.69 and -0.651 values between -0.65 and -0.602 values between -0.60 and -0.561 values between -0.56 and -0.521 values between -0.52 and -0.483 values between -0.48 and -0.442 values between -0.44 and -0.401 values between -0.40 and -0.361 values between -0.36 and -0.326 values between -0.32 and -0.2813 values between -0.28 and -0.2417 values between -0.24 and -0.2034 values between -0.20 and -0.1665 values between -0.16 and -0.12166 values between -0.12 and -0.07377 values between -0.07 and -0.031708 values between -0.03 and 0.01889 values between 0.01 and 0.05358 values between 0.05 and 0.09125 values between 0.09 and 0.1357 values between 0.13 and 0.1734 values between 0.17 and 0.2110 values between 0.21 and 0.2510 values between 0.25 and 0.297 values between 0.29 and 0.3310 values between 0.33 and 0.383 values between 0.38 and 0.421 values between 0.46 and 0.504 values between 0.62 and 0.661 values between 0.91 and 0.95offnegativepositiveoften

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. much
  2. dx
  3. thèse
  4. interchangeable
  5. ��
  6. half
  7. Compare
  8. 一致
  9. plement
  10. ần

Suppresses

  1. onium
  2. 到现在
  3. utenberg
  4. ования
  5. ocene
  6. undert
  7. 的动作
  8. omba
  9. 桷
  10. ullah