Neuron Explorer

Qwen3-0.6B

'of' after 'died' (cause of death)

A neuron in layer 1 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' after 'died', in phrases like 'died of plague' that name what someone died from. Turning it up did not make the model talk about illness, so the effect seems limited to that phrase.

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

    . A civil war then erupted. In the midst of this war, King John died of dysentery on October 18, 1216. His

  2. #2 peak

    , Lyon and his apprentice left town, taking a ship to Delaware. Two days after their arrival, the apprentice died of yellow fever. Reading the newspaper while he was away, Lyon was interested in accounts

  3. #3 peak

    She carried test tubes of radioactive materials in her pocket and kept them in her desk drawer. She often wrote about the pretty blue light it gave off. She died of aplastic anaemia – a direct result of her work.⏎William Bullock invented

  4. #4 peak

    quality of life of children. Early detection and treatment would greatly increase the survival chances of children with birth defects, diseases and other health deficiencies.⏎“If a child dies of a preventable cause, it is a huge tragedy and the nation has to get

  5. #5 peak

    tough one, hitting people over 65 the hardest but also putting many young adults and children into the hospital. So far, 37 children have died of flu, the Centers for Disease Control and Prevention reports. Half of the people hospitalized

  6. #6 peak

    Reformers during the French religious wars, Mercier was obliged to take refuge in Venice in 1567 and, after returning to France, he died of the plague. One of his best-known works was the Libellus de abbrev

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 can't be sure to get the right. But I can't be sure to get the right. But I can't be sure to get the right. But I can't be sure to get the right. But I can't be

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

    D level is above 30-40 ng/ml.⏎Pneumonia is yet another bacterial infection that vitamin D can help prevent. For example, during the 1918-19 epidemic influenza in the United States,

  2. #2 peak

    more than one edge. So when I draw this graph on my 2D screen, it can happen that edges

  3. #3 peak

    This was a pencilled entry only, the vehicle became replacement "N 3".⏎Some N wagons were passed onto line contractors as replacement vehicles. This occurred when contractors wagons were running on government lines and were destroyed in derailments or collisions

  4. #4 peak

    electronic devices are becoming more powerful while shedding or maintaining size. These smaller gadgets require significant amounts of energy, which current batteries don't exactly efficiently provide yet, especially after extended use. Scientists from the University of Central Florida (U

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. L0:1224 suppresses
  2. L0:1301 suppresses
  3. L0:178 suppresses
  4. L0:1468 weakly suppresses
  5. L0:253 weakly suppresses
  6. L0:2998 weakly suppresses
  7. L0:1632 weakly excites
  8. L0:1492 weakly excites
  9. L0:1857 weakly excites
  10. L0:288 weakly excites
  11. L0:977 weakly suppresses "like" after "you'd" or "something"
  12. L0:2917 weakly excites
  13. L0:139 weakly excites
  14. L0:149 weakly excites
  15. L0:978 weakly excites
  16. L0:684 weakly suppresses
  17. L0:2543 weakly suppresses
  18. L0:2782 weakly suppresses
  19. L0:2727 weakly suppresses
  20. L0:1599 weakly suppresses
  21. L0:1121 weakly excites
  22. L0:2169 weakly excites
  23. L0:408 weakly excites
  24. L0:2923 weakly suppresses

Feeds later layers

  1. L19:2031 weakly excites line breaks in web footers and metadata
  2. L21:276 weakly excites
  3. L3:1176 weakly suppresses
  4. L7:488 weakly excites
  5. L18:2025 weakly suppresses
  6. L22:246 weakly suppresses
  7. L12:1580 weakly excites
  8. L18:1816 weakly excites
  9. L23:2420 weakly suppresses
  10. L9:1152 weakly excites
  11. L21:659 weakly excites end of document boundary
  12. L8:2265 weakly suppresses
  13. L13:405 weakly suppresses
  14. L9:1438 weakly excites
  15. L10:266 weakly suppresses
  16. L16:909 weakly excites
  17. L18:1938 weakly excites
  18. L4:805 weakly excites
  19. L8:2757 weakly suppresses
  20. L10:1234 weakly suppresses
  21. L12:2769 weakly excites
  22. L11:2988 weakly excites
  23. L5:2959 weakly suppresses
  24. L20:70 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 -0.86 and -0.748 values between -0.62 and -0.5018 values between -0.50 and -0.3858 values between -0.38 and -0.26222 values between -0.26 and -0.141420 values between -0.14 and -0.011946 values between -0.01 and 0.11198 values between 0.11 and 0.2317 values between 0.23 and 0.352 values between 0.35 and 0.477 values between 0.47 and 0.591 values between 0.59 and 0.711 values between 0.95 and 1.071 values between 1.07 and 1.191 values between 1.19 and 1.312 values between 1.31 and 1.431 values between 1.43 and 1.561 values between 2.16 and 2.281 values between 2.88 and 3.001 values between 3.00 and 3.121 values between 3.85 and 3.97offnegativepositiveoften

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. abal
  2. arton
  3. aje
  4. 除此
  5. hift
  6. 縮
  7. estre
  8. ages
  9. age
  10. ứ

Suppresses

  1. inas
  2. DCF
  3. SOR
  4. cocks
  5. 日在
  6. pray
  7. liest
  8. ưới
  9. tabpanel
  10. iners