Neuron Explorer

Qwen3-0.6B

the word COVID before -19

A neuron in layer 10 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 lights up on the word COVID when it comes right before '-19', as in news about the pandemic. Turning it up made the writing wander into odd capitalised words and repeated phrases.

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

    their essay and learned to focus on points in their lives, rather than the entirety of it.<|endoftext|>COVID-19 Vaccine Enters Final Test Phase⏎The COVID-19 pandemic seemingly has no expiration date. At its onset, we thought

  2. #2 peak

    public – as social distancing restrictions are eased – can help us further reduce the spread of COVID-19 that we’ve achieved by staying at home,” said Bill Hayes,

  3. #3 peak

    pandemics such as COVID-19. Suppression can work if the outbreak is small and you’re

  4. #4 peak

    ADA), to protect workers who contract COVID-19. These guidelines were set because the impact of COVID-19 on the mind and body can meet the ADA's definition of a

  5. #5 peak

    countries and we expect it to soon be the dominant COVID-19 strain circulating worldwide.⏎“The world is watching in real-time as the COVID-19 virus continues to change and become more transmissible.”⏎Kindly follow

  6. #6 peak

    that people should be well informed about the disease-causing virus and the ways through which it spreads the disease. The only easy and effective way to protect yourself from COVID-19 is by using an alcohol-based massage or washing your hands frequently,

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 a 5 of 5, 5, 5, 5, 5, 5. and 5, 5, , and, 5. 5, and, 5,

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

    audio⏎OAK RIDGE, Tenn., Dec. 7, 2005 —

  2. #2 peak

    a maximum of 100 mg/day. The American Congress of Obstetricians and Gynecologists recommends that first-line treatment of nausea and<|endoftext|>PITTSBURGH — Do you have a daily indulgence in your diet? Even in

  3. #3 peak

    was reduced to a province within the Middle Assyrian Empire.⏎ Some historians place the Kassites and Hurrians within the bounds of Mitanni.⏎The KASSITES – c. 1184-763⏎When Matanni

  4. #4 peak

    images from the receiving ends of the bundled fibers to the surface (the eyeballs), creating the illusion of animated eyes. In addition to moving eyes, PAPILLON can project expressive symbols (such as dollar signs) in place of pupils 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

  1. L4:533 strongly suppresses
  2. L5:35 strongly suppresses
  3. L7:128 strongly suppresses
  4. L3:627 strongly suppresses
  5. L6:2269 strongly suppresses
  6. L9:1737 suppresses
  7. L1:1488 suppresses
  8. L0:76 excites
  9. L1:1260 suppresses
  10. L1:1124 excites
  11. L0:704 excites
  12. L0:2979 suppresses
  13. L1:2221 excites
  14. L3:2329 excites
  15. L5:281 excites
  16. L3:1479 suppresses
  17. L6:430 suppresses
  18. L2:2051 excites
  19. L2:1830 suppresses
  20. L0:1904 suppresses
  21. L6:861 excites
  22. L7:524 suppresses
  23. L0:2229 suppresses
  24. L2:2174 excites

Feeds later layers

  1. L16:584 strongly excites
  2. L11:631 strongly excites
  3. L18:373 strongly excites
  4. L22:651 strongly excites
  5. L20:896 excites mid-word fragments in capitalised titles
  6. L22:1126 excites
  7. L15:257 excites
  8. L11:332 suppresses
  9. L11:322 excites
  10. L24:1991 excites
  11. L18:1733 excites
  12. L20:562 excites
  13. L18:599 excites
  14. L25:1750 excites
  15. L24:980 excites
  16. L20:159 suppresses
  17. L26:1531 excites
  18. L21:533 excites
  19. L27:782 excites
  20. L17:1747 suppresses split tech names and acronyms
  21. L27:1360 excites
  22. L22:212 excites
  23. L17:1622 excites
  24. L14:642 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 -2.46 and -2.331 values between -1.57 and -1.441 values between -1.19 and -1.0611 values between -1.06 and -0.9358 values between -0.93 and -0.81185 values between -0.81 and -0.68359 values between -0.68 and -0.55489 values between -0.55 and -0.43574 values between -0.43 and -0.30574 values between -0.30 and -0.17475 values between -0.17 and -0.04345 values between -0.04 and 0.08248 values between 0.08 and 0.21181 values between 0.21 and 0.34117 values between 0.34 and 0.4786 values between 0.47 and 0.5950 values between 0.59 and 0.7247 values between 0.72 and 0.8522 values between 0.85 and 0.9719 values between 0.97 and 1.1017 values between 1.10 and 1.2315 values between 1.23 and 1.3615 values between 1.36 and 1.486 values between 1.48 and 1.613 values between 1.61 and 1.742 values between 1.74 and 1.863 values between 1.86 and 1.992 values between 1.99 and 2.122 values between 2.50 and 2.63offnegativepositiveoften

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. _BEGIN
  2. MonoBehaviour
  3. solid
  4. linux
  5. hipster
  6. peninsula
  7. addtogroup
  8. uga
  9. ',⏎
  10. ushort

Suppresses

  1. 第八
  2. arith
  3. -o
  4. 第五
  5. ение
  6. ervative
  7. fäll
  8. 侬
  9. 制造
  10. encial