Neuron Explorer

Qwen3-0.6B

commas closing a list item

A neuron in layer 3 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 a comma that comes right after a single item in a list of examples, such as "a house, a doctor" or "a phone number or photograph,". Turned up, the model shifted to quiz and question-style 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

    ubs, trees, ferns, gymnosperms, medicinal and scented plants in Fern Garden, Medicinal Herb Garden,

  2. #2 peak

    PV array, which then adjusts the incoming voltage in order to maintain the most efficient level of power for the system in place.⏎<|endoftext|>Serious diseases, including diabetes, cancer, and eating disorders, can manifest as signs and symptoms inside of the mouth

  3. #3 peak

    efforts will help to address some of the current challenges in space security.⏎By entering this website, you consent to the use of technologies, such as cookies and analytics, to customise content, advertising and provide social media features. This will be used

  4. #4 peak

    the party machine. Immigrants lined up at their respective Democratic party committee headquarters and waited their turn for the bosses to find them a job, a house, a doctor. It was an unfair system that did not work the way the civ

  5. #5 peak

    "undo" or "select all," running within apps at all times. It creates a virtual cloud-based clipboard where any data, like a phone number or photograph, can temporarily live until it's "pasted" to

  6. #6 peak

    question should be enough for your answers. Worth 10 points each.⏎1. Evaluate the following statement. "A temporarily high government spending, like a war, should be financed by temporarily high taxes."⏎2. Is the following statement true,

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 need to get back to the point of a few more. I want to talk about it. I want to talk about it. I need to get back to the point of a few more. I want to talk about it. I

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

    all of our lives, but if we have an increased propensity to become forgetful -- perhaps to ask a question that has already been answered or increasingly misplace items -- forget appointments, forget medications. If these are new changes, then that's a

  2. #2 peak

    North Pacific Gyre.⏎Twice the size of Texas – though some scientists have argued that its mass has extended to twice the size of the United States – a majority of the garbage is comprised of tiny bits of plastic, barely visible to

  3. #3 peak

    -menopausal women chosen from the Framingham Heart Study - which has tracked the medical history of the residents of Framingham, Massachusetts since 1948 - to see whether factors like height, weight or cholesterol impacted the number of children women

  4. #4 peak

    interviews back to back (Len Irish, who took our cover photo of Collins, says she was one of the nicest people he ever photographed – “no fuss”), there’s a television crew waiting for her to put on her spacesuit, and

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. L2:845 excites
  2. L1:379 weakly suppresses
  3. L2:308 weakly suppresses
  4. L2:651 weakly suppresses
  5. L2:265 weakly excites
  6. L2:1272 weakly suppresses the word 'wide' in 'a wide range'
  7. L1:536 weakly excites
  8. L0:225 weakly suppresses
  9. L0:804 weakly suppresses
  10. L2:840 weakly excites
  11. L0:450 weakly excites
  12. L1:1275 weakly suppresses
  13. L2:877 weakly suppresses
  14. L1:2336 weakly excites
  15. L1:1197 weakly suppresses
  16. L2:452 weakly excites
  17. L2:1966 weakly suppresses
  18. L2:288 weakly excites
  19. L0:1338 weakly excites
  20. L1:2293 weakly excites
  21. L2:358 weakly suppresses
  22. L1:93 weakly suppresses
  23. L0:2548 weakly excites
  24. L0:1464 weakly suppresses

Feeds later layers

  1. L4:238 suppresses
  2. L4:69 suppresses
  3. L4:91 weakly suppresses
  4. L11:1604 weakly excites
  5. L23:805 weakly excites
  6. L8:2503 weakly suppresses
  7. L19:2035 weakly excites
  8. L24:3022 weakly suppresses
  9. L15:1459 weakly excites
  10. L14:1195 weakly suppresses closing bracket ending an aside
  11. L14:2596 weakly excites
  12. L4:535 weakly suppresses
  13. L20:966 weakly suppresses
  14. L5:2085 weakly excites
  15. L11:3018 weakly excites
  16. L15:1505 weakly suppresses
  17. L8:2319 weakly suppresses
  18. L7:1352 weakly suppresses
  19. L10:1572 weakly suppresses
  20. L19:17 weakly excites
  21. L17:1616 weakly suppresses
  22. L4:539 weakly suppresses
  23. L15:2628 weakly excites
  24. L13:2864 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 -2.05 and -1.931 values between -1.58 and -1.461 values between -1.46 and -1.342 values between -1.10 and -0.982 values between -0.86 and -0.744 values between -0.62 and -0.504 values between -0.50 and -0.3856 values between -0.38 and -0.26299 values between -0.26 and -0.141220 values between -0.14 and -0.021771 values between -0.02 and 0.10404 values between 0.10 and 0.2269 values between 0.22 and 0.3430 values between 0.34 and 0.469 values between 0.46 and 0.588 values between 0.58 and 0.696 values between 0.69 and 0.813 values between 0.81 and 0.934 values between 0.93 and 1.052 values between 1.05 and 1.171 values between 1.17 and 1.292 values between 1.29 and 1.411 values between 1.53 and 1.652 values between 2.13 and 2.252 values between 2.25 and 2.371 values between 2.37 and 2.492 values between 2.61 and 2.73offnegativepositiveoften

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. 了多少
  2. main
  3. 旁
  4. rozen
  5. чив
  6. seen
  7. ousands
  8. _fa
  9. /main
  10. чат

Suppresses

  1. evity
  2. .itemView
  3. rollers
  4. iện
  5. stellen
  6. "));⏎
  7. agger
  8. ughter
  9. 械
  10. venge