Neuron Explorer

Qwen3-0.6B

'and' or comma between paired items

A neuron in layer 13 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

Fires on the word 'and' or a comma that sits between two parallel names or items, such as 'officers and non-officers', 'David and Michael' or 'modelspace and paperspace'. It tends to go quiet on the space inside a number, such as 'an hour and 15 minutes'. Turning it up produced repetitive text, so the steering does not clearly confirm it.

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

    the Tata Employees Union (TEU) formed in 1935. There are two ranks of the workforce at Tata Steel, i.⏎e., officers and non-officers (workers who work in mines and plants and staff members). TW

  2. #2 peak

    d.1759 Madison County, Virginia⏎Facts and Events⏎About Mathias Cristler⏎Matthais and his two younger sons, David and Michael removed to the land grant located in the center of the present city of Madison

  3. #3 peak

    under power relays, directional or reverse current relays differential relays, and distance relays.⏎According to timing characteristics the relays can be divided into instantaneous, definite time lag, inverse time lag and inverse definite minimum time (IDMT)

  4. #4 peak

    it in the START menu on the bottom left corner of your computer screen.Ad⏎2Go to modelspace. There are two views in AutoCAD: modelspace and paperspace. Your drawing should always be made in modelspace, and dimensions that

  5. #5 peak

    long list should show up, displaying various toolbars containing different commands. The most popular toolbars that are used for AutoCAD 2D drawings are the DRAW, MODIFY, and OBJECT PROPERTIES toolbars. Select these toolbars<|endoftext|>February

  6. #6 peak

    RNA-protein complexes called small nuclear ribonucleoprotein⏎particles (snRNPs). The RNAs found in snRNPs are identified as U1, U2, U4, U5 and U6 (snRNAs).

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 community, but I have no means of doing it, but I can't do this, but this means is a community, but I have no means of doing this, but this is a community, but this is a

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

    adults get at least 2 1/2 hours of moderate-intensity cardio exercise each week. Alternately, you can perform an hour and 15 minutes of vigorous cardio exercise. By regularly performing cardio exercise, you can

  2. #2 peak

    quartz crystal, and measured 5 inches high, 7 inches long, and 5 inches wide, weighing 11 pounds, 7 ounces. It closely corresponds in size to a small human cranium, with

  3. #3 peak

    health care provider if⏎Call 911 or go to an emergency room if:⏎- You think vomiting is from poisoning⏎- You notice blood or dark, coffee-colored material in the vomit⏎Call a health care provider right away or

  4. #4 peak

    Holiday Crowd At The Zoo⏎- The Child Is Father Of The Man (Es⏎Past Perfect Tense Interrogative⏎Had + subject + verb 3rd from + object.⏎Past Perfect Interrogative T

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. L12:51 excites
  2. L10:838 excites comma in a list of paired nouns
  3. L6:427 suppresses
  4. L5:1469 suppresses
  5. L12:2808 suppresses
  6. L12:1505 suppresses
  7. L9:92 suppresses "and" or "or" joining two items
  8. L7:1223 suppresses "and" joining two decades or times
  9. L12:1786 excites
  10. L7:519 suppresses
  11. L8:1238 suppresses
  12. L6:1052 suppresses
  13. L6:1868 excites
  14. L6:1706 suppresses
  15. L8:70 excites 'and' between stacked superlatives
  16. L12:690 excites
  17. L8:1046 excites
  18. L12:1336 suppresses
  19. L1:1946 suppresses
  20. L8:2014 excites
  21. L9:1449 suppresses
  22. L10:1053 suppresses
  23. L8:131 weakly suppresses
  24. L7:498 weakly suppresses

Feeds later layers

  1. L14:976 suppresses
  2. L14:672 excites
  3. L16:2094 suppresses
  4. L15:2737 suppresses
  5. L22:1403 weakly suppresses
  6. L17:46 weakly excites
  7. L16:2048 weakly excites
  8. L14:745 weakly excites
  9. L14:3060 weakly suppresses
  10. L16:229 weakly excites
  11. L22:1029 weakly suppresses
  12. L15:1459 weakly suppresses
  13. L15:752 weakly suppresses
  14. L15:2598 weakly excites
  15. L15:670 weakly excites
  16. L17:1604 weakly suppresses
  17. L16:1179 weakly excites
  18. L14:1426 weakly suppresses
  19. L15:2201 weakly suppresses
  20. L16:316 weakly excites
  21. L16:300 weakly excites commas inside lists of nouns
  22. L19:2735 weakly suppresses
  23. L16:2521 weakly excites
  24. L16:485 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 -3.23 and -3.041 values between -2.84 and -2.652 values between -2.65 and -2.451 values between -2.45 and -2.253 values between -2.06 and -1.862 values between -1.86 and -1.662 values between -1.66 and -1.476 values between -1.47 and -1.278 values between -1.27 and -1.0712 values between -1.07 and -0.8822 values between -0.88 and -0.6833 values between -0.68 and -0.4871 values between -0.48 and -0.29541 values between -0.29 and -0.092504 values between -0.09 and 0.11533 values between 0.11 and 0.3065 values between 0.30 and 0.5023 values between 0.50 and 0.7013 values between 0.70 and 0.8922 values between 0.89 and 1.097 values between 1.09 and 1.294 values between 1.29 and 1.485 values between 1.48 and 1.686 values between 1.68 and 1.883 values between 1.88 and 2.072 values between 2.07 and 2.275 values between 2.27 and 2.475 values between 2.47 and 2.661 values between 2.66 and 2.861 values between 3.06 and 3.251 values between 3.45 and 3.651 values between 3.65 and 3.841 values between 4.43 and 4.63offnegativepositiveoften

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. m
  2. m
  3. ра
  4. M
  5. M
  6. oga
  7. ht
  8. hm
  9. ствовать
  10. hb

Suppresses

  1. condem
  2. rooting
  3. setMessage
  4. anness
  5. seksi
  6. disciplined
  7. setName
  8. atism
  9. elib
  10. decency