Neuron Explorer

Qwen3-0.6B

'set' as in 'set up'

A neuron in layer 25 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 'set' when it means to establish or arrange something, as in 'set up an office' or 'set up a warehouse'. When it was turned down, the model got stuck repeating 'establishment' and 'established', which are close in meaning to the trigger.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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 cores back on campus.⏎A new laboratory, the Cal-Dendro Tree Ring Laboratory, has been set up to conduct research on the cores, which are expected to reveal valuable data going

  2. #2 peak

    later known as the Cape Dutch) did not trek eastward, a small number did.⏎The first colonists, who arrived in 1652 to set

  3. #3 peak

    warehouse:⏎- Choose the right location to set up your warehouse by considering the transportation costs and areas of highest demand.⏎- Make sure that you have sufficient capital to set up the warehouse.⏎- Have an optimized floor plan to increase storage space, prevent

  4. #4 peak

    nations that have tense relations to continue some kind of working contact. The U.S. has operated “interest sections” in places like Cuba and Iran.⏎South Korea set up a liaison office in the North Korean city of Kaesong last year after

  5. #5 peak

    has remained mostly unused for more than 20 years. But it was once used as an office to handle U.S.-North Korean matters. The office was set up after an agreement signed in 1995 that gave Sweden permission to

  6. #6 peak

    to. The Trustee then holds that property for the trust’s beneficiaries. They also invest the trust assets and administer the trust terms according to the terms created by the Settlor.⏎Trusts all fall into one

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 falliversigoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoupoup

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

    use of epoxy-based resins and may be classified as the thermosetting epoxide polymer. Exteriors that receive this coating are resistant to ultra

  2. #2 peak

    weeks 5 days ago fairst put some water in a air tight box and put salt seasalt works best and put water in the container then let the salt desolve the more salt the better and put ure bait in a zipper bag and pop

  3. #3 peak

    million jobs being provided by the manufacturing industry, which employs 12 percent of the nation’s women and girls. That steady hum of progress in Dhaka--despite its unscrupulousness--has helped the very poorest break free of

  4. #4 peak

    to users worldwide. The company has developed a series of networks since it was founded in 1979. The withdrawal of RBGAN occurred because it was superseded by the newer BGAN technology.⏎BGAN signal acquisition requires line-of-site

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. L24:1292 strongly suppresses
  2. L24:1472 strongly excites
  3. L24:591 strongly suppresses
  4. L24:2351 strongly excites
  5. L23:1075 strongly suppresses
  6. L22:907 strongly suppresses
  7. L24:881 strongly suppresses
  8. L24:2511 strongly excites
  9. L23:1136 excites
  10. L22:2938 suppresses
  11. L24:1607 suppresses
  12. L24:1231 suppresses
  13. L23:2853 suppresses
  14. L23:698 suppresses
  15. L22:1476 excites
  16. L24:1693 excites
  17. L22:976 suppresses
  18. L20:810 suppresses
  19. L22:1531 suppresses
  20. L23:2430 weakly excites
  21. L22:1793 weakly excites
  22. L24:1276 weakly suppresses
  23. L23:573 weakly excites
  24. L24:1820 weakly suppresses

Feeds later layers

  1. L27:624 strongly suppresses
  2. L27:1741 strongly suppresses
  3. L26:2332 suppresses
  4. L26:2576 weakly suppresses
  5. L27:1233 weakly suppresses
  6. L27:1148 weakly suppresses
  7. L27:2138 weakly suppresses
  8. L27:2365 weakly suppresses
  9. L27:1225 weakly suppresses
  10. L27:2773 weakly suppresses
  11. L27:1158 weakly suppresses
  12. L26:2889 weakly suppresses
  13. L26:323 weakly suppresses
  14. L27:2507 weakly suppresses
  15. L26:2118 weakly suppresses
  16. L27:2569 weakly suppresses
  17. L26:2144 weakly suppresses
  18. L26:1296 weakly suppresses word-start 'mult' fragment
  19. L26:1626 weakly excites the word 'I' as subject
  20. L27:1093 weakly suppresses
  21. L26:66 weakly excites
  22. L27:971 weakly excites
  23. L26:1253 weakly suppresses
  24. L26:1798 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 -19.20 and -18.272 values between -7.12 and -6.192 values between -6.19 and -5.262 values between -5.26 and -4.335 values between -4.33 and -3.404 values between -3.40 and -2.4729 values between -2.47 and -1.54109 values between -1.54 and -0.612770 values between -0.61 and 0.32924 values between 0.32 and 1.2535 values between 1.25 and 2.189 values between 2.18 and 3.117 values between 3.11 and 4.042 values between 4.04 and 4.972 values between 4.97 and 5.901 values between 5.90 and 6.831 values between 7.76 and 8.691 values between 8.69 and 9.621 values between 9.62 and 10.551 values between 17.06 and 17.98offnegativepositiveoften

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. orners
  2. adamente
  3. itzer
  4. ウォー
  5. ič
  6. уд
  7. ARING
  8. Ames
  9. amide
  10. ط

Suppresses

  1. set
  2. Set
  3. Set
  4. (set
  5. sets
  6. set
  7. SET
  8. SET
  9. Sets
  10. 设置