Neuron Explorer

Qwen3-0.6B

'but' joining contrasting clauses

A neuron in layer 11 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 'but' when it links two contrasting ideas, such as 'a small but obvious distinction' or 'the percentage differences were minor, but still noteworthy'. Turning it up did not produce a clear theme in the continuations.

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

    new st<|endoftext|>Can a small camera possibly have a large lens in the front?⏎There are really "large" lenses. This search is not the biggest one, but the look of it is great.⏎But you are referring to a large diameter compared

  2. #2 peak

    have an X-Ray? Luckily, it has been over a few years since I last had one, but I always remember the strange experience seeing those X-Ray photos hung up on display.

  3. #3 peak

    of references, but its sources remain unclear because it has insufficient inline citations. (May 2009)|⏎Lexicography is divided into two separate but equally important groups:⏎- Practical lexicography is the

  4. #4 peak

    . More important, but sometimes less emphasized (pun intended), is Rhythm.” For further<|endoftext|>By Robert

  5. #5 peak

    may be a slight but obvious distinction between New Years in North Korea and with the rest of the world.

  6. #6 peak

    had increased from 8% to 84%, and that the number of reported head injuries and potentially serious head injuries decreased.⏎The percentage differences were minor, but still noteworthy.⏎A similar analysis was conducted by the NHS in the United Kingdom.

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 future of the future of the future, the future of the future, the future of the future, the future of the future, the future of the future, the future of the future, the future of the future, the future of

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

    humans, activity levels vary widely from couch-potato-style inactivity to highly active athletic endeavors.⏎"We have a huge epidemic of obesity in Western society, and yet we have little understanding of what determines variation among individuals for voluntary exercise levels," Garland

  2. #2 peak

    medicine. Especially in India, Neem has been used for thousands of years.⏎Not many CPG companies in the US food industry are currently using Neem.⏎However, many hygiene products – such as

  3. #3 peak

    house quickly if a fire starts.⏎However, fewer people know how to prevent fires in a motor vehicle. Vehicle fires are among the most terrifying types of accidents, yet few people can name ways to prevent them or address them if they happen. Experienced

  4. #4 peak

    area and increase property values.⏎The students found 74 percent of people in the Millers Bay neighborhood had no knowledge of the benefits of shoreline restoration. However, after students educated them on the topic, more than 80 percent of homeowners

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. L9:1755 suppresses
  2. L10:339 suppresses
  3. L4:659 excites
  4. L3:919 excites
  5. L6:1054 excites
  6. L7:1380 weakly excites
  7. L9:2802 weakly suppresses
  8. L5:2775 weakly suppresses
  9. L3:2553 weakly excites
  10. L9:2516 weakly excites
  11. L8:1784 weakly excites
  12. L0:2223 weakly excites
  13. L9:1579 weakly suppresses
  14. L10:743 weakly suppresses 'therefore' or 'thus' after a subject
  15. L10:3039 weakly excites
  16. L8:2563 weakly suppresses
  17. L10:2742 weakly suppresses
  18. L4:1319 weakly excites the 'B' in vitamin B
  19. L1:2957 weakly suppresses
  20. L5:1859 weakly excites
  21. L5:314 weakly excites
  22. L2:219 weakly excites
  23. L8:487 weakly excites
  24. L2:2706 weakly excites

Feeds later layers

  1. L17:1730 strongly excites
  2. L12:387 excites
  3. L15:1791 excites
  4. L12:2005 excites
  5. L16:1488 excites
  6. L15:1422 excites
  7. L12:2622 excites
  8. L15:1948 excites
  9. L18:1682 weakly suppresses
  10. L19:2213 weakly excites
  11. L16:2408 weakly excites
  12. L14:2621 weakly excites
  13. L17:2269 weakly suppresses
  14. L24:1976 weakly excites
  15. L17:2707 weakly excites
  16. L12:2965 weakly suppresses
  17. L15:1345 weakly excites
  18. L17:2081 weakly excites
  19. L16:1470 weakly excites
  20. L17:337 weakly suppresses
  21. L16:2212 weakly excites
  22. L25:2067 weakly suppresses
  23. L20:1684 weakly excites
  24. L20:1908 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.

34 values between -1.01 and -0.69735 values between -0.69 and -0.372614 values between -0.37 and -0.05479 values between -0.05 and 0.2715 values between 0.27 and 0.585 values between 0.58 and 0.905 values between 0.90 and 1.223 values between 1.22 and 1.543 values between 1.86 and 2.173 values between 2.17 and 2.493 values between 2.81 and 3.131 values between 3.44 and 3.761 values between 5.35 and 5.671 values between 6.62 and 6.942 values between 8.85 and 9.171 values between 9.80 and 10.121 values between 10.44 and 10.761 values between 10.76 and 11.081 values between 11.39 and 11.71offnegativepositiveoften

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. ilda
  2. efs
  3. Mathematical
  4. lef
  5. hrs
  6. emd
  7. ämpfe
  8. íveis
  9. 并不代表
  10. lige

Suppresses

  1. asti
  2. ylon
  3. gtk
  4. ?”
  5. atism
  6. izione
  7. .DriverManager
  8. fl
  9. !”
  10. 或者