Neuron Explorer

Qwen3-0.6B

name pieces inside a multi-part word

A neuron in layer 1 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on the middle or second half of a name or phrase that is split up, like 'Web' in 'Amazon Web Services' or 'ville' in 'Bartlesville'. It is weak overall, and turning it up did not produce a clear theme in the output.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    sites or client-side applications. Most of these services are not exposed directly to end users, but instead offer functionality that other developers can use in their applications. Amazon Web Services’ offerings are accessed over HTTP, using REST and SOAP protocols. All services

  2. #2 peak

    . The principal producing horizon is the Bartlesville sand, found at an average depth of 1,150 feet.

  3. #3 peak

    , services, and supports. Lifetime costs to care for an individual with ASD have been estimated to be $3.2 million. ⏎With funding from the American Recovery and Reinvestment Act (Recovery Act), the U.S. Department of Health

  4. #4 peak

    from the boundaries of asia to places as far away as Spain, New Zealand and the Americas. From the Migyur Dorje Stupa in Poolesville, Maryland to the Mahabodhi Stupa in Stokes Valley, New Zealand

  5. #5 peak

    Amazon Web Services⏎|Type of site||Web service, cloud computing|⏎Amazon Web Services (AWS) is a collection of remote computing services (also called web services

  6. #6 peak

    constant air exchange regardless of weather conditions,⏎- filtration of supplied air by means of special filters,⏎- savings on heating as a result of recovering heat<|endoftext|>In the beginning, they rallied behind Hitler in the national interest of Germany; in the end,

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: with the to. it. and. and. and. and. and. and. and. in. and. and. in. and. and. and. and.. and. for. and. up. and..

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

    of giant ice made up of frozen nitrogen mixed with carbon monoxide and methane.⏎It covers an area four kilometers deep and a thousand kilometers wide in the Sputnik Plain of Pluto.⏎To understand the origins of this

  2. #2 peak

    reputation that I don't think it really deserved," says Neufeld. "The Vanguard launch came after the success of Sputnik 1 and Sputnik 2,

  3. #3 peak

    .aug.25 » We have the Russians to thank for the Internet. In 1957 they won the race into space by launching Sputnik (satellite with a cool name), which remained in orbit for some 3

  4. #4 peak

    , has been unveiled in the Oeiras municipality in the Lisbon metropolitan area, the Russian embassy in Portugal told Sputnik.⏎“We will always remember the names of the first daredevils who made the

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. L0:1301 suppresses
  2. L0:1224 suppresses
  3. L0:178 suppresses
  4. L0:283 excites
  5. L0:1817 suppresses
  6. L0:851 weakly suppresses
  7. L0:1279 weakly excites
  8. L0:305 weakly excites
  9. L0:794 weakly suppresses
  10. L0:2093 weakly excites
  11. L0:632 weakly suppresses
  12. L0:1845 weakly suppresses
  13. L0:1940 weakly suppresses
  14. L0:1730 weakly suppresses
  15. L0:84 weakly suppresses
  16. L0:2599 weakly suppresses
  17. L0:1043 weakly excites
  18. L0:2634 weakly suppresses
  19. L0:591 weakly suppresses
  20. L0:2303 weakly excites
  21. L0:2124 weakly excites
  22. L0:2009 weakly suppresses
  23. L0:1069 weakly suppresses
  24. L0:1329 weakly suppresses

Feeds later layers

  1. L2:85 suppresses
  2. L3:94 suppresses
  3. L2:642 suppresses
  4. L2:325 suppresses
  5. L4:367 suppresses
  6. L2:302 suppresses
  7. L2:670 suppresses
  8. L2:1566 suppresses
  9. L2:93 excites
  10. L2:913 excites
  11. L3:676 suppresses
  12. L6:819 excites
  13. L2:427 excites
  14. L3:529 excites
  15. L4:218 excites
  16. L3:311 suppresses
  17. L2:1504 suppresses
  18. L5:305 suppresses
  19. L3:320 weakly suppresses
  20. L5:2680 weakly excites
  21. L2:1750 weakly suppresses
  22. L6:87 weakly suppresses
  23. L2:1467 weakly suppresses
  24. L19:1667 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.

1 values between -0.52 and -0.492 values between -0.46 and -0.433 values between -0.41 and -0.384 values between -0.38 and -0.354 values between -0.35 and -0.3215 values between -0.32 and -0.2913 values between -0.29 and -0.2639 values between -0.26 and -0.2448 values between -0.24 and -0.2154 values between -0.21 and -0.1885 values between -0.18 and -0.15162 values between -0.15 and -0.12188 values between -0.12 and -0.09344 values between -0.09 and -0.07497 values between -0.07 and -0.04513 values between -0.04 and -0.01386 values between -0.01 and 0.02299 values between 0.02 and 0.05247 values between 0.05 and 0.08220 values between 0.08 and 0.10148 values between 0.10 and 0.13148 values between 0.13 and 0.16121 values between 0.16 and 0.1998 values between 0.19 and 0.2280 values between 0.22 and 0.2557 values between 0.25 and 0.2731 values between 0.27 and 0.3027 values between 0.30 and 0.3319 values between 0.33 and 0.3613 values between 0.36 and 0.3912 values between 0.39 and 0.428 values between 0.42 and 0.446 values between 0.44 and 0.476 values between 0.47 and 0.503 values between 0.50 and 0.534 values between 0.53 and 0.561 values between 0.56 and 0.592 values between 0.59 and 0.61offnegativepositiveoften

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. sätze
  2. periment
  3. эм
  4. 'à
  5. filer
  6. 娶
  7. ramid
  8. ��
  9. ласт
  10. pler

Suppresses

  1. .HandlerFunc
  2. serviceName
  3. angelo
  4. Egypt
  5. eking
  6. RequiredMixin
  7. ittel
  8. izio
  9. ence
  10. acco