Neuron Explorer

Qwen3-0.6B

word fragments inside longer words

A neuron in layer 7 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron weakly fires on the middle or end of a longer word that is split up, such as 'ware' in 'Mind Ransomware' or 'istics' in 'Ballistics', and on digits in book ISBN numbers. The activations are small, so it is a weak signal. Turning it up gave unrelated text.

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

    not block you from accessing all of your files. However, it can do a lot of other things that can be quite annoying. In fact, Mind Ransomware can basically block you from operating your computer because it really wants you to pay attention

  2. #2 peak

    size and of a finite energy⏎- Infinitely massive, finitely old, of degenerate energy and uniform⏎1 of 20⏎Other questions in this quiz⏎2. Olbers' Paradox disproved Newton's model by

  3. #3 peak

    remove this infection from your system. It should be possible to decrypt your files with the key left by the application itself.⏎It is not clear how Mind Ransomware gets distributed around, but since the ransomware is “educational,” it is

  4. #4 peak

    5⏎Paperback ISBN: 9781503603844⏎Ebook ISBN:

  5. #5 peak

    2014|⏎|ISBN: ||9781312744967 (DRM-EPUB

  6. #6 peak

    the comparison of 45-70 vs 444 Marlin, which can be found when clicking that link.⏎45-70 Ballistics Chart:⏎You can use the following ballistics charts as a .45-

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 have been so busy writing this as a way of life, but I don't know how to let go of it for myself, and I am so used to the idea that the world is perfect and so I feel that I am in

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

    and the Ballotechnic model21 for the limits of complete reaction .⏎- Curious Case of Red Mercury⏎- “The new material is known as a 'ballotechnic' explosive, even though it does

  2. #2 peak

    that to snowpack,” said Matheus. “Looking at snow, rather than runoff, is an important distinction because it is the source of our water supply.⏎“When we have droughts, they might seem severe on a local time scale,

  3. #3 peak

    en their back.⏎Dr Lund said: "Scoliosis is a painful illness, and Richard would have been in quite a lot of pain on a daily basis.⏎"These methods could also

  4. #4 peak

    crossed out with a pen, and the word made department.⏎The rival company acting Torture got away with it successfully, especially, complains our Yorkshire priest "as 'ure' was pronounced 'yaw' in the best southern

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. L6:120 strongly excites
  2. L6:599 suppresses
  3. L3:92 excites
  4. L6:174 excites
  5. L2:240 weakly suppresses
  6. L3:1388 weakly excites
  7. L0:364 weakly suppresses
  8. L4:1124 weakly suppresses
  9. L5:2548 weakly excites
  10. L0:1067 weakly suppresses
  11. L5:2685 weakly excites
  12. L3:2005 weakly suppresses
  13. L3:103 weakly excites
  14. L0:2726 weakly suppresses
  15. L2:642 weakly excites
  16. L6:362 weakly suppresses
  17. L5:2725 weakly suppresses
  18. L1:197 weakly suppresses
  19. L1:1792 weakly excites
  20. L4:2091 weakly suppresses
  21. L0:2209 weakly suppresses
  22. L4:15 weakly suppresses
  23. L2:461 weakly suppresses
  24. L5:2113 weakly suppresses

Feeds later layers

  1. L8:2615 suppresses
  2. L19:1801 weakly suppresses
  3. L16:957 weakly suppresses
  4. L21:1246 weakly excites
  5. L9:915 weakly excites
  6. L23:2878 weakly excites
  7. L26:103 weakly excites blank space before a number
  8. L24:2827 weakly suppresses 'as' in 'as far as' and 'as many as'
  9. L26:1027 weakly excites
  10. L19:1406 weakly excites
  11. L13:1149 weakly suppresses
  12. L14:238 weakly suppresses
  13. L21:1189 weakly excites
  14. L27:1453 weakly excites
  15. L17:449 weakly suppresses
  16. L26:1268 weakly excites
  17. L26:2975 weakly excites
  18. L24:656 weakly excites
  19. L14:498 weakly suppresses
  20. L27:2879 weakly suppresses
  21. L26:2551 weakly suppresses
  22. L20:1428 weakly excites
  23. L26:2629 weakly excites
  24. L27:739 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 -2.29 and -2.211 values between -1.19 and -1.111 values between -1.03 and -0.951 values between -0.95 and -0.881 values between -0.88 and -0.801 values between -0.80 and -0.725 values between -0.72 and -0.644 values between -0.64 and -0.562 values between -0.56 and -0.486 values between -0.48 and -0.4022 values between -0.40 and -0.3346 values between -0.33 and -0.25124 values between -0.25 and -0.17310 values between -0.17 and -0.09678 values between -0.09 and -0.01964 values between -0.01 and 0.07798 values between 0.07 and 0.14570 values between 0.14 and 0.22232 values between 0.22 and 0.30108 values between 0.30 and 0.3820 values between 0.38 and 0.466 values between 0.46 and 0.542 values between 0.54 and 0.622 values between 0.62 and 0.691 values between 0.69 and 0.772 values between 0.77 and 0.85offnegativepositiveoften

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. *)⏎⏎
  2. ertiary
  3. handleMessage
  4. 们的
  5. [e
  6. mpz
  7. )["
  8. [res
  9. 省教育
  10. ("?

Suppresses

  1. ?,?,?,?,
  2. صل
  3. leaflet
  4. kiên
  5. fabric
  6. worm
  7. imension
  8. réalisé
  9. 共
  10. fold