Neuron Explorer

Qwen3-0.6B

unclear: scattered technical token positions

A neuron in layer 7 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 does not have a clear pattern. Its strongest moments are on odd spots in technical text, like a split word in plantar fasciitis or a symbol in a formula. Turning it up gave wandering, unfocused 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

    etween generations, particle differ by their quantum number and mass). The first generation consist of the electron (e−⏎) and electron-neutrino (νe⏎). The second generation consist of the muon (μ−⏎) and mu

  2. #2 peak

    program-flow order. This greatly reduces delays based on data dependencies within a certain range of code. For example, the following fragment R1 <= mem\[R0\]; R2 <= R1+R2; R5 <= R5+

  3. #3 peak

    the ground. The following are interesting facts about your feet:⏎What Causes Plantar Fasciitis to Flare Up?⏎The best way to prevent plantar fasciitis from flaring up is by wearing your custom orthotic insoles regularly and

  4. #4 peak

    are as follows:⏎① The angular velocity of the gear blank is ω 1 and ω 1, and the angular velocity of the gear blank is

  5. #5 peak

    .photo.rectify can be run directly, specifying options in the command line or the GUI, or it can be invoked as OPTION 8 through i.ortho.photo. If invoked though i.ortho.photo, an interactive terminal is used to determine the

  6. #6 peak

    Whether you are gaining body fat, muscle mass, or healthy weight from pregnancy, the added pounds put extra strain on your feet. This can cause plantar fasciitis for the first time, or trigger a new bout once you have already healed

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 thinking about the most I can do to you to make you feel is something to do, I have been thinking about the most I can do you to make you feel is something to do, I have been thinking about the most

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

    http://www.tandfonline.com/pricing/journal/VECE20|⏎When requesting a correction, please mention this item's handle: RePEc:taf:jeduce:v:42:y:2011

  2. #2 peak

    itoneal cancers. Whether peritoneal cancer starts in the peritoneum or spreads from somewhere else (colon⏎Ovarian Epithelial, Fallopian Tube, and Primary Peritoneal Cancer …⏎(7 days ago) Treatment

  3. #3 peak

    68k-palmos-coff-gcc -v should report its version number to be (at time of writing) 22.214.171.124-kgpd-07109

  4. #4 peak

    CRIN website but might not be easily retrieveable.⏎See the Wiki here: http://wiki.crin.org/mediawiki/index.php?title=Main_Page⏎We are launching the Wiki with an initial batch of 41 country pages

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. L4:269 excites
  2. L5:1178 excites
  3. L5:1026 excites
  4. L6:2642 suppresses
  5. L5:443 weakly suppresses
  6. L4:313 weakly excites
  7. L0:1384 weakly suppresses
  8. L3:248 weakly suppresses
  9. L6:661 weakly excites
  10. L0:2445 weakly excites
  11. L0:1471 weakly excites
  12. L4:1415 weakly excites
  13. L2:2428 weakly suppresses
  14. L3:1930 weakly excites
  15. L6:2415 weakly suppresses
  16. L3:2666 weakly suppresses
  17. L6:2728 weakly excites
  18. L6:427 weakly excites
  19. L5:379 weakly suppresses
  20. L1:234 weakly suppresses
  21. L4:23 weakly excites
  22. L5:794 weakly suppresses
  23. L5:2278 weakly excites
  24. L4:1262 weakly excites

Feeds later layers

  1. L13:727 excites
  2. L9:392 excites "and" inside organisation names
  3. L8:2231 weakly excites
  4. L8:198 weakly suppresses
  5. L12:2058 weakly suppresses
  6. L10:1528 weakly excites
  7. L10:1677 weakly excites
  8. L9:722 weakly suppresses
  9. L11:1794 weakly excites
  10. L12:815 weakly suppresses
  11. L8:405 weakly excites
  12. L10:2413 weakly excites
  13. L19:2468 weakly excites
  14. L21:2773 weakly suppresses
  15. L11:127 weakly suppresses
  16. L10:21 weakly excites
  17. L11:1381 weakly excites
  18. L9:509 weakly excites
  19. L12:2972 weakly excites
  20. L20:2939 weakly suppresses
  21. L14:347 weakly excites
  22. L9:554 weakly suppresses
  23. L9:976 weakly excites
  24. L25:474 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 -0.56 and -0.531 values between -0.53 and -0.512 values between -0.43 and -0.413 values between -0.41 and -0.3811 values between -0.38 and -0.366 values between -0.36 and -0.3318 values between -0.33 and -0.3121 values between -0.31 and -0.2929 values between -0.29 and -0.2653 values between -0.26 and -0.2452 values between -0.24 and -0.2169 values between -0.21 and -0.19120 values between -0.19 and -0.16140 values between -0.16 and -0.14208 values between -0.14 and -0.11287 values between -0.11 and -0.09343 values between -0.09 and -0.06428 values between -0.06 and -0.04505 values between -0.04 and -0.01577 values between -0.01 and 0.01387 values between 0.01 and 0.04227 values between 0.04 and 0.06153 values between 0.06 and 0.0987 values between 0.09 and 0.1153 values between 0.11 and 0.1435 values between 0.14 and 0.1627 values between 0.16 and 0.1912 values between 0.19 and 0.2110 values between 0.21 and 0.2310 values between 0.23 and 0.2611 values between 0.26 and 0.2815 values between 0.28 and 0.312 values between 0.31 and 0.333 values between 0.33 and 0.361 values between 0.36 and 0.381 values between 0.41 and 0.43offnegativepositiveoften

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. лим
  2. iforn
  3. 幸
  4. _.
  5. cente
  6. _',
  7. __;⏎
  8. Liberation
  9. leveled
  10. _->

Suppresses

  1. 众人
  2. ately
  3. \xff
  4. ati
  5. atz
  6. azy
  7. rite
  8. 混合
  9. 全员
  10. 的世界