Neuron Explorer

Qwen3-0.6B

the 'lo' fragment in loom and loaves

A neuron in layer 26 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 small chunk 'lo' inside words like 'loom', 'loins', 'loaves' and 'loo'. The opposite side responds to the letter 'b' starting words like 'bumpy' and 'bemoan'. Steering did not show a clear theme.

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

    the second best with the red cordon, and the third with the green. The thread is girt about their loins, and no ribbon of the Legion of Honour, or Knight of the G

  2. #2 peak

    wright thought for one moment of the hardship and misery he was about to unleash on a vulnerable section of society, he would never have sat down to invent the power loom.⏎Cartwright (1743-1823) was

  3. #3 peak

    Museum’s Textile Collection contains over four thousand patent models. The collection includes many examples of carding machines, spinning machines, knitting machines, rope making machines, looms, baskets, carpets, fabrics, and sewing machines. Even the simple clothes

  4. #4 peak

    doesn’t like to launch things across the room?!⏎Little engineers can use rubber bands to solve a STEM paintbrush engineering challenge!⏎Put together a quick rubber or loom band busy bag to take with you on travels or elsewhere and keep your toddler

  5. #5 peak

    14, 2004). To the right is a set of rocks referred to as "Rotten Rocks" for their resemblance to rotting loaves of bread. The insides of these rocks appear to have been eroded

  6. #6 peak

    a surplus of storable food, a system of writing, a more complex social organization, and technological advances such as the plough, potter's wheel, loom, and metallurgy.⏎Cities exist for many reasons, and the diversity of

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: there are some people who make money withlesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylesslylessly

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

    . Israel Sadovnik Socratus. =. P.S. ' They would play a greater<|endoftext|>The big house at the end of the long and bumpy road is rarely quiet. If the house’s occupants are outside, a visitor

  2. #2 peak

    frequently have the opposite phenomenon, dysphemism, where a relatively neutral word is replaced with a harsher, more offensive one.⏎Such as calling a cemetery a boney

  3. #3 peak

    taken to complaining about the lack of AM radio. Consumers took to forums to bemoan the omission, a few even saying they wouldn’t buy an i3

  4. #4 peak

    tois brothers were a mere hundred miles apart; Sévère had set up an independent government on Providencia Island, in the western Caribbean, and Joseph was embarking on a political career of his own in Port-au-Prince, Haiti.

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. L21:754 strongly excites
  2. L19:986 strongly excites
  3. L13:44 suppresses inside web addresses and citations
  4. L18:2403 excites
  5. L17:1061 excites
  6. L19:400 excites
  7. L10:39 suppresses
  8. L21:525 suppresses
  9. L20:541 excites
  10. L10:428 excites
  11. L20:54 excites
  12. L10:155 excites
  13. L18:178 excites
  14. L9:17 suppresses
  15. L19:134 suppresses
  16. L15:390 excites
  17. L18:550 suppresses
  18. L4:342 suppresses
  19. L12:158 suppresses
  20. L16:2766 excites
  21. L18:251 excites
  22. L23:126 suppresses
  23. L10:669 suppresses
  24. L18:233 excites

Feeds later layers

  1. L27:1916 weakly excites capital P starting a word
  2. L27:1402 weakly excites
  3. L27:418 weakly excites
  4. L27:1786 weakly excites
  5. L27:1347 weakly excites
  6. L27:709 weakly excites
  7. L27:930 weakly excites
  8. L27:1022 weakly excites
  9. L27:2148 weakly excites
  10. L27:2880 weakly excites
  11. L27:1333 weakly suppresses
  12. L27:2461 weakly excites
  13. L27:365 weakly excites
  14. L27:1984 weakly suppresses
  15. L27:1583 weakly excites
  16. L27:2448 weakly excites
  17. L27:1655 weakly excites
  18. L27:14 weakly excites
  19. L27:2080 weakly excites first piece of a split name
  20. L27:349 weakly suppresses
  21. L27:461 weakly excites
  22. L27:1007 weakly suppresses
  23. L27:2016 weakly excites
  24. L27:1454 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 -14.09 and -13.441 values between -10.22 and -9.571 values between -9.57 and -8.931 values between -5.70 and -5.061 values between -4.41 and -3.772 values between -3.12 and -2.486 values between -2.48 and -1.8341 values between -1.83 and -1.19301 values between -1.19 and -0.542014 values between -0.54 and 0.101331 values between 0.10 and 0.75176 values between 0.75 and 1.3917 values between 1.39 and 2.046 values between 2.04 and 2.682 values between 2.68 and 3.332 values between 3.33 and 3.971 values between 3.97 and 4.621 values between 5.91 and 6.551 values between 6.55 and 7.201 values between 7.84 and 8.491 values between 11.07 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. akest
  2. олод
  3. �
  4. enuity
  5. plets
  6. emark
  7. ilateral
  8. redirected
  9. uali
  10. ertas

Suppresses

  1. """⏎⏎⏎
  2. lessly
  3. hood
  4. 称之为
  5. gravy
  6. 市中心
  7. archy
  8. …”⏎⏎
  9. (QObject
  10. وجب