Neuron Explorer

Qwen3-0.6B

'and' starting a place or person name

A neuron in layer 22 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 letters 'And' when they start a place or name, such as Andes, Andaman, Andhra or Andromeda. Turning it up made the model repeat the word 'yet', which the neuron does not clearly relate to its trigger.

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

    . So, for example, China with its population of 1.3 billion people gets 1,900 electoral-college votes, but even tiny Andorra with less than 500,000 people gets three electoral

  2. #2 peak

    their reflections off the serene lake. Visible in the distant background sky is the band of our Milky Way Galaxy, the Pleiades open clusters of stars and the Andromeda galaxy.”⏎A NASA original, this galaxy is appropriately named for it’s

  3. #3 peak

    the geology department, gave an account of the activities of his department.⏎Later, Dr. Bhushan told reporters that after the Sumatra earthquake, the Andaman and Nicobar islands experienced 200 aftershocks of the magnitude

  4. #4 peak

    ta⏎called a jubba. Both of these clothes are called "kurtas." In the state⏎of Andhra Pradesh, both of these options are common. Either of these shirts⏎would

  5. #5 peak

    emphasis on downwind new fires occuring.⏎The highest spots are in south central, north central and northeast South Dakota. Just this week, the area near Lake Andes is also considered at 800.⏎It's an important number to

  6. #6 peak

    largest consumer of cocaine and has been leading efforts to eradicate coca production in the Andes for decades.⏎Bolivia is the world's first biggest producer of cocaine

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 don't *f * *f * *f * *f * *f * *f * *f * *f * *f * *f * *f * *f * *f * *f * *f *

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 feeling?⏎Piper urges us to think for the glory of God. He demonstrates from Scripture that glorifying God with our minds and hearts is not either-or, but both-and. Thinking carefully about God fuels passion and affections for God.

  2. #2 peak

    it was written before the political philosophy known as Zionism was invented.⏎It is perhaps a pity that Lloyd George did not live long enough to see the fruition of the policy in the middle east that he and B

  3. #3 peak

    value of their hips, they include Rosa Rugosa (Japanese Rose), Sweetbriar (a.k.a. Shakespeare's Eglantine), and Rosa Blanda (Prairie Rose).⏎Probably originated by Dutch growers in 17

  4. #4 peak

    phalia and Bavaria continue to reverberate. They have once again raised our awareness of the fact that climate change is not just a vision of the future, but has already long since begun – that we can no longer afford to continue our

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. L21:478 suppresses
  2. L20:1764 suppresses
  3. L19:2294 suppresses
  4. L19:2080 suppresses ampersand joining two items
  5. L16:788 excites
  6. L20:1194 suppresses
  7. L21:870 excites
  8. L19:1667 excites
  9. L20:54 weakly suppresses
  10. L21:1380 weakly suppresses
  11. L19:214 weakly suppresses
  12. L21:583 weakly suppresses
  13. L4:57 weakly suppresses
  14. L12:3057 weakly excites
  15. L13:2192 weakly excites
  16. L17:473 weakly excites
  17. L18:1167 weakly excites
  18. L11:2733 weakly suppresses
  19. L6:2867 weakly excites
  20. L7:92 weakly suppresses
  21. L17:2269 weakly excites
  22. L10:39 weakly excites
  23. L21:15 weakly excites
  24. L10:797 weakly suppresses

Feeds later layers

  1. L27:2009 strongly excites
  2. L25:637 excites
  3. L23:89 weakly excites
  4. L25:801 weakly excites
  5. L27:1647 weakly suppresses
  6. L27:1280 weakly excites
  7. L23:292 weakly excites
  8. L23:1549 weakly suppresses
  9. L23:1127 weakly excites
  10. L27:1357 weakly suppresses
  11. L26:3045 weakly suppresses
  12. L26:1592 weakly suppresses
  13. L23:2480 weakly excites
  14. L23:1543 weakly suppresses 'up' in 'up to' limits
  15. L25:2882 weakly suppresses
  16. L27:776 weakly excites
  17. L25:2975 weakly suppresses
  18. L23:1665 weakly suppresses
  19. L25:1969 weakly excites
  20. L27:2824 weakly suppresses
  21. L24:278 weakly suppresses
  22. L27:884 weakly excites
  23. L26:3071 weakly suppresses
  24. L27:951 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.

2 values between -10.48 and -9.491 values between -9.49 and -8.501 values between -8.50 and -7.516 values between -7.51 and -6.525 values between -6.52 and -5.527 values between -5.52 and -4.5313 values between -4.53 and -3.5423 values between -3.54 and -2.5565 values between -2.55 and -1.56209 values between -1.56 and -0.572879 values between -0.57 and 0.42445 values between 0.42 and 1.4194 values between 1.41 and 2.4054 values between 2.40 and 3.3932 values between 3.39 and 4.3827 values between 4.38 and 5.376 values between 5.37 and 6.368 values between 6.36 and 7.3513 values between 7.35 and 8.345 values between 8.34 and 9.332 values between 9.33 and 10.324 values between 10.32 and 11.311 values between 11.31 and 12.301 values between 12.30 and 13.292 values between 14.28 and 15.281 values between 16.27 and 17.261 values between 17.26 and 18.251 values between 28.15 and 29.14offnegativepositiveoften

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. yet
  2. yet
  3. Yet
  4. Yet
  5. 且
  6. ainda
  7. yeti
  8. aún
  9. 而又
  10. 而且

Suppresses

  1. anny
  2. insic
  3. aps
  4. utter
  5. arest
  6. [:-
  7. uty
  8. NS
  9. azo
  10. ury