Neuron Explorer

Qwen3-0.6B

the 'w' in shipwreck

A neuron in layer 27 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 mostly fires on the letters 'w' in the word shipwreck, where the word is split into pieces. It goes strongly negative on a 'w' that starts other words, such as wax, wick and warts. Turning it up made the model drift into repeated words like 'product' and 'prediction'.

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

    /proclivities, etc. Psychometric profiles of civilian<|endoftext|>Recently we were called out to investigate a remnant shipwreck that had materialized on the shoreline near Georgetown, Prince Edward Island. The

  2. #2 peak

    alike. As for us, we were delighted and slightly surprised at the surge of interest. While we, too, were excited by the chance to investigate a shipwreck, for us it presented a unique archaeological site that required systematic documentation in order

  3. #3 peak

    m) wide. Among all of the 37 shipwrecks discovered at Yenikapi, there were a total of six Byz

  4. #4 peak

    found at a single location," said study author Cemal Pulak, of the Institute of Nautical Archaeology at Texas A&M University.⏎The eight shipwrecks highlighted in the new report, which date from the seventh to 1

  5. #5 peak

    , and even being raised from the sea.⏎Students could conduct a Rapid Research topic where groups look into 17 famous shipwrecks:⏎- The Mary Rose, 1545⏎- The

  6. #6 peak

    1821, through military action, foundering, grounding, fire and explosion, deliberate sinking to create artificial reefs, and other causes. Such shipwrecks are the major focus of the Florida Navy Legacy Shipwreck Project.⏎

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: since my own ownaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesaclesacles

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

    phonon charge-density wave coupling.<|endoftext|>Soy Wax vs. Paraffin Wax:⏎Candles can be made with different types of wax. The most common waxes used are paraffin and soy. Paraffin wax is a refined

  2. #2 peak

    exhibit. He has more prominent “warts” on his face, and his tusks are more apparent. Spartacus is still a pretty young guy, so those warts and tusks will get bigger as he gets older. He is also

  3. #3 peak

    -floating wick. Instead, it used a flame-enclosing burner and a wick bent into a cylinder shape, which provided the fire with just enough air.

  4. #4 peak

    on that migration and hence every child of Adam (as), weeps when he opens his eyes in this world. If this is true, then this reason for wailing certainly supports our point of view that in the grief of a departure, w

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. L22:2646 excites
  2. L20:417 excites
  3. L18:533 suppresses
  4. L23:400 excites
  5. L21:463 suppresses
  6. L0:3043 suppresses
  7. L23:206 suppresses
  8. L7:421 excites
  9. L18:365 excites
  10. L18:569 excites
  11. L16:67 weakly suppresses
  12. L2:2142 weakly suppresses
  13. L8:13 weakly suppresses
  14. L19:1015 weakly excites
  15. L26:1564 weakly excites
  16. L8:7 weakly excites
  17. L22:1511 weakly suppresses
  18. L6:13 weakly excites
  19. L13:810 weakly suppresses
  20. L19:196 weakly excites
  21. L7:287 weakly excites
  22. L8:4 weakly excites
  23. L22:1550 weakly suppresses
  24. L6:16 weakly excites

Feeds later layers

None: this is the last layer.

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 -32.41 and -29.862 values between -22.22 and -19.681 values between -17.13 and -14.583 values between -14.58 and -12.041 values between -12.04 and -9.491 values between -9.49 and -6.957 values between -4.40 and -1.853841 values between -1.85 and 0.6932 values between 0.69 and 3.244 values between 3.24 and 5.795 values between 5.79 and 8.333 values between 8.33 and 10.881 values between 10.88 and 13.421 values between 18.52 and 21.061 values between 23.61 and 26.152 values between 26.15 and 28.701 values between 59.25 and 61.801 values between 66.89 and 69.44offnegativepositiveoften

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. age
  2. -product
  3. yz
  4. product
  5. 味
  6. -products
  7. edd
  8. orn
  9. 效力
  10. ep

Suppresses

  1. ähnlich
  2. utation
  3. oose
  4. まして
  5. 쟁
  6. acles
  7. oteca
  8. 饵
  9. endwhile
  10. ница