Neuron Explorer

Qwen3-0.6B

unclear: book and report titles

A neuron in layer 17 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 fires on the words inside titles and descriptions of books, reports and academic topics, but the pattern is scattered and it also fires weakly on random number and symbol fragments. Turning it up made the model drift into citation-like, academic-sounding text, though the other outputs were mostly unrelated.

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

    and 1930s, a time when avant-garde movements transformed writers' and readers' conceptions of literature. Urban Chroniclers in Modern Latin America: The Shared Intimacy of Everyday

  2. #2 peak

    in his country and intentionally gave the refugees who passed through Portugal during the war a hard—but not impossible—time.⏎In Hitler’s Jewish Refugees: Hope and Anxiety in Portugal, the distinguished historian of modern German Jewry Marion Kaplan is less interested

  3. #3 peak

    , please check our FAQs, and if you can't find the answer there, please contact us.<|endoftext|>Indigenous Peoples and Climate Change: from victims to change agents through decent work⏎The present report analyses the situation of indigenous peoples in the context

  4. #4 peak

    , often without organized funding or direct support from the church hierarchy. A comprehensive history of Roman Catholic nuns and sisters in the American West, Across God's Frontiers reveals Catholic sisters as dynamic and creative architects of civic and religious institutions in western communities

  5. #5 peak

    and Modern Ireland, 1782-1992 (1994).⏎Chapter 1 The Easter Rising 1⏎Chapter

  6. #6 peak

    s and now seeing renewed interest, uses liquid fuel made of fluoride thorium and a salt. There are safety advantages<|endoftext|>The Path from Discontentment to Contentment⏎Shakespeare’s play Richard III, Act-I, Scene-I, Lines

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: the idea of, but it's not, the idea of, but it's not, the idea of, but it's not, the idea of, but it's not, the idea of, but it's not, the idea of

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

    04) 13: 2511. doi:10.1023/B:BIOC.0000048451.07820.9c⏎- 1

  2. #2 peak

    ://mynursingpaper.com/wp-content/uploads/2020/06/cropped-155972031988413967-1-768x146

  3. #3 peak

    pp. 125–135. doi:10.1112/S0025579300008718.<|endoftext|>The Air We Breathe: Are Newer

  4. #4 peak

    mynursingpaper.com/wp-content/uploads/2020/06/cropped-155972031988413967-1-768x146

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. L13:1325 excites
  2. L14:1909 excites
  3. L15:1066 suppresses
  4. L15:2532 suppresses
  5. L10:270 suppresses
  6. L12:996 suppresses
  7. L16:2398 excites
  8. L13:852 excites
  9. L12:2494 suppresses
  10. L14:895 excites
  11. L13:1632 suppresses words inside headlines and titles
  12. L14:790 suppresses
  13. L14:2061 excites
  14. L16:2471 suppresses
  15. L12:1956 suppresses
  16. L16:1047 excites
  17. L15:2591 excites
  18. L15:2459 suppresses
  19. L16:590 excites
  20. L7:1656 excites
  21. L13:1797 suppresses
  22. L15:655 suppresses
  23. L6:234 suppresses
  24. L12:2654 excites

Feeds later layers

  1. L22:1712 excites colon before a subtitle
  2. L18:830 excites opening quote marks before a coined term
  3. L18:2412 excites
  4. L18:2706 excites
  5. L19:2461 excites
  6. L19:260 excites
  7. L23:526 excites
  8. L21:1078 excites
  9. L21:1067 excites
  10. L18:100 weakly excites
  11. L27:1103 weakly excites
  12. L21:2216 weakly excites
  13. L19:2780 weakly excites
  14. L19:2458 weakly excites
  15. L18:2611 weakly excites
  16. L20:36 weakly suppresses
  17. L19:950 weakly excites
  18. L18:2933 weakly excites
  19. L18:113 weakly excites
  20. L26:1721 weakly excites
  21. L21:2020 weakly excites
  22. L20:758 weakly excites
  23. L19:1896 weakly excites
  24. L18:1287 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 -2.24 and -1.835 values between -1.83 and -1.4279 values between -1.42 and -1.01767 values between -1.01 and -0.602102 values between -0.60 and -0.19773 values between -0.19 and 0.2268 values between 0.22 and 0.6436 values between 0.64 and 1.0514 values between 1.05 and 1.469 values between 1.46 and 1.878 values between 1.87 and 2.2811 values between 2.28 and 2.695 values between 2.69 and 3.107 values between 3.10 and 3.525 values between 3.52 and 3.931 values between 3.93 and 4.344 values between 4.75 and 5.161 values between 5.16 and 5.571 values between 5.57 and 5.983 values between 5.98 and 6.401 values between 6.81 and 7.221 values between 7.63 and 8.041 values between 8.45 and 8.861 values between 8.86 and 9.271 values between 10.10 and 10.511 values between 11.74 and 12.151 values between 12.98 and 13.391 values between 13.80 and 14.21offnegativepositiveoften

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. -part
  2. 部分
  3. PART
  4. ernel
  5. 》,
  6. SEMB
  7. 第九
  8. (parts
  9. ichte
  10. Part

Suppresses

  1. …"⏎⏎
  2. ..."⏎
  3. 啬
  4. )">⏎
  5. ;"⏎
  6. ..."⏎⏎
  7. clap
  8. ...">⏎
  9. urat
  10. =num