Neuron Explorer

Qwen3-0.6B

'in' after 'published' or 'appears'

A neuron in layer 16 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

The neuron fires on the word 'in' when it follows a verb like 'published' or 'appears' and comes before the name of a journal or publication, such as 'published in Nature'. Turning it up made the model write about books, journals and publications, so this one is fairly clear.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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 study, “Differential roles for DNAJ isoforms in HTT-polyQ and FUS aggregation modulation revealed by chaperone screens,” was published in Nature Communications.⏎Neurodegenerative diseases, including ALS and Huntington’s,

  2. #2 peak

    .⏎"We could do it very simply, with a very fast and economic process that is easily incorporated in patient care," Venneti says. The study is published in Science Translational Medicine.⏎Traditional prognostic markers based on tumor grade have

  3. #3 peak

    23(2), 1976 pp. 125–135. doi:10.1112/S0025579300008718.<|endoftext|>

  4. #4 peak

    physiography,"⏎says George Zandt, a geophysicist with the University of Arizona in Tucson,⏎and lead author of the study, which appears in today's Nature.⏎"Normally, you have the greatest thickness underneath the tallest mountains

  5. #5 peak

    and life science director at Diamond Light Source, and Dr Bryan Charleston of The Pirbright Institute. The findings are published in the journal PLOS Pathogens.⏎'What we have achieved here is close to

  6. #6 peak

    6 miles long and 5 feet wide wedged between the two plates. A paper detailing the event will be published Wednesday (Dec. 11) in Nature.⏎Dyking events have been reported in the thin oceanic crust but

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 more the past and the future, the more the past and the future, and the more the past and the future, forever, for the past and the future, and forever, and the infinite, and the infinite, and the infinite

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

    : Daroff RB, Fenichel GM, Jankovic J, Mazziotta JC, eds. Bradley's Neurology in Clinical Practice. 6th ed. Philadelphia, PA: Elsevier Saunders; 2012:chap

  2. #2 peak

    activity disorder (ADHD) in North American children compared to placebo. A 2015 study by M.A. Stein and colleagues in the journal CNS Drugs extended this research, determining that guanfacine also improved academic and social functioning

  3. #3 peak

    .⏎Floral humidity, an area of elevated humidity within the headspace of the flower, has been demonstrated to occur in a number of flower species (Corbet et al., 1979

  4. #4 peak

    a submerged late Pleistocene shoreline feature that formed during a sea-level stillstand from 13.8 to 14.5 ka (Jarrett et al. 2005). The reef is currently 60–

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. L12:920 excites
  2. L5:1502 excites Bible and book words
  3. L10:2818 suppresses
  4. L6:2570 suppresses
  5. L13:2821 excites
  6. L13:1585 excites
  7. L10:2328 excites
  8. L11:2930 suppresses
  9. L10:2707 excites
  10. L9:1690 suppresses
  11. L8:1609 suppresses
  12. L10:2309 excites
  13. L11:2531 excites
  14. L7:2025 excites
  15. L9:824 excites
  16. L8:923 excites
  17. L8:2908 excites
  18. L15:1751 excites
  19. L9:2297 excites
  20. L9:2834 suppresses
  21. L15:2889 excites
  22. L7:1426 weakly suppresses
  23. L12:1794 weakly suppresses
  24. L15:2072 weakly suppresses

Feeds later layers

  1. L18:2293 excites
  2. L22:1468 excites
  3. L17:2231 excites
  4. L19:1896 excites
  5. L24:575 excites
  6. L17:1045 excites
  7. L19:2961 excites
  8. L20:1413 excites
  9. L21:1911 excites
  10. L27:2028 excites
  11. L17:2693 excites
  12. L20:1693 excites
  13. L23:2656 excites
  14. L20:223 excites
  15. L26:819 excites
  16. L17:2636 excites
  17. L25:3017 excites
  18. L19:2053 excites
  19. L21:612 suppresses
  20. L24:1365 suppresses
  21. L19:1426 weakly excites
  22. L20:1136 weakly excites
  23. L25:1854 weakly excites
  24. L19:2878 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.

2 values between -1.75 and -1.2316 values between -1.23 and -0.711100 values between -0.71 and -0.182615 values between -0.18 and 0.3491 values between 0.34 and 0.8627 values between 0.86 and 1.3821 values between 1.38 and 1.904 values between 1.90 and 2.424 values between 2.42 and 2.943 values between 2.94 and 3.466 values between 3.46 and 3.984 values between 3.98 and 4.503 values between 4.50 and 5.022 values between 5.02 and 5.541 values between 5.54 and 6.061 values between 6.06 and 6.581 values between 6.58 and 7.101 values between 7.10 and 7.631 values between 7.63 and 8.151 values between 8.15 and 8.671 values between 9.71 and 10.231 values between 10.23 and 10.751 values between 17.52 and 18.041 values between 18.56 and 19.08offnegativepositiveoften

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. published
  2. publications
  3. literature
  4. books
  5. .books
  6. 期刊
  7. textbooks
  8. /books
  9. 出版
  10. 刊

Suppresses

  1. /tags
  2. pios
  3. miss
  4. stration
  5. яд
  6. masked
  7. trust
  8. MISS
  9. legation
  10. uai