Neuron Explorer

Qwen3-0.6B

mid-sentence words in web boilerplate

A neuron in layer 19 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on ordinary words in the middle of a sentence, such as 'or', 'is' and 'indicates', most often in book-site or paywall boilerplate like 'download or read it online'. It turned negative on words like 'see' and 'print' in access notices. The steering results did not show a clear theme, so its job is hard to pin down.

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

    stanz. The team of scientists headed by...⏎Yersiniae cause severe intestinal infections. Studies using Yersinia pseudotuberculosis as a model organism aim to elucidate the infection mechanisms of these...⏎Researchers from the

  2. #2 peak

    24 years old on Friday April 5th in 1974. On what day⏎of the week was she born?⏎A tetromino is made up of four squares joined edge to edge. Can⏎this tetrom<|endoftext|>

  3. #3 peak

    ?⏎Based on our data, we think this problem is relevant for Professor Finnegan's class at WSU.⏎What textbook is this problem found in?⏎Our data indicates that this problem or a close variation was asked in Chemistry: An Atoms First

  4. #4 peak

    measured unless subject to both ethical and commercial consideration.⏎Home page url⏎Download or read it online for free here:⏎by Flora Klickmann - G.

  5. #5 peak

    Major Characters Chapter Four: Complete Summary Chapter Five: Critical Analysis⏎To view this DRM protected ebook on your desktop or laptop you will need to have Adobe Digital Editions installed. It is a free

  6. #6 peak

    Physiology in Health and Illness 12e⏎Goodreads helps you keep track of books you want to read. Want to Read saving…. Want to Read Currently Reading Read. Other editions. Enlarge cover. Error rating book.⏎Browse more

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 have to be a man in a * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

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

    status as electors in August 1920 by the Nineteenth Amendment to the federal Constitution, the...⏎Users without a subscription are not able to see the full⏎to access all content.<|endoftext|>Great Slaty becomes World's

  2. #2 peak

    able solution for polygonal transects remains elusive.⏎Keywords: Boundary overlap, Inclusion region, Reflection method, Transect configuration⏎- We recommend that you also print this page and attach it to the printout of the article

  3. #3 peak

    2007. The result is encouraging. The model provides an objective basis for the development of acid rain forecasting operation in the province.⏎This is a preview of subscription content

  4. #4 peak

    .25 new (22% off) $21.95 direct from Amazon Amazon page|⏎Setup an account with your affiliations in order to access resources via your University's proxy server⏎Configure custom proxy (use this if your

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. L11:29 excites
  2. L18:396 suppresses
  3. L16:1543 suppresses
  4. L11:17 suppresses
  5. L16:248 excites
  6. L15:1317 excites
  7. L18:1077 excites
  8. L17:892 suppresses
  9. L10:61 suppresses
  10. L10:46 suppresses
  11. L0:2643 weakly suppresses
  12. L18:218 weakly excites
  13. L7:22 weakly suppresses
  14. L12:1115 weakly excites
  15. L10:124 weakly suppresses
  16. L16:132 weakly suppresses
  17. L15:1135 weakly excites
  18. L18:819 weakly excites
  19. L1:491 weakly excites
  20. L15:23 weakly suppresses
  21. L16:314 weakly excites
  22. L11:349 weakly excites 'of' or 'for' linking a place or group
  23. L10:333 weakly excites
  24. L17:682 weakly excites

Feeds later layers

  1. L26:1082 excites
  2. L20:543 suppresses
  3. L27:2134 weakly excites
  4. L24:1707 weakly excites
  5. L22:157 weakly suppresses
  6. L25:1914 weakly excites
  7. L22:304 weakly suppresses
  8. L21:81 weakly suppresses
  9. L24:1047 weakly excites
  10. L25:913 weakly excites
  11. L26:1653 weakly suppresses
  12. L25:1890 weakly excites
  13. L25:1668 weakly excites
  14. L25:741 weakly excites
  15. L20:1232 weakly suppresses
  16. L23:1886 weakly excites
  17. L20:71 weakly excites
  18. L25:440 weakly excites
  19. L22:108 weakly suppresses
  20. L27:990 weakly excites
  21. L20:128 weakly suppresses
  22. L22:1894 weakly excites
  23. L20:1719 weakly suppresses
  24. L25:2069 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 -6.08 and -5.664 values between -1.81 and -1.3862 values between -1.38 and -0.96341 values between -0.96 and -0.531011 values between -0.53 and -0.102361 values between -0.10 and 0.32103 values between 0.32 and 0.7510 values between 0.75 and 1.184 values between 1.18 and 1.613 values between 1.61 and 2.032 values between 2.89 and 3.311 values between 3.74 and 4.172 values between 5.02 and 5.451 values between 5.88 and 6.301 values between 10.15 and 10.571 values between 10.57 and 11.00offnegativepositiveoften

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. $b
  2. =query
  3. $max
  4. =value
  5. (vs
  6. :b
  7. 棣
  8. losion
  9. bles
  10. ном

Suppresses

  1. rop
  2. ro
  3. board
  4. ạnh
  5. Britann
  6. 要
  7. api
  8. Sole
  9. fo
  10. тя