Neuron Explorer

Qwen3-0.6B

blank spot where a number goes

A neuron in layer 17 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 responds to the empty space just before a number in citations, statistics and counts, such as "volume 7" or "about 30 per day." Its opposite side fires on the spot before a second number in a list. Turning it up or down gave writing that did not clearly relate to numbers.

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

    narica and Procyon lotor (Carnivora: Procyonidae): evidence of infection control?⏎Parasites & Vectors volume 7, Article number: 405 (2014)⏎A

  2. #2 peak

    might also be interested in:⏎About 70<|endoftext|>- freely available⏎Int. J. Environ. Res. Public Health 2012, 9(11), 4170-4186;

  3. #3 peak

    BMC Musculoskeletal Disorders volume 4, Article number: 4 (2003)⏎The purpose of this study is to present an

  4. #4 peak

    like the one examined in the current study. This could be because cancer of the salivary gland is normally quite rare, accounting, for example, for only 0.23 percent of all cancers diagnosed in the United Kingdom each year.

  5. #5 peak

    and 100 million Americans had received their COVID-19 shots, there were 3,544 reported deaths following COVID vaccination, or about 30 per day⏎- In just four months, the COVID-19

  6. #6 peak

    2) of it (0.67%) is water.⏎As of the census of 2000, there were 112 people, 46 households, and 32 families residing in the town. The population

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 a 300% change in my work I am doing in my lab. My work is only possible because my work I have been doing in my head is changed, and my research I have I made up all of my

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

    the goals of their academic programs.⏎Films and film clips⏎The site features 120 films from the NFB collection (60 in English; 60 in French in the French part). Of those, 48 (

  2. #2 peak

    in French in the French part) are available with closed-captioning. Users will also find 164 film clips (80 in English; 84 in French in the French part).⏎These 120 films are

  3. #3 peak

    December 2015. Overall, 99 sources from 62 countries were retrieved: 53 from 22 developed countries, and 46 from 40 developing countries. Representative guidelines from America (the USA

  4. #4 peak

    9 square miles (1,710 km2), of which 576 square miles (1,490 km2) is land and 83 square miles (210 km2) (13%) is

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. L16:256 excites number right after 'age of' or 'ages'
  2. L10:868 suppresses
  3. L10:220 suppresses hyphen inside compound words
  4. L13:1372 excites number before an age or week count
  5. L14:471 weakly excites
  6. L7:110 weakly suppresses
  7. L15:1285 weakly suppresses
  8. L16:372 weakly excites
  9. L6:756 weakly suppresses
  10. L3:122 weakly excites
  11. L2:75 weakly suppresses
  12. L15:2091 weakly excites
  13. L8:1745 weakly excites
  14. L14:1793 weakly suppresses
  15. L15:167 weakly excites
  16. L7:1524 weakly suppresses
  17. L6:1167 weakly excites
  18. L10:1184 weakly excites
  19. L3:59 weakly excites
  20. L15:1909 weakly suppresses blank slot before a number
  21. L10:2839 weakly suppresses
  22. L12:1420 weakly excites digits starting a percentage or rate
  23. L16:903 weakly excites
  24. L9:183 weakly suppresses

Feeds later layers

  1. L20:2749 weakly excites
  2. L18:2388 weakly excites
  3. L19:2500 weakly excites
  4. L18:2694 weakly excites
  5. L22:1436 weakly suppresses
  6. L22:1118 weakly suppresses
  7. L19:2011 weakly excites
  8. L26:237 weakly suppresses
  9. L26:1270 weakly excites
  10. L18:2844 weakly excites
  11. L21:1172 weakly suppresses
  12. L25:2783 weakly excites
  13. L18:1728 weakly excites
  14. L18:983 weakly excites
  15. L19:3018 weakly excites
  16. L18:659 weakly suppresses
  17. L25:2104 weakly excites
  18. L21:771 weakly suppresses
  19. L18:2454 weakly excites
  20. L20:1791 weakly suppresses
  21. L21:558 weakly excites
  22. L20:2470 weakly suppresses
  23. L21:2133 weakly excites
  24. L18:2188 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 -5.14 and -4.943 values between -3.32 and -3.121 values between -2.72 and -2.521 values between -2.52 and -2.322 values between -2.32 and -2.125 values between -2.12 and -1.911 values between -1.91 and -1.714 values between -1.71 and -1.512 values between -1.51 and -1.314 values between -1.31 and -1.118 values between -1.11 and -0.9110 values between -0.91 and -0.7117 values between -0.71 and -0.50109 values between -0.50 and -0.30599 values between -0.30 and -0.101950 values between -0.10 and 0.10929 values between 0.10 and 0.30211 values between 0.30 and 0.5031 values between 0.50 and 0.706 values between 0.70 and 0.913 values between 0.91 and 1.113 values between 1.11 and 1.313 values between 1.31 and 1.511 values between 1.51 and 1.711 values between 2.11 and 2.322 values between 2.52 and 2.721 values between 2.72 and 2.92offnegativepositiveoften

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. 芥
  2. 不方便
  3. -next
  4. 疤
  5. �
  6. Interstate
  7. 覚
  8. olutely
  9. ictionaries
  10. _inline

Suppresses

  1. rior
  2. ода
  3. inker
  4. 研究成果
  5. 果汁
  6. nio
  7. sky
  8. ratio
  9. 所所
  10. ertest