Neuron Explorer

Qwen3-0.6B

'and' inside organisation or paired names

A neuron in layer 19 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 fires on the word 'and' when it joins parts of a named organisation or subject, such as 'Health and Medical Sciences'. It goes strongly negative on 'and' or 'or' linking a pair of contrasting words, and steering produced repetitive lists of 'and' or random words.

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

    , it is important to understand where people are acquiring information about the virus, whom they trust, and how to overcome mistrust and information. The Center for Health and Research Transformation (CHRT) and the Department of Learning Health Sciences (LHS

  2. #2 peak

    ongoing concern. And the challenges brought on by the unprecedented COVID-19 pandemic created even more roadblocks for many of us.⏎According to a survey from Environment and Climate Change Canada, just

  3. #3 peak

    Day, Dr Dan Horton and Professors Susan Lanham-New, Christian Heiss and Roberto La Ragione from the Faculty of Health and Medical Sciences explore the factors that would influence a healthier world for all.⏎What would

  4. #4 peak

    Education Level: N/A⏎Audience: Teachers; Practitioners⏎Sponsor: Pennsylvania State Dept. of Education, Harrisburg. Bureau of Adult Basic and Literacy Education.⏎Authoring Institution: Lutheran Children and Family Service, Philadelphia,

  5. #5 peak

    . 2005. Edge Effects in Line Intersect Sampling With. Journal Agricultural, Biological, and Environmental Statistics 10:460-477⏎Get the

  6. #6 peak

    higher education, particularly in respect of emerging models of teaching and learning.⏎Professor Laura Czerniewicz, director of UCT’s Centre for Innovation in Learning and Teaching (CILT), who was a principal investigator on the study along with Professor

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: a person who a person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person person

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

    whether it is practising gratitude yourself, with your family, partner, or finding ways to cultivate gratitude at work. The book explores why gratitude is different for men and women and the impacts this has on relationships, romantic or not. It provides practical

  2. #2 peak

    the amount of random error involved in measuring student performance. The risk is that teachers and schools may be wrongfully rewarded or punished because value-added techniques either over- or underestimated their students’ learning gains. However, no intelligent users of standardized testing would

  3. #3 peak

    , and blood pressure cuffs for village health workers to identify high-risk pregnancies and life-threatening post partum and neonatal infections.⏎Terma’s training on pre- and post-natal care (birthing practices, lactation and infant health) has

  4. #4 peak

    a pin), they are transferred to a tank in a building closer to the water, where water from the Peconic Bay (the water body between the North and South Forks of Long Island) is pumped through their tanks and they feed on

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. L18:2863 strongly suppresses
  2. L18:1413 strongly excites
  3. L16:1659 excites
  4. L17:2886 excites
  5. L12:51 excites
  6. L18:2238 suppresses
  7. L8:70 excites 'and' between stacked superlatives
  8. L9:92 suppresses "and" or "or" joining two items
  9. L16:2498 excites
  10. L18:104 suppresses
  11. L17:2250 suppresses
  12. L14:440 suppresses
  13. L16:520 suppresses
  14. L13:977 suppresses
  15. L16:1388 suppresses
  16. L16:376 suppresses 'or' joining alternatives
  17. L4:57 suppresses
  18. L16:492 suppresses
  19. L14:745 suppresses
  20. L18:1738 excites
  21. L17:1575 excites
  22. L17:1740 suppresses
  23. L16:2577 excites
  24. L14:706 excites

Feeds later layers

  1. L20:955 suppresses
  2. L20:439 suppresses
  3. L20:160 suppresses
  4. L21:167 weakly suppresses
  5. L22:202 weakly excites
  6. L20:165 weakly suppresses
  7. L20:1200 weakly suppresses 'and' in 'more and more' doubling
  8. L23:107 weakly excites
  9. L23:319 weakly suppresses
  10. L20:156 weakly excites
  11. L20:80 weakly excites
  12. L21:247 weakly excites colon after a said or explains verb
  13. L21:647 weakly suppresses
  14. L27:2150 weakly excites
  15. L20:331 weakly excites
  16. L20:969 weakly excites
  17. L20:337 weakly suppresses
  18. L21:1074 weakly excites
  19. L24:216 weakly excites
  20. L27:2544 weakly excites
  21. L20:300 weakly excites
  22. L26:980 weakly excites
  23. L22:168 weakly suppresses
  24. L22:255 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 -9.52 and -9.091 values between -9.09 and -8.661 values between -5.65 and -5.223 values between -4.79 and -4.363 values between -4.36 and -3.934 values between -3.93 and -3.504 values between -3.50 and -3.0710 values between -3.07 and -2.646 values between -2.64 and -2.2113 values between -2.21 and -1.7818 values between -1.78 and -1.3519 values between -1.35 and -0.9263 values between -0.92 and -0.49850 values between -0.49 and -0.062637 values between -0.06 and 0.37206 values between 0.37 and 0.8023 values between 0.80 and 1.2318 values between 1.23 and 1.667 values between 1.66 and 2.098 values between 2.09 and 2.526 values between 2.52 and 2.951 values between 2.95 and 3.381 values between 3.38 and 3.811 values between 3.81 and 4.241 values between 4.67 and 5.101 values between 5.10 and 5.531 values between 7.25 and 7.68offnegativepositiveoften

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. MaxY
  2. defensively
  3. DUCT
  4. lessly
  5. Davis
  6. ,{"
  7. vided
  8. 崖
  9. .plus
  10. ,Q

Suppresses

  1. ätze
  2. inion
  3. aska
  4. pired
  5. paced
  6. oding
  7. }`,
  8. ized
  9. ehen
  10.