Neuron Explorer

Qwen3-0.6B

'of' inside organization names

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

This neuron fires on the word 'of' when it appears inside the name of an institution, like 'Department of Natural Resources' or 'Georgia Institute of Technology'. It switches off on 'of' in phrases like 'Moore's law' or 'Newton's law of'. Turning it up or down did not give a clear theme in the writing.

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

    benefit to the host via reduction of antigen load from the gut.”⏎Intermittent Fasting May Provide Comparable Health Benefits to Calorie Restriction<|endoftext|>The Minnesota Department of Natural Resources announced Tuesday that an eaglet has hatched in the nest of two

  2. #2 peak

    and 4. Shell damage may be due to predation by fish (i.e. common carp or freshwater drum). Top photo by Mike Davis, Minnesota Department of Natural Resources; bottom photo by Gary Wege, U.S. Fish and Wildlife

  3. #3 peak

    learning that are essential for all students. They are the ideas, concepts, and skills that provide a foundation on which subsequent learning may be built. The Arizona Department of Education’s Virtual Learning Hub is a resource for teachers and families to assist them as

  4. #4 peak

    processing:⏎3. Masks that have to be subjected to a test for temperature resistance at 70°C over 24 hours before reprocessing:⏎Federal Ministry of Health. Use of protective masks in healthcare facilities. Paper from the crisis team with

  5. #5 peak

    of the Georgia Institute of Technology said in a video about the work.⏎Kemp and his team programmed a

  6. #6 peak

    paper is featured this month by the American Psychological Association on the Office of Ethnic Minority Affairs' African American Heritage Month website, found at: http://www.apa.org/pi/oema/resources/ethnicity-health/african-american/index.aspx⏎"So

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: it's the one with the three different states. It's the one with the three states of the one way? Or how exactly? So, in case, I can be divided into three states? Or how? Also, how does that

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

    and is known only to the Butonian people on the Pulau Makassar. The study used the symbolic interaction theory by Herbert Blumer. The identification of six informers was deemed viable. A

  2. #2 peak

    the mathematical tools needed to compute the motions of these bodies. Given initial positions and velocities, we can calculate the forces acting on each object, using Newton’s law of universal gravitation. From the forces we can determine accelerations, and then update

  3. #3 peak

    chip performance—measured by the number of transistors on a chip—doubles every 18 months.⏎It is no secret that Moore’s law, which has driven progress in the computer industry and the economy for decades, no

  4. #4 peak

    ⏎Your credit report is your window into your ID security. The Fair and Accurate Credit Transactions Act, passed by the Federal government in 2003, mandates that each of the major credit bureaus supply consumers with a free copy of

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. L8:2050 suppresses
  2. L3:616 suppresses
  3. L3:2793 excites
  4. L12:3035 excites
  5. L10:2129 weakly excites
  6. L12:708 weakly suppresses
  7. L15:1786 weakly suppresses
  8. L6:2394 weakly suppresses
  9. L10:448 weakly suppresses
  10. L3:1835 weakly excites
  11. L5:1132 weakly excites
  12. L4:1418 weakly suppresses
  13. L11:2177 weakly suppresses
  14. L4:1200 weakly suppresses
  15. L4:2616 weakly suppresses
  16. L7:1885 weakly suppresses
  17. L14:2147 weakly suppresses
  18. L11:1268 weakly excites
  19. L7:1063 weakly excites
  20. L15:1427 weakly suppresses
  21. L10:1070 weakly suppresses
  22. L15:718 weakly suppresses
  23. L11:1874 weakly suppresses
  24. L10:1755 weakly suppresses

Feeds later layers

  1. L19:1630 excites
  2. L20:706 excites
  3. L23:1696 weakly suppresses
  4. L19:2840 weakly excites
  5. L18:2599 weakly excites
  6. L20:2012 weakly excites
  7. L24:1241 weakly excites
  8. L17:779 weakly excites
  9. L17:110 weakly excites
  10. L26:1629 weakly excites
  11. L21:2450 weakly excites
  12. L18:2941 weakly excites
  13. L22:2320 weakly suppresses
  14. L17:3014 weakly excites
  15. L18:2868 weakly excites
  16. L17:1136 weakly excites
  17. L17:2491 weakly suppresses
  18. L21:1344 weakly excites
  19. L23:2585 weakly excites
  20. L20:2809 weakly suppresses
  21. L22:1813 weakly excites
  22. L20:606 weakly suppresses
  23. L20:1247 weakly excites
  24. L17:352 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 -7.95 and -7.692 values between -6.90 and -6.641 values between -6.11 and -5.851 values between -5.59 and -5.331 values between -4.80 and -4.542 values between -3.49 and -3.231 values between -2.97 and -2.713 values between -2.71 and -2.443 values between -2.44 and -2.187 values between -2.18 and -1.925 values between -1.92 and -1.6611 values between -1.66 and -1.3913 values between -1.39 and -1.1317 values between -1.13 and -0.8737 values between -0.87 and -0.61111 values between -0.61 and -0.35676 values between -0.35 and -0.081941 values between -0.08 and 0.18605 values between 0.18 and 0.44266 values between 0.44 and 0.70104 values between 0.70 and 0.9746 values between 0.97 and 1.2331 values between 1.23 and 1.4910 values between 1.49 and 1.756 values between 1.75 and 2.024 values between 2.02 and 2.283 values between 2.28 and 2.54offnegativepositiveoften

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. 用户提供
  2. 尊敬
  3. 用户的
  4. 箭
  5. mounted
  6. jumper
  7. 兴趣
  8. finish
  9. 击杀
  10. 👤

Suppresses

  1. plies
  2. og
  3. ocate
  4. icles
  5. ulate
  6. ulates
  7. ogi
  8. ibbean
  9. uento
  10. Framework