Neuron Explorer

Qwen3-0.6B

parenthesis introducing an acronym

A neuron in layer 25 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 bracket that opens a short form right after its full name, such as the "(" in "Federal Emergency Management Agency (FEMA)". Turning it up made the writing fill up with the sound "pe" over and over, but the negative side also hits these brackets, so its role is not clean.

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

    , and gave glycerin-supplemented poultry feed to broilers that were 7 to 45 days old.⏎Dozier evaluated the apparent metabolizable energy (AME) intake—a standard measure of energy—for the study group and for a control

  2. #2 peak

    , for example, mentioned above as contributing to the dissolution of boundaries in education, can also be instrumental in reinforcing existing boundaries. The concept of Personal Learning Environments (PLEs), whether they extend control of existing learning platforms or enable the independence of

  3. #3 peak

    verse Childhood Experiences (ACEs). This team has developed a child ACE screening instrument in an effort to

  4. #4 peak

    with persistent breathing problems and other post-COVID complications.⏎The new method developed by KAUST — known as Deep-Lung Parenchyma-Enhancing (DLPE) — overlays

  5. #5 peak

    The Australian Government is making it easier for new migrants to learn and improve their English language skills by increasing the number of available hours through the Adult Migrant English Program (AMEP).⏎From today, migrants in Australia will

  6. #6 peak

    ussions, a multicenter, prospective diagnostic study was launched and led by the Advanced Research Core (ARC) of the Concussion Assessment, Research, and Education (CARE) Consortium. The consortium is a joint effort of the National Collegiate Athletics

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: what if the public input into the data and the data into the public input was dual? That is, the public input would carry the data and the data would carry the public input, such that both ends have a dual input and dual output

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

    do Fogo stratovolcano - photo © 2013 Carole and Frederic Hardy⏎Fogo - Cha das Caldeiras - Pico Pequeno - photo © 2013 Carole and Frederic

  2. #2 peak

    potentially predict a disaster, usually in regards to weather, and how to measure the impact of a disaster immediately after its impact is felt. The Federal Emergency Management Agency (FEMA) often has questions like these. One indicator they informally use is

  3. #3 peak

    the National Cancer Institute and the CDC analyzed cancer incidence and follow-up information from nine cancer registries that are part of the Surveillance, Epidemiology, and End Results (SEER) program, which covers approximately 10% of the U.S

  4. #4 peak

    lets mostly smooth; plants 2-6 m tall var. arborescens (T.& G.) A. Gray. Nutlets slightly wrinkled or pebbly; plants 0.5-3 m tall⏎Moist

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. L22:2948 strongly suppresses
  2. L22:109 strongly excites
  3. L24:2225 excites
  4. L23:1707 suppresses
  5. L22:285 suppresses
  6. L24:1585 suppresses
  7. L12:2194 excites
  8. L12:472 weakly excites
  9. L24:1623 weakly suppresses
  10. L9:337 weakly excites
  11. L4:2513 weakly excites
  12. L23:2829 weakly excites
  13. L24:235 weakly excites
  14. L21:44 weakly suppresses
  15. L13:249 weakly excites
  16. L23:54 weakly suppresses
  17. L6:2896 weakly excites
  18. L8:609 weakly excites
  19. L13:2600 weakly suppresses
  20. L8:1458 weakly excites
  21. L17:2930 weakly suppresses
  22. L22:2849 weakly suppresses
  23. L22:168 weakly excites
  24. L11:2138 weakly excites

Feeds later layers

  1. L27:178 strongly suppresses
  2. L27:273 excites
  3. L27:1188 excites
  4. L27:2083 excites
  5. L27:621 excites
  6. L26:2865 excites
  7. L27:796 excites
  8. L27:1059 weakly excites
  9. L27:1243 weakly excites
  10. L27:1525 weakly excites
  11. L26:869 weakly excites word before him or he
  12. L27:1563 weakly excites
  13. L27:1085 weakly excites
  14. L27:1334 weakly excites
  15. L26:1101 weakly excites
  16. L26:109 weakly suppresses
  17. L26:1356 weakly excites
  18. L27:2404 weakly excites
  19. L26:2102 weakly excites
  20. L27:1994 weakly excites
  21. L27:2268 weakly excites
  22. L27:909 weakly excites
  23. L27:2325 weakly suppresses
  24. L26:504 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 -9.20 and -8.572 values between -7.93 and -7.306 values between -6.02 and -5.393 values between -4.75 and -4.127 values between -4.12 and -3.482 values between -3.48 and -2.8415 values between -2.84 and -2.2121 values between -2.21 and -1.57100 values between -1.57 and -0.94529 values between -0.94 and -0.302825 values between -0.30 and 0.34328 values between 0.34 and 0.9725 values between 0.97 and 1.6111 values between 1.61 and 2.246 values between 2.24 and 2.882 values between 2.88 and 3.529 values between 3.52 and 4.153 values between 4.15 and 4.795 values between 4.79 and 5.421 values between 5.42 and 6.062 values between 6.06 and 6.701 values between 6.70 and 7.332 values between 7.97 and 8.601 values between 9.88 and 10.511 values between 15.60 and 16.23offnegativepositiveoften

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. (pe
  2. PE
  3. pe
  4. _pe
  5. Pe
  6. Pe
  7. pe
  8. PE
  9. -pe
  10. CE

Suppresses

  1. 公司章程
  2. Pandora
  3. 盱
  4. ImageSharp
  5. forth
  6. 工商银行
  7. AxisAlignment
  8. 磺
  9. istrator
  10. 打球