Neuron Explorer

Qwen3-0.6B

'reasons' or 'permitting' after a qualifier

A neuron in layer 1 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 the word that finishes a short phrase giving a justification, such as 'for health reasons', 'weather permitting' or 'if time allows'. When turned up, the model's writing did not show a clear link to this idea.

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

    " is commonly used for distribution systems with voltages above 1 kV and generally applied up to and including 52 kV . For technical and economic reasons, the service voltage of medium voltage distribution networks rarely exceeds 35 kV

  2. #2 peak

    world around them, but that they can make choices that reduce their impacts on the land, water, air, and other living and/or natural systems. If time allows, ask the students to write about some of the connections that come to mind in

  3. #3 peak

    an exception to copyright protection called fair use. This important part of copyright law supersedes the author's exclusive right to duplicate and distribute his works for public policy reasons, recognizing the tension between an author's ownership rights and society's desire to use

  4. #4 peak

    (EGR) systems are evaluated, as such systems will probably be applied on future engines due to expected new NOx emission regulations, as well as for economical reasons of the engine production.⏎To confirm the potential, a reduced-air-flow test has

  5. #5 peak

    an individual and physical examination must be performed in a respectful and culturally sensitive manner. Privacy is essential and the nurse must be aware of their own ...⏎You

  6. #6 peak

    D.C., at age 67.⏎In 1930, the chief justice of the United States, William Howard Taft, resigned for health reasons. (He died just over a month later.)⏎In 1943

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 need to take more time to get a good look, but also to take care about the time I have spent on my work. I will take a look at the photos of the best fashion brands, and try to find the right

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

    & Aid⏎- Economy & Trade⏎- Human Rights⏎- Global Governance⏎- Civil Society⏎Sunday, September 19, 2021⏎MONTE

  2. #2 peak

    asters,”UN News Service, 13 June 2006.<|endoftext|>On Thursday, December 15, 2016 the Civil Society Working Group on Synthetic Biology circulated a short fact sheet to help clear up four of

  3. #3 peak

    who have studied climate change: Human activities since the Industrial Revolution have been pumping greenhouse gases into the atmosphere, and this is not good.⏎"Warming of the climate system is unequivocal," stated the leading authority on climate change, the Intergovernment

  4. #4 peak

    -forms on it evolving slowly over hundreds of millions of years.⏎That seems to contradict the biblical version of creation, with God making the earth and its many life-forms rapidly.and probably not all that long ago.⏎I brought my boys Austin and

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. L0:1224 suppresses
  2. L0:178 suppresses
  3. L0:1301 suppresses
  4. L0:2067 weakly suppresses
  5. L0:924 weakly suppresses
  6. L0:921 weakly excites
  7. L0:1571 weakly excites
  8. L0:1369 weakly suppresses
  9. L0:794 weakly excites
  10. L0:1834 weakly suppresses
  11. L0:1299 weakly suppresses
  12. L0:1799 weakly excites
  13. L0:225 weakly suppresses
  14. L0:1118 weakly suppresses
  15. L0:909 weakly excites
  16. L0:1203 weakly excites
  17. L0:2591 weakly excites
  18. L0:1582 weakly suppresses
  19. L0:2229 weakly excites
  20. L0:659 weakly suppresses
  21. L0:176 weakly excites
  22. L0:1213 weakly suppresses
  23. L0:837 weakly suppresses
  24. L0:1674 weakly excites

Feeds later layers

  1. L2:1254 excites
  2. L9:1655 weakly suppresses
  3. L15:2193 weakly excites
  4. L9:199 weakly excites
  5. L21:1004 weakly excites
  6. L2:2494 weakly excites
  7. L11:604 weakly excites
  8. L10:783 weakly suppresses
  9. L16:1604 weakly excites
  10. L13:2951 weakly excites
  11. L11:558 weakly suppresses
  12. L4:1387 weakly excites
  13. L10:1040 weakly excites
  14. L6:413 weakly suppresses
  15. L21:258 weakly excites
  16. L15:2203 weakly suppresses
  17. L10:2317 weakly suppresses
  18. L24:569 weakly suppresses
  19. L5:2634 weakly suppresses
  20. L8:3037 weakly suppresses
  21. L5:767 weakly suppresses
  22. L15:296 weakly excites
  23. L8:2264 weakly suppresses
  24. L12:1147 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 -0.87 and -0.803 values between -0.80 and -0.722 values between -0.72 and -0.641 values between -0.64 and -0.571 values between -0.57 and -0.492 values between -0.49 and -0.415 values between -0.41 and -0.3421 values between -0.34 and -0.2657 values between -0.26 and -0.18230 values between -0.18 and -0.11770 values between -0.11 and -0.031688 values between -0.03 and 0.05771 values between 0.05 and 0.12257 values between 0.12 and 0.2074 values between 0.20 and 0.2811 values between 0.28 and 0.352 values between 0.35 and 0.432 values between 0.43 and 0.514 values between 0.51 and 0.582 values between 0.58 and 0.661 values between 1.12 and 1.191 values between 1.58 and 1.651 values between 1.96 and 2.041 values between 2.11 and 2.19offnegativepositiveoften

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. stood
  2. Jens
  3. skl
  4. ần
  5. ijing
  6. Devin
  7. (so
  8. mad
  9. Subview
  10. (Edit

Suppresses

  1. ste
  2. omencl
  3. Caval
  4. lyph
  5. 走路
  6. 复古
  7. ughters
  8. 当时的
  9. itudes
  10. 巡回