Neuron Explorer

Qwen3-0.6B

'a' after prepositions like 'for' or 'by'

A neuron in layer 7 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 small word "a" when it comes after prepositions like "for", "by", or "of", as in "hit by a car" or "for a while". Its negative side fires on "is" and "'s" inside longer noun phrases. Turning it up gave repetitive or unrelated text, so the steering does not confirm the pattern.

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

    —a 500-meter stretch of it, anyway—transforms. Its markings glow in the dark.⏎It's an experiment that's been in the works for a couple of years at this point—an attempt to solve the problem of too-dark roads

  2. #2 peak

    are a testament to the brutality of Afghan religious fundamentalists. In the case of Palmyra,<|endoftext|>Let's admit it- we all try losing weight at some point in our lives. The primary step towards shedding those extra kilos is controlling

  3. #3 peak

    (Dubbele.com; Gr 9 Up; Free) app, which will bring users to the website. The site, which has been tracking elections for a number of years, includes detailed maps and

  4. #4 peak

    English teacher, is really nice, is going to give me a letter for college.⏎- The boy, that was playing in the street, that got hit by a car was my brother.⏎- My neighbor, that lives on the 2nd

  5. #5 peak

    , 2016<|endoftext|>What is Eczema?⏎Eczema is a form of dermatitis. The term Eczema, is given to a range of persistent skin conditions. Eczema is a recurring, non-infect

  6. #6 peak

    Vein treatment⏎- Scar removal⏎- Skin tightening⏎Cosmetic dermatology - Surgical procedures⏎Most dermatologists and dermasurgeons also offer treatments for a variety of medical skin conditions, in addition to cosmetic treatments. Although most skin problems

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: do you know a person well? I have the opportunity to be involved in your life. I am very interested in what the society, the country, and the people in it think of the country." I have the opportunity to be involved in

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

    2013. But Ratto and coons aren’t simply putting their designs, software and tutorials online and calling it a day: Just because something is available in theory doesn’t automatically make it accessible to all. (Linux, anyone

  2. #2 peak

    which leads to excessive consumption of critical nutrients such as added sugar, said INSP.⏎Interpreting the information presented in the GDA format to understand a product’s healthiness also requires consumers to do confusing mathematical operations, it said.⏎Undergraduate

  3. #3 peak

    species that exist in a given area. The healthier an area is, the more biodiversity it has. Different forms of wildlife and plants inhabit areas

  4. #4 peak

    and Facts about Value-Added Analysis November 11, 2011 > ˅⏎Value-added analysis uses standardized test scores to determine how much a teacher helped her students learn over the course of the year. Here are the facts

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. L4:49 excites
  2. L6:2818 excites
  3. L5:2842 suppresses
  4. L2:434 excites
  5. L1:427 excites
  6. L4:19 suppresses
  7. L5:94 weakly suppresses
  8. L2:426 weakly suppresses
  9. L4:137 weakly excites
  10. L4:2386 weakly excites
  11. L6:54 weakly suppresses
  12. L6:426 weakly excites
  13. L6:1579 weakly suppresses
  14. L4:567 weakly excites
  15. L5:1509 weakly excites 'is' in definitions
  16. L3:691 weakly excites
  17. L3:187 weakly excites
  18. L3:464 weakly excites
  19. L2:286 weakly suppresses
  20. L6:1449 weakly excites
  21. L6:263 weakly excites
  22. L0:146 weakly excites
  23. L6:2804 weakly excites
  24. L3:520 weakly suppresses

Feeds later layers

  1. L8:868 suppresses
  2. L10:1897 suppresses
  3. L12:462 suppresses
  4. L8:1056 suppresses
  5. L9:160 excites
  6. L11:2108 excites
  7. L10:802 suppresses
  8. L12:2312 suppresses
  9. L8:323 weakly excites
  10. L13:2508 weakly suppresses
  11. L8:2193 weakly suppresses
  12. L21:509 weakly excites
  13. L9:1905 weakly suppresses
  14. L17:2666 weakly suppresses
  15. L13:1694 weakly suppresses
  16. L9:931 weakly suppresses
  17. L8:89 weakly suppresses
  18. L12:1458 weakly suppresses
  19. L10:1425 weakly suppresses
  20. L10:1851 weakly suppresses
  21. L14:400 weakly excites
  22. L9:1272 weakly suppresses
  23. L10:694 weakly suppresses
  24. L12:1093 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 -4.59 and -4.431 values between -2.34 and -2.181 values between -1.86 and -1.702 values between -1.70 and -1.544 values between -1.54 and -1.387 values between -1.38 and -1.221 values between -1.22 and -1.063 values between -1.06 and -0.892 values between -0.89 and -0.739 values between -0.73 and -0.5722 values between -0.57 and -0.41171 values between -0.41 and -0.25884 values between -0.25 and -0.091706 values between -0.09 and 0.07836 values between 0.07 and 0.23201 values between 0.23 and 0.3940 values between 0.39 and 0.556 values between 0.55 and 0.713 values between 0.71 and 0.871 values between 0.87 and 1.031 values between 1.03 and 1.203 values between 1.20 and 1.361 values between 1.36 and 1.522 values between 1.68 and 1.84offnegativepositiveoften

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. 之心
  2. ございます
  3. Dys
  4. ritos
  5. ernote
  6. STDOUT
  7. -Token
  8. -cols
  9. ocate
  10. +offset

Suppresses

  1. poi
  2. 成龙
  3. forms
  4. uses
  5. uses
  6. ondon
  7. unda
  8. passe
  9. must
  10. use