Neuron Explorer

Qwen3-0.6B

abbreviations before a period (Dr, Mr, e.g)

A neuron in layer 20 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron lights up on short abbreviations that are followed by a full stop, such as 'Dr', 'Mr', 'e.g' and 'Fig'. Turning it up did not produce a clear change in what the model wrote.

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 had absolutely nothing worry but fear itself” (pg. 6) c) “absolutely nothing to purchase and no money to purchase with it” (pg.⏎6) d) “nothing to see outside the boundaries of Maycomb County

  2. #2 peak

    said first and second control signals for input into said power management control circuit;⏎means connected to the output of said AC generator for developing a second error signal (e.g. e4) proportional to the difference in frequency between said AC generator and an

  3. #3 peak

    a uniform magnetic field as shown in Fig. (a).⏎(2) by rotating a magnetic field at a constant angular velocity within a stationary coil as shown in Fig. (a).⏎In either case, the generated voltage will be of sinusoidal

  4. #4 peak

    English. Others arrive with behaviour problems. Parents’ work schedules are often so inflexible that Federal Hill must cover basic logistics beyond school pick-up and drop-off. Ms Fernandez says she had to use her own car

  5. #5 peak

    two-fold; (1) to have students work with two contrasting design materials- one common and easily manipulated (e.g. wood) and one less common e.g plaster. (2) To cultivate hands-on material skills with the minimum of rapid

  6. #6 peak

    served in the cavalry in WWI. The flag flew over CDC for about 3 weeks, until the new one arrived. In 1995, Mr. Paine presented the flag to the museum. The 48-star,

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: [·····]*(‮ · × ×°× ·°··°×°°°°°°°°°°°°°°°°°°°°

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

    been given to you?⏎October 3rd, 2010, 12:50 PM⏎i started the homework⏎i dont know how to finish it here what i

  2. #2 peak

    -835-5322 (TTY) or at email@example.com.<|endoftext|>Table of Contents :⏎Top Suggestions Senior Kindergarten Writting Worksheet :⏎Senior Kindergarten Writting Worksheet In this worksheet your students will practice their handwriting

  3. #3 peak

    run using Run command. after selecting the path click on the next button.⏎Step 3: Set Run command⏎Now This is important to set run command.⏎Here you have to type the command name which you want to set.⏎Remember you can

  4. #4 peak

    value was changed to 20 in another(),so this prints 20. Thus final answer is 10 0 20. I hope i have been clear enough. ------------------ Come on in !! Drinks are on the house

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. L19:184 strongly excites
  2. L18:454 strongly suppresses
  3. L18:580 excites
  4. L15:735 excites
  5. L18:232 excites
  6. L10:58 suppresses
  7. L16:704 suppresses
  8. L15:2567 excites
  9. L11:1820 suppresses
  10. L16:2959 suppresses
  11. L17:581 excites
  12. L14:160 suppresses
  13. L10:1050 excites end of document marker token
  14. L15:1287 suppresses
  15. L16:269 excites
  16. L10:2374 suppresses
  17. L17:119 suppresses
  18. L16:926 suppresses
  19. L19:268 excites
  20. L11:308 suppresses
  21. L15:2788 excites
  22. L18:804 suppresses
  23. L19:710 suppresses
  24. L9:1544 suppresses

Feeds later layers

  1. L22:120 excites
  2. L26:732 suppresses
  3. L25:643 excites
  4. L23:225 excites
  5. L25:409 suppresses
  6. L25:171 excites
  7. L23:291 excites
  8. L22:201 suppresses
  9. L21:1291 excites
  10. L24:190 excites
  11. L27:279 excites
  12. L27:243 excites
  13. L27:215 excites
  14. L26:146 excites
  15. L24:363 suppresses
  16. L26:89 excites
  17. L25:114 excites
  18. L25:522 excites
  19. L23:749 weakly excites
  20. L25:932 weakly excites
  21. L21:359 weakly excites
  22. L26:134 weakly excites
  23. L27:134 weakly excites
  24. L27:250 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 -3.39 and -3.071 values between -2.75 and -2.432 values between -1.80 and -1.481 values between -1.48 and -1.1618 values between -1.16 and -0.84180 values between -0.84 and -0.52954 values between -0.52 and -0.201826 values between -0.20 and 0.12470 values between 0.12 and 0.44158 values between 0.44 and 0.7692 values between 0.76 and 1.0855 values between 1.08 and 1.4042 values between 1.40 and 1.7228 values between 1.72 and 2.0419 values between 2.04 and 2.3510 values between 2.35 and 2.6711 values between 2.67 and 2.995 values between 2.99 and 3.318 values between 3.31 and 3.635 values between 3.63 and 3.952 values between 3.95 and 4.278 values between 4.27 and 4.594 values between 4.59 and 4.911 values between 4.91 and 5.233 values between 5.23 and 5.551 values between 5.55 and 5.861 values between 5.86 and 6.181 values between 8.10 and 8.421 values between 9.06 and 9.38offnegativepositiveoften

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. ..
  2. l
  3. fie
  4. nn
  5. `
  6. .."
  7. s
  8. cls
  9. n
  10. amp

Suppresses

  1. onian
  2. ertime
  3. mPid
  4. ([[
  5. ��
  6. 鸰
  7. Seah
  8. 燧
  9. Pandora
  10. REW