Neuron Explorer

Qwen3-0.6B

space between month and day number

A neuron in layer 18 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 gap between a month name and the day of the month, as in 'October 31' or 'September 26'. Turning it up or down did not produce anything obviously related to dates, so the effect on the model's writing is unclear.

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 part of a series of features highlighting freshman seminars offered at Princeton this semester. Other features:⏎Exploring the science and nuance of facial perception⏎Posted October 31, 2011; 12:00 p

  2. #2 peak

    -economic status, ethnicity, and religion?⏎- Are they conscious of themselves as having a “self?”⏎Submit your assignment to the Discussion Area by Saturday, November 22, 2014.<|endoftext|>Latest posts by Jessica Edwards (

  3. #3 peak

    Seehofer has blamed Merkel's refugee policy for the AfD's success.<|endoftext|>Posted by Jayme Blaschke⏎University News Service⏎September 26, 2011⏎Victor Frankenstein’s infamous monster

  4. #4 peak

    Annap. Harb. Ctr.: 410/266-1447.⏎Tiny Tots Fall for Nature Tues. Oct 12 (10:30am-noon) Three- to five

  5. #5 peak

    , plays, poetry and non-fiction texts.⏎All our teachers are English specialists.⏎For more information contact the head of department, Miss S Beadle.⏎In Years 10 and 11, all pupils study for the AQA GCSE

  6. #6 peak

    notes reflecting on the moment. Listen to learn what inspired the project and hear some of the writing that hikers left behind.⏎Current Event October 20, 2017⏎Middle and high school students can spend

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: is the * ( or ) and the the- is, in the is- or - to- the, or - is to, or the- is, or - to, or is, or is-

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

    , Serbia, Singapore, Spain, Sri Lanka, Sudan, Tajikistan, Tunisia, Turkey, Ukraine, Yemen⏎The Wiki is a web-based, multi-lingual and interactive project - this means we need your input to ensure the pages

  2. #2 peak

    16/18 vaccine may prevent cervical cancer just as effectively as three doses (Kreimer et al., 2011). A one- or two-dose program could lower the cost of vaccination and encourage the global dissemination

  3. #3 peak

    ]⏎The World As It Could Be Curriculum and Resource Guide for Teachers & Organization Leaders is now available through the Education Resources Information Center (ERIC) at http://www.eric.ed.gov/ERICWebPortal/detail?accno=ED532

  4. #4 peak

    is the most common example of a 'drying oil' but rapeseed, cottonseed, peanut, corn oil and safflower oil may also self-heat. Fish oils are notorious for their self-heating properties. Mineral oils like

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. L13:923 suppresses word right before a number or year
  2. L4:53 excites blank slot before a number
  3. L6:715 excites
  4. L10:220 suppresses hyphen inside compound words
  5. L14:140 suppresses
  6. L16:1407 excites
  7. L9:7 excites
  8. L7:193 suppresses space before day after month
  9. L2:404 excites
  10. L1:217 suppresses
  11. L3:421 suppresses
  12. L14:348 suppresses
  13. L4:185 weakly excites
  14. L16:396 weakly suppresses
  15. L6:17 weakly suppresses
  16. L3:342 weakly excites
  17. L4:490 weakly suppresses decade digit in years like 1930s
  18. L1:1609 weakly excites
  19. L4:628 weakly suppresses
  20. L8:350 weakly suppresses blank space before a year
  21. L5:97 weakly suppresses
  22. L0:1301 weakly suppresses
  23. L2:75 weakly suppresses
  24. L17:426 weakly suppresses

Feeds later layers

  1. L22:2384 suppresses
  2. L20:163 suppresses
  3. L23:239 suppresses
  4. L25:422 suppresses
  5. L21:302 suppresses
  6. L25:304 suppresses
  7. L27:553 suppresses
  8. L27:317 suppresses
  9. L27:981 suppresses
  10. L26:418 weakly suppresses
  11. L21:29 weakly suppresses
  12. L27:527 weakly suppresses
  13. L20:1779 weakly suppresses
  14. L26:526 weakly suppresses
  15. L21:1358 weakly excites
  16. L27:431 weakly suppresses
  17. L22:265 weakly suppresses
  18. L20:1067 weakly suppresses
  19. L22:372 weakly suppresses empty spot before a number or "3D"
  20. L26:459 weakly suppresses opening bracket before an acronym
  21. L27:344 weakly suppresses
  22. L25:513 weakly suppresses
  23. L27:2114 weakly excites
  24. L27:220 weakly suppresses end-of-text token between documents

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 -22.47 and -21.551 values between -10.49 and -9.572 values between -6.81 and -5.891 values between -5.89 and -4.961 values between -4.96 and -4.043 values between -4.04 and -3.126 values between -2.20 and -1.2828 values between -1.28 and -0.363751 values between -0.36 and 0.5696 values between 0.56 and 1.495 values between 1.49 and 2.413 values between 2.41 and 3.334 values between 3.33 and 4.252 values between 4.25 and 5.171 values between 5.17 and 6.091 values between 6.09 and 7.011 values between 8.86 and 9.781 values between 13.46 and 14.38offnegativepositiveoften

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. казан
  2. name
  3. Leap
  4. Pract
  5. practice
  6. innings
  7. nak
  8. Practice
  9. ứng
  10. Lent

Suppresses

  1. individual
  2. ǧ
  3. automatic
  4. →→⏎→→⏎→→⏎
  5. ugging
  6. aggregate
  7. uitar
  8. utar
  9. auty
  10. ipelines