Neuron Explorer

Qwen3-0.6B

period after abbreviations and dates

A neuron in layer 17 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 a full stop that follows an abbreviation or a date, such as "B.C.E.", "U.S.", "A.D." or "Jr.", where the sentence may or may not be over. Turned up, the model got stuck repeating phrases about time and days.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    .⏎English is Germanic and therefore based on Proto-Indo-European (PIE) language. PIE was spoken around 3500 B.C.E. by a group of nomads called the Aryans living in what is now Southern

  2. #2 peak

    the seeds of spruce and tamarack, so White-winged Crossbills live and thrive mainly in the conifer forests of the far northern U.S. as well as Canada. They also eat insects, fir seeds, and the occasional

  3. #3 peak

    by heft. Nonetheless, this particular book is the definitive guide to descents from the Magna Carta Barons of 1215 A.D. for over 200 individuals who emigrated from the British Isles to the

  4. #4 peak

    14% increase during the 1980s. The population density in 2000 was 41.5 per sq mi. The estimated

  5. #5 peak

    965, the white high school student traveled to Selma, Alabama, to participate in the march there, led by the Rev. Martin Luther King Jr. He stayed on as a volunteer to help register African-Americans to vote.⏎An invitation

  6. #6 peak

    War; interrelationships of economic, social, political, and intellectual developments.⏎This class covers the political, economic, social, and cultural history of the U.S. from the end of Reconstruction to the late 20th Century. It will

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 have always been working with the people their family... they are their family... their family... their... their family... their family... their... their... the family... their family... their... their... their...their...their... their

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

    thermal Community College⏎P.O. Box 804, Spindale NC 28160⏎(828) 395-1307 fax (828) 286-

  2. #2 peak

    9, 12). The exam will be comprehensive and will include material from the entire course. To get the most recent copies of the study guides click at left.⏎The field trip to Great Falls (MD) took place on Saturday April

  3. #3 peak

    is part of the UPMC family.⏎Be safe anytime, anywhere.⏎To find a pediatrician or pediatric specialist, please call 412-692-7337 or search our directory.⏎A resource for our network of

  4. #4 peak

    Missouri Extension⏎Published: Monday, March 18, 2013⏎David H. Trinklein, 573-882-9631⏎COLUMBIA, Mo. – It

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. L16:1939 strongly suppresses line break mid-sentence in wrapped text
  2. L15:286 strongly suppresses
  3. L14:409 suppresses
  4. L14:348 excites
  5. L16:1608 suppresses
  6. L8:517 excites
  7. L13:132 suppresses "of" after a noun
  8. L7:6 excites
  9. L13:1046 excites 'in' or 'on' ending a phrase
  10. L13:212 excites
  11. L15:316 excites
  12. L14:2677 suppresses
  13. L6:101 excites
  14. L10:217 excites
  15. L16:314 suppresses
  16. L14:1857 suppresses
  17. L15:1601 suppresses
  18. L6:1752 suppresses
  19. L14:917 suppresses
  20. L14:453 excites
  21. L10:272 excites
  22. L7:8 suppresses
  23. L12:792 excites
  24. L14:404 suppresses

Feeds later layers

  1. L19:1363 excites
  2. L18:93 suppresses
  3. L18:90 suppresses
  4. L19:223 suppresses
  5. L20:383 excites
  6. L20:2552 excites
  7. L18:1270 suppresses
  8. L19:1122 excites
  9. L19:219 suppresses
  10. L19:1935 excites
  11. L18:186 excites
  12. L19:413 excites
  13. L19:257 suppresses
  14. L19:1506 suppresses
  15. L18:1731 excites
  16. L19:1611 excites
  17. L18:1479 excites
  18. L18:602 excites
  19. L18:635 excites
  20. L21:81 weakly excites
  21. L20:1639 weakly excites
  22. L19:1204 weakly excites
  23. L20:405 weakly excites
  24. L18:1094 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.

2 values between -4.35 and -3.973 values between -3.97 and -3.597 values between -3.59 and -3.229 values between -3.22 and -2.8417 values between -2.84 and -2.4617 values between -2.46 and -2.0849 values between -2.08 and -1.70108 values between -1.70 and -1.32244 values between -1.32 and -0.94480 values between -0.94 and -0.56544 values between -0.56 and -0.19483 values between -0.19 and 0.19359 values between 0.19 and 0.57322 values between 0.57 and 0.95303 values between 0.95 and 1.33240 values between 1.33 and 1.71178 values between 1.71 and 2.09151 values between 2.09 and 2.47122 values between 2.47 and 2.8479 values between 2.84 and 3.2271 values between 3.22 and 3.6032 values between 3.60 and 3.9834 values between 3.98 and 4.3614 values between 4.36 and 4.7417 values between 4.74 and 5.124 values between 5.12 and 5.502 values between 5.50 and 5.875 values between 5.87 and 6.254 values between 6.25 and 6.633 values between 6.63 and 7.011 values between 7.01 and 7.391 values between 7.39 and 7.771 values between 8.15 and 8.531 values between 8.53 and 8.901 values between 10.42 and 10.80offnegativepositiveoften

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. ffset
  2. .Absolute
  3. uable
  4. mbH
  5. umed
  6. ategy
  7. iates
  8. ировка
  9. 硌
  10. ocities

Suppresses

  1. **
  2. ⏎
  3. [assembly
  4. [
  5. perfect
  6. [
  7. ####⏎
  8. ...
  9. [X
  10. و