Neuron Explorer

Qwen3-0.6B

line break inside a sentence

A neuron in layer 6 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 hard line break that splits a sentence in the middle, as happens in text copied from older printed or wrapped sources. It is negative on words like "temperatures" after "record high." Turning it up gave unrelated text, so the steering doesn't confirm this.

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

    . - I was there too, but didn't⏎see you. I also will be working there from 3 to 5 tomorrow⏎7. When she⏎returns from Moscow, she'll tell us

  2. #2 peak

    uffed at the knees, a style that⏎involves folding the garment in half and tucking it under the waist. This style⏎of draping the⏎dhoti is known as the half-fold. The dhoti can be

  3. #3 peak

    all. But finally,⏎after years of waiting, and generally giving up I have run across a passage⏎which may give us the answer.⏎"The formality of⏎the marriage consists in drinking sake after a particular manner. The sake⏎is poured

  4. #4 peak

    too⏎You can have mutable values in the IO or (ST s) monads, if you need them.⏎> That is, I can't change a⏎> component of somethingNew without creating y. The point is there's⏎>

  5. #5 peak

    any of the other auberges were situated on a piazza, is not⏎certain but shall be discussed later on.⏎Unfortunately, only three of⏎Cassar's seven auberges have survived. The German aub

  6. #6 peak

    withdrawn temporarily from Fontaine-les-Croiselles⏎and neutralized fire from Fontaine Wood.⏎No. 2 Company (Ramsay) was⏎with the 6th C.I.B. less the 29th Battalion

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: 11,22,23,24,25,26,27,28,29,30. 30. Sum: 21,22,23,2

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

    Climate change has an impact on extreme heat waves and record high temperatures, making them more frequent. For example, summer of 2022 brought record high temperatures in Europe. One can compare maximum LST for summer months of 20

  2. #2 peak

    mph at Austin, TX, and gusts to 75 mph at Tulsa, OK.⏎Many cities in the central and eastern U.S. reported record high temperatures for the date. Highs of 100° at Erie, PA

  3. #3 peak

    feel tired after use at a computer; vision that is filmy; eyes that are bloodshot; eyes that are irritated or hurt.⏎Recent research has shown that Dry Eye Syndrome (DES) is a chronic inflammatory condition (like arthritis).⏎

  4. #4 peak

    “Dry Eye” is a term used to describe a group of conditions that cause irritation and damage to the ocular surface. There are many different forms of dry eye, each with a unique cause, and each with a unique set of treatment options

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. L3:604 excites
  2. L3:86 excites
  3. L4:1863 weakly suppresses
  4. L2:1494 weakly suppresses
  5. L3:182 weakly suppresses
  6. L4:863 weakly suppresses
  7. L4:2241 weakly suppresses
  8. L5:128 weakly excites
  9. L3:1563 weakly suppresses
  10. L3:1198 weakly suppresses
  11. L1:1464 weakly excites
  12. L0:1071 weakly excites
  13. L2:1666 weakly excites
  14. L5:414 weakly suppresses
  15. L4:2643 weakly suppresses
  16. L2:2229 weakly suppresses
  17. L4:1429 weakly suppresses
  18. L0:745 weakly excites
  19. L4:397 weakly excites
  20. L3:2737 weakly excites
  21. L2:935 weakly excites
  22. L3:1558 weakly excites
  23. L2:1873 weakly excites
  24. L4:133 weakly excites

Feeds later layers

  1. L16:151 weakly suppresses
  2. L17:1242 weakly excites
  3. L9:2594 weakly suppresses
  4. L10:1529 weakly suppresses
  5. L9:2808 weakly suppresses
  6. L18:863 weakly suppresses
  7. L16:2337 weakly excites
  8. L7:133 weakly excites line break inside a sentence
  9. L15:2995 weakly suppresses
  10. L11:777 weakly excites
  11. L7:1728 weakly excites
  12. L16:735 weakly excites
  13. L14:2385 weakly suppresses
  14. L15:446 weakly excites
  15. L13:1217 weakly excites page and volume numbers in citations
  16. L16:2437 weakly suppresses
  17. L24:2236 weakly excites
  18. L18:1473 weakly suppresses
  19. L10:3005 weakly suppresses
  20. L17:1650 weakly excites
  21. L16:2347 weakly suppresses
  22. L7:2233 weakly suppresses
  23. L16:793 weakly suppresses
  24. L17:2883 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 -1.13 and -1.071 values between -0.95 and -0.891 values between -0.71 and -0.651 values between -0.65 and -0.596 values between -0.53 and -0.4713 values between -0.47 and -0.4126 values between -0.41 and -0.3590 values between -0.35 and -0.29197 values between -0.29 and -0.23282 values between -0.23 and -0.17499 values between -0.17 and -0.11637 values between -0.11 and -0.05750 values between -0.05 and 0.01552 values between 0.01 and 0.07368 values between 0.07 and 0.13202 values between 0.13 and 0.20146 values between 0.20 and 0.2678 values between 0.26 and 0.3232 values between 0.32 and 0.3810 values between 0.38 and 0.4412 values between 0.44 and 0.501 values between 0.50 and 0.561 values between 0.56 and 0.621 values between 0.62 and 0.681 values between 1.22 and 1.28offnegativepositiveoften

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. 蘼
  3. ><![
  4. heels
  5. �
  6. ichi
  7. igrate
  8. PermissionsResult
  9. ämpfe
  10. реб

Suppresses

  1. 有效
  2. 蕉
  3. hana
  4. ヽ
  5. REFER
  6. 石榴
  7. 蓬
  8. GOODMAN
  9. erase
  10. ayas