Neuron Explorer

Qwen3-0.6B

line breaks inside a sentence

A neuron in layer 4 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 when a line of text ends in the middle of a sentence, where the writing has been wrapped by hand or by a website. Turning it up did not produce a clear theme in the text it 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

    . - 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

    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

  3. #3 peak

    that changes how the⏎cd command works. Normally when you try and cd⏎into a directory, it

  4. #4 peak

    greeting us at the top of the⏎steps. Adorning each of the vessels are the origami butterflies, both⏎male and female.⏎Several years ago I⏎wrote a Shinto scholar who teaches at a major American university. I asked

  5. #5 peak

    that it⏎bridges them. Using a straw, blow air underneath⏎the paper. This

  6. #6 peak

    ," Professor Hacker said. "We want to understand⏎the physics behind it, so we can forecast it."⏎Russell White, of Byron Bay, was one⏎of two glider pilots who pioneered the art of soaring on a Morning Glory⏎

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 been thinking about myself. I have been thinking of me. I have been thinking of me. I have been thinking of me. I have been thinking of me. I have been thinking of me. I have been thinking of me

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

    1922, have been preserved and donated to Olympic National Park by Doreen Taylor, the wife of Fannie's grandson. In her diaries Fannie describes the day to day activities of community life, such as interactions with road

  2. #2 peak

    strict discipline. The treatments were tested at UA’s Disruptive Behavior Clinic in advance of Salekin’s partnership with DYS.⏎The teens and pre-teens Salekin and his staff treat have a variety of behavioral problems, like oppositional defiance

  3. #3 peak

    story of a young prince who’s father recently died. Hamlets uncle Claudius marries his mother the queen and takes the throne. As the play is told Hamlet finds out his father was murdered by the recently crowned king. The theme that

  4. #4 peak

    you to know about how long it will take before your skin begins to redden while using a product that features it, such as sunscreen. (2)⏎How UPF Is Measured⏎Typ

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. L3:131 suppresses
  2. L3:0 excites
  3. L3:155 suppresses
  4. L3:182 weakly suppresses
  5. L3:555 weakly suppresses
  6. L2:139 weakly excites
  7. L3:254 weakly suppresses
  8. L2:25 weakly suppresses
  9. L3:80 weakly suppresses
  10. L2:557 weakly excites
  11. L2:216 weakly excites
  12. L3:1034 weakly suppresses
  13. L2:270 weakly excites
  14. L2:464 weakly suppresses
  15. L3:1449 weakly suppresses
  16. L2:868 weakly suppresses line breaks inside lists
  17. L3:303 weakly excites
  18. L3:439 weakly suppresses
  19. L2:87 weakly suppresses
  20. L2:932 weakly excites
  21. L3:115 weakly suppresses
  22. L1:211 weakly excites
  23. L3:472 weakly suppresses
  24. L0:45 weakly excites

Feeds later layers

  1. L5:61 weakly excites
  2. L18:1012 weakly suppresses
  3. L7:264 weakly suppresses unclear: 'as' after a verb phrase
  4. L19:73 weakly excites
  5. L22:461 weakly suppresses
  6. L27:1866 weakly suppresses
  7. L11:2559 weakly suppresses
  8. L11:1130 weakly suppresses
  9. L18:42 weakly excites
  10. L23:268 weakly suppresses
  11. L17:370 weakly suppresses
  12. L20:2566 weakly suppresses
  13. L6:146 weakly excites
  14. L8:475 weakly suppresses
  15. L11:1571 weakly suppresses
  16. L15:11 weakly excites
  17. L9:1802 weakly suppresses
  18. L19:171 weakly excites
  19. L18:2141 weakly excites
  20. L22:215 weakly excites
  21. L22:1223 weakly suppresses
  22. L14:352 weakly excites
  23. L25:2529 weakly suppresses
  24. L18:70 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 -1.15 and -1.041 values between -0.93 and -0.821 values between -0.82 and -0.724 values between -0.72 and -0.6110 values between -0.61 and -0.5015 values between -0.50 and -0.3952 values between -0.39 and -0.29245 values between -0.29 and -0.18602 values between -0.18 and -0.071790 values between -0.07 and 0.04695 values between 0.04 and 0.15215 values between 0.15 and 0.2591 values between 0.25 and 0.3651 values between 0.36 and 0.4735 values between 0.47 and 0.5828 values between 0.58 and 0.6816 values between 0.68 and 0.796 values between 0.79 and 0.9012 values between 0.90 and 1.019 values between 1.01 and 1.116 values between 1.11 and 1.223 values between 1.22 and 1.334 values between 1.33 and 1.443 values between 1.44 and 1.543 values between 1.54 and 1.652 values between 1.76 and 1.871 values between 1.87 and 1.973 values between 1.97 and 2.081 values between 2.08 and 2.191 values between 2.40 and 2.511 values between 3.05 and 3.15offnegativepositiveoften

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. 而去
  2. 켓
  3. \Unit
  4. edian
  5. ęk
  6. nee
  7. ออกไป
  8. าศ
  9. LETTE
  10. ulle

Suppresses

  1. oling
  2. oled
  3. duino
  4. ruk
  5. efe
  6. '));⏎
  7. omo
  8. :');⏎
  9. ;</
  10. Dota