Neuron Explorer

Qwen3-0.6B

line breaks mid-sentence

A neuron in layer 3 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 invisible line break inside a sentence, where a line of text has been wrapped or split in the middle of a thought. The examples mostly come from text copied from PDFs or web pages with hard line wraps. Turning it up did not produce a clear theme in the replies.

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

    Understanding antonyms is part of vocabulary development and this worksheet offers good practice with antonyms there are some synonyms and other words mixed in too so it s important that This video first appeared as a facebook live on 15th october download a

  2. #2 peak

    actual practice that difficulties could⏎be resolved by discussion as they could not be resolved by force. Moreover,⏎the British Commonwealth was founded on the conception that war between⏎

  3. #3 peak

    as the uprising⏎of the British Empire. But we have no right to this position of special⏎and honorable responsibility unless we are determined to use it to the⏎full, not for our own

  4. #4 peak

    has said⏎that he would not have one Christian missionary in India if he rose to⏎power. And if the Zionist Jews became the governors of Palestine all missionary⏎work there

  5. #5 peak

    4. While the crops⏎could produce incredible yields when properly managed and with plenty of⏎water, doing so reduced the water table, added salt deposits to the⏎land, and made portions of the land unsuitable for

  6. #6 peak

    policy shifts in member states,⏎such as Greece, Italy the United Kingdom and Ireland. The Multireader⏎project examined complex navigation through digital materials, and the⏎TEDUB project covered a similar area for graphical information

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: the importance of the teacher., and a student, ready to help me with. Now, I need help here. Because of the the hard fact, I need. So, I have to help with. This is a very

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

    it was her own foolish choice and actions that lead to her and Romeo's death. Though the death of Romeo and Juliet was a tragic love story of two star-crossed lovers, caused by their families and friends, more specifically Friar Lawrence,

  2. #2 peak

    580.3 m (1,904 ft), bears the name of the Scot who discovered it in 1880. William Sutherland was one of several Scots who explored the country in the 19th century

  3. #3 peak

    for those that don’t have the ability to move to another state, the college they wish to attend is just a few mouse clicks away.⏎Is this where grade-school level education is headed? I can’t say for certain that is the case,

  4. #4 peak

    shapes - single, semi-double and double-flowered - is enormous. There are geraniums that can reach a height of one metre, but also creeping groundcover which only grows to a height of 10cm and hanging varieties which can

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. L2:2324 suppresses
  2. L1:1279 suppresses
  3. L2:323 excites
  4. L2:835 suppresses
  5. L2:1580 suppresses
  6. L2:2665 suppresses
  7. L1:557 excites
  8. L0:37 excites
  9. L1:999 weakly excites
  10. L2:1933 weakly suppresses
  11. L2:605 weakly suppresses
  12. L2:204 weakly excites
  13. L2:2786 weakly excites
  14. L1:428 weakly suppresses
  15. L1:1723 weakly suppresses
  16. L2:962 weakly excites
  17. L2:772 weakly suppresses
  18. L0:602 weakly suppresses words right before a number
  19. L0:538 weakly suppresses
  20. L0:2996 weakly suppresses
  21. L1:1960 weakly excites
  22. L2:2553 weakly suppresses
  23. L1:198 weakly suppresses
  24. L2:1486 weakly excites

Feeds later layers

  1. L4:67 suppresses
  2. L7:1666 weakly suppresses
  3. L8:2086 weakly excites
  4. L4:11 weakly excites
  5. L21:2755 weakly suppresses
  6. L13:2826 weakly excites
  7. L19:180 weakly excites
  8. L19:2223 weakly excites
  9. L22:1162 weakly suppresses
  10. L18:3005 weakly suppresses
  11. L9:1207 weakly suppresses unclear: mid-sentence tokens in web text
  12. L17:2460 weakly excites
  13. L8:2520 weakly suppresses
  14. L11:1847 weakly suppresses
  15. L7:2140 weakly suppresses
  16. L6:2546 weakly excites
  17. L13:226 weakly suppresses
  18. L6:1487 weakly suppresses
  19. L6:1799 weakly suppresses
  20. L9:2479 weakly suppresses
  21. L6:429 weakly suppresses
  22. L23:2777 weakly suppresses
  23. L16:2175 weakly excites
  24. L11:1369 weakly suppresses

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 -1.00 and -0.881 values between -0.76 and -0.651 values between -0.65 and -0.537 values between -0.53 and -0.4133 values between -0.41 and -0.29323 values between -0.29 and -0.171058 values between -0.17 and -0.051703 values between -0.05 and 0.07633 values between 0.07 and 0.19126 values between 0.19 and 0.3014 values between 0.30 and 0.423 values between 0.42 and 0.541 values between 0.90 and 1.022 values between 1.37 and 1.491 values between 1.73 and 1.851 values between 3.63 and 3.75offnegativepositiveoften

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. unting
  2. 所
  3. nię
  4. 权
  5. _structure
  6. usher
  7. uture
  8. uktur
  9. 位
  10. SIZE

Suppresses

  1. 一听
  2. ypi
  3. 鸨
  4. ¥
  5. /GL
  6. Thief
  7. raid
  8. 杀手
  9. Decay
  10. reator