Neuron Explorer

Qwen3-0.6B

unclear: mixed, mostly everyday health and advice prose

A neuron in layer 12 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

The neuron fires on a grab bag of words in ordinary sentences, such as 'status' in a web error, 'you' in 'glad you did', and 'is' after 'a healthy diet'. There is no single clear trigger. Turning it up gave unrelated text, so the meaning is not clear.

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

    not index a URL with 406 Not Acceptable response status, and consequently, URLs that have been indexed in the past but are now returning this HTTP status code will be removed from

  2. #2 peak

    . This can make land unsuitable for agriculture, lead to crop stress and attract new pests that thrive in warmer temperatures⏎- Warmer temperatures and drought are also contributing to animal stress, decreasing production and making animal welfare an issue.⏎These impacts are

  3. #3 peak

    your caving library. -- Danny A. Brass Underground Movement If you are at all interested in bats, get a copy of this book. You will be glad you did! Book Bargains and Previews A well-written, interesting primer on bats

  4. #4 peak

    ) in 1873 and into medical literature by Babington's translation of Hecker's Der Schwarze Tod in 1833.⏎The source of the infection of bubonic plague was traced in the 1890

  5. #5 peak

    scientific name keeps it easier to classify. Scientific names dont change and show relationships and classification.⏎Why is it better for scientist to

  6. #6 peak

    ," Powell says. "It's that significant. He is one of the most important figures that we've seen in the last 25 years or so."⏎Powell interviewed Tupac a number of times for Vibe magazine, including while the

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 this I have been thinking about this I have been thinking about this I have been thinking about this I have been thinking about this I have been thinking about this I have been thinking about this

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

    3) diplomatic activities, (4) intelligence activities, and (5) cryptology. The latter two areas might be considered to be special parts of the first three areas. That is, intelligence and cryptology are "service" functions for the

  2. #2 peak

    ARD syndrome and Noonan syndrome are both caused by mutations in the PTPN11 and RAF1 genes.⏎The symptoms and physical characteristics associated with LEOPARD syndrome are highly variable. However, most affected individuals tend to exhibit characteristic abnormalities of

  3. #3 peak

    anagrams for endoplasm found in this word list.⏎Words formed by adding one letter before or after endoplasm (in bold), or to adelmnops in any order:⏎s - endoplasms⏎All words formed from end

  4. #4 peak

    come to speak to her classes about their careers. Across the country, just 10 percent of high schools offer computer science classes, according to a website promoting Computer Science Education Week, which began Monday.⏎And yet, experts predict more than

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. L11:990 suppresses
  2. L11:1782 excites
  3. L11:1377 suppresses
  4. L8:105 suppresses
  5. L8:1377 excites
  6. L11:1001 excites
  7. L9:557 weakly excites
  8. L5:985 weakly suppresses
  9. L2:768 weakly excites
  10. L5:197 weakly suppresses
  11. L1:1435 weakly excites 'as' in 'just as'
  12. L10:279 weakly suppresses
  13. L11:1705 weakly suppresses
  14. L11:170 weakly suppresses
  15. L11:341 weakly excites
  16. L11:401 weakly excites
  17. L11:163 weakly excites
  18. L10:1452 weakly suppresses
  19. L8:621 weakly suppresses
  20. L10:2061 weakly suppresses
  21. L10:2416 weakly excites
  22. L1:2596 weakly excites
  23. L4:1052 weakly suppresses
  24. L0:1483 weakly excites

Feeds later layers

  1. L13:801 excites
  2. L13:1547 suppresses
  3. L15:1734 suppresses
  4. L13:1923 suppresses
  5. L13:1665 excites
  6. L13:1123 suppresses
  7. L13:2538 suppresses
  8. L13:143 excites
  9. L13:1680 suppresses
  10. L14:1768 suppresses
  11. L13:2002 suppresses
  12. L16:1398 suppresses
  13. L13:1716 suppresses
  14. L14:1199 suppresses
  15. L14:228 excites
  16. L13:1018 suppresses
  17. L14:108 excites
  18. L13:1761 suppresses
  19. L13:1913 suppresses
  20. L13:650 suppresses
  21. L13:644 suppresses
  22. L15:2673 suppresses
  23. L15:2526 weakly suppresses
  24. L17:807 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.

3 values between -1.38 and -1.271 values between -1.27 and -1.162 values between -1.16 and -1.046 values between -1.04 and -0.937 values between -0.93 and -0.8216 values between -0.82 and -0.7115 values between -0.71 and -0.6038 values between -0.60 and -0.4857 values between -0.48 and -0.3791 values between -0.37 and -0.26159 values between -0.26 and -0.15251 values between -0.15 and -0.03432 values between -0.03 and 0.08451 values between 0.08 and 0.19412 values between 0.19 and 0.30419 values between 0.30 and 0.42374 values between 0.42 and 0.53288 values between 0.53 and 0.64244 values between 0.64 and 0.75165 values between 0.75 and 0.86105 values between 0.86 and 0.98114 values between 0.98 and 1.0968 values between 1.09 and 1.2053 values between 1.20 and 1.3143 values between 1.31 and 1.4323 values between 1.43 and 1.5415 values between 1.54 and 1.6523 values between 1.65 and 1.765 values between 1.76 and 1.886 values between 1.88 and 1.997 values between 1.99 and 2.101 values between 2.10 and 2.214 values between 2.21 and 2.322 values between 2.32 and 2.443 values between 2.44 and 2.551 values between 2.55 and 2.661 values between 2.66 and 2.772 values between 2.77 and 2.891 values between 3.00 and 3.11offnegativepositiveoften

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. left
  2. shared
  3. set
  4. 共享
  5. thư
  6. library
  7. SET
  8. public
  9. legal
  10. إليه

Suppresses

  1. igers
  2. 碣
  3. �回
  4. 朋友们对
  5. 遑
  6. ี้
  7. displayName
  8. getProperty
  9. PageIndex
  10. IFICATIONS