Neuron Explorer

Qwen3-0.6B

unclear: pieces of longer words and analogies

A neuron in layer 15 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

This neuron fires on odd spots inside words or phrases, such as 'jigsaw', 'corollary' and 'parallel lives'. The negative side seems to show up in comparisons like 'a kind of doorman' or 'a teacher supervising'. No clear pattern, and turning it up gave unrelated maths and rambling.

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

    form of extractive capitalism is no different from industries that take physical resources and is reminiscent of colonial attitudes and activities.⏎Financial exclusion is a large part of the jigsaw of global poverty. But in

  2. #2 peak

    Austronesian seafarers who entered Near Oceania mixed with diverse groups already there, eventually giving birth to the distinct Lapita people, from which arose Polynesian cultures.⏎It is this sheer amount of genetic diversity that destroyed

  3. #3 peak

    life, the more you will find you use it.⏎A degree of experimentation is necessary to get exact cooking times for<|endoftext|>Here are the 15 types of distorted thinking:⏎1. Filtering: You take the negative details and magnify them,

  4. #4 peak

    taking the U.K. government to court over how the country has failed to reach its own targets over climate change.⏎Backed by crowdsourcing legal fees, Plan B’s argument that the government is discriminating towards the young by failing to make a

  5. #5 peak

    rights solely through historical precedents.⏎“Will the Court’s approach permit judges to reach the outcomes they prefer and<|endoftext|>Team first to record key event that breaks continents apart⏎(PhysOrg.com) -- Researchers have captured for the first time a geological

  6. #6 peak

    -called “resettlers” (Aussiedler and Spätaussiedler) from Eastern Europe. In this contribution, I will sketch these “parallel lives” and reflect

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 a city in the state of New York, and I have had a lot of friends, families, and communities that I have had a lot of friends, families and communities in the state of New, which I have had a

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

    land instead of falling down into the fires beneath where everything was burned. Smith's Bible Dictionary gives a description of this valley — which was something like one of our city dumps today where trash and rubbish are burned up.⏎If anything, especially a dead

  2. #2 peak

    the input-output pair. It is called supervised learning because the process of an algorithm learning from the training dataset can be thought of as a teacher supervising the learning process.⏎Some use cases that are based on supervised machine learning

  3. #3 peak

    ‘wrong’ kinds of carbs contribute to it.⏎Insulin resistance occurs when the body stops responding to this critical hormone. Think of insulin as a kind of doorman. Normally, it triggers a reaction that allows glucose to pass through cell walls,

  4. #4 peak

    Fraud⏎Mail fraud describes any action meant to defraud somebody that makes use of the United States Postal Service or any other interstate mail carrier. This would include any action in

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. L14:969 excites
  2. L12:2690 suppresses
  3. L12:2520 suppresses
  4. L14:1275 suppresses
  5. L12:890 excites
  6. L11:67 excites
  7. L13:2037 weakly excites
  8. L13:1714 weakly excites
  9. L7:204 weakly suppresses
  10. L13:2062 weakly excites
  11. L13:142 weakly excites
  12. L10:2079 weakly excites
  13. L12:1152 weakly suppresses
  14. L14:1872 weakly suppresses
  15. L14:1539 weakly suppresses
  16. L5:2278 weakly excites
  17. L13:2461 weakly excites
  18. L8:1878 weakly suppresses
  19. L13:1344 weakly excites
  20. L12:2570 weakly suppresses
  21. L14:145 weakly suppresses
  22. L11:2403 weakly excites
  23. L10:1849 weakly excites
  24. L13:785 weakly suppresses

Feeds later layers

  1. L19:443 excites
  2. L19:746 excites
  3. L18:336 excites
  4. L19:313 suppresses
  5. L20:2189 suppresses
  6. L16:3008 excites
  7. L22:177 excites
  8. L20:409 excites
  9. L19:253 excites
  10. L26:140 suppresses
  11. L20:253 excites
  12. L20:657 excites
  13. L16:1403 excites
  14. L21:372 suppresses
  15. L22:383 suppresses
  16. L21:3052 suppresses
  17. L20:432 excites
  18. L20:1810 suppresses
  19. L26:101 suppresses
  20. L19:428 excites
  21. L21:2752 suppresses
  22. L21:401 excites
  23. L20:893 suppresses
  24. L18:839 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 -2.91 and -2.741 values between -2.40 and -2.231 values between -2.23 and -2.061 values between -2.06 and -1.891 values between -1.89 and -1.733 values between -1.73 and -1.562 values between -1.56 and -1.397 values between -1.39 and -1.2213 values between -1.22 and -1.0521 values between -1.05 and -0.8833 values between -0.88 and -0.7163 values between -0.71 and -0.54108 values between -0.54 and -0.37212 values between -0.37 and -0.20487 values between -0.20 and -0.03978 values between -0.03 and 0.14613 values between 0.14 and 0.31468 values between 0.31 and 0.47277 values between 0.47 and 0.64223 values between 0.64 and 0.81149 values between 0.81 and 0.9884 values between 0.98 and 1.1550 values between 1.15 and 1.3235 values between 1.32 and 1.4917 values between 1.49 and 1.6621 values between 1.66 and 1.8312 values between 1.83 and 2.0011 values between 2.00 and 2.174 values between 2.17 and 2.334 values between 2.33 and 2.503 values between 2.50 and 2.672 values between 2.67 and 2.842 values between 3.01 and 3.181 values between 3.69 and 3.86offnegativepositiveoften

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. CellValue
  2. Substring
  3. ecimal
  4. Multiply
  5. sin
  6. 痧
  7. (sin
  8. adecimal
  9. cosine
  10. sin

Suppresses

  1. 聘请
  2. bankrupt
  3. personnel
  4. headquarters
  5. staff
  6. year
  7. rented
  8. outreach
  9. 国家级
  10. 年度