Neuron Explorer

Qwen3-0.6B

numbers and ranges in data and statistics

A neuron in layer 9 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 spots just before a number in statistical or data-heavy text, like age ranges or years in a study. Turning it up did not produce a clear, consistent change in the writing.

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

    See<|endoftext|>Data Set Overview⏎|Data set||LBA-ECO LC-09 Land Cover Transitions Maps for Study Sites in Para, Brazil: 1970-2001|⏎|Project||

  2. #2 peak

    of a flash<|endoftext|>Undated (WXOW) - In 2009, the US Preventative Task force recommended against breast cancer screening for women aged 40 to 49.⏎It said the risks outweigh the small potential benefit

  3. #3 peak

    live, feed and breed.⏎The report also divides the Arctic Ocean into 12 regions. Using daily sea ice concentration data from satellite images taken by NASA between 1979 and 2013, changes in the dates of

  4. #4 peak

    ) than were nonveterans (19 percent).⏎Among Gulf War-era II veterans in 2009, 31 percent of veterans aged 18 to 24 years were enrolled in school. As for older

  5. #5 peak

    Skip to Search Results⏎Identifying and analyzing spatial and temporal patterns of lightning-ignited wildfires in Western Canada from 1981-2018.Download⏎This study looked at the spatial and temporal patterns of lightning-ignited

  6. #6 peak

    , 2,019 waterbodies had 5 or more years of data. When the University of Florida data are added, the number jumps to 4,800 waterbodies. Of these, 17%

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: How many full 3-digit numbers are there that can be used to describe a person's state of being in a fair and peaceful situation? A person has 1000000 characters in their system. You have 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

    blank sheet. You can make as many columns as<|endoftext|>Johann Christian Heyl⏎Johann Christian Heyl (1

  2. #2 peak

    ?⏎Use the alphabetical glossary listing to find a short definition to a medical or dental term. If you would like to know more, please remember to call us on (02) 9420 5050. Our

  3. #3 peak

    treat serious damage, where an endodontist does.⏎The type of issues an endodontist treats⏎While minor issues such as plaque buildup or even a minor chip in the tooth may only need the services of a general dentist, there are many

  4. #4 peak

    light is off.⏎Mirrors and Windshield⏎- Mirrors should be clean and adjusted properly from the inside.⏎- Windshield should be clean with no illegal

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. L7:1111 excites
  2. L7:34 excites
  3. L7:2773 suppresses
  4. L7:277 suppresses
  5. L8:969 excites
  6. L7:275 suppresses
  7. L6:337 suppresses
  8. L6:263 suppresses
  9. L6:391 suppresses
  10. L8:1027 suppresses
  11. L5:151 excites
  12. L7:1550 suppresses
  13. L7:468 excites
  14. L6:262 weakly suppresses
  15. L8:383 weakly excites
  16. L7:231 weakly excites
  17. L2:2088 weakly excites
  18. L7:322 weakly suppresses
  19. L6:762 weakly suppresses
  20. L2:2320 weakly suppresses
  21. L6:665 weakly excites
  22. L7:619 weakly suppresses
  23. L5:2013 weakly excites
  24. L7:140 weakly excites

Feeds later layers

  1. L10:391 excites
  2. L11:924 excites
  3. L10:2012 excites
  4. L11:1330 excites
  5. L10:347 suppresses
  6. L11:2343 excites
  7. L11:330 suppresses
  8. L11:1030 excites
  9. L10:1436 suppresses
  10. L12:1172 excites
  11. L10:935 excites
  12. L11:2196 excites
  13. L11:1193 excites
  14. L10:948 suppresses
  15. L10:2998 excites
  16. L11:2825 excites
  17. L12:70 suppresses
  18. L10:736 excites
  19. L11:497 suppresses
  20. L11:3020 suppresses
  21. L21:1826 excites
  22. L17:2826 weakly excites
  23. L12:30 weakly suppresses
  24. L11:1828 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 -2.52 and -2.401 values between -2.27 and -2.151 values between -1.90 and -1.782 values between -1.78 and -1.662 values between -1.66 and -1.535 values between -1.53 and -1.415 values between -1.41 and -1.296 values between -1.29 and -1.1616 values between -1.16 and -1.0415 values between -1.04 and -0.9242 values between -0.92 and -0.7937 values between -0.79 and -0.6760 values between -0.67 and -0.5598 values between -0.55 and -0.42143 values between -0.42 and -0.30256 values between -0.30 and -0.18519 values between -0.18 and -0.051285 values between -0.05 and 0.07630 values between 0.07 and 0.20342 values between 0.20 and 0.32194 values between 0.32 and 0.44105 values between 0.44 and 0.5761 values between 0.57 and 0.6934 values between 0.69 and 0.8114 values between 0.81 and 0.9417 values between 0.94 and 1.067 values between 1.06 and 1.182 values between 1.18 and 1.312 values between 1.31 and 1.433 values between 1.43 and 1.551 values between 1.92 and 2.051 values between 2.17 and 2.291 values between 2.29 and 2.42offnegativepositiveoften

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. auté
  2. Seeder
  3. .INSTANCE
  4. 箔
  5. Wildcard
  6. instance
  7. Installation
  8. HeaderValue
  9. Gundam
  10. 种子

Suppresses

  1. attery
  2. tery
  3. db
  4. uzzi
  5. =db
  6. eleg
  7. ��
  8. proprietary
  9. manage
  10. _db