Neuron Explorer

Qwen3-0.6B

"at least" and minimum-quantity phrases

A neuron in layer 20 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 lights up on the words "at least" when they set a minimum amount, like "at least one" or "at least 70%". When it was turned up, the model kept producing "at least" and "least" over and over.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    of dog food, but all of them must meet certain requirements to be considered “real” dog food. For example, the food must have a meat content of at least 70% and it must not contain any artificial colors or flavors.

  2. #2 peak

    across disparate operating systems.⏎Kubernetes clusters are actually a set of nodes categorized into master nodes and worker nodes. Every cluster needs to have at least one of each. While the master node is meant to control the state of

  3. #3 peak

    its evolution into an all-engulfing process, why it is so difficult to stop once it has begun, and what our best hopes are for ending war or at least limiting its dangers.⏎Several important developments have taken place since Dr. Dyer

  4. #4 peak

    or disorder is the genotype with two recessive alleles. So I guess in order for the normal genotype, or what is considered normal, a person has to have at least

  5. #5 peak

    using it.⏎Shared libraries can also help you get around the 32K code size limit; you can break up your code into a main portion (of at most 32K), and a number of libraries (each of at most

  6. #6 peak

    1 cm diameter at 1.3 m from the rooting point) they hosted. Overall, 57.9% of trees in the study sites carried at least one liana. The infested trees hosted an average of 1.

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 two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of the two of

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

    ibility.⏎To reduce reverberation and noise levels, acoustical ceiling panels should have a noise reduction coefficient (NRC) of 0.70 or greater. (The NRC indicates the average percentage of sound a material absorbs;

  2. #2 peak

    that meet minimum population density requirements. These urbanized areas are determined by the U.S. Census Bureau as having a population of 50,000 or more people.⏎- 2010 Census Urban Area FAQs⏎- Missouri

  3. #3 peak

    Physical Security Measures in K–12 Schools⏎Download eBook for Free⏎|PDF file||0.6 MB||⏎Use Adobe Acrobat Reader version 10 or higher for the best experience.⏎- What challenges do LEAs face related to school

  4. #4 peak

    it manifests and the consequences for people and society.<|endoftext|>Fact 1: The world’s population is rapidly ageing⏎The number of people aged 60 years or older will rise from 900 million to 2 billion between 2

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. L19:1171 strongly excites
  2. L16:2990 excites
  3. L19:1461 suppresses
  4. L18:1920 suppresses
  5. L12:2102 suppresses
  6. L18:2965 excites
  7. L17:2409 suppresses
  8. L18:504 suppresses
  9. L18:2197 suppresses
  10. L19:1211 excites
  11. L18:1760 suppresses
  12. L19:2623 excites
  13. L19:1222 suppresses
  14. L15:2308 weakly excites
  15. L15:1910 weakly excites
  16. L15:1212 weakly excites
  17. L19:2653 weakly excites "most" meaning "very" or "the most"
  18. L12:294 weakly excites
  19. L10:2816 weakly excites
  20. L19:2969 weakly excites
  21. L11:245 weakly suppresses
  22. L18:569 weakly excites
  23. L8:1710 weakly excites
  24. L7:846 weakly excites

Feeds later layers

  1. L23:1589 strongly excites 'at' in 'at least' and 'minimum'
  2. L23:1543 excites 'up' in 'up to' limits
  3. L27:599 excites
  4. L26:910 excites
  5. L24:380 excites the word 'at' in place phrases
  6. L26:2537 excites
  7. L26:835 excites
  8. L24:2610 excites
  9. L25:2301 excites
  10. L24:1826 excites
  11. L21:2199 excites
  12. L27:1465 excites
  13. L25:501 suppresses
  14. L22:615 excites
  15. L27:863 excites
  16. L23:380 excites
  17. L27:1086 excites
  18. L21:1256 excites
  19. L27:1060 weakly excites
  20. L25:2378 weakly excites
  21. L23:2966 weakly suppresses
  22. L27:778 weakly excites
  23. L22:969 weakly excites
  24. L24:1585 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 -4.97 and -4.402 values between -2.71 and -2.1524 values between -2.15 and -1.58299 values between -1.58 and -1.021483 values between -1.02 and -0.451923 values between -0.45 and 0.11140 values between 0.11 and 0.6812 values between 0.68 and 1.2410 values between 1.24 and 1.805 values between 1.80 and 2.371 values between 2.37 and 2.931 values between 2.93 and 3.501 values between 3.50 and 4.061 values between 4.62 and 5.192 values between 5.19 and 5.751 values between 6.88 and 7.441 values between 9.13 and 9.701 values between 17.03 and 17.59offnegativepositiveoften

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. 至少
  2. least
  3. 最少
  4. least
  5. 起码
  6. 少なく
  7. atleast
  8. minimum
  9. minimum
  10. Least

Suppresses

  1. ét
  2. 还有什么
  3. elé
  4. éric
  5. 配上
  6. manipulate
  7. 配
  8. hormones
  9. 真的
  10. 才发现