Neuron Explorer

Qwen3-0.6B

'whole' in 'a whole lot'

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 word 'whole' when it comes after 'a', as in 'a whole lot' or 'a whole host of'. It is the strongest on that word and quiet elsewhere. Turning it up did not visibly shift 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

    is a lot of work. She commented that sometimes memorization work makes a poem "fall apart" and lose meaning or beauty.⏎If you love the poem a whole lot, spending more time and attention on it through memorization just creates additional lo

  2. #2 peak

    how many petals then had fallen?<|endoftext|>At EdPlace we are passionate about planning and preparation to avoid undue stress to students, whatever their age.⏎There are a whole host of reasons why a 6-week learning plan may be useful to your child

  3. #3 peak

    to time during the semester.<|endoftext|>June 1st 1584⏎Perth's HammermenThe Hammermen’s Incorporation of Perth embraced a whole collection of crafts including silversmiths, goldsmiths, clocksmiths

  4. #4 peak

    isn't going to prematurely age your brain, but regularly consuming added sugar in large quantities may contribute to complications with your brain and cognitive health (not to mention a whole host of other health issues).⏎Read on to learn more about some of the drinks

  5. #5 peak

    generations, DNA changes at a presumably steady rate. For birds, that's about 2% change (for mitochondrial DNA) in a million years -- not a whole lot! But it's enough to figure out their evolutionary history. What have we

  6. #6 peak

    cancer.⏎Water is the essence of life, but not just human life. Almost every living thing on the planet must have water to survive, and that includes a whole host of bacteria and microorganisms that can wreak havoc on human health.

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 more I have been able to get to 20% from my savings. So, this is a common indicator of how much I have been saving, and the more I have been able to get to 20% from my

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

    health care system.⏎Other species, too, suffer as a result of the way we eat. Industrial food production pollutes rivers and streams, creating massive offshore dead zones. It kills off bee colonies, consigns millions of animals to inhum

  2. #2 peak

    in the 1980s. The scientists used computer modeling for their prediction and will take shipboard measurements in July throughout the gulf. The dead zone is caused by an abundance of nutrients from sources

  3. #3 peak

    felsic composition, and is typically black in color but may also be rust-red or display whiteish ‘snowflake’ inclusions. It often displays conchoidal fracturing with very sharp edges. Obsidian is made from the rapid cooling

  4. #4 peak

    -red, dark green, or iridescent. Inclusions of white ‘snowflakes’ are common. Obsidian is very smooth, with scoop-shaped conchoidal fractures that often result in sharp edges when broken.⏎Pro Tip: I have

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. L1:951 weakly excites
  2. L1:2483 weakly excites
  3. L0:2129 weakly suppresses
  4. L0:4 weakly suppresses
  5. L2:2909 weakly excites
  6. L0:571 weakly excites
  7. L0:976 weakly excites
  8. L0:2583 weakly suppresses
  9. L2:2001 weakly excites
  10. L1:2967 weakly suppresses
  11. L1:2724 weakly excites
  12. L0:2230 weakly excites
  13. L2:1321 weakly suppresses
  14. L1:718 weakly suppresses
  15. L1:1799 weakly suppresses
  16. L0:1316 weakly suppresses
  17. L1:1293 weakly excites
  18. L0:1 weakly suppresses
  19. L0:178 weakly suppresses
  20. L2:235 weakly suppresses
  21. L0:814 weakly excites
  22. L1:2559 weakly excites
  23. L2:2532 weakly suppresses
  24. L0:1826 weakly suppresses

Feeds later layers

  1. L4:1369 weakly excites
  2. L24:2959 weakly excites
  3. L9:2811 weakly excites
  4. L23:566 weakly suppresses
  5. L20:909 weakly suppresses
  6. L7:2155 weakly excites
  7. L10:1323 weakly suppresses
  8. L6:1096 weakly suppresses
  9. L9:2408 weakly excites
  10. L5:1852 weakly suppresses
  11. L23:1507 weakly excites
  12. L18:2803 weakly excites
  13. L23:1685 weakly suppresses
  14. L8:2646 weakly suppresses
  15. L21:497 weakly suppresses
  16. L11:2884 weakly excites
  17. L11:1007 weakly excites
  18. L26:3018 weakly excites
  19. L15:2255 weakly suppresses
  20. L5:2903 weakly excites
  21. L17:362 weakly suppresses
  22. L14:2117 weakly suppresses
  23. L14:933 weakly excites
  24. L8:785 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 -2.27 and -2.112 values between -1.32 and -1.161 values between -1.16 and -1.002 values between -0.85 and -0.695 values between -0.69 and -0.539 values between -0.53 and -0.37145 values between -0.37 and -0.211110 values between -0.21 and -0.062166 values between -0.06 and 0.10410 values between 0.10 and 0.2651 values between 0.26 and 0.424 values between 0.42 and 0.581 values between 1.53 and 1.691 values between 3.90 and 4.06offnegativepositiveoften

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. 废话
  2. 不由
  3. crap
  4. Fortune
  5. 命周期
  6. ensive
  7. orable
  8. ertificate
  9. lime
  10. 盆地

Suppresses

  1. func
  2. $log
  3. oster
  4. odor
  5. ador
  6. SB
  7. diá
  8. (symbol
  9. ued
  10. stellen