Neuron Explorer

Qwen3-0.6B

'to' in 'according to'

A neuron in layer 7 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 word 'to' when it comes right after 'according', as in 'according to the CDC'. When it was turned up, the model's writing did not visibly change, so the word pattern is the only clear signal.

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

    of Americans who have a history of cancer was 11.7 million in 2007, or nearly 4% of the population, according to new figures from the Centers for Disease Control and Prevention (CDC).⏎That's up

  2. #2 peak

    you why…⏎Public health is awesome (and I’m not just saying that). Thanks to the achievements of public health, we live longer and healthier lives. According to the National Center for Health Statistics, “life expectancy at birth among U.S.

  3. #3 peak

    . coli before it enters a newly protected stream. Estimated costs for installing disinfection equipment and operating it for the first year are about $95 million, according to department figures.⏎Without major improvements and more costs, though, disinfection could be

  4. #4 peak

    about cybersecurity and how it affects your SMB:⏎Why do hackers target small businesses?⏎The size of your corporation doesn’t guarantee your safety against cybercriminals. According to Stephen Cobb, a senior security researcher at the antivirus company ESET, small businesses

  5. #5 peak

    ). Those fires resulted in 155 deaths, 1,100 injuries and $176.1 million in direct property damage, according to NFPA.⏎In addition, NFPA reports that twice as many home candle fires

  6. #6 peak

    to minimize symptoms.⏎Why are so many alcoholics skinny?⏎More likely, it results from interference with the body’s ability to derive energy from other foods. According to Lieber’s report, experiments in laboratory animals and in heavy drinkers found that alcohol

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: 13, 12, 12, 12, 12. I need to find the six 1s. The 12 and the 12 and the 12. I think I

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

    after contracting influenza was pneumonia. However, those living in the sunnier communities had lower death rates after developing influenza than those living in the less sunny communities. Vitamin D helped in two ways: first by reducing the aggressive immune response that among other things

  2. #2 peak

    and farm managers with a global perspective on a range of potential environmental effects and their relative intensity,” said Dr. Michael Rubino, director of NOAA’s Fisheries Office of Aquaculture. “It is a tool that can be used when evaluating proposed

  3. #3 peak

    0 mg per day⏎1 to 3 years old: 700 mg per day⏎Your diet should also include adequate amounts of vitamin D. Vitamin D is essential for the body to absorb calcium. A lack of vitamin D inhibits

  4. #4 peak

    in this SPM can be found in the glossary of the SRREN (Annex I).Conventions and methodologies for determining costs,

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. L5:2208 excites
  2. L6:2182 excites
  3. L6:2759 excites
  4. L6:773 suppresses
  5. L6:1530 suppresses
  6. L6:2193 weakly suppresses
  7. L6:1637 weakly excites
  8. L6:2285 weakly suppresses
  9. L6:2692 weakly suppresses
  10. L6:2842 weakly excites
  11. L6:1720 weakly suppresses
  12. L6:1425 weakly suppresses
  13. L6:2977 weakly suppresses
  14. L6:2349 weakly excites
  15. L4:2679 weakly excites
  16. L3:2043 weakly excites
  17. L6:961 weakly excites
  18. L6:1975 weakly excites
  19. L5:1276 weakly suppresses
  20. L4:2979 weakly excites
  21. L6:1222 weakly suppresses
  22. L6:2758 weakly suppresses
  23. L0:2804 weakly suppresses
  24. L2:2751 weakly excites spiral and screw words

Feeds later layers

  1. L8:127 excites
  2. L8:941 suppresses
  3. L18:2681 weakly suppresses
  4. L9:283 weakly suppresses
  5. L10:1224 weakly suppresses
  6. L20:759 weakly excites
  7. L24:1175 weakly suppresses
  8. L13:2413 weakly excites
  9. L9:1392 weakly suppresses
  10. L8:586 weakly suppresses
  11. L20:638 weakly excites
  12. L16:1291 weakly suppresses
  13. L23:2060 weakly suppresses
  14. L8:455 weakly suppresses
  15. L13:2633 weakly excites
  16. L8:1572 weakly suppresses
  17. L9:2952 weakly suppresses
  18. L21:542 weakly suppresses
  19. L15:830 weakly excites
  20. L19:2322 weakly suppresses
  21. L19:1160 weakly suppresses
  22. L10:1178 weakly suppresses
  23. L9:2280 weakly suppresses
  24. L11:361 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 -2.02 and -1.821 values between -1.61 and -1.401 values between -1.40 and -1.203 values between -1.20 and -0.998 values between -0.99 and -0.7813 values between -0.78 and -0.5835 values between -0.58 and -0.37232 values between -0.37 and -0.172424 values between -0.17 and 0.04942 values between 0.04 and 0.25127 values between 0.25 and 0.4543 values between 0.45 and 0.6623 values between 0.66 and 0.8715 values between 0.87 and 1.074 values between 1.07 and 1.287 values between 1.28 and 1.497 values between 1.49 and 1.695 values between 1.69 and 1.905 values between 1.90 and 2.101 values between 2.31 and 2.521 values between 2.52 and 2.723 values between 2.72 and 2.931 values between 2.93 and 3.141 values between 3.14 and 3.341 values between 3.34 and 3.551 values between 3.76 and 3.961 values between 4.17 and 4.371 values between 4.99 and 5.201 values between 6.02 and 6.23offnegativepositiveoften

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. 中共
  2. Together
  3. iero
  4. conjunction
  5. AY
  6. azio
  7. ferr
  8. ause
  9. ay
  10. 殿

Suppresses

  1. |x
  2. MENTS
  3. uds
  4. antry
  5. llx
  6. ancy
  7. reeNode
  8. -proof
  9. %x
  10. hotmail