Neuron Explorer

Qwen3-0.6B

physical measurements and dimensions

A neuron in layer 22 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 words and numbers that describe the size of something, like '4 inches on each side', 'a diameter of 5,150 km', or '10 ft in diameter'. Turned down, the text got stuck repeating words like 'area' and 'volume', and turned up it wrote about numeric problems.

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

    reach can trap nestlings inside a house even when they need to leave and enter the outside world.⏎- Interior Floor Space: A square interior floor measuring four inches on each side is the best spacing

  2. #2 peak

    calculate a different measurement. Ask students what additional information they need to find the total amount of cardboard in the box. When students recognize that they need the lengths of the diagonal sides of the box give them the measurements for those sides (2.2

  3. #3 peak

    |• Land||96.6 sq mi (250.2 km2)|⏎|• Water||0.7 sq mi (1.7 km2)|⏎|Elevation||1,148 ft (

  4. #4 peak

    vertical orientation of the sculpture references the old-growth trees that once covered the landscape near the museum. Its tapered shape—roughly 10 ft in diameter at its base and approximately 4 ft in diameter at its top— recalls a ship

  5. #5 peak

    to calculate and return the volume of a sphere based on the given radius.⏎- Your second subroutine should be called getVolumeOfCuboid(), take three parameters called width, length and height and apply the relevant formula to calculate and return the volume

  6. #6 peak

    by the thickness of the membrane. At steady state, Fickian diffusion can be approximated by:⏎J = D (CWaterlooEmitter - groundwater) / membrane wall thickness⏎where D is the diffusion coefficient for the membrane material and is

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 area around the volume and surface area of the volume, the volume and surface area volumes, and the volume and surface area volumes, and the volume and area volumes, and the area and volume volumes, and the volume and volume volumes,

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

    take one parameter called radius and apply the relevant formula to calculate and return the volume of a sphere based on the given radius.⏎- Your second subroutine should be called getVolumeOfCuboid(), take three parameters called width, length and height and

  2. #2 peak

    , they calculate the surface area of the prism. This provides students with the opportunity to work with complex shapes to find surface area in a given context.⏎Display the image for all to see throughout the

  3. #3 peak

    : Revisiting the Box of Chocolates (10 minutes)⏎In this activity students are presented with a prism that was used in a previous lesson to calculate volume. Here

  4. #4 peak

    out (performed) one after another.⏎We will use the following steps:⏎- Convert all measurement to the same unit (e.g. mm)⏎- Calculate the volume of planet Earth⏎- Calculate the volume

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. L9:2347 suppresses
  2. L11:1255 excites
  3. L20:2483 excites
  4. L20:815 suppresses
  5. L20:1203 excites
  6. L6:1887 suppresses
  7. L10:825 suppresses 'B' in vitamin B names
  8. L9:2044 suppresses
  9. L8:1466 excites
  10. L7:1001 weakly suppresses
  11. L6:2569 weakly excites
  12. L10:1369 weakly excites
  13. L20:402 weakly excites
  14. L8:2079 weakly excites
  15. L12:1533 weakly excites
  16. L5:783 weakly suppresses
  17. L10:1640 weakly excites
  18. L6:779 weakly suppresses
  19. L15:1586 weakly excites
  20. L9:628 weakly suppresses
  21. L17:2890 weakly excites
  22. L9:1335 weakly excites
  23. L10:2277 weakly excites
  24. L20:1079 weakly suppresses

Feeds later layers

  1. L27:1910 suppresses
  2. L26:2464 suppresses
  3. L23:904 suppresses
  4. L25:1853 suppresses
  5. L23:2849 suppresses
  6. L27:1729 suppresses
  7. L24:621 suppresses
  8. L27:2698 weakly suppresses
  9. L24:1235 weakly suppresses
  10. L26:2349 weakly suppresses
  11. L26:2384 weakly suppresses
  12. L24:795 weakly suppresses
  13. L24:1664 weakly suppresses
  14. L24:550 weakly suppresses surname split after first letter D
  15. L25:3061 weakly suppresses
  16. L27:1805 weakly excites
  17. L24:3035 weakly suppresses
  18. L23:2167 weakly suppresses
  19. L24:1708 weakly excites
  20. L25:1803 weakly suppresses
  21. L25:2790 weakly excites
  22. L23:2263 weakly suppresses
  23. L27:1430 weakly suppresses
  24. L24:3003 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 -14.63 and -14.121 values between -9.98 and -9.471 values between -6.89 and -6.372 values between -5.85 and -5.341 values between -4.82 and -4.302 values between -3.27 and -2.751 values between -2.75 and -2.243 values between -2.24 and -1.724 values between -1.72 and -1.2038 values between -1.20 and -0.69503 values between -0.69 and -0.172836 values between -0.17 and 0.35476 values between 0.35 and 0.8634 values between 0.86 and 1.384 values between 1.38 and 1.901 values between 5.51 and 6.03offnegativepositiveoften

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. batim
  2. 的时代
  3. ibili
  4. 相亲
  5. bearer
  6. iddles
  7. 仆
  8. 非常高
  9. ating
  10. Bender

Suppresses

  1. area
  2. Area
  3. areas
  4. area
  5. Area
  6. _area
  7. -area
  8. areas
  9. (area
  10. 面积