Neuron Explorer

Qwen3-0.6B

unclear: sun angles, geometry and road-safety terms

A neuron in layer 9 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 about angles, slopes and directions, like 'perpendicular to' or 'low-angle sun', and it is quiet or pushed well below zero on text about visiting or travelling to places. When it was turned up, the model drifted into maths problems and word puzzles.

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

    long shadows produced by the low-angle sun to lead the viewer's eye into the photo.⏎Shadows extending toward the

  2. #2 peak

    See also single-crystal silicon and polycrystalline silicon.⏎Angle of Incidence — The angle that a ray of sun makes with a line perpendicular to the surface. For example, a surface that directly faces the sun has a solar

  3. #3 peak

    light is off.⏎Mirrors and Windshield⏎- Mirrors should be clean and adjusted properly from the inside.⏎- Wind

  4. #4 peak

    different designs will work better in different spaces. However, some basic tips include making sure the stairs are wide enough for comfortable use, and that they have a gradual slope rather than a steep incline<|endoftext|>Teaching in any Setting: Participation Project Environment

  5. #5 peak

    coming headlights, rather look toward the right side of the road and use the white fog line as a guide. When headlights from vehicles following you reflect in your rearview mirror, use the “day-night” feature on the mirror or adjust your mirror

  6. #6 peak

    ed from iCoachMath]⏎Focus:Focus of a parabola lies on the axis of symmetry.⏎Directrix:Directrix is a line that is perpendicular to the axis of symmetry of a parabola.⏎Geometric figure and Formulas

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: San Francisco, San Francisco, (4x3x) San Francisco – San Francisco on the way, which is the best way to get to the city. I have been staying in a 5-star hotel, which is the best way

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

    The lessons take place on the school soccer field, where groups of students practice setting up tents, learn how to use a compass, are introduced to "Leave No Trace" ethics, and learn the basic skills needed to plan a trip or overnight stay

  2. #2 peak

    strong currents and winds make landing on the few steep rocky shores a treacherous task, even in the zodiac rafts. We were fortunate that one of our cruises had been chartered and with several descendants/relatives from the

  3. #3 peak

    .⏎Still an architectural marvel, the Seaside is today a complicated monument to both genuine attempts at medical innovation.⏎Know Before You Go⏎Vehicles should be left at the parking lot or grassy area before the gate. The only way to enter

  4. #4 peak

    th-18th, 2017, Vicenza, Italy)<|endoftext|>- Welcome to Botswana⏎- Where To Go⏎- What To Do⏎- Where To Stay⏎- Flora and Fauna⏎- Visitor Information

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. L6:2660 excites
  2. L6:3062 suppresses
  3. L6:3001 suppresses
  4. L8:1720 suppresses
  5. L7:529 suppresses
  6. L6:2824 suppresses
  7. L6:2196 excites
  8. L6:2816 suppresses
  9. L7:1957 suppresses
  10. L7:102 suppresses
  11. L5:3030 weakly suppresses
  12. L4:2271 weakly excites
  13. L8:2850 weakly excites
  14. L3:659 weakly suppresses
  15. L8:775 weakly excites
  16. L6:1768 weakly suppresses
  17. L5:1155 weakly suppresses
  18. L7:676 weakly suppresses
  19. L7:1687 weakly excites
  20. L8:1456 weakly excites
  21. L7:581 weakly excites
  22. L4:2013 weakly suppresses
  23. L1:329 weakly excites
  24. L2:45 weakly excites

Feeds later layers

  1. L10:2368 suppresses
  2. L16:959 suppresses
  3. L22:2121 suppresses
  4. L22:1490 suppresses
  5. L23:2050 suppresses
  6. L22:2670 suppresses
  7. L10:1996 suppresses
  8. L21:3021 suppresses
  9. L19:1834 suppresses
  10. L23:1733 suppresses
  11. L20:2189 suppresses
  12. L17:2803 suppresses
  13. L10:2758 suppresses
  14. L11:3012 suppresses
  15. L10:2774 suppresses
  16. L21:1513 suppresses
  17. L21:3052 suppresses
  18. L22:665 suppresses
  19. L21:965 excites
  20. L26:2866 suppresses
  21. L19:2263 suppresses
  22. L10:2245 weakly suppresses
  23. L20:2226 weakly suppresses
  24. L22:1998 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 -3.67 and -3.541 values between -2.53 and -2.411 values between -2.41 and -2.281 values between -2.03 and -1.903 values between -1.65 and -1.523 values between -1.52 and -1.401 values between -1.40 and -1.277 values between -1.27 and -1.153 values between -1.15 and -1.026 values between -1.02 and -0.906 values between -0.90 and -0.773 values between -0.77 and -0.6412 values between -0.64 and -0.5220 values between -0.52 and -0.3944 values between -0.39 and -0.27142 values between -0.27 and -0.14460 values between -0.14 and -0.011152 values between -0.01 and 0.111099 values between 0.11 and 0.24648 values between 0.24 and 0.37236 values between 0.37 and 0.4947 values between 0.49 and 0.6211 values between 0.62 and 0.741 values between 1.25 and 1.37offnegativepositiveoften

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. 孀
  2. Bundesliga
  3. 酪
  4. .createClass
  5. 榄
  6. centage
  7. ificial
  8. parate
  9. inputData
  10. newData

Suppresses

  1. visit
  2. visits
  3. visited
  4. 游览
  5. visitor
  6. tours
  7. visiting
  8. visit
  9. 访
  10. tour