Neuron Explorer

Qwen3-0.6B

spiral and screw words

A neuron in layer 2 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron responds to words about spirals and screws, such as 'spiral', 'lead screw' and 'spiral fracture'. Turning it up or down did not change the writing in a clear, related way.

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

    pipetting systems.⏎One of the most common applications of lead screws is for positioning the print head in 3D printers. They often use 8mm lead screws as demonstrated in the video below. The concepts explained above are all shown in the

  2. #2 peak

    alcohol, bi-carbonate of soda, vaseline, or burnt with a match in an attempt to remove. Such action can induce a tick to release spirochetes and/or other pathogens known to cause

  3. #3 peak

    terms used to identify the types of broken bones, including:⏎TransverseA bone break is considered transverse when the fracture extends across the bone.⏎SpiralAs you may have guessed, a spiral fracture spirals around the bone.⏎

  4. #4 peak

    es, CNC machines or complex robotics, but they can just as commonly be found around the home, in conventional products such as printers or DVD drives.⏎A Lead Screw’s main purpose is to rotate, using its teeth (or single tooth) to

  5. #5 peak

    your home Child Friendly!⏎For more information about how you can reduce risks in your home and what to do if your child has an accident, visit⏎MySpira is the world's first augmented reality asthma training game, developed in collaboration with

  6. #6 peak

    the two sums are in general different, and the above formula does not necessarily hold.⏎It was shown by Lehmer that the exponential Gauss walk always has the spiral-type shape observed in the visualization. On the other hand, the shape of the

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: I am very much like the first of the first of the first of the first of the first of the first of the first of the first of the first of the first of the first of the first of the first of the first of the

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

    -Caulle however a larger VEI3 eruption occurred in 1960. This has been attributed in various sources to the Valdivia earthquake of 1960 the most powerful earthquake ever recorded on the

  2. #2 peak

    of the Leopold Leadership Program at Stanford University. He is also the director of the Tree Biosafety

  3. #3 peak

    : Stepped stage⏎Learn about this topic in these articles:⏎invention by Jessner⏎...be attributed to the Expressionist movement. The director Leopold Jessner capitalized on the earlier innovations in stage design. His use of

  4. #4 peak

    ademia del Cimento in 1657. In the 18th century the Lorraine family kept collecting and the Grand Duke Peter Leopold of Habsburg-Lorraine founded in Torrigiani Palace a veritable Museum

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 excites
  2. L0:1224 suppresses
  3. L0:1301 weakly suppresses
  4. L0:178 weakly suppresses
  5. L1:1960 weakly excites
  6. L1:2657 weakly excites
  7. L1:189 weakly suppresses
  8. L1:1460 weakly excites
  9. L0:2004 weakly suppresses
  10. L1:1965 weakly suppresses
  11. L0:653 weakly suppresses
  12. L1:2946 weakly suppresses
  13. L1:2728 weakly suppresses
  14. L1:1692 weakly excites
  15. L1:2143 weakly suppresses
  16. L1:545 weakly suppresses
  17. L1:2816 weakly excites
  18. L0:2942 weakly suppresses
  19. L1:1610 weakly excites
  20. L1:2902 weakly suppresses
  21. L1:492 weakly excites
  22. L1:1857 weakly suppresses
  23. L1:2164 weakly excites
  24. L1:5 weakly suppresses

Feeds later layers

  1. L9:1988 weakly suppresses
  2. L12:908 weakly excites
  3. L11:1319 weakly excites
  4. L7:502 weakly suppresses unclear: words after 'less than' or 'of'
  5. L14:3062 weakly excites
  6. L20:663 weakly excites
  7. L19:1081 weakly excites
  8. L9:1431 weakly excites
  9. L4:2181 weakly suppresses
  10. L4:1148 weakly suppresses
  11. L20:1092 weakly suppresses
  12. L17:1917 weakly excites
  13. L7:2471 weakly suppresses
  14. L23:2421 weakly excites
  15. L12:678 weakly excites
  16. L10:1436 weakly excites
  17. L3:1222 weakly excites
  18. L11:2493 weakly suppresses
  19. L23:1865 weakly suppresses
  20. L5:2381 weakly excites
  21. L7:341 weakly excites 'to' in 'according to'
  22. L8:2928 weakly excites
  23. L20:1644 weakly suppresses
  24. L13:2671 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 -0.91 and -0.872 values between -0.65 and -0.611 values between -0.61 and -0.572 values between -0.57 and -0.521 values between -0.48 and -0.443 values between -0.40 and -0.354 values between -0.35 and -0.319 values between -0.31 and -0.2712 values between -0.27 and -0.2237 values between -0.22 and -0.1872 values between -0.18 and -0.14150 values between -0.14 and -0.09385 values between -0.09 and -0.051059 values between -0.05 and -0.011294 values between -0.01 and 0.04447 values between 0.04 and 0.08214 values between 0.08 and 0.12118 values between 0.12 and 0.1654 values between 0.16 and 0.2122 values between 0.21 and 0.2516 values between 0.25 and 0.291 values between 0.29 and 0.342 values between 0.38 and 0.421 values between 0.72 and 0.771 values between 0.77 and 0.81offnegativepositiveoften

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. [value
  2. [count
  3. ments
  4. //{⏎
  5. itchens
  6. clud
  7. cts
  8. //{⏎
  9. [vi
  10. رب

Suppresses

  1. STRUCTOR
  2. ệt
  3. privileged
  4. kB
  5. okit
  6. ham
  7. nore
  8. puck
  9. actory
  10. 杉