Neuron Explorer

Qwen3-0.6B

'static' and 'static'-like words

A neuron in layer 8 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 reacts most to the word "static", especially in its programming sense (a static inner class), and also to words that sound or look similar, like "eggs" or "passive". Its strongest negative cases are "static" meaning radio noise. Turning it up or down gave unrelated text, so the link to the trigger is only loose.

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

    behavior. Here's a class called CarParts that declares a static inner class called Wheel. If we consider the real-world, we can find many

  2. #2 peak

    5 and others.⏎> Let us start from some anecdotal evidence: isinstance() is one of the most called functions in large scale Python code-bases (by static call count). In particular, when analyzing some multi-million line production code base,

  3. #3 peak

    :37⏎Exercise7. Read the following sentences, underline the main verb and state the verb: Active or Passive.⏎1.Railroads were developed in the 19 century and became the

  4. #4 peak

    streaked with yellow. Male painted turtles have longer front toenails than the females, apparently to "tickle" the female's head during courtship and stimulate egg production.⏎Only a few sightings of the rubber boa have been made in the park

  5. #5 peak

    home floral arrangements. "New weed infestations can occur when people transport the pretty flowers," he says.⏎Yellow toadflax, sometimes called butter and eggs, was actually introduced as an ornamental plant. "It's a member of

  6. #6 peak

    participlePast perfect (had been) Will + be + past participle Can + be + past participle Will + be Can + be⏎Active voicePassive voice Helen cleans the house The house is cleaned by Helen Some people collect pictures Pictures

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 user, you, is the one who said, "2011". So, the first part of this is the... for this 2011 The you are the one who... so, the first

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

    cars, the BMW i3 and Tesla Model X, are ditching terrestrial AM radio because electromagnetic noise from the electric motor interferes with the broadcast reception, causing static, as the blog Music 3.0 recently pointed out.⏎Electronics have

  2. #2 peak

    thousands of times a second, basically chopping up energy. This process causes electromagnetic interference that gets picked up by the radio.⏎AM radio has always been more susceptible to static than its partner on the dial. "AM" stands for amplitude modulation, which

  3. #3 peak

    21st, 2019⏎Overview of chronic pelvic pain⏎Pelvic pain is a serious form of non-communicable disease, which directly interferes with healthy living. Acute pelvic pain conditions can

  4. #4 peak

    less severe limitations in functioning.Fact 3: The most common health conditions in older age are noncommunicable diseases⏎Older people in low- and middle-income countries carry a greater disease

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. L4:2345 excites
  2. L7:76 excites
  3. L7:122 weakly excites
  4. L6:572 weakly suppresses
  5. L6:1852 weakly suppresses
  6. L7:545 weakly excites
  7. L0:1301 weakly suppresses
  8. L6:379 weakly suppresses
  9. L6:8 weakly suppresses
  10. L4:1080 weakly suppresses
  11. L0:1224 weakly suppresses
  12. L6:16 weakly excites
  13. L4:339 weakly excites
  14. L2:2471 weakly suppresses
  15. L0:1 weakly suppresses
  16. L0:178 weakly suppresses
  17. L7:2187 weakly excites
  18. L1:1864 weakly suppresses
  19. L0:2768 weakly suppresses
  20. L6:2888 weakly suppresses
  21. L4:1 weakly excites
  22. L3:1339 weakly excites
  23. L3:2019 weakly excites
  24. L6:1961 weakly suppresses

Feeds later layers

  1. L10:1686 excites
  2. L9:3054 excites
  3. L12:2634 excites
  4. L23:2616 weakly suppresses
  5. L14:1421 weakly suppresses
  6. L10:231 weakly excites
  7. L25:2652 weakly suppresses
  8. L9:2987 weakly suppresses
  9. L14:1668 weakly excites
  10. L14:970 weakly suppresses
  11. L16:972 weakly suppresses
  12. L9:2942 weakly suppresses
  13. L11:1589 weakly suppresses
  14. L18:2025 weakly excites
  15. L9:692 weakly excites
  16. L11:3016 weakly excites slash in fractions
  17. L24:2137 weakly suppresses
  18. L11:2442 weakly excites
  19. L19:2822 weakly suppresses
  20. L15:2376 weakly excites
  21. L14:967 weakly excites
  22. L14:2144 weakly excites
  23. L11:2009 weakly suppresses
  24. L19:3064 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 -0.61 and -0.581 values between -0.58 and -0.552 values between -0.55 and -0.523 values between -0.52 and -0.492 values between -0.49 and -0.464 values between -0.46 and -0.432 values between -0.43 and -0.405 values between -0.40 and -0.378 values between -0.37 and -0.3419 values between -0.34 and -0.3219 values between -0.32 and -0.2932 values between -0.29 and -0.2649 values between -0.26 and -0.2367 values between -0.23 and -0.2085 values between -0.20 and -0.17108 values between -0.17 and -0.14172 values between -0.14 and -0.11250 values between -0.11 and -0.08366 values between -0.08 and -0.05533 values between -0.05 and -0.02704 values between -0.02 and 0.01469 values between 0.01 and 0.04301 values between 0.04 and 0.06219 values between 0.06 and 0.09163 values between 0.09 and 0.1294 values between 0.12 and 0.1580 values between 0.15 and 0.1846 values between 0.18 and 0.2131 values between 0.21 and 0.2419 values between 0.24 and 0.2715 values between 0.27 and 0.3012 values between 0.30 and 0.3310 values between 0.33 and 0.365 values between 0.36 and 0.383 values between 0.38 and 0.413 values between 0.41 and 0.443 values between 0.44 and 0.472 values between 0.47 and 0.501 values between 0.53 and 0.56offnegativepositiveoften

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. ire
  2. IR
  3. ayscale
  4. polled
  5. IQ
  6. ir
  7. Toll
  8. ULT
  9. 在全球
  10. èle

Suppresses

  1. expense
  2. 工程机械
  3. 你以为
  4. 达不到
  5. 玑
  6. 需要用到
  7. .StoredProcedure
  8. buster
  9. 需要用
  10. 罟