Neuron Explorer

Qwen3-0.6B

"if not" in "X, if not Y" phrases

A neuron in layer 2 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 fires on the word "not" when it comes right after "if" in phrases like "if not all" or "if not millions". When it was turned up, the model's writing drifted into odd, loose text, with no clear shared theme.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    tanks, large flags, chest thumping marches, discipline-for-itself and martyr walls help inspire nationalism and patriotism? This becomes a particularly good question, if not a surprising one, because, historically speaking, ideas of nationalism and patriotism have never

  2. #2 peak

    nothing until it is achieved, and with the initial focus often projected at how to pass the entry exams, the aftermath of admission has been a nightmare for thousands if not millions of students.⏎English is one the most widely spoken international languages and many would

  3. #3 peak

    is one of the most important, if not the most important, metrics in valuing the bond.⏎Now, let's look

  4. #4 peak

    as cabinet minister Michael Chong recognized, is a recognition of "ethnic nationalism".⏎In the U.S., we have a number of ethnic nations, most, if not all, being Amerindian in nature. But, citizens of the Navajo

  5. #5 peak

    high-presure steam riding rocket'. In the early part of the 19th century, steam propulsion was regarded by some as something of an unreliable, if not downright dangerous, fad, with the risk of boiler explosions particularly to be feared

  6. #6 peak

    eating plan.⏎Beyond promoting every nutrition expert’s mantra of “eat more vegetables and fruits”, the standard raw food diet is true to its name – most, if not all, foods need to be “raw”. This means dieters are prohibited from

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: that I have a lot of time to be, and to do so to do so to do so to do so to do so to do so to do so to do so to do so to do so to do so to do so to

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

    Commission (EEOC) authority to sue in federal courts when it finds reasonable cause to believe that

  2. #2 peak

    of modern physics--the empirical basis of quantum mechanics, wave-particle duality and the uncertainty principle, and the action-at-a-distance theory of Wheeler and Feynman. 1961 ed.|⏎|Keywords||Physics History Physics

  3. #3 peak

    , when he was a postdoctoral fellow. (This was his first return visit since then.) He had just gotten his PhD at Caltech under Richard Feynman, and was busy trying to understand the “zoo” of hadronic

  4. #4 peak

    ’m a fan of physicist Richard Feynman, and one of his most memorable quotes explains that “Science is the belief

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. L1:189 weakly suppresses
  2. L1:1581 weakly excites
  3. L0:492 weakly excites
  4. L1:367 weakly suppresses
  5. L0:178 weakly suppresses
  6. L0:165 weakly excites
  7. L1:1464 weakly excites
  8. L1:2930 weakly excites
  9. L0:1224 weakly suppresses
  10. L0:1301 weakly suppresses
  11. L1:1923 weakly excites
  12. L1:1724 weakly excites
  13. L1:673 weakly excites
  14. L0:778 weakly suppresses
  15. L0:2268 weakly suppresses
  16. L0:1 weakly suppresses
  17. L0:312 weakly suppresses
  18. L0:3000 weakly suppresses
  19. L1:331 weakly suppresses
  20. L1:2143 weakly suppresses
  21. L1:951 weakly excites
  22. L0:2167 weakly suppresses
  23. L0:1075 weakly excites
  24. L0:2914 weakly suppresses

Feeds later layers

  1. L11:1519 weakly suppresses
  2. L9:1892 weakly excites
  3. L4:265 weakly excites
  4. L10:2223 weakly excites
  5. L22:136 weakly excites
  6. L19:2886 weakly excites
  7. L11:1975 weakly excites
  8. L20:969 weakly excites
  9. L4:1873 weakly suppresses
  10. L10:1055 weakly suppresses
  11. L22:1749 weakly suppresses
  12. L9:929 weakly suppresses 'each' in 'each other'
  13. L7:2499 weakly excites
  14. L15:1929 weakly suppresses
  15. L15:142 weakly suppresses
  16. L7:1921 weakly excites
  17. L14:2659 weakly excites
  18. L10:1499 weakly excites
  19. L23:2556 weakly suppresses
  20. L21:1748 weakly excites
  21. L16:621 weakly excites
  22. L6:2531 weakly excites
  23. L13:364 weakly excites
  24. L22:2705 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 -1.02 and -0.844 values between -0.65 and -0.4617 values between -0.46 and -0.28406 values between -0.28 and -0.093065 values between -0.09 and 0.10392 values between 0.10 and 0.2816 values between 0.28 and 0.472 values between 0.47 and 0.662 values between 0.66 and 0.841 values between 0.84 and 1.031 values between 1.40 and 1.591 values between 6.26 and 6.45offnegativepositiveoften

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. ugo
  2. ìm
  3. enge
  4. ilm
  5. �
  6. 钢筋
  7. ung
  8. 实战
  9. mode
  10. eng

Suppresses

  1. CallCheck
  2. eness
  3. antly
  4. bons
  5. egra
  6. worthy
  7. panion
  8. ibly
  9. criptions
  10. ativity