Neuron Explorer

Qwen3-0.6B

mixed, numbers and list items

A neuron in layer 16 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 a scattered mix of words and digits, such as a digit inside a number range, a word in a list of topics, or a part of a longer word. No single theme stands out. Turning it up made the model repeat itself, so it does not reveal much on its own.

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

    6. He got very depressed after his failure.⏎7. Jessica’s failure was due to her idleness.⏎8. Jessica’s failure was due to her idleness.⏎9

  2. #2 peak

    That is, it would be discounted from the final price.⏎- How to calculate percentage in Excel⏎- Find the difference between two numbers⏎- Calculate percentage in Excel based on two quantities⏎- How to Calculate Excel Percentage in Multiple Cells⏎-

  3. #3 peak

    30⏎- Roman Numerals 1-50⏎- Roman Numerals 1-100⏎- Roman Numerals 1-200⏎- Roman Numerals 1-500⏎- Roman Numer

  4. #4 peak

    55% humidity to manage health effects and illnesses.⏎Comfortable: 30% – 60%⏎Recommended: 45% – 55%⏎High: 55% – 80%⏎Contact Air Care

  5. #5 peak

    our numbers, but the injury to Jewish...⏎Become a Member⏎Read all member content⏎Get exclusive in-depth reports from Israel.⏎Get exclusive in-depth reports from Israel⏎Connect with Israel, right from your home.⏎Lift up the voice of

  6. #6 peak

    .⏎For thirty winters Theodric ruled⏎the Mćring stronghold with an iron hand;⏎many acknowledged his mastery and moaned.⏎That passed away; this also may.⏎We have heard too of Ermanaric's wolfish ways,⏎

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 have been thinking about lately... I have been thinking about lately... I have been thinking about lately... I have been thinking about lately... I have been thinking about lately... I have been thinking about lately... I have been thinking about lately

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

    each with one of the following statements written in large letters: “Strongly Agree,” “Agree Somewhat,” “Disagree Somewhat,” “Strongly Disagree”⏎Activities / Procedures:⏎1. WARM-UP/DO-NOW

  2. #2 peak

    ays differential relays, and distance relays.⏎According to timing characteristics the relays can be divided into instantaneous, definite time lag, inverse time lag and inverse definite minimum time (IDMT) lag relays.⏎In induction type of relays

  3. #3 peak

    of these tenses has four sub-categories (or aspects) which better describe the tense used within the writing. These aspects are simple, perfect, progressive, perfect progressive. Confused yet?⏎Simple Past: Simple past is used to describe an action

  4. #4 peak

    direction⏎B) equal in magnitude but have opposite direction⏎C) not equal in magnitude but have the same direction⏎D) not equal in magnitude but have opposite direction⏎E) None of the above⏎In a uniform circular motion, which

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. L15:968 excites
  2. L15:474 excites
  3. L15:1081 excites
  4. L15:532 suppresses
  5. L12:123 weakly suppresses
  6. L15:121 weakly excites
  7. L14:1904 weakly excites
  8. L15:2915 weakly excites
  9. L14:580 weakly suppresses
  10. L7:56 weakly suppresses
  11. L15:1547 weakly excites "also" before a verb
  12. L15:1884 weakly excites
  13. L15:92 weakly suppresses
  14. L12:689 weakly excites
  15. L7:2589 weakly excites
  16. L15:99 weakly excites
  17. L14:2617 weakly excites
  18. L14:3069 weakly suppresses
  19. L6:1258 weakly suppresses "mean by" in questions and definitions
  20. L15:459 weakly excites
  21. L6:1781 weakly excites
  22. L10:3061 weakly excites
  23. L7:167 weakly suppresses
  24. L0:450 weakly suppresses

Feeds later layers

  1. L17:1497 strongly excites
  2. L17:2570 excites
  3. L17:1709 excites
  4. L17:1149 excites
  5. L18:812 excites
  6. L17:2419 excites
  7. L19:1530 excites
  8. L18:1218 excites
  9. L17:1586 excites
  10. L17:1606 suppresses
  11. L19:742 suppresses
  12. L18:421 suppresses
  13. L17:1009 suppresses
  14. L18:1573 excites
  15. L19:320 suppresses
  16. L20:2843 excites
  17. L19:458 suppresses
  18. L18:351 suppresses
  19. L18:981 excites
  20. L17:1826 excites
  21. L17:731 excites
  22. L22:219 suppresses
  23. L19:200 excites
  24. L18:1272 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 -2.93 and -2.841 values between -2.44 and -2.341 values between -2.34 and -2.241 values between -2.15 and -2.053 values between -2.05 and -1.952 values between -1.85 and -1.752 values between -1.75 and -1.651 values between -1.65 and -1.563 values between -1.56 and -1.463 values between -1.36 and -1.265 values between -1.26 and -1.164 values between -1.16 and -1.066 values between -1.06 and -0.9610 values between -0.96 and -0.8716 values between -0.87 and -0.7727 values between -0.77 and -0.6724 values between -0.67 and -0.5733 values between -0.57 and -0.4741 values between -0.47 and -0.3780 values between -0.37 and -0.27144 values between -0.27 and -0.18251 values between -0.18 and -0.08820 values between -0.08 and 0.02991 values between 0.02 and 0.12713 values between 0.12 and 0.22433 values between 0.22 and 0.32197 values between 0.32 and 0.4267 values between 0.42 and 0.5119 values between 0.51 and 0.617 values between 0.61 and 0.712 values between 0.91 and 1.01offnegativepositiveoften

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. лан
  2. :',⏎
  3. pecially
  4. ')),⏎
  5. ATURE
  6. ')</
  7. )'),⏎
  8. itunes
  9. ropolitan
  10. ')):⏎

Suppresses

  1. !
  2. !⏎⏎
  3. !
  4. !⏎
  5. !⏎⏎
  6. ?
  7. irth
  8. ?⏎⏎
  9. 0
  10. ضبط