Neuron Explorer

Qwen3-0.6B

the word 'made'

A neuron in layer 24 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 lights up on the word 'made' when it means 'produced or constructed', as in 'is made of', 'made from' or 'made-up'. Turning it up made the model repeat the word 'up' endlessly, and turning it down produced looping 'right' and 'community' text.

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

    not require the last check, but none of real⏎"I have come to believe that the whole world is an enigma, a harmless enigma⏎that is made terrible by our own mad attempt to interpret it as though it had⏎an underlying

  2. #2 peak

    way: “The law is not made for a righteous man” (1Tim. 1:9). Paul was

  3. #3 peak

    case or motion being tried or heard and the laws applicable to it.⏎(2)Either the outline of a case made in preparation for a class, or a written argument for Legal Writing and Research.⏎

  4. #4 peak

    .⏎A type of game in which one person, usually called the DungeonMaster or GameMaster (also written DM and GM), directs one or more players through a made-up world. Typically, the GM will describe a situation and it is up to

  5. #5 peak

    stream of air against a sharp edge. The term has been applied to various parts of this, including the block and the channel.⏎- ‘The fipple is made from a piece of 5/16th hardwood dowel about one

  6. #6 peak

    play Richard III, Act-I, Scene-I, Lines 1-4, begins with the following famous stanza:⏎“Now is the winter of our discontent⏎Made glorious summer by this son of York;⏎And all the clouds that

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 right to the free the community the right to the right to the community the right to the community. The community here is the right to the community. I think the community here is the right to the community. What do you think?

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

    answer the original question, its seems the NT favors the LXX.⏎Of the places where the New Testament quotes the Old, the great majority is from the Septuagint version. Protestant authors Archer and Chirichigno list 3

  2. #2 peak

    , 1374). Anglicans emphatically reject this teaching as “repugnant to the plain words of Scripture” and superstitious (Articles of Religion, Article 28), but it is the greatest curiosity that some still

  3. #3 peak

    almost the whole of Targum Jonathan b. Uzziel on the Prophets; and he wrote annotations to Santes *Pagnini's Thesaurus (Oẓar Leshon ha-Kodesh; Lyons,

  4. #4 peak

    Because on our view of “real presence” in the sacrament hinges all the other biblical core matters that were recovered at the time of the English Reformation: Holy Scripture as our authority, the high priesthood of Christ (and universal priesthood of all

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. L0:1293 weakly excites
  2. L20:1074 weakly excites
  3. L2:1852 weakly suppresses
  4. L7:92 weakly suppresses
  5. L6:227 weakly excites
  6. L10:537 weakly suppresses
  7. L20:357 weakly suppresses
  8. L22:100 weakly suppresses
  9. L19:119 weakly suppresses
  10. L9:207 weakly excites
  11. L4:76 weakly suppresses
  12. L2:1301 weakly suppresses
  13. L16:933 weakly excites
  14. L6:1245 weakly suppresses
  15. L18:967 weakly suppresses
  16. L20:1489 weakly excites
  17. L4:2457 weakly excites
  18. L4:342 weakly excites
  19. L13:1528 weakly suppresses
  20. L5:445 weakly excites
  21. L3:137 weakly excites
  22. L19:679 weakly suppresses
  23. L6:3007 weakly excites
  24. L4:1833 weakly suppresses

Feeds later layers

  1. L26:1637 excites
  2. L25:1136 weakly suppresses
  3. L26:1582 weakly excites
  4. L27:770 weakly excites
  5. L26:1798 weakly excites
  6. L27:1008 weakly excites
  7. L26:2782 weakly suppresses
  8. L27:1296 weakly suppresses
  9. L26:1353 weakly excites
  10. L27:2059 weakly excites
  11. L25:2623 weakly excites
  12. L26:2679 weakly excites
  13. L26:2101 weakly excites
  14. L25:1568 weakly suppresses
  15. L27:2843 weakly excites
  16. L25:1485 weakly suppresses
  17. L26:1943 weakly suppresses
  18. L27:3000 weakly suppresses
  19. L25:263 weakly excites
  20. L25:1204 weakly suppresses
  21. L26:1121 weakly suppresses
  22. L26:2786 weakly suppresses
  23. L26:2357 weakly excites
  24. L26:813 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 -25.36 and -23.931 values between -22.51 and -21.081 values between -21.08 and -19.651 values between -19.65 and -18.231 values between -16.80 and -15.372 values between -15.37 and -13.952 values between -13.95 and -12.523 values between -12.52 and -11.095 values between -11.09 and -9.677 values between -9.67 and -8.2413 values between -8.24 and -6.8113 values between -6.81 and -5.3933 values between -5.39 and -3.9651 values between -3.96 and -2.53142 values between -2.53 and -1.112858 values between -1.11 and 0.32579 values between 0.32 and 1.7598 values between 1.75 and 3.1753 values between 3.17 and 4.6014 values between 4.60 and 6.0310 values between 6.03 and 7.456 values between 7.45 and 8.882 values between 8.88 and 10.313 values between 10.31 and 11.733 values between 11.73 and 13.162 values between 13.16 and 14.582 values between 14.58 and 16.011 values between 26.00 and 27.421 values between 30.28 and 31.70offnegativepositiveoften

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. up
  2. Up
  3. Up
  4. -Up
  5. upt
  6. iders
  7. -up
  8. bang
  9. igators
  10. igator

Suppresses

  1. oire
  2. .getRight
  3. 疟
  4. inium
  5. managedType
  6. outer
  7. irth
  8. xDB
  9. .pageX
  10. лу