Neuron Explorer

Qwen3-0.6B

last word before end of document

A neuron in layer 19 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 final word or punctuation mark just before a text ends, where the next token is the end-of-document marker. The steering results are mostly unrelated, with repeated dots, numbers or stray words, so they do not add much.

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

    takes longer than mechanical disruption.⏎Can't find the product you've used to perform this experiment? It would be great if you can help us by Adding a product<|endoftext|>In 1800s, structural steels have started to use in

  2. #2 peak

    460 degrees Celsius<|endoftext|>For the Union Dead⏎by Robert Lowell⏎Stanza 10 Summary⏎

  3. #3 peak

    of balloons.⏎Submit a review<|endoftext|>1 year ago⏎Verified Sherpa Tutor⏎If a liquid has a definite volume

  4. #4 peak

    officer dies in the line of duty.⏎- The peace officer mourning ribbon does not take the place of the poppy on Remembrance Day.⏎- Date modified:<|endoftext|>Research mines to discover minerals and the best ways to get them out.⏎Engineers

  5. #5 peak

    limestone and cullet⏎Information provided by the Glass Packaging Institute http://www.gpi.org.⏎Answers: C, A, B, B, C<|endoftext|>This article is about the Great American Novel as a concept. For other uses,

  6. #6 peak

    how the common language runtime uses the machine configuration file for assembly binding, see How the Runtime Locates Assemblies.⏎An application configuration file contains settings that are specific<|endoftext|>Mercury is used in thermometers, fluorescent light bulbs, and some electrical switches

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 a 30 years old man, 3 years ago I was 1

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

    composition. The basic⏎source of all minerals is the hot magma in the⏎interior of the earth.⏎(more content follows the advertisement below) A D V E R T I S E M E N T⏎When magma cools,⏎

  2. #2 peak

    Extended Library Functions, Volume 2 Oracle Solaris 11.1 Information Library|⏎- load exponent of a floating point number⏎c99 [ flag... ] file... -lm [ library... ] #include <math.h> double

  3. #3 peak

    the refracted angle ( r ).⏎- The light beam can bend towards the normal or away from the normal.

  4. #4 peak

    the outer loop test in step 4 will execute ( n + 1 )⏎> times (note that an extra step is required to terminate the for loop

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. L18:795 suppresses
  2. L18:246 suppresses
  3. L16:795 suppresses
  4. L16:317 excites
  5. L18:955 excites
  6. L17:15 suppresses
  7. L6:2543 excites
  8. L18:48 weakly suppresses
  9. L9:234 weakly suppresses line break inside hard-wrapped text
  10. L18:2566 weakly excites
  11. L7:1240 weakly suppresses
  12. L16:455 weakly excites
  13. L10:527 weakly excites
  14. L17:2095 weakly suppresses
  15. L0:2059 weakly excites
  16. L0:22 weakly excites
  17. L18:76 weakly suppresses blank spot before a number
  18. L9:1768 weakly suppresses
  19. L17:1568 weakly excites
  20. L11:20 weakly excites
  21. L16:170 weakly suppresses
  22. L6:2157 weakly excites
  23. L9:2943 weakly excites
  24. L16:2945 weakly excites

Feeds later layers

  1. L22:16 strongly suppresses comma in dates and birth lines
  2. L26:21 strongly suppresses
  3. L20:59 strongly excites
  4. L27:11 strongly suppresses unclear: mixed fragments in lists and pronunciations
  5. L27:33 strongly excites
  6. L25:109 suppresses
  7. L22:14 excites
  8. L24:39 excites
  9. L25:17 excites
  10. L27:97 excites
  11. L21:418 excites
  12. L24:51 excites
  13. L20:489 excites
  14. L24:33 suppresses
  15. L25:30 excites
  16. L23:80 excites blank slot before a year or number
  17. L21:70 excites
  18. L27:101 excites
  19. L20:193 suppresses
  20. L23:60 excites
  21. L26:103 excites blank space before a number
  22. L23:19 suppresses
  23. L27:71 suppresses
  24. L24:94 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 -2.00 and -1.743 values between -1.49 and -1.232 values between -1.23 and -0.978 values between -0.97 and -0.7178 values between -0.71 and -0.45628 values between -0.45 and -0.191758 values between -0.19 and 0.071029 values between 0.07 and 0.33215 values between 0.33 and 0.5938 values between 0.59 and 0.8519 values between 0.85 and 1.1119 values between 1.11 and 1.3713 values between 1.37 and 1.638 values between 1.63 and 1.8912 values between 1.89 and 2.159 values between 2.15 and 2.4115 values between 2.41 and 2.6710 values between 2.67 and 2.936 values between 2.93 and 3.197 values between 3.19 and 3.456 values between 3.45 and 3.705 values between 3.70 and 3.963 values between 3.96 and 4.224 values between 4.22 and 4.481 values between 4.48 and 4.741 values between 4.74 and 5.001 values between 5.00 and 5.263 values between 5.26 and 5.522 values between 5.52 and 5.782 values between 6.04 and 6.301 values between 7.86 and 8.121 values between 8.12 and 8.38offnegativepositiveoften

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. igor
  2. adal
  3. ibus
  4. éis
  5. -strokes
  6. atos
  7. they
  8. aday
  9. 循
  10. 性的

Suppresses

  1. "
  2. L
  3. '
  4. “
  5. ‘
  6. US
  7. A
  8. +
  9. UN
  10. American