Neuron Explorer

Qwen3-0.6B

"it" after whether or when

A neuron in layer 21 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 "it" when it follows "whether" or "when", as in "whether it's" or "when it comes to". Turned up, the model stuck in loops of "it's it it's".

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

    but if anyone deserved the description of polymath, it was Cartwright. Born into a landed family in Marnham, Nottinghamshire

  2. #2 peak

    .⏎Though the results are promising<|endoftext|>In today’s digital era, apps have become an integral part of our lives, catering to our every need and desire. Whether it’s a gaming app or an investing app like Acorn, one cannot underestimate the

  3. #3 peak

    Abstract: Data science has long considered learning better models and making better decisions based on existing data. With data science systems penetrating into every aspect of daily life, whether it be navigation systems that influence which route you pick for your commute or personalized news delivery

  4. #4 peak

    readers."⏎—Andrew Piper, McGill University⏎"If there is one book to assign Humanities and Digital Humanities students that provides a wide range of knowledge about text technologies, it is this one. It introduces principal concepts along with ample historical examples, five diverse

  5. #5 peak

    , and draw in the hands on an analog clock.⏎- I can write the time and draw in the hands on an analog clock when someone tells me what time it is to the nearest

  6. #6 peak

    to understand the required knowledge that goes into the planning of a house. First Step: The Process: Understanding the fundamentals of the design process is the most important when it comes to designing your own home.⏎Designing a house is the process of shaping

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;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;

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

    8211; Basic Chemistrynbsp;nbsp;Chemistry Questions 8211; Learn Basic Chemistry Part 1 (1-25)nbsp;1) When a beam of white rays is dispersed by a prism which colour

  2. #2 peak

    Children in those communities are at risk of developing severe tooth enamel fluorosis, a condition that can cause tooth enamel loss and pitting. A majority of the report\u0092s authoring committee also concluded that people who drink

  3. #3 peak

    ner]Associated Press Psychologist Urie Bronfenbrennerdeveloped the ecological model ofhuman development. Urie Bronfenbrenner8217;s ecological model of human development has had essay junk food should be banned enduring impact

  4. #4 peak

    students across the toronto district school board have been Children who attend preschool enter kindergarten with improved skills but that advantage is nearly halved by the end of the year quot ensuring that young children enter kindergarten

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. L20:1427 excites
  2. L20:2563 suppresses
  3. L20:2581 suppresses
  4. L20:594 suppresses
  5. L20:2717 excites
  6. L6:6 weakly excites
  7. L17:240 weakly suppresses
  8. L19:2288 weakly suppresses
  9. L20:2508 weakly suppresses
  10. L13:986 weakly excites
  11. L17:2216 weakly excites
  12. L20:3007 weakly excites
  13. L9:1491 weakly excites
  14. L19:1767 weakly suppresses
  15. L12:21 weakly excites
  16. L14:492 weakly suppresses
  17. L6:152 weakly excites
  18. L19:2866 weakly suppresses
  19. L18:2798 weakly suppresses
  20. L15:1492 weakly excites
  21. L18:2701 weakly excites
  22. L18:565 weakly suppresses
  23. L10:527 weakly excites
  24. L14:1833 weakly suppresses

Feeds later layers

  1. L25:2637 suppresses
  2. L22:2514 excites
  3. L24:378 suppresses
  4. L27:347 suppresses
  5. L27:607 excites
  6. L26:908 excites
  7. L26:421 suppresses
  8. L23:974 excites
  9. L22:1349 weakly excites
  10. L24:569 weakly suppresses
  11. L24:82 weakly suppresses
  12. L24:792 weakly suppresses
  13. L25:296 weakly suppresses
  14. L27:1495 weakly excites
  15. L23:1291 weakly suppresses semicolons and list-like citations
  16. L25:923 weakly excites
  17. L24:342 weakly suppresses unit conversions in parentheses
  18. L26:2676 weakly excites
  19. L22:2851 weakly excites
  20. L22:134 weakly suppresses
  21. L27:958 weakly suppresses
  22. L23:915 weakly suppresses
  23. L24:264 weakly suppresses
  24. L26:1255 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 -13.58 and -13.081 values between -9.09 and -8.592 values between -3.10 and -2.604 values between -2.10 and -1.6046 values between -1.60 and -1.10296 values between -1.10 and -0.601524 values between -0.60 and -0.101882 values between -0.10 and 0.40112 values between 0.40 and 0.9025 values between 0.90 and 1.405 values between 1.40 and 1.904 values between 1.90 and 2.392 values between 2.39 and 2.891 values between 2.89 and 3.391 values between 4.89 and 5.392 values between 5.89 and 6.39offnegativepositiveoften

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. it
  2. (it
  3. →it
  4. 它
  5. It
  6. zyć
  7. _it
  8. eties
  9. [it
  10. казалось

Suppresses

  1. ;
  2. ;amp
  3. ;.
  4. ;x
  5. ;/
  6. ;;
  7. ;⏎
  8. ;-
  9. ;\
  10. ;left