Neuron Explorer

Qwen3-0.6B

cutting and dividing in instructions

A neuron in layer 7 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 lights up on words about cutting or splitting things apart, like 'cut out', 'break up' or 'divided between', mostly in craft and how-to text. Its opposite side fires on phrases about breaking a job into small steps. Turning it down made the model drift into step-by-step planning, which fits that opposite side.

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

    level, such as transplanting seedlings.⏎Hoes and Mattocks⏎Hoes are tools used to create furrows in the garden for planting; to break up hard, clumpy soil; and to remove weeds. Hoes come in

  2. #2 peak

    glue the printed pages to cardboard (like old cereal boxes) and then cut out the cards and boards.⏎The Rules: To play, you need to print out the deck

  3. #3 peak

    to trace the person's shadow onto the⏎poster board. After the image is traced, turn the lights on and cut around⏎the tracing. Using the cut-out as a pattern, trace the person's shape onto⏎a dark piece of paper

  4. #4 peak

    and Craft Projects for Presidents Day⏎Copy silhouettes of Washington and Lincoln onto sheets of cardstock. Give⏎each child a sheet. Let children cut the sheet into 4-12 pieces (depending⏎on ability level), then re

  5. #5 peak

    the students. We recommend reading this ceremony using a narrator and two student readers; their parts are marked below. The texts titled "Student" are to be divided between the two readers. Solemn background music is appropriate. There are two notes in

  6. #6 peak

    um first – not just skinning the poor animal, removing the hair, stretching, drying and processing the skin into parchment, but also preparing the sheets, pricking

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 a thought." So, for example, the first step is to create something that can be done in smaller pieces to get the most out of the best. Then, the first step is to create a piece of work that is

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

    time to market and minimizes post-release bugs. Instead of a complex implementation of global tasks, the team following the principles of the Agile development cycle performs tasks in small sprints lasting 1-4 weeks, depending on the specifics of the project

  2. #2 peak

    student during in-class assignments.⏎- Use more than one way to demonstrate or explain information.⏎- Break information into small steps when teaching many new tasks in one lesson (state objectives, review previous lesson

  3. #3 peak

    landscape may seem daunting, but by breaking the process down into smaller steps, organizations can leverage their security program to achieve business objectives. One critical component

  4. #4 peak

    know how to handle unlike denominators or how to regroup or to borrow with mixed numbers? Take a deep breath and let us take it one step at a time. Click and begin.⏎GCF and LCM by Prime Factorization⏎A

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. L6:2746 weakly excites
  2. L6:1724 weakly excites
  3. L3:1413 weakly excites
  4. L4:1598 weakly excites
  5. L4:2131 weakly suppresses
  6. L6:1465 weakly excites
  7. L6:386 weakly suppresses
  8. L6:2793 weakly excites
  9. L2:657 weakly excites
  10. L6:1799 weakly suppresses
  11. L2:2974 weakly excites
  12. L2:1368 weakly excites course and business document nouns
  13. L6:833 weakly suppresses
  14. L5:1968 weakly suppresses
  15. L0:2402 weakly excites
  16. L5:3013 weakly suppresses
  17. L5:448 weakly excites
  18. L4:1545 weakly suppresses
  19. L6:2451 weakly suppresses
  20. L3:2159 weakly suppresses
  21. L2:1606 weakly excites
  22. L2:3029 weakly excites
  23. L6:1390 weakly excites
  24. L6:1481 weakly excites

Feeds later layers

  1. L12:2621 suppresses
  2. L22:1856 suppresses
  3. L19:1924 suppresses
  4. L9:1106 suppresses
  5. L8:2688 suppresses
  6. L14:2426 suppresses
  7. L21:828 suppresses
  8. L19:2500 suppresses
  9. L21:2067 suppresses
  10. L9:2505 suppresses
  11. L10:2059 suppresses
  12. L12:2840 suppresses
  13. L10:1559 suppresses
  14. L23:292 weakly excites
  15. L9:1739 weakly excites
  16. L11:1993 weakly suppresses
  17. L20:2112 weakly suppresses
  18. L8:2798 weakly suppresses
  19. L23:1621 weakly suppresses
  20. L17:1168 weakly suppresses
  21. L18:1168 weakly excites
  22. L9:2292 weakly suppresses
  23. L13:2374 weakly suppresses
  24. L11:2976 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 -1.77 and -1.701 values between -1.57 and -1.502 values between -1.50 and -1.432 values between -1.08 and -1.011 values between -1.01 and -0.951 values between -0.88 and -0.811 values between -0.81 and -0.742 values between -0.74 and -0.674 values between -0.67 and -0.604 values between -0.60 and -0.5312 values between -0.53 and -0.4618 values between -0.46 and -0.4051 values between -0.40 and -0.33100 values between -0.33 and -0.26210 values between -0.26 and -0.19352 values between -0.19 and -0.12644 values between -0.12 and -0.05943 values between -0.05 and 0.02762 values between 0.02 and 0.09428 values between 0.09 and 0.16208 values between 0.16 and 0.23102 values between 0.23 and 0.2935 values between 0.29 and 0.368 values between 0.36 and 0.435 values between 0.43 and 0.504 values between 0.50 and 0.572 values between 0.57 and 0.641 values between 0.64 and 0.711 values between 0.71 and 0.781 values between 0.78 and 0.852 values between 0.91 and 0.98offnegativepositiveoften

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. argar
  2. omen
  3. aptor
  4. вести
  5. FieldType
  6. 公网
  7. ären
  8. ialis
  9. ivar
  10. !.⏎⏎

Suppresses

  1. 分割
  2. divisions
  3. split
  4. partitions
  5. ProgressDialog
  6. цы
  7. 磊
  8. .Concurrent
  9. fragments
  10. 分成