Neuron Explorer

Qwen3-0.6B

"make" or "do" before an object noun

A neuron in layer 4 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 the verb right before a phrase like "make a difference", "make sense", "make a list" or "do some research". When it was turned up, the writing drifted into vague, general talk about everyday topics.

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

    the long-term benefits it can provide. It's vital to make a difference and discover your life's purpose at any age and in any circumstance.⏎Vol

  2. #2 peak

    wisdom to the young. This sharing of information is a way to nurture and protect our children as it empowers them to make sense of the past, to make wise decisions in the present, and to cultivate a

  3. #3 peak

    point of view and brainstorm for solutions. Put yourself in the other person’s shoes (empathy - the ability to feel with -)⏎5. Never make a complaint or air a griev

  4. #4 peak

    a weakness in your infrastructure that could potentially expose you to threats.⏎Measuring Organizational Risk⏎The first step to measuring risk is identifying threats. Start by making a list of threats to your organization’s digital assets. From there,

  5. #5 peak

    it’s time to do some serious watering.⏎If your succulent is outdoor, it’s a good idea to

  6. #6 peak

    afford it, you can try it on your own. Don’t embark on any new diet restrictions without a doctor’s supervision if you are ill, pregnant or breastfeeding

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 a research project to do, and I have been the effort of this statement. I have been the effort of this statement. I have been the effort of this statement. I have been the effort of this statement. I have

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

    the analyst, the developer and the tester – have a common understanding of the scope of the work. From this they can agree a common definiton of “done

  2. #2 peak

    prolong the cycle of life, human or non. Sustainable decisions and actions are those that cause no harm and consider the long term effects.⏎So depending on the entire scope of a project’s conception and construction, a building with green parts isn't necessarily

  3. #3 peak

    website.<|endoftext|>After reading this chapter you will…⏎- understand graphs as a mathematical structure.⏎- be able to traverse graphs using well-known algorithms.⏎- learn the scope of problems that can be addressed using graphs.⏎Graphs are perhaps the most versatile

  4. #4 peak

    outside by identifying business outcomes, and then drills down into the feature set that will achieve those outcomes. Each feature is captured as a “story”, which defines the scope of the feature along with its acceptance criteria. This article introduces the BDD approach

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. L2:2692 excites
  2. L3:50 weakly excites
  3. L3:456 weakly suppresses
  4. L0:194 weakly suppresses
  5. L2:1710 weakly suppresses
  6. L2:111 weakly suppresses
  7. L3:1247 weakly suppresses
  8. L3:2326 weakly excites
  9. L3:464 weakly excites
  10. L1:342 weakly excites
  11. L0:467 weakly suppresses
  12. L1:3017 weakly excites
  13. L1:828 weakly suppresses
  14. L2:49 weakly suppresses
  15. L2:2997 weakly excites
  16. L3:421 weakly suppresses
  17. L1:2891 weakly suppresses
  18. L1:2380 weakly excites
  19. L3:1756 weakly suppresses
  20. L3:2639 weakly suppresses
  21. L3:1266 weakly excites
  22. L2:1608 weakly suppresses
  23. L1:1593 weakly suppresses
  24. L0:1784 weakly suppresses

Feeds later layers

  1. L5:2825 weakly excites
  2. L5:2552 weakly suppresses
  3. L6:1556 weakly suppresses
  4. L8:2897 weakly suppresses
  5. L19:2132 weakly suppresses
  6. L16:2683 weakly excites
  7. L8:2797 weakly suppresses
  8. L10:299 weakly suppresses
  9. L9:1015 weakly excites
  10. L12:911 weakly suppresses
  11. L10:2430 weakly suppresses
  12. L16:1642 weakly suppresses
  13. L22:1196 weakly suppresses
  14. L22:2511 weakly excites
  15. L7:2179 weakly excites
  16. L22:2856 weakly suppresses
  17. L7:2292 weakly suppresses
  18. L26:2859 weakly suppresses
  19. L9:892 weakly suppresses
  20. L6:2425 weakly suppresses
  21. L13:2402 weakly suppresses
  22. L16:1297 weakly excites unclear: document ends and answer colons
  23. L12:213 weakly excites
  24. L24:2779 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 -3.54 and -3.421 values between -1.56 and -1.441 values between -1.21 and -1.091 values between -1.09 and -0.981 values between -0.98 and -0.865 values between -0.86 and -0.743 values between -0.74 and -0.635 values between -0.63 and -0.5111 values between -0.51 and -0.3970 values between -0.39 and -0.28304 values between -0.28 and -0.161180 values between -0.16 and -0.041773 values between -0.04 and 0.07431 values between 0.07 and 0.1980 values between 0.19 and 0.3123 values between 0.31 and 0.427 values between 0.42 and 0.542 values between 0.54 and 0.664 values between 0.66 and 0.772 values between 0.77 and 0.893 values between 1.01 and 1.12offnegativepositiveoften

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. ложения
  2. aments
  3. ))/(
  4. pdata
  5. eos
  6. amente
  7. ся
  8. 直至
  9. junto
  10. -rating

Suppresses

  1. ��
  2. Scope
  3. scope
  4. şık
  5. SCO
  6. ła
  7. OfWork
  8. Scope
  9. otre
  10. .Scope