Neuron Explorer

Qwen3-0.6B

"that" before a need or must clause

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

The neuron fires on "that" after a noun when the description that follows says something is needed, required or must be handled, like "issues that are common" or "a risk that needs to be managed." Turned up, it produced a stream of "need to have to", which fits its job of anticipating that kind of wording.

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

    for sale.⏎- Land auctions.⏎- Local authorities who may have parcels of land they are prepared to sell for development purposes.⏎- Utility companies may have surplus land that is for sale

  2. #2 peak

    to ensure that the symptoms are not being caused by other types of stressors or conditions. This way, the patient can be sure that there are no underlying causes that could become worse.⏎Ignoring the symptoms associated with dizziness and blurry vision can be

  3. #3 peak

    lorinated biphenyl (PCB). While these compounds have helped to protect the personal health and safety of humans, they create risks and concerns to the environment that are constantly being researched.”⏎An investigative study by the Associated Press in 20

  4. #4 peak

    material heating to its ignition temperature. Since the balance of oxygen flow and heat loss needs to be carefully balanced, spontaneous ignition is not common but is still a risk that needs to be managed Some businesses have had unsafe work practices for many years before they

  5. #5 peak

    Every data-driven trend comes with vulnerabilities that, if left unchecked and unprotected, can cause great damage.⏎In such cases,

  6. #6 peak

    and presented in the Global Report on Communication for Empowerment (2010). The findings from each of the five pilot countries reveal a number of issues that are common to more than one pilot country, some highlighting emerging trends, and others

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 going in the 205.50.50.55.5.5.5.5.5.5.5.5.5.5.5.5.5.5.5

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

    tools for electrical engineers, technicians and hobbyists. They help to visualize how a circuit works and provide essential information about its components and interconnections. But understanding these diagrams isn't always easy. To get the most out of them, it pays to

  2. #2 peak

    that it uses so many items from recycle bin!⏎Guitar craft is a great alternative to purchasing low-quality plastic toy guitars. And cheaper for sure! All you need are few simple items you probably already have at home.⏎Make an amazing ABC DIY

  3. #3 peak

    added. … Baking powder is a combination of sodium bicarbonate and powdered acid (like cream of tartar). Because it already contains an acid, all it needs is moisture and heat added to activate it.⏎Can vinegar and baking soda be mixed

  4. #4 peak

    . I cover more reasons in this post on why seeds are better than buying seedlings.⏎Seed starting can sound intimidating! All the supplies and tools you think you need, when and where to start seeds and how to take care of them until they

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. L17:2255 excites
  2. L19:2648 suppresses
  3. L18:522 suppresses
  4. L18:1162 suppresses
  5. L20:1015 suppresses
  6. L19:1298 suppresses
  7. L17:2087 suppresses
  8. L3:1914 suppresses
  9. L19:945 weakly suppresses
  10. L20:2035 weakly suppresses
  11. L12:1678 weakly excites
  12. L17:257 weakly suppresses
  13. L15:2624 weakly suppresses
  14. L20:1205 weakly suppresses
  15. L20:386 weakly suppresses
  16. L18:2301 weakly excites
  17. L17:2643 weakly suppresses
  18. L19:2676 weakly excites
  19. L19:2168 weakly suppresses
  20. L17:2818 weakly suppresses
  21. L19:1681 weakly excites
  22. L18:2474 weakly excites
  23. L14:2208 weakly suppresses
  24. L18:2719 weakly excites

Feeds later layers

  1. L24:764 strongly excites
  2. L23:1876 strongly excites
  3. L23:1500 excites
  4. L25:1678 excites
  5. L22:1422 excites
  6. L27:1123 excites
  7. L24:1041 excites
  8. L26:1276 excites
  9. L22:2551 excites
  10. L24:2481 excites
  11. L24:2268 excites
  12. L23:103 suppresses
  13. L23:1542 excites
  14. L22:2891 excites
  15. L24:1571 excites
  16. L27:1493 excites
  17. L22:2322 excites
  18. L26:1955 weakly excites
  19. L24:2503 weakly suppresses
  20. L25:1181 weakly excites
  21. L22:2148 weakly excites
  22. L25:1211 weakly excites
  23. L22:3032 weakly excites
  24. L24:2784 weakly 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 -7.62 and -7.271 values between -7.27 and -6.921 values between -6.58 and -6.231 values between -5.54 and -5.191 values between -5.19 and -4.843 values between -4.49 and -4.156 values between -3.80 and -3.454 values between -3.45 and -3.116 values between -3.11 and -2.767 values between -2.76 and -2.4112 values between -2.41 and -2.0713 values between -2.07 and -1.7221 values between -1.72 and -1.3741 values between -1.37 and -1.0365 values between -1.03 and -0.68249 values between -0.68 and -0.331370 values between -0.33 and 0.021486 values between 0.02 and 0.36419 values between 0.36 and 0.71107 values between 0.71 and 1.0623 values between 1.06 and 1.4023 values between 1.40 and 1.7510 values between 1.75 and 2.1012 values between 2.10 and 2.456 values between 2.45 and 2.794 values between 2.79 and 3.142 values between 3.14 and 3.495 values between 3.49 and 3.833 values between 3.83 and 4.182 values between 4.18 and 4.531 values between 4.87 and 5.221 values between 5.22 and 5.571 values between 5.57 and 5.921 values between 5.92 and 6.26offnegativepositiveoften

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. need
  2. 需要
  3. 需
  4. 必须
  5. needs
  6. needs
  7. must
  8. must
  9. need
  10. 需要用

Suppresses

  1. omite
  2. <Type
  3. anesthesia
  4. 新闻网
  5. KeyName
  6. enna
  7. ocre
  8. rew
  9. درجة
  10. **)