Neuron Explorer

Qwen3-0.6B

"terms" in "in terms of"

A neuron in layer 10 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 reacts very strongly to the word "terms" in the phrase "in terms of", as in "advantages in terms of speed". Turning it up made the model's text wander into repetitive loops instead of anything about that phrase.

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

    approval – especially in matters concerning finance and law. As the world changes, most of our financial matters have moved seamlessly online. Our digital identity is lagging in terms of adoption but it is only due to the sensitivity of the matter at hand.⏎

  2. #2 peak

    from five healthy volunteers harbored a large and<|endoftext|>Although a Last Will and Testament remains the most common estate planning tool, a living trust is not far behind in terms of popularity. If you find yourself to be the beneficiary of a living trust,

  3. #3 peak

    a site of memory, including the ‘ESMA Memory Site Museum,’ dedicated to telling the stories about captivity in the building.⏎“Argentina had a great advance in terms of truth trials, but there’s a lot left to do for women,” said

  4. #4 peak

    If equity isn’t at the center of the decisions that are made to open up outdoor spaces, “we risk perpetuating existing inequalities that have a real cost in terms of the health of communities of color as well as others who have historically lacked equitable

  5. #5 peak

    and multilingual society. “It was after having been in the USA, where I studied Library Science, that I saw that Malaysia has a lot to offer in terms of its diversity. You grow up aware of differences but being comfortable with it.”

  6. #6 peak

    the highest quality in terms of natural flavor and taste, and which are free from additives and preservatives.

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 trying to create a system that will help users build a virtual reality environment. What is the first thing I think about doing? I need to create a system that will allow me to build a virtual environment, what I need to do

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

    batch of 41 country pages, with more to follow. They are:⏎Afghanistan, Angola, Argentina, Bahrain, Belarus, Belgium, Burkina Faso, Burundi, Cambodia, Cameroon, Colombia, Costa Rica, Czech Republic

  2. #2 peak

    .⏎The passages in classical literature which are to be considered, are the following: —⏎(1) Dionysius II 65: οὔτε γὰπ τὸ χωρίον τοῦ

  3. #3 peak

    States and Mexico; The opposite occurs with Cubans and Puerto Ricans living in the United States.⏎This is reflected in a survey carried out between

  4. #4 peak

    the presence of Paleozoic rocks among the field of cainosoic rocks. Twenty years ago B.Pilya found some strange glasses here and P.V.Florensky began to study them. With a point of view of meteoritical

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. L6:293 strongly suppresses
  2. L8:62 strongly suppresses
  3. L4:1572 strongly excites
  4. L7:9 strongly suppresses
  5. L5:2525 strongly suppresses
  6. L6:296 strongly excites
  7. L5:418 excites
  8. L8:79 suppresses
  9. L5:667 suppresses
  10. L4:1045 excites
  11. L6:1031 excites
  12. L6:1058 suppresses
  13. L6:2400 suppresses
  14. L4:2029 excites
  15. L4:1781 suppresses
  16. L5:1127 excites
  17. L4:2382 suppresses
  18. L8:1546 suppresses
  19. L6:108 suppresses
  20. L2:87 suppresses
  21. L5:34 excites
  22. L5:58 excites
  23. L7:701 excites
  24. L4:1854 excites

Feeds later layers

  1. L17:70 strongly excites
  2. L12:50 strongly excites
  3. L19:204 strongly excites
  4. L20:131 strongly excites
  5. L17:152 excites mid-word fragments, often 'intric'
  6. L18:513 excites
  7. L14:118 excites
  8. L14:248 excites
  9. L14:254 excites
  10. L16:414 excites
  11. L14:126 excites
  12. L16:269 excites
  13. L16:283 excites
  14. L14:280 excites
  15. L15:953 excites
  16. L21:189 excites document break token mid-sentence
  17. L19:541 excites
  18. L24:39 excites
  19. L14:830 suppresses
  20. L22:164 excites
  21. L18:417 excites
  22. L14:421 suppresses
  23. L25:17 excites
  24. L17:259 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 -1.14 and -0.978 values between -0.97 and -0.80105 values between -0.80 and -0.63371 values between -0.63 and -0.46692 values between -0.46 and -0.30668 values between -0.30 and -0.13570 values between -0.13 and 0.04383 values between 0.04 and 0.21299 values between 0.21 and 0.37189 values between 0.37 and 0.54131 values between 0.54 and 0.71110 values between 0.71 and 0.8889 values between 0.88 and 1.0464 values between 1.04 and 1.2156 values between 1.21 and 1.3839 values between 1.38 and 1.5542 values between 1.55 and 1.7212 values between 1.72 and 1.8817 values between 1.88 and 2.0515 values between 2.05 and 2.227 values between 2.22 and 2.3913 values between 2.39 and 2.553 values between 2.55 and 2.722 values between 2.72 and 2.897 values between 2.89 and 3.063 values between 3.06 and 3.225 values between 3.22 and 3.391 values between 3.89 and 4.063 values between 4.06 and 4.232 values between 4.90 and 5.071 values between 5.40 and 5.57offnegativepositiveoften

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. itary
  2. udio
  3. 我又
  4. lider
  5. 目的
  6. -flex
  7. imately
  8. OfWork
  9. entially
  10. aised

Suppresses

  1. ningar
  2. jag
  3. }{⏎
  4. j
  5. rational
  6. <j
  7. pert
  8. \Db
  9. gra
  10. vido