Neuron Explorer

Qwen3-0.6B

'from' after prevent or stop

A neuron in layer 7 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 "from" when it follows verbs like "prevent" or "stop", as in "preventing small issues from escalating". Its negative side fires on "to" after words like "expect" or "plan". Turning it up made the model write unrelated text about maths and weather.

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

    ricated. Regular maintenance helps detect wear and tear early, preventing small issues from escalating into major repairs. This proactive approach to maintenance extends the life of the HVAC

  2. #2 peak

    material that you put on a chair or the floor to make it more comfortable [↪ pillow]:⏎a velvet cushion⏎a cushion cover⏎something that stops one thing from hitting another thing:⏎Good sports shoes should provide a cushion when running.⏎3 [

  3. #3 peak

    reactor fuel). To prevent the chain reaction from moving too rapidly, control mechanisms between the fuel containers absorb extra neutrons. As

  4. #4 peak

    prevented in most people by suitable antiepileptic medication. Oxcarbazepine works by stabilising the electrical activity of your brain, which prevents the seizures from occurring.⏎Before taking ox

  5. #5 peak

    her quite atypical was her chosen field of art specialization in sculpture rather than the standard popular domain of painting.⏎William Blair Bruce’s⏎formative years could not have been more different from Caroline’s. William was born in 1859

  6. #6 peak

    <|endoftext|>There are so many reasons to adapt to a more sustainable lifestyle.⏎Helping to cut down on the amount of plastic produced, reducing food waste, preventing rubbish from going to our landfills, and helping to save our oceans are just some of

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 a new student, I have a new course, I have a new teacher. I have a new student. I have a new course. I have a new teacher. I have a new student. I have a new teacher.

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

    itia Barbauld and her brother, John Aikin, Sarah Hale explains that the texts were written ?for English children. You must, therefore, expect to find the scenes mostly laid in England, and English character and manners described.?1

  2. #2 peak

    available for you to download for free, either on e-reader or desktop, until December 5 at https://www.smashwords.com/<|endoftext|>If you expect to have any hope of building wealth you need to have a firm grasp on basic financial

  3. #3 peak

    tools for cultivation don't have to be expensive, and the size of each tool may differ depending upon the dimension of your garden space and what plants you plan to grow.⏎Shovels, Spades, and Trowels⏎Shov

  4. #4 peak

    . We had lots of fun decorating our tins in lots of way. It was lots of fun to experiment with paint and mixing colours. Some of us decided to glue pom poms, sequin and tissue paper onto our tins too.

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. L6:2747 strongly suppresses
  2. L6:914 strongly excites
  3. L6:1133 excites "to" before an infinitive verb
  4. L6:2882 suppresses
  5. L6:439 suppresses
  6. L6:1621 suppresses
  7. L6:2064 suppresses
  8. L6:1880 excites "by" introducing a deadline or means
  9. L6:2359 excites
  10. L4:113 suppresses
  11. L4:518 suppresses
  12. L4:650 suppresses
  13. L4:2890 suppresses
  14. L5:1755 excites
  15. L5:2026 excites "to" after made or need-type verbs
  16. L6:2057 suppresses
  17. L4:312 excites 'that' after 'the way' or 'the first time'
  18. L3:1128 suppresses
  19. L6:2544 suppresses
  20. L6:2407 suppresses
  21. L5:2217 suppresses
  22. L5:1387 suppresses
  23. L3:682 excites
  24. L4:1391 excites

Feeds later layers

  1. L8:369 strongly suppresses 'of' after 'learned' or 'because'
  2. L16:1153 strongly excites
  3. L19:385 suppresses
  4. L11:429 excites
  5. L8:403 excites
  6. L8:94 excites
  7. L14:659 excites
  8. L9:226 excites 'to' starting an infinitive
  9. L15:1870 excites
  10. L19:776 excites
  11. L8:333 excites
  12. L18:452 excites
  13. L27:626 excites
  14. L24:208 excites
  15. L19:617 excites
  16. L17:1994 excites
  17. L8:500 excites 'to' after verbs like continue or help
  18. L15:1176 excites
  19. L17:1592 excites
  20. L16:559 excites weak, mixed: verbs and adverbs in claims
  21. L18:715 excites
  22. L26:436 excites
  23. L20:480 excites
  24. L18:141 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 -4.58 and -4.321 values between -2.48 and -2.221 values between -2.22 and -1.964 values between -1.96 and -1.703 values between -1.70 and -1.443 values between -1.44 and -1.188 values between -1.18 and -0.9115 values between -0.91 and -0.6570 values between -0.65 and -0.39939 values between -0.39 and -0.132598 values between -0.13 and 0.13187 values between 0.13 and 0.4026 values between 0.40 and 0.667 values between 0.66 and 0.9214 values between 0.92 and 1.183 values between 1.18 and 1.4413 values between 1.44 and 1.704 values between 1.70 and 1.974 values between 1.97 and 2.231 values between 2.23 and 2.491 values between 2.49 and 2.751 values between 3.54 and 3.801 values between 3.80 and 4.061 values between 4.84 and 5.111 values between 5.37 and 5.631 values between 5.63 and 5.89offnegativepositiveoften

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. jam
  2. ğu
  3. sville
  4. getArguments
  5. asonry
  6. artment
  7. 在北京
  8. пут
  9. anych
  10. odby

Suppresses

  1. little
  2. weakest
  3. least
  4. knowledge
  5. principal
  6. top
  7. context
  8. 桌上
  9. false
  10. derivative