Neuron Explorer

Qwen3-0.6B

unclear: sentence ends in how-to text

A neuron in layer 15 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 tends to fire on the full stop or line break that closes a sentence in advice, instructions or educational text. It also fires on a few unrelated spots, and its negative side seems to dislike the subject and verb in grammar examples about passive voice. Turning it up made the model get stuck repeating words, and turning it down gave plainer prose, so its job is hard to pin down.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    . Its high fiber content will also help you feel fuller with fewer calories. Try adding broccoli to your Asian-inspired stir-fries for a quick, nutrient-packed meal.⏎One last note: buy organic produce when you can to avoid

  2. #2 peak

    solution A) isotonic B) hypertonic C) hypotonic D) none of these I think it's B. I'm not too sure though.⏎November 27, 2007<|endoftext|>Hopefully your child will not experience

  3. #3 peak

    in San Mateo County compared to previous years; however, ticks are present and the diseases they carry can be very serious. Learn how to prevent tick bites here.⏎Click on the tick below to see a video of a tick 'quest

  4. #4 peak

    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 straight forward way to find

  5. #5 peak

    honesty⏎Start early to show your child that you trust him and that honesty is important to you, and you'll lay the groundwork for rearing an honest adult.⏎How to teach your child responsibility⏎To instill a sense of responsibility in your

  6. #6 peak

    2017 | Max-Planck-Institut für Bildungsforschung⏎A sudden drop in outdoor temperature increases the risk of respiratory infections⏎11.0

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 want to, as a computer, make this part of the world, which is my computer, to make it more efficiently and, as a computer, to use some other parts, which are available in the world, to develop some other

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

    by Rita. Future (or modal verbs) : A letter will be written by Rita.⏎Example of Present Simple: Active: Tom cleans the house. Passive: The house is cleaned by Tom. Example of Past

  2. #2 peak

    to describe actions that began and were completed in the past. E.g. I had gone to the shop and bought some chips.⏎Other examples include.⏎After she had read it from start to finish, Sophie ripped the letter in half.⏎The frog

  3. #3 peak

    dog killed a ratA rat was killed by the dog The teacher will explain the story The story will be explained by the teacher⏎Examples Simple present: A letter is written by Rita. Simple past: A letter was written by Rita. Present perfect

  4. #4 peak

    geyser and found that the distribution is symmetric. Draw a boxplot of these data. Use the boxplot and quartiles to ...The following data represent the ages of chief justices of the U.S. Supreme Court when they were appointed.

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. L13:2182 suppresses
  2. L14:1257 excites
  3. L13:2363 suppresses
  4. L13:1154 suppresses
  5. L13:1860 excites
  6. L12:215 excites
  7. L13:2454 suppresses
  8. L13:1407 excites
  9. L13:1620 suppresses
  10. L14:2434 excites
  11. L14:993 suppresses
  12. L13:646 excites
  13. L13:381 excites
  14. L13:51 excites
  15. L14:1117 suppresses
  16. L11:1567 excites
  17. L13:680 excites
  18. L12:757 excites
  19. L14:2631 excites
  20. L14:1828 weakly suppresses
  21. L14:1125 weakly excites
  22. L14:1569 weakly suppresses
  23. L14:2143 weakly suppresses
  24. L12:1932 weakly excites

Feeds later layers

  1. L16:1132 strongly excites
  2. L16:2525 excites
  3. L18:497 excites
  4. L16:1733 excites
  5. L16:2916 suppresses
  6. L16:1498 suppresses
  7. L16:1904 excites
  8. L16:1034 excites
  9. L16:510 suppresses
  10. L20:204 excites
  11. L17:251 excites
  12. L17:2726 excites
  13. L16:1445 excites
  14. L16:613 weakly excites
  15. L16:1688 weakly excites
  16. L17:2629 weakly excites
  17. L18:379 weakly excites
  18. L16:767 weakly suppresses
  19. L18:664 weakly excites
  20. L16:330 weakly excites
  21. L16:472 weakly excites
  22. L16:2717 weakly suppresses
  23. L16:1227 weakly excites
  24. L16:962 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 -4.40 and -4.201 values between -3.80 and -3.601 values between -3.60 and -3.401 values between -2.01 and -1.811 values between -1.62 and -1.422 values between -1.42 and -1.226 values between -1.22 and -1.023 values between -1.02 and -0.8210 values between -0.82 and -0.6225 values between -0.62 and -0.42108 values between -0.42 and -0.221212 values between -0.22 and -0.022120 values between -0.02 and 0.18228 values between 0.18 and 0.3785 values between 0.37 and 0.5737 values between 0.57 and 0.7728 values between 0.77 and 0.9711 values between 0.97 and 1.1710 values between 1.17 and 1.3710 values between 1.37 and 1.574 values between 1.57 and 1.771 values between 1.77 and 1.961 values between 1.96 and 2.161 values between 2.16 and 2.361 values between 3.36 and 3.56offnegativepositiveoften

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. another
  2. 又
  3. fox
  4. pac
  5. Another
  6. ito
  7. ebenfalls
  8. 璧
  9. 又有
  10. issa

Suppresses

  1. rine
  2. ucks
  3. iences
  4. atisation
  5. ulist
  6. ophysical
  7. recru
  8. plevel
  9. -scripts
  10. aises