Neuron Explorer

Qwen3-0.6B

"or" in either-or choices

A neuron in layer 20 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 "or" when the sentence offers two alternatives, like "true or false", "whether it is X or not" or "one way or another". Turned up, the model started repeating "not not not".

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

    foods at the Jimmy Dean Foods division of Sara Lee Corporation discovered that breakfast for more than 60% of Americans consists of cold foods. Whether that’s true or not is debatable but what nutritionists do agree on is that it’s a

  2. #2 peak

    course antique wood and furniture restoration and the application of traditional restoration techniques and materials. In this context, one might ask: Does restoration increase the value of an object or diminish

  3. #3 peak

    the art appreciation and there have been discussions concerning its future.⏎Today, lots of people dispute whether art still has an essential duty in the social background of a nation or society. Numerous experts agree that art and also the creation of artworks is extremely essential

  4. #4 peak

    , doctors need to know if you are taking these products so they can accurately judge whether particular symptoms are related to a medical illness or a side effect from an herbal medicine. Those who take herbal products should also keep

  5. #5 peak

    ers,” he says, “So the next step is to refine these idealisations and relax the initial conditions to make it more general.” [Nature]⏎To confirm or refute this model, astronomers

  6. #6 peak

    all kinds has been going on to determine if there’s danger to people or wildlife.⏎“Air monitors in the community, throughout the area. We were testing

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 a 324,000} * 0.000324 + 0.000324 * 00.000324 + 00.

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

    ?⏎a demon is a long-running background process that responds to requests for services. The term originated with Unix, but most operating systems use daemons in one form or another. On Unix, daemon names conventionally end in "d". Some examples

  2. #2 peak

    lasted in the UK until 200<|endoftext|>Technology has become a defining force in our lives, and its relevance is spreading to virtually every field in one way or another. While some people may go their entire lives never needing to recall the knowledge

  3. #3 peak

    are examples of creating markers, setting markers, and moving point to markers:⏎;; Make a new marker that initially does not point anywhere: (setq m1 (make-marker)) ⇒ #<marker in no buffer>⏎m1to point between the

  4. #4 peak

    undeserved . They will also suspect that<|endoftext|>Constipation can cause difficult and painful bowel movements of dry, hard stool. Most people deal with it at one time or another. The most common causes of the problem are poor diet and a lack of

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. L17:2638 suppresses
  2. L17:1480 excites
  3. L19:2864 excites
  4. L13:2939 excites
  5. L18:1302 excites
  6. L19:2686 suppresses
  7. L18:1582 weakly suppresses
  8. L19:2544 weakly suppresses
  9. L8:3 weakly excites
  10. L18:5 weakly excites
  11. L13:1603 weakly suppresses
  12. L18:1203 weakly excites end of document marker
  13. L8:7 weakly excites
  14. L18:1586 weakly suppresses
  15. L19:1571 weakly excites
  16. L14:249 weakly excites
  17. L17:1407 weakly excites "and" joining questions (wh-question context)
  18. L2:126 weakly excites
  19. L18:2036 weakly excites
  20. L8:818 weakly suppresses
  21. L19:835 weakly suppresses
  22. L9:1098 weakly suppresses
  23. L18:584 weakly suppresses
  24. L9:2055 weakly suppresses

Feeds later layers

  1. L21:1044 strongly excites
  2. L24:334 excites
  3. L26:406 excites
  4. L22:2227 excites
  5. L23:434 excites
  6. L24:2628 excites
  7. L21:1538 excites
  8. L27:714 excites
  9. L21:1839 excites
  10. L24:776 excites
  11. L23:773 excites
  12. L25:541 excites
  13. L25:2646 excites 'there' after a because or if clause
  14. L27:838 excites
  15. L24:402 weakly excites
  16. L27:1334 weakly suppresses
  17. L23:223 weakly excites
  18. L27:233 weakly excites
  19. L21:749 weakly excites
  20. L21:744 weakly excites
  21. L27:138 weakly excites
  22. L22:539 weakly excites
  23. L25:1270 weakly excites
  24. L21:360 weakly excites 'such' introducing examples

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.

3 values between -0.95 and -0.851 values between -0.85 and -0.747 values between -0.74 and -0.636 values between -0.63 and -0.5314 values between -0.53 and -0.4228 values between -0.42 and -0.3168 values between -0.31 and -0.20193 values between -0.20 and -0.101842 values between -0.10 and 0.011332 values between 0.01 and 0.12244 values between 0.12 and 0.2290 values between 0.22 and 0.3332 values between 0.33 and 0.4416 values between 0.44 and 0.5414 values between 0.54 and 0.657 values between 0.65 and 0.761 values between 0.76 and 0.872 values between 0.87 and 0.975 values between 0.97 and 1.081 values between 1.40 and 1.511 values between 3.11 and 3.221 values between 3.22 and 3.32offnegativepositiveoften

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. not
  2. 不是
  3. ли
  4. not
  5. 否
  6. Not
  7. Not
  8. 不是一个
  9. không
  10. 不在

Suppresses

  1. }*
  2. (CC
  3. 纽
  4. ⽅
  5. »
  6. 趸
  7. .dds
  8. ;c
  9. tré
  10. ístico