Neuron Explorer

Qwen3-0.6B

"it" as a subject pronoun

A neuron in layer 8 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

Fires on the word "it" when it starts a clause and refers back to a thing or organisation named earlier, such as a program, a stamp scheme or a satellite. Turning it up did not produce any clear theme in the writing.

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

    history in this sense `scientific history'.⏎Although the object of research and the subject matter of scientific history are the events and episodes of the past, the laws which it deduces are not

  2. #2 peak

    Five miles south of here is Egan Canyon, where one of the oldest gold mines in Nevada was located. Indians mined gold there. A stage station was located by Major Howard Egan in 185

  3. #3 peak

    An element can be represented on several diagrams. You can quickly locate which diagrams it occurs in throughout the model, using the Element Usage dialog. Having identified a diagram

  4. #4 peak

    to purchase wetlands and wildlife habitat for inclusion into the National Wildlife Refuge System — the only federal lands where wildlife comes first.⏎Since the stamp program was created, it has generated over $700 million and protected over 5.2 million

  5. #5 peak

    aviolet background—that transform it into a mere dark-matter shadow of its former self.⏎Simultaneously, as the satellite galaxy moved closer to the massive system, it encountered the overwhelming gravitational force of the much larger mass. This force wrenched lumin

  6. #6 peak

    of color on the moon for the first time.⏎The vehicle the astronauts ride in will look different as well. Rather than a rocket and capsule built by NASA, it will be constructed by SpaceX. The private firm pitched its massive Starship rocket for

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: The user might be interested in using a different approach to find the solution. Let me think about some of the possible approaches. First, I will try to use the given equation $ f(x) = \frac{f(x)}{x

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

    modified file (here called source.cpp) and the non-modified version of the file (here called source.cpp.orig.)⏎diff -u -p source.cpp.orig source.cpp⏎That lists the difference between the two files in the unified diff

  2. #2 peak

    Mr Kamesh.Let's walk through the program once again. 1) ValHold v = new ValHold(); v.i=30; another(v,i);⏎This code creates a obj. reference v and passes that reference to the

  3. #3 peak

    can be the ideal vehicle for children to begin to develop some insight into the experiences of individuals escaping from conflict situations, and empathy for their plight. Despite the theme these books all contain threads of hope and indicate the power of kindness.⏎Azzi in

  4. #4 peak

    sample configuration file which resides under Drupal directory ‘sites/default/default.settings.php’.⏎# cd /var/www/html/drupal⏎# cp sites/default/default.settings.php sites/default/settings.php⏎Now make a “default” directory writable by installer.⏎#

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. L7:312 strongly excites
  2. L6:142 excites
  3. L3:528 excites
  4. L6:270 excites
  5. L4:1088 suppresses
  6. L7:401 suppresses
  7. L6:324 excites
  8. L7:193 suppresses space before day after month
  9. L6:59 suppresses
  10. L7:23 suppresses
  11. L6:61 suppresses
  12. L6:29 weakly suppresses
  13. L0:1926 weakly excites
  14. L2:1739 weakly excites
  15. L6:45 weakly suppresses
  16. L7:396 weakly excites
  17. L4:1527 weakly excites
  18. L7:59 weakly suppresses
  19. L3:472 weakly suppresses
  20. L4:260 weakly excites
  21. L4:319 weakly suppresses
  22. L7:13 weakly suppresses 'the' after 'member of' or 'one of'
  23. L5:8 weakly excites
  24. L6:152 weakly excites

Feeds later layers

  1. L14:254 suppresses
  2. L9:1557 suppresses 'the' after 'disease of' or 'infection of'
  3. L17:609 suppresses
  4. L13:536 suppresses
  5. L10:67 suppresses
  6. L9:145 suppresses
  7. L11:55 excites
  8. L16:342 suppresses
  9. L10:342 suppresses
  10. L12:72 suppresses
  11. L11:577 suppresses
  12. L12:71 suppresses
  13. L9:682 suppresses
  14. L18:598 suppresses
  15. L15:376 excites
  16. L9:172 excites
  17. L14:382 suppresses
  18. L11:329 weakly suppresses number or citation fragment split
  19. L10:121 weakly suppresses
  20. L11:2805 weakly suppresses "to" or "for" linking into a following phrase
  21. L15:2427 weakly suppresses
  22. L16:844 weakly suppresses
  23. L9:270 weakly excites
  24. L13:317 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 -3.00 and -2.851 values between -2.00 and -1.861 values between -1.57 and -1.431 values between -1.43 and -1.291 values between -1.29 and -1.141 values between -1.14 and -1.004 values between -1.00 and -0.865 values between -0.86 and -0.723 values between -0.72 and -0.5721 values between -0.57 and -0.4357 values between -0.43 and -0.29314 values between -0.29 and -0.151348 values between -0.15 and -0.001555 values between -0.00 and 0.14433 values between 0.14 and 0.2898 values between 0.28 and 0.4331 values between 0.43 and 0.5714 values between 0.57 and 0.715 values between 0.71 and 0.851 values between 0.85 and 1.007 values between 1.00 and 1.141 values between 1.14 and 1.282 values between 1.28 and 1.421 values between 1.57 and 1.711 values between 1.71 and 1.851 values between 2.56 and 2.71offnegativepositiveoften

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. eral
  2. ดวง
  3. хват
  4. 网讯
  5. pan
  6. ẳng
  7. notation
  8. compet
  9. 离
  10. cribing

Suppresses

  1. aukee
  2. Georg
  3. Augusta
  4. urent
  5. ftware
  6. Benedict
  7. iled
  8. .getOwnPropertyDescriptor
  9. aurant
  10. Dank