Neuron Explorer

Qwen3-0.6B

"the" or "of" before a noun phrase

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

This neuron fires on small connecting words like "the" and "of" that come right before the thing being talked about, for example "bracing of the leg" or "the continuous effort of the foundation". On the negative side it fires on "of" after a noun, as in "floor of the mouth". Steering did not give a clear theme, so the pattern is only a rough guess.

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

    do this:⏎- Start by analyzing the message you want to convey to the reader. Before presenting it, define the problem described in the text, the expected reaction of the reader to the text, the purpose of the text, as well as the

  2. #2 peak

    software, novel materials, more dexterous robots, new processes (notably three-dimensional printing) and a whole range of web-based services. The factory of the past was based on cranking out zillions of identical products: Ford famously said

  3. #3 peak

    he spotted a tub of yeast in the room. Recalling an old tradition that rotting meat suspended over yeast would become pure and sweet again, he dosed the youngster with it – and cured him.⏎He treated several other parishioners in

  4. #4 peak

    test method of the improved Godron M. The plant community stability was closest to the point of stability

  5. #5 peak

    0.93.3⏎To perform the test, the patient is instructed to stand on one leg without support of the upper extremities or bracing of the unweighted leg against the stance leg. The patient begins the test with the eyes

  6. #6 peak

    double strand break DNA break (DSB) most often results in the activation, by phosphorylation (indicated in this figure by a circled p), of the PI3 kinase ATM. Multiple pathways are downstream of ATM including

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 number of the days you are in the room, the number of the days you are in the room, the number of the days you are in the room, the number of the days you are in the room... the number of the

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

    Despite being the largest pair of glands, only approximately 25% of saliva is produced by the glands.⏎The are a pair of glands located beneath the floor of the mouth, superior to the . The secretion produced is a mixture of both and

  2. #2 peak

    triangle is equal to the sum of the squares of the sides containing the right angle." But Pythagoras had an all-inclusive

  3. #3 peak

    gen cyanide is thought to be the main anticancer compound formed from laetrile when broken down in the body. Amygdalin is found in the pits of many fruits and in raw nuts. It is also found in other plants such as

  4. #4 peak

    of the Prophet Jonah, “As Jonah was in the belly of the whale three days and three nights, so also shall the Son of Man be in the heart of the earth three days and three nights” (Mt. 12:

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. L4:1686 strongly excites
  2. L6:2664 strongly excites
  3. L5:2454 excites
  4. L6:357 suppresses
  5. L6:388 excites
  6. L6:487 suppresses
  7. L6:1735 excites
  8. L5:363 excites
  9. L6:1114 excites
  10. L1:1565 excites
  11. L6:2326 suppresses
  12. L4:4 suppresses
  13. L7:8 excites
  14. L7:23 suppresses
  15. L6:925 suppresses
  16. L5:453 suppresses
  17. L1:752 excites
  18. L6:1637 excites
  19. L6:1534 suppresses
  20. L3:315 excites 'of' after a noun phrase
  21. L6:2785 excites
  22. L5:38 suppresses
  23. L5:1389 weakly excites
  24. L7:751 weakly suppresses

Feeds later layers

  1. L10:272 excites
  2. L10:362 suppresses
  3. L18:520 suppresses
  4. L14:1519 suppresses
  5. L10:469 suppresses 'of' after a noun phrase
  6. L14:1659 suppresses
  7. L18:183 excites
  8. L16:2612 suppresses
  9. L15:632 suppresses
  10. L13:203 suppresses
  11. L15:349 suppresses
  12. L9:283 excites
  13. L9:335 suppresses 'of' after nouns like anniversary or onset
  14. L14:655 suppresses
  15. L17:865 suppresses
  16. L9:592 suppresses
  17. L18:381 suppresses
  18. L15:1028 suppresses
  19. L16:1551 suppresses
  20. L20:482 suppresses
  21. L12:2849 suppresses '-based' after a place name
  22. L12:850 excites
  23. L14:126 excites
  24. L13:210 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 -5.11 and -4.911 values between -4.09 and -3.891 values between -3.89 and -3.684 values between -3.48 and -3.282 values between -3.28 and -3.072 values between -3.07 and -2.874 values between -2.87 and -2.662 values between -2.66 and -2.462 values between -2.46 and -2.263 values between -2.26 and -2.053 values between -2.05 and -1.855 values between -1.85 and -1.658 values between -1.65 and -1.446 values between -1.44 and -1.246 values between -1.24 and -1.0315 values between -1.03 and -0.8325 values between -0.83 and -0.6344 values between -0.63 and -0.4292 values between -0.42 and -0.22632 values between -0.22 and -0.011852 values between -0.01 and 0.19922 values between 0.19 and 0.39229 values between 0.39 and 0.6031 values between 0.60 and 0.806 values between 0.80 and 1.011 values between 1.01 and 1.212 values between 1.21 and 1.411 values between 1.41 and 1.623 values between 1.62 and 1.821 values between 2.02 and 2.231 values between 2.23 and 2.431 values between 2.84 and 3.04offnegativepositiveoften

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. ][-
  2. 在香港
  3. 是我的
  4. 让消费者
  5. 凭什么
  6. depuis
  7. 让她
  8. 是韩国
  9. 帮我
  10. нстру

Suppresses

  1. mankind
  2. implements
  3. ELSE
  4. lic
  5. adar
  6. (savedInstanceState
  7. vik
  8. ETO
  9. usra
  10. unicorn