Neuron Explorer

Qwen3-0.6B

'a' before a noun phrase

A neuron in layer 17 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 "a" when it comes between a verb or preposition and the noun that follows, like "has a long history" or "a good idea". Turned up, the model produced vague, repetitive phrases with "a lot" and "a kind".

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

    which you will read and write about for Exercise 1), Havelock launches his main argument about the shift from oral culture to literate culture, and is a good way for us to start wrapping up the oral unit.⏎Read the full article

  2. #2 peak

    do this by using a moisture meter. If your succulent is indoors, it’s a good idea to get a small spout watering can. It’s also a good idea to avoid using high pressure streams of water as this can damage the roots

  3. #3 peak

    beautiful discourse of the Holy Quran and Sunnah, Muslims come to learn about the various rights that Allah (SWT) has enjoined upon His creation – in a practical and detailed manner. Without the fulfillment of these basic rights, the whole institution

  4. #4 peak

    , and has evolved over time to reflect changing artistic styles and techniques. In this blog post, we'll explore the fascinating world of Japanese art.⏎Japanese art has a long and rich history, dating back to prehistoric times. Early examples of Japanese

  5. #5 peak

    & Conditions here⏎This set is designed to get students thinking and talking about math. With the 120 color-coded task cards, you will always have a problem at your fingertips. Use the cards on the carpet or in small groups—even

  6. #6 peak

    's a "citation?"⏎When you write a research paper, at the end of the paper you include a list of the resources you used to write the paper on a page titled "Works Cited" or "References." Each entry in that list

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 same of a one of the the other, but the one of the one of the one of the one... one of the. The one of the, but the one of the, but the one of the... one 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

    A growing body of evidence suggests that increasing poor people’s access to better financial tools can help

  2. #2 peak

    one place, and every day was exactly the same, and nothing that you did mattered?" Phil asks two local guys at the<|endoftext|>Courtesy of EarthSky⏎A Clear Voice for Science⏎Here is the view northward on April evenings. Now the

  3. #3 peak

    need time. In fact, Rahner thought of an entire century! Therefore, it is our task to deal with the globalization of society and the Church.⏎A Globalized World and a Universal Church⏎In 2013, the Roman

  4. #4 peak

    a 2013 study, Peter Strelan of the University of Adelaide looked at<|endoftext|>November 19, 1998⏎A weekly feature provided by scientists at the Hawaiian Volcano Observatory.⏎Hawaiian honeycre

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. L16:880 strongly suppresses
  2. L16:2248 strongly excites
  3. L4:137 strongly excites
  4. L9:59 strongly suppresses "a" meaning per (a day, a year)
  5. L11:331 strongly suppresses
  6. L16:1309 suppresses
  7. L4:49 excites
  8. L11:2859 suppresses
  9. L11:875 excites
  10. L16:2062 suppresses "other" and contrast words
  11. L0:173 excites
  12. L1:1062 suppresses
  13. L10:365 suppresses
  14. L16:2611 excites
  15. L3:2559 suppresses
  16. L3:1810 excites
  17. L15:1848 excites
  18. L3:597 suppresses
  19. L13:1234 excites
  20. L5:225 suppresses
  21. L10:1162 suppresses
  22. L6:640 excites "the" introducing a noun phrase after a name or preposition
  23. L16:1949 suppresses
  24. L16:342 excites

Feeds later layers

  1. L20:901 strongly excites
  2. L22:1881 strongly excites
  3. L18:635 strongly excites
  4. L21:592 strongly excites
  5. L19:2088 excites
  6. L18:1084 excites
  7. L23:218 excites
  8. L20:835 excites
  9. L18:216 suppresses
  10. L19:2115 suppresses
  11. L22:260 excites
  12. L27:387 excites 'a' right after 'played' or 'play'
  13. L22:211 excites
  14. L22:2381 excites "a" before "bit" or "little"
  15. L18:314 suppresses
  16. L27:2632 excites
  17. L25:1586 suppresses
  18. L25:390 excites
  19. L18:2389 excites
  20. L25:320 excites
  21. L18:2608 excites
  22. L19:751 excites the article 'a' before a noun
  23. L18:602 excites
  24. L24:489 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.

544 values between -2.15 and -1.153012 values between -1.15 and -0.16250 values between -0.16 and 0.838 values between 0.83 and 1.822 values between 1.82 and 2.821 values between 2.82 and 3.812 values between 3.81 and 4.801 values between 4.80 and 5.792 values between 14.72 and 15.712 values between 16.70 and 17.692 values between 17.69 and 18.694 values between 18.69 and 19.682 values between 19.68 and 20.678 values between 20.67 and 21.666 values between 21.66 and 22.655 values between 22.65 and 23.658 values between 23.65 and 24.646 values between 24.64 and 25.638 values between 25.63 and 26.626 values between 26.62 and 27.614 values between 27.61 and 28.608 values between 28.60 and 29.606 values between 30.59 and 31.583 values between 31.58 and 32.573 values between 32.57 and 33.561 values between 33.56 and 34.561 values between 34.56 and 35.551 values between 35.55 and 36.542 values between 36.54 and 37.53offnegativepositiveoften

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. lot
  2. few
  3. very
  4. certain
  5. solution
  6. variety
  7. combination
  8. ikip
  9. number
  10. wide

Suppresses

  1. 的一种
  2. slashes
  3. 这三个
  4. 的一个
  5. 的一款
  6. languages
  7. lard
  8. anches
  9. 有一天
  10. plings