Neuron Explorer

Qwen3-0.6B

'because' introducing a reason

A neuron in layer 13 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 words like 'because' or 'is' right after a claim that is about to be explained, such as 'the reason for this is'. When turned up, the model drifted into odd, unrelated text, so the steering does not clearly match.

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

    are several reasons why people may choose to eat white turkey meat. Some people believe that it is more tender and flavorful than dark meat. Others prefer the white meat because it is lower in fat and calories. And some people simply like the way it

  2. #2 peak

    out of principle alone. For starters, he refused to ever short a stock. That’s because shorting a stock is basically betting on someone else to fail, meaning you profit

  3. #3 peak

    way the decision goes.⏎Those are the ‘easy’ situations<|endoftext|>You hear it all the time, “don’t write your picture book in rhyme.” That’s because many writers don’t do it well. The story suffers to fit the rhyme;

  4. #4 peak

    their nests and babies.⏎● To help them fly. Geese honk while they’re flying to help them maintain their V formation. They fly in this formation because it helps them cut down on wind resistance, allowing them to save energy.⏎●

  5. #5 peak

    In addition, 50 percent more men are affected than women, according to the National Institute of Neurological Disorders and Stroke. However, the reason for this is unclear.⏎Family history is another important risk factor. Individuals with a parent or sibling

  6. #6 peak

    relor.⏎Counseling Point for Anti-Platelets:⏎Aspirin and warfarin are sometimes prescribed together, and this is because one is anti platelet and the other is anticoagulant. Please see

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: 10000000000000000000000000000000000000000000000

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

    ex-ante consumption expenditure is Rs. 250 Cr. and the ex-ante investment expenditure is Rs. 100 Cr. in an economy without a government.⏎21. Differentiate between foreign trade surplus and

  2. #2 peak

    osacral portion) or lower extremities.⏎Causes of chronic pelvic pain⏎Chronic pelvic pain may originate due to multiple etiological factors. The commonly studied causes of chronic pelvic pain are as follows-⏎- Central and peripheral nervous systems:

  3. #3 peak

    the occasion will be celebrated as a sign of the boy's passage into maturity⏎and promising future. A significant event in the culture and history of South⏎India is known as the Dhoti Ceremony. This is a momentous day in the

  4. #4 peak

    joint project of the non-profit National Association of County and City Health Officials (NACCHO) and the Centers for Disease Control and Prevention (CDC), the U.S. government agency that “serves as the national focus for developing and applying disease

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. L12:2880 suppresses words that expect an of, to or for next
  2. L12:124 suppresses
  3. L12:2340 excites
  4. L11:1793 excites
  5. L10:614 suppresses
  6. L11:2997 suppresses
  7. L12:1760 suppresses
  8. L7:439 excites
  9. L12:2402 excites
  10. L11:305 excites
  11. L12:811 excites
  12. L7:2811 excites
  13. L9:2052 excites 'because' and 'that' after a reason
  14. L12:2779 weakly excites
  15. L11:1465 weakly suppresses
  16. L6:431 weakly excites
  17. L11:1986 weakly suppresses
  18. L10:1977 weakly excites
  19. L11:383 weakly suppresses
  20. L12:292 weakly suppresses
  21. L7:1307 weakly suppresses
  22. L11:2087 weakly excites
  23. L12:347 weakly suppresses
  24. L10:331 weakly excites

Feeds later layers

  1. L14:1096 suppresses
  2. L15:34 excites
  3. L14:1503 excites
  4. L17:24 excites
  5. L14:377 excites
  6. L17:2264 weakly suppresses
  7. L16:1474 weakly suppresses
  8. L22:612 weakly suppresses
  9. L22:1730 weakly suppresses
  10. L19:545 weakly excites
  11. L15:2765 weakly excites
  12. L14:2032 weakly excites
  13. L17:905 weakly suppresses
  14. L14:555 weakly excites
  15. L16:2677 weakly excites
  16. L14:735 weakly suppresses
  17. L16:1067 weakly excites
  18. L15:2589 weakly suppresses
  19. L19:865 weakly excites
  20. L16:1599 weakly suppresses
  21. L14:3009 weakly suppresses
  22. L14:1741 weakly suppresses
  23. L17:110 weakly excites
  24. L14:2003 weakly 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.

2 values between -2.47 and -2.291 values between -2.12 and -1.945 values between -1.94 and -1.779 values between -1.77 and -1.599 values between -1.59 and -1.4113 values between -1.41 and -1.2423 values between -1.24 and -1.0644 values between -1.06 and -0.8967 values between -0.89 and -0.71137 values between -0.71 and -0.54217 values between -0.54 and -0.36371 values between -0.36 and -0.19825 values between -0.19 and -0.011142 values between -0.01 and 0.17463 values between 0.17 and 0.34244 values between 0.34 and 0.52130 values between 0.52 and 0.6979 values between 0.69 and 0.8748 values between 0.87 and 1.0423 values between 1.04 and 1.2222 values between 1.22 and 1.408 values between 1.40 and 1.577 values between 1.57 and 1.754 values between 1.75 and 1.924 values between 1.92 and 2.104 values between 2.10 and 2.271 values between 2.27 and 2.453 values between 2.45 and 2.622 values between 2.80 and 2.981 values between 4.38 and 4.56offnegativepositiveoften

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. ери
  2. VERT
  3. Jenner
  4. STER
  5. Vance
  6. 元左右
  7. Piper
  8. mer
  9. 往
  10. ницы

Suppresses

  1. %=
  2. MonoBehaviour
  3. 髑
  4. ksi
  5. bases
  6. .Generated
  7. jede
  8. extends
  9. 蹇
  10. spins