Neuron Explorer

Qwen3-0.6B

opening quote marks before a coined term

A neuron in layer 18 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires on the opening quotation mark that introduces a nickname or coined term, such as the quote before "Great Migration" or "mastiche". It is negative on the word "to" in menu-style link text like "Books to Print" or "Email to a Friend". Turning it up or down produced unrelated text, so the steering doesn't confirm this.

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

    economy?⏎By the 1950s, the flow of Puerto Ricans to the mainland United States had increased so drastically that historians dubbed the phenomenon the “Great Migration.” An estimated 470,000 people—or

  2. #2 peak

    culture, we have to start with ancient Greece, where citizens routinely chewed on the resin (or sap) from the Mastic tree. This substance was dubbed “mastiche

  3. #3 peak

    Prior to the restructuring of vineyards of 1955-60, one of the areas in which it was divided Sicily was called "Nero d'Avola and Frappato", coinciding with the provinces of Sir

  4. #4 peak

    in their September 1915 issue. The use of aerial detection to spot and report forest fires—dubbed the “Wisconsin Plan”—soon became a vital tool in fighting<|endoftext|>Books and Movies⏎

  5. #5 peak

    the fitness industry. As does the “no pain, no gain” myth, which came to prominence in the early 1980

  6. #6 peak

    training as part of ongoing efforts to implement NCA.⏎ONS recently published estimates of the total value of all the assets held in the UK, known as the ‘National Balance

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 ****************************************************************************************

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

    00 Web Pages⏎Sample Pages for Prospective Subscribers, or click below⏎Overview of Site|⏎Enchanted Learning Home⏎Monthly Activity Calendar⏎Books to Print⏎Parts of Speech⏎The Test of Time⏎TapQuiz Maps - free

  2. #2 peak

    1587). Among those who studied under Mercier was the Huguenot leader and author Philippe<|endoftext|>|Back to Main Print This Page Email to a Friend|⏎Pleural fluid analysis is a test that examines a sample of

  3. #3 peak

    works to give pride of place to those of high heritage value.⏎How the site is organized⏎The Across Cultures Web site is divided into five sections: Themes to Explore; Cinema and Representation; Points of View; See Everything, Hear Everything;

  4. #4 peak

    1:30 AM (CANBERRA TIME) 24/04/2001⏎|Page tools: Print Page RSS Search this Product|⏎This variable identifies holders of Australian citizenship.⏎Citizenship data are used

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. L13:1632 strongly suppresses words inside headlines and titles
  2. L14:790 suppresses
  3. L15:1865 suppresses
  4. L16:1652 suppresses
  5. L14:895 excites
  6. L15:2459 suppresses
  7. L17:1362 excites unclear: book and report titles
  8. L17:2414 excites
  9. L12:2654 excites
  10. L16:2398 excites
  11. L14:2061 excites
  12. L16:1086 suppresses
  13. L10:270 suppresses
  14. L15:2591 excites
  15. L4:887 suppresses
  16. L17:114 suppresses
  17. L14:2130 suppresses
  18. L14:1311 excites
  19. L12:2494 suppresses
  20. L8:311 excites
  21. L17:743 excites
  22. L14:2884 excites
  23. L14:853 suppresses
  24. L11:45 suppresses

Feeds later layers

  1. L19:950 suppresses
  2. L19:2883 suppresses
  3. L20:495 suppresses
  4. L22:268 suppresses
  5. L21:647 suppresses
  6. L21:1287 excites
  7. L20:36 weakly excites
  8. L19:1707 weakly excites
  9. L19:1110 weakly suppresses
  10. L20:739 weakly suppresses
  11. L21:1133 weakly suppresses
  12. L27:100 weakly suppresses
  13. L22:748 weakly suppresses 'and so on' list endings
  14. L22:431 weakly suppresses
  15. L19:2022 weakly suppresses
  16. L20:755 weakly excites
  17. L25:487 weakly suppresses
  18. L20:570 weakly suppresses
  19. L19:1946 weakly suppresses
  20. L26:1394 weakly excites
  21. L20:1981 weakly excites
  22. L19:438 weakly suppresses
  23. L21:511 weakly excites
  24. L20:1726 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.

1 values between -7.36 and -6.951 values between -6.55 and -6.141 values between -6.14 and -5.731 values between -5.32 and -4.914 values between -4.50 and -4.096 values between -4.09 and -3.6812 values between -3.68 and -3.276 values between -3.27 and -2.8625 values between -2.86 and -2.4531 values between -2.45 and -2.0433 values between -2.04 and -1.6384 values between -1.63 and -1.22167 values between -1.22 and -0.82310 values between -0.82 and -0.41961 values between -0.41 and 0.001444 values between 0.00 and 0.41630 values between 0.41 and 0.8298 values between 0.82 and 1.2320 values between 1.23 and 1.6424 values between 1.64 and 2.059 values between 2.05 and 2.4611 values between 2.46 and 2.876 values between 2.87 and 3.287 values between 3.28 and 3.696 values between 3.69 and 4.101 values between 4.10 and 4.512 values between 4.51 and 4.921 values between 4.92 and 5.322 values between 5.32 and 5.731 values between 5.73 and 6.141 values between 6.14 and 6.551 values between 6.55 and 6.961 values between 8.60 and 9.01offnegativepositiveoften

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. CreateTable
  2. idth
  3. uran
  4. doctype
  5. isters
  6. (styles
  7. SupportedContent
  8. 方式进行
  9. identally
  10. 樘

Suppresses

  1. Others
  2. onomy
  3. 城
  4. Ratio
  5. Fact
  6. حة
  7. &A
  8. حت
  9. ception
  10. &