Neuron Explorer

Qwen3-0.6B

'such' after 'no' or 'any'

A neuron in layer 3 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 'such' when it follows phrases like 'no such', 'any such' or 'such an', which are common in negative or quantity statements. Turning it up produced rambling, generic text without a clear link to the idea.

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

    9’s negative impacts on the cardiovascular system highlights a great need for identifying COVID-19 patients at risk for heart problems, the researchers say. However, no such predictive capabilities currently exist.⏎“This project will provide clinicians with early warning signs and ensure

  2. #2 peak

    a prince’s brain knows what his mouth’s doing?” Bertie: “You’re obviously not well acquainted with many royal princes.”⏎No one could have imagined any such dialogue involving Archduke Otto von Habsburg, who died on July

  3. #3 peak

    by inserting a verse about the Exodus into the third paragraph of the Shema], thus ensuring that we never forget this important historical event and its attendant lessons, no such obligation exists (according to most Halachic authorities, with the possible exception of

  4. #4 peak

    Piet has dreamed of proving that he is a skater as brave and strong as Pim Mulier -- but he had never imagined that his test would fall under such dangerous circumstances. Louise Borden's moving text captures all the tension, excitement,

  5. #5 peak

    ing, it is likely to subside into quiescence as soon as its object is fulfilled by the supramental being's appearance. But there is no such stress in reality: the idea of human progress itself is very probably an illusion,

  6. #6 peak

    ⏎The Museu Civico in Sassoferrato is fortunate to hold such an important and rare relic of the Palaiologan Renaissance, the final period

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 have been thinking about lately, and I have been thinking about lately. So I want to make a difference, and I want to make a difference. I have been thinking about lately, and I have been thinking about lately. So 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

    comes as the maverick Australian speech therapist, Lionel Logue, is just getting to know His Royal Highness Prince Albert, the stammering Duke of York:⏎Logue

  2. #2 peak

    Diddle Diddle, All Around the Mulberry Bush, I’m a Little Tea Pot, Hickety Pickety Bumble Bee, The Grand Old Duke of York, Ride a Cock-horse to Banbury Cross, Pat-a-cake,

  3. #3 peak

    her daughters’ safety and arrange appropriate marriages for them and so she permitted them to leave sanctuary and attend his court.⏎Henry VII married her eldest daughter, Elizabeth of York

  4. #4 peak

    to ascertain the nature of humanity and of God and to facilitate the experience of the presence of divine love and wisdom in the world.⏎Sufism, or Taṣawwuf ( الْتَّصَوُّف;

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. L2:2293 suppresses
  2. L1:1056 weakly excites
  3. L2:122 weakly excites
  4. L2:1404 weakly excites
  5. L0:2929 weakly suppresses
  6. L2:1036 weakly excites
  7. L1:1200 weakly suppresses
  8. L2:79 weakly excites
  9. L2:934 weakly suppresses
  10. L0:1862 weakly suppresses
  11. L1:1091 weakly suppresses
  12. L0:1115 weakly suppresses
  13. L0:1472 weakly excites
  14. L2:1134 weakly suppresses
  15. L2:2227 weakly suppresses
  16. L0:246 weakly excites
  17. L0:1700 weakly suppresses
  18. L1:1862 weakly excites
  19. L2:2346 weakly excites
  20. L2:2424 weakly excites
  21. L2:2285 weakly excites
  22. L2:2254 weakly suppresses
  23. L0:2833 weakly excites
  24. L2:1497 weakly excites

Feeds later layers

  1. L4:1498 excites
  2. L4:2688 excites
  3. L5:1364 excites
  4. L4:468 excites "that" after "than" or "like"
  5. L17:2988 excites
  6. L6:1892 excites
  7. L4:653 excites "that" after "such" or "such a way"
  8. L5:2308 weakly excites
  9. L11:2331 weakly excites 'so' before much or many
  10. L11:1890 weakly excites
  11. L4:1873 weakly excites
  12. L6:1458 weakly excites
  13. L7:2434 weakly excites
  14. L18:2248 weakly suppresses
  15. L11:229 weakly excites
  16. L13:2092 weakly excites
  17. L5:2175 weakly excites
  18. L10:2918 weakly suppresses
  19. L22:2159 weakly suppresses
  20. L7:2380 weakly excites
  21. L11:1039 weakly suppresses
  22. L5:2567 weakly excites
  23. L4:1955 weakly suppresses
  24. L5:1372 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 -1.14 and -0.931 values between -0.72 and -0.5159 values between -0.51 and -0.30819 values between -0.30 and -0.092641 values between -0.09 and 0.12356 values between 0.12 and 0.3315 values between 0.33 and 0.542 values between 0.54 and 0.751 values between 0.75 and 0.961 values between 1.38 and 1.583 values between 1.58 and 1.791 values between 2.00 and 2.211 values between 2.21 and 2.421 values between 2.63 and 2.841 values between 3.47 and 3.681 values between 3.68 and 3.891 values between 4.10 and 4.311 values between 4.52 and 4.731 values between 5.57 and 5.781 values between 7.04 and 7.25offnegativepositiveoften

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. ied
  2. deployment
  3. severity
  4. ielding
  5. Hispan
  6. CALLBACK
  7. ields
  8. .Callback
  9. 椅子
  10. ิง

Suppresses

  1. acf
  2. awl
  3. aos
  4. apolis
  5. ôm
  6. омер
  7. adder
  8. ooth
  9. uell
  10. abox