Neuron Explorer

Qwen3-0.6B

'of' after 'those' or 'that' in comparisons

A neuron in layer 2 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 reacts to the word 'of' when it comes right after 'that' or 'those' in a comparison, as in 'different from those of the national sample'. Turning it up did not produce any obvious change in topic or style.

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

    adjacent districts of the northern lowlands. Smaller agglomerations of this group are also found along the eastern littoral where their settlements are juxtaposed with those of the Muslims.⏎The Sri Lankan Muslims trace their origin back to Arab traders of

  2. #2 peak

    ulating layers and one or more copper layers that contain the signal traces and the powers and grounds; the design of the layout of PCBs can be compared to that of the electrical circuit.⏎- Single Layer PCB –

  3. #3 peak

    .⏎San Antonio has a unique and strategic location. It is on the edge of the Great Plains. The topography, rainfall and water supply were different from that of the East. The Easterners had a forest background. The traditions, laws,

  4. #4 peak

    G. maedai sp. nov. (middle to ?upper Cenomanian). The habitats of Goshoraia have been extensively compared with those of common Cretaceous, nonsiphonate burrowers, such

  5. #5 peak

    Indian students reported significantly higher lifetime and last 30-day substance use rates compared with the national sample. Lifetime use rates for American Indian students were higher than those of the national sample for all substance measures except tranquilizers and amphetamines (all

  6. #6 peak

    , right? See it in action below, and check out the project’s elaborate overview here.<|endoftext|>Our system of government is different from those of all other nations, because part of the political power is vested in the State,

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 need to know what the number of the number of the number of the number of the number of the number of the number of the number of the number of the number...? -4444444444

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

    of Ireland 1911⏎Household Return (Form A) for house No. 15⏎In 1911, James Grealish (

  2. #2 peak

    molecules with complementary rather than identical structures in juxtaposition. We see here a direct connection to molecular packing in molecular crystals. Aleksandr Isaakovich Kitaigorodsky (left) published a brief paper in 1945, “

  3. #3 peak

    William Sutherland was one of several Scots who explored the country in the 19th century. He ended a wandering life (he had earlier fought under Garibaldi in Italy) when he settled at Milford Sound in

  4. #4 peak

    and signed by James. It was witnessed as a truthful return and collected by Constable Thomas Fitzgerald on April 11th.⏎10. James Michael Grealish, 11. James (Jas) Grealish,

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. L1:1055 weakly excites
  2. L1:1857 weakly suppresses
  3. L1:933 weakly excites
  4. L1:2413 weakly suppresses
  5. L1:92 weakly excites
  6. L0:3039 weakly excites
  7. L1:74 weakly excites
  8. L0:2951 weakly excites
  9. L1:10 weakly excites
  10. L1:912 weakly suppresses
  11. L1:2242 weakly suppresses
  12. L0:501 weakly suppresses
  13. L1:2449 weakly excites
  14. L0:2836 weakly suppresses
  15. L1:1541 weakly suppresses
  16. L1:2822 weakly suppresses
  17. L0:2523 weakly excites
  18. L1:303 weakly suppresses
  19. L0:1951 weakly suppresses
  20. L0:1042 weakly suppresses "This" starting a labelled sentence
  21. L1:892 weakly excites
  22. L0:1271 weakly excites
  23. L1:2825 weakly excites
  24. L1:338 weakly excites

Feeds later layers

  1. L18:2664 suppresses
  2. L6:1001 weakly suppresses
  3. L15:275 weakly excites
  4. L14:173 weakly excites
  5. L14:2673 weakly excites
  6. L11:2817 weakly excites
  7. L15:2608 weakly suppresses
  8. L13:801 weakly excites
  9. L26:1228 weakly excites
  10. L24:1362 weakly excites
  11. L14:2845 weakly suppresses
  12. L14:2983 weakly suppresses
  13. L16:320 weakly excites
  14. L27:796 weakly excites
  15. L5:361 weakly excites
  16. L3:2605 weakly excites
  17. L19:1475 weakly suppresses
  18. L3:2372 weakly suppresses
  19. L9:2262 weakly suppresses
  20. L6:1722 weakly excites
  21. L15:1107 weakly suppresses
  22. L26:2424 weakly suppresses
  23. L23:1469 weakly excites
  24. L21:883 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 -1.28 and -1.142 values between -1.14 and -1.003 values between -0.86 and -0.723 values between -0.72 and -0.572 values between -0.57 and -0.4339 values between -0.43 and -0.29328 values between -0.29 and -0.151676 values between -0.15 and -0.011571 values between -0.01 and 0.14231 values between 0.14 and 0.2831 values between 0.28 and 0.429 values between 0.42 and 0.565 values between 0.56 and 0.701 values between 0.70 and 0.841 values between 0.84 and 0.982 values between 1.13 and 1.272 values between 1.98 and 2.121 values between 4.24 and 4.38offnegativepositiveoften

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. ycop
  2. ahir
  3. ʏ
  4. 原著
  5. ès
  6. 导入
  7. ro
  8. Ts
  9. 대로
  10. ays

Suppresses

  1. politic
  2. lich
  3. Mailer
  4. lix
  5. plode
  6. $
  7. dzi
  8. hibit
  9. 当たり
  10. .false