Neuron Explorer

Qwen3-0.6B

opening parenthesis after a measurement

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 lights up on the opening bracket that comes right after a measurement, like the "(" in "80 metres (295 ft)", where the same amount is repeated in another unit. Turning it up made the model's writing wander into rambling, unrelated sentences, so the effect on ordinary text was not clear.

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

    edited for style and length.⏎Cite This Page:<|endoftext|>Briers, Mississippi facts for kids⏎Quick facts for kids⏎|Elevation||62 ft (19 m)|⏎|Time zone||UTC-6 (Central (C

  2. #2 peak

    since apps have made their way onto most electronic devices, you might think there would be dozens of worthwhile products available on the topic designed for students. Think again.⏎Sure, there are plenty of apps devoted to November’s election, there just aren’t

  3. #3 peak

    Norman cathedral. It is the seat of the Bishop of Coutances and Avranches, previously the Bishops of Coutances. Standing 80 metres (295 ft) tall, it dominates the town and can be seen from

  4. #4 peak

    (66 feet) and weighed in at roughly 50 tonnes (55 tons), or equal to around eight African bush elephants.⏎The dinosaur is

  5. #5 peak

    centimetres). Temperatures are high, and they change little during the year. The coolest month has an average temperature no lower than 64 degrees F (18 degrees C). The temperature difference between day and night is greater than the

  6. #6 peak

    perianth segments. There can be as many as half a dozen flowers on a stalk. The flowering stalk may stand up to 3 ft (1 m) tall. The sword shaped leaves are mostly basal, in two ranks

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 new development in the country's new development of the country's new policy of the country's new success of the country's successful in the country's success of the country's overall success of the country's success to achieve the overall success of

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

    -Brava⏎platform with 500 m bathymetric contours and 200m topographic contours on Fogo. Dotted areas mark⏎seismically active regions in the channel between Fogo and Brava (se

  2. #2 peak

    in⏎fructooligosaccharides. Levrat et al. (1991) demonstrated that even low⏎levels of inulin are effective in decreasing plasma cholesterol in rats.⏎DESIGN:

  3. #3 peak

    OMG). OMG has called for information on object-oriented⏎methodologies, that might create a rigorous software modelling language. Many⏎industry leaders have responded in earnest to help create the standard.⏎See also: ST

  4. #4 peak

    . Ordained in 1868, his success as a “pastor,⏎preacher, liberal theological teacher and social reformer” led to his⏎appointment as head of Scots Church, Collins Street (1875-8

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. L11:2928 excites
  2. L12:2666 suppresses
  3. L10:553 suppresses list-style essay phrasing, weak
  4. L11:2971 excites
  5. L12:852 weakly excites
  6. L10:157 weakly suppresses end of document and web boilerplate
  7. L8:13 weakly suppresses
  8. L5:209 weakly suppresses
  9. L6:472 weakly excites
  10. L0:964 weakly suppresses
  11. L9:511 weakly suppresses
  12. L3:2510 weakly suppresses the word 'operation' and its forms
  13. L9:99 weakly suppresses
  14. L2:61 weakly suppresses
  15. L10:58 weakly excites
  16. L10:1715 weakly suppresses
  17. L9:1560 weakly excites
  18. L7:1188 weakly suppresses
  19. L11:163 weakly suppresses
  20. L12:2553 weakly suppresses
  21. L1:307 weakly suppresses
  22. L10:2049 weakly excites
  23. L11:784 weakly suppresses
  24. L7:276 weakly suppresses

Feeds later layers

  1. L17:814 suppresses
  2. L14:385 excites
  3. L14:194 suppresses
  4. L15:1229 excites
  5. L15:1785 excites
  6. L15:2635 suppresses
  7. L17:496 suppresses
  8. L15:2979 weakly excites
  9. L14:209 weakly excites
  10. L18:847 weakly suppresses
  11. L14:1080 weakly suppresses
  12. L14:564 weakly suppresses
  13. L25:456 weakly suppresses
  14. L14:2351 weakly suppresses
  15. L20:1190 weakly excites
  16. L21:2026 weakly suppresses
  17. L14:2869 weakly suppresses
  18. L26:288 weakly excites
  19. L26:373 weakly suppresses
  20. L17:2384 weakly excites
  21. L15:1570 weakly excites
  22. L17:1718 weakly excites
  23. L14:905 weakly suppresses
  24. L17:933 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.

2 values between -0.64 and -0.611 values between -0.57 and -0.543 values between -0.54 and -0.517 values between -0.51 and -0.477 values between -0.47 and -0.4416 values between -0.44 and -0.4114 values between -0.41 and -0.3718 values between -0.37 and -0.3435 values between -0.34 and -0.3151 values between -0.31 and -0.2862 values between -0.28 and -0.24101 values between -0.24 and -0.2185 values between -0.21 and -0.18110 values between -0.18 and -0.14158 values between -0.14 and -0.11203 values between -0.11 and -0.08235 values between -0.08 and -0.04279 values between -0.04 and -0.01366 values between -0.01 and 0.02319 values between 0.02 and 0.06277 values between 0.06 and 0.09285 values between 0.09 and 0.12251 values between 0.12 and 0.16229 values between 0.16 and 0.19201 values between 0.19 and 0.22152 values between 0.22 and 0.25129 values between 0.25 and 0.2988 values between 0.29 and 0.3269 values between 0.32 and 0.3556 values between 0.35 and 0.3935 values between 0.39 and 0.4223 values between 0.42 and 0.4516 values between 0.45 and 0.4913 values between 0.49 and 0.524 values between 0.52 and 0.552 values between 0.55 and 0.593 values between 0.59 and 0.621 values between 0.62 and 0.652 values between 0.65 and 0.69offnegativepositiveoften

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. ⏎⏎
  2. 流水
  3. An
  4. RNA
  5. :⏎⏎
  6. Höhe
  7. lòng
  8. 高的
  9. isi
  10. VALID

Suppresses

  1. ",(
  2. �建
  3. ',(
  4. _QMARK
  5. ncias
  6. NSBundle
  7. ?(
  8. reak
  9. #__
  10. ')</