Neuron Explorer

Qwen3-0.6B

capital 'I' starting an acronym

A neuron in layer 22 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 a capital 'I' that begins a word or acronym in the middle of a sentence, such as 'ICREA' or 'IUCN' in the text. It is strongly negative on the start 'Di' of words like 'Diabetic' and 'Diagnostics'. Turning it up or down did not produce a clear theme.

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

    . The study, conducted on Drosophila melanogaster by scientists at the Institute for Research in Biomedicine (IRB Barcelona) and led by ICREA researcher Marco Milán, has reproduced each of the steps known

  2. #2 peak

    (1986); the APA’s Award for Outstanding Lifetime Contributions to Psychology (1993)⏎- ACM fellow (1994)⏎- IJCAI Award for Research Excellence (1995)⏎- Katherine,

  3. #3 peak

    , where lung injury is caused by severe immune reactions, that could be a beneficial combination, said Sakshi Dua, a pulmonologist and professor at the Ica

  4. #4 peak

    a Russian MiG-15 and went in pursuit.⏎Captain Semyon Alekseyevich Fedorets, commanding the 913th IAP (Istrebitel’nyy Aviatsionnyy Pol

  5. #5 peak

    coal to back up their intermittent renewable technologies and to ensure that they have adequate power to satisfy electricity demand.”⏎This claim is based on a previous paper published by the IER itself (PDF). But as we demonstrate in our study German Coal Conund

  6. #6 peak

    the species.⏎The Great Slaty Woodpecker is a vulnerable species on the IUCN Red List, due to habitat loss. Without proper conservation measures, its

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: If we can, perhaps, open up the open door, and perhaps, perhaps, open up the open door and perhaps open open... (replaying every one of us as themselves) and perhaps open open every one of us as themselves,

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

    , weight gain, and frothy urine.⏎What are the causes of Kidney Failure?⏎- Kidney stones⏎- Polycystic kidney disease⏎- Diabetic Nephropathy⏎- Renal tubular acidosis⏎What are

  2. #2 peak

    or internal metal pumps, these patients cannot get an MRI scan at all, as the magnet may dangerously shift these objects in the patient's body.Learn more about Diagnostics & Imaging<|endoftext|>Sudoku Puzzle #11⏎Education

  3. #3 peak

    , light sensitivity, or generalized blur from a young age. Their conditions are often progressive.⏎- Eye trauma⏎- Certain systemic diseases, like diabetes or hypertension⏎Diagnosing retinal disease:⏎Your eye doctor can check for chronic conditions like mac

  4. #4 peak

    common site of endometriosis than the small bowel. Some symptoms are due to adhesions constricting, twisting, or pulling on the bowel.⏎Diagnosis relies a great deal on the woman's description of her symptoms and menstrual history

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. L19:1241 strongly excites
  2. L17:367 suppresses
  3. L17:366 suppresses
  4. L20:1964 suppresses
  5. L2:1463 suppresses
  6. L1:45 excites
  7. L16:118 suppresses
  8. L0:47 suppresses
  9. L16:503 suppresses
  10. L21:1492 excites
  11. L20:342 suppresses
  12. L0:425 excites
  13. L19:193 suppresses
  14. L2:7 excites
  15. L20:374 suppresses
  16. L10:1641 suppresses
  17. L14:392 weakly suppresses
  18. L16:711 weakly excites
  19. L4:221 weakly suppresses
  20. L6:483 weakly suppresses
  21. L19:311 weakly excites
  22. L18:1442 weakly excites unclear: word fragments and digits inside numbers
  23. L19:316 weakly excites
  24. L20:2877 weakly excites

Feeds later layers

  1. L27:2113 strongly suppresses
  2. L24:932 weakly suppresses
  3. L26:729 weakly excites
  4. L27:1716 weakly suppresses
  5. L27:2021 weakly suppresses
  6. L27:183 weakly excites
  7. L23:678 weakly suppresses
  8. L23:2087 weakly suppresses
  9. L27:1042 weakly excites
  10. L27:657 weakly excites
  11. L25:754 weakly suppresses
  12. L25:2307 weakly excites
  13. L26:1028 weakly suppresses
  14. L23:1903 weakly excites
  15. L25:2011 weakly suppresses linking words like 'further' and 'worse'
  16. L26:1634 weakly excites
  17. L23:2823 weakly suppresses
  18. L25:2906 weakly excites
  19. L23:1029 weakly suppresses
  20. L26:3042 weakly suppresses
  21. L27:1856 weakly suppresses
  22. L24:2219 weakly excites
  23. L25:2621 weakly suppresses
  24. L27:1814 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 -101.88 and -97.591 values between -37.61 and -33.331 values between -24.76 and -20.473 values between -16.19 and -11.901 values between -11.90 and -7.624 values between -7.62 and -3.333814 values between -3.33 and 0.9579 values between 0.95 and 5.232 values between 5.23 and 9.521 values between 13.80 and 18.091 values between 65.22 and 69.50offnegativepositiveoften

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. lay
  2. ging
  3. chin
  4. vy
  5. van
  6. alogy
  7. ISIBLE
  8. gen
  9. 真心
  10. a

Suppresses

  1. angles
  2. empo
  3. enen
  4. oren
  5. ennial
  6. ore
  7. ene
  8. enal
  9. ềm
  10. orest