Neuron Explorer

Qwen3-0.6B

"one" as a pronoun after a verb

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 "one" when it stands in for a noun, as in "getting one" or "find one of the few". Turning it up made the text loop and repeat, so the steering does not show much.

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

    have a word processing or spreadsheet program for your computer, consider getting one to help you be responsible in school. Once you learn how to use the program

  2. #2 peak

    izations, and memories with a computer. An Imagination Interface, perhaps. By any name, it doesn’t take much time to realize that you have always wanted one. Think of a world where you could selectively download your life’s memories- sights

  3. #3 peak

    as cancer and heart disease? Next best is salad, then fruit.⏎Remember that if you have a health insurance or life insurance policy, or are thinking about taking one out, good health can help to lower your premiums - another reason to reach for

  4. #4 peak

    .⏎It can quickly seem like nothing we do matters all that much. "What would you do if you were stuck in one place, and every day was exactly the same, and nothing that you did mattered

  5. #5 peak

    so rare that generations of men and women never live to see one, and so significant it enabled astronomers armed with telescopes and trigonometry to

  6. #6 peak

    seen. This giant rat also seemed very tame and approachable towards the scientist. If you’re not afraid of rats then you’re probably asking yourself where to get one as a pet? Well, this extremely rare animal can only be found in one

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 first time I saw the light of the world in a new way. But that's not to be gone. What's interesting in the first place is the way to the north of the mind. But that's not to be gone.

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

    to guide the treatment of children and adolescents with schizophrenia.⏎“Although the U.S. Food and Drug Administration (FDA) has recently approved the use of aripiprazole and risperidone for adolescents with schizophrenia, few controlled data are

  2. #2 peak

    such as olanzapine (Zyprexa), risperidone (Risperdal), ziprasidone (Geodon), or aripiprazole (Abilify) might have less side effects than haloperid

  3. #3 peak

    of the explosive eruption of Hecates Tholus provides more evidence of recent Mars rumblings. In December, members of the same research team revealed that calderas on five major Mars volcanoes were repeatedly active as little as 2 million

  4. #4 peak

    iusLeadership | Messenger birds | Lugh Long-arm | Odin | Rigveda | Thor | Odin and Tyr | Balder myth | Trinity | AS-ideas |Goddesses | indexBR | index

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. L2:70 weakly excites
  2. L0:396 weakly excites
  3. L2:2917 weakly suppresses unclear: 'ardu' and 'axi' word fragments
  4. L1:2267 weakly suppresses
  5. L0:156 weakly suppresses
  6. L2:2545 weakly suppresses
  7. L0:2989 weakly suppresses
  8. L0:1609 weakly suppresses
  9. L1:778 weakly suppresses
  10. L0:1071 weakly excites
  11. L0:1 weakly suppresses
  12. L2:1494 weakly suppresses
  13. L0:2759 weakly excites
  14. L0:1301 weakly suppresses
  15. L2:506 weakly suppresses
  16. L2:1160 weakly excites
  17. L1:2198 weakly excites
  18. L0:95 weakly suppresses
  19. L2:1430 weakly suppresses
  20. L2:1213 weakly excites
  21. L2:2328 weakly suppresses
  22. L1:94 weakly suppresses
  23. L0:312 weakly suppresses
  24. L2:1786 weakly excites

Feeds later layers

  1. L4:323 excites
  2. L5:860 weakly excites
  3. L10:1875 weakly excites
  4. L19:2967 weakly suppresses
  5. L4:2589 weakly suppresses
  6. L12:2888 weakly suppresses
  7. L7:1109 weakly excites
  8. L25:2429 weakly excites
  9. L8:1873 weakly excites
  10. L10:1278 weakly excites
  11. L15:1065 weakly excites
  12. L9:802 weakly suppresses
  13. L16:2822 weakly suppresses
  14. L17:2482 weakly suppresses
  15. L19:2804 weakly suppresses 'the' after 'in' or 'use'
  16. L12:2637 weakly suppresses
  17. L24:653 weakly suppresses
  18. L12:1582 weakly excites
  19. L8:1776 weakly excites
  20. L16:297 weakly suppresses
  21. L25:2246 weakly suppresses
  22. L9:679 weakly excites
  23. L18:2909 weakly excites
  24. L23:1867 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 -3.65 and -3.321 values between -1.98 and -1.642 values between -1.31 and -0.975 values between -0.97 and -0.6421 values between -0.64 and -0.302540 values between -0.30 and 0.031305 values between 0.03 and 0.3717 values between 0.37 and 0.704 values between 0.70 and 1.044 values between 1.04 and 1.371 values between 3.05 and 3.381 values between 3.72 and 4.051 values between 4.39 and 4.721 values between 4.72 and 5.061 values between 5.06 and 5.391 values between 7.06 and 7.401 values between 8.07 and 8.401 values between 9.41 and 9.74offnegativepositiveoften

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. ax
  2. Ax
  3. 有限公司
  4. sew
  5. preached
  6. ester
  7. ций
  8. spar
  9. 出现
  10. all

Suppresses

  1. EMON
  2. atoire
  3. Berm
  4. itemap
  5. körper
  6. cano
  7. 金字塔
  8. usat
  9. bove
  10. urname