Neuron Explorer

Qwen3-0.6B

'in' in 'bear in mind' phrases

A neuron in layer 21 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

Fires on the word 'in' in set phrases like 'bear in mind', 'keep in mind' and 'with this in mind', where 'mind' is about to follow. When it was turned up, the model kept producing 'in mind' over and over.

Written by Claude from this neuron's activations and steering outputs. Confidence: high. 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

    them fascinating. After you have mastered some fundamental activities, you’ll be able to try new ones and incorporate a brand new twist for the previous kinds. And bear in mind, it’s in no way too early to start creating new memories. >>>

  2. #2 peak

    The⏎Halacha section will primarily be a review of the lessons from the previous⏎cycles while the Aggadah section will be new material. Please bear in mind⏎that the purpose of learning the laws of Shmiras haloshon

  3. #3 peak

    forest for one use if it's going to compromise other uses? Our forests are incredibly important for harvesting but also for tourism [and] we want to always bear in mind the possible tradeoffs.⏎Tourism is a huge asset to Maine

  4. #4 peak

    ? Please include any references to your response that is indicated in the Torah.⏎Answer: Before I answer your question, as a word of background, you need to keep in mind that Judaism, ever since the destruction of the second Temple in 7

  5. #5 peak

    for the business⏎- Clear goals for the management of livestock⏎The management process should involve:⏎- Reviewing these strategies regularly⏎– Making decision with these goals in mind⏎– Monitoring the health of your land, your business and the people

  6. #6 peak

    of the text, as well as the reason why the text will be of interest to the reader.⏎- Many people involved with writing hate to sketch. With this in mind, one simple piece of advice can be given: tackle the text as soon

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 am now, on the current, to be a new, which is to say, to be a man on the world's, to be a man, to be a man, to be a man, to be a man, to

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

    materials is creating big data.⏎Take, for example, the emergence of Moocs – massive open online courses. While some universities

  2. #2 peak

    : maintain a tree board or department, having a community tree ordinance, spending at least $2 per capita on urban forestry and celebrating Arbor Day.⏎Alison Oelschl

  3. #3 peak

    eat is the main ingredient in your diet.⏎Food is food insofar as our tastes are concerned. But, when it comes to our stomachs, and, by extension, our whole body is involved, food is more than just food. It

  4. #4 peak

    lenses and how to use them. It’s a very useful book and provides a constructive way of viewing problems.⏎A third<|endoftext|>Mecklenburg-Western Pomerania, also known as Mecklenburg-Vorpommern, is a

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. L19:2394 excites
  2. L17:1368 excites
  3. L7:1494 suppresses
  4. L18:2180 suppresses
  5. L1:1244 weakly suppresses "of" after "board" or "off"
  6. L12:175 weakly suppresses
  7. L18:2372 weakly excites
  8. L12:2949 weakly excites
  9. L16:1297 weakly suppresses unclear: document ends and answer colons
  10. L9:1483 weakly excites
  11. L20:2186 weakly suppresses
  12. L20:30 weakly excites
  13. L5:1029 weakly suppresses
  14. L18:1257 weakly suppresses
  15. L10:1611 weakly excites
  16. L14:152 weakly suppresses
  17. L17:1691 weakly suppresses
  18. L2:2649 weakly excites
  19. L16:2418 weakly excites
  20. L12:2422 weakly excites
  21. L20:1539 weakly excites
  22. L9:2590 weakly suppresses
  23. L16:1059 weakly excites
  24. L10:2439 weakly excites

Feeds later layers

  1. L23:1788 strongly excites
  2. L27:1619 excites
  3. L23:2944 excites
  4. L22:2230 excites
  5. L26:2125 excites
  6. L27:1297 weakly excites
  7. L22:2468 weakly excites
  8. L27:1383 weakly excites
  9. L24:2218 weakly excites
  10. L25:23 weakly suppresses
  11. L24:2573 weakly excites
  12. L22:2909 weakly excites
  13. L22:2562 weakly excites 'to' after 'things' or 'factors'
  14. L23:2056 weakly excites
  15. L25:1776 weakly excites
  16. L23:1979 weakly excites blank before a number word
  17. L23:2597 weakly excites
  18. L23:2659 weakly excites
  19. L25:1008 weakly suppresses
  20. L26:2993 weakly excites
  21. L24:441 weakly suppresses
  22. L22:687 weakly excites
  23. L23:510 weakly suppresses
  24. L26:2348 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 -5.97 and -5.082 values between -3.29 and -2.4017 values between -2.40 and -1.50292 values between -1.50 and -0.613356 values between -0.61 and 0.29196 values between 0.29 and 1.1819 values between 1.18 and 2.078 values between 2.07 and 2.974 values between 2.97 and 3.861 values between 3.86 and 4.751 values between 4.75 and 5.652 values between 5.65 and 6.541 values between 6.54 and 7.431 values between 8.33 and 9.223 values between 9.22 and 10.111 values between 10.11 and 11.011 values between 11.01 and 11.901 values between 27.98 and 28.871 values between 28.87 and 29.77offnegativepositiveoften

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. remember
  2. mind
  3. recall
  4. Remember
  5. memory
  6. recalling
  7. remembering
  8. Recall
  9. 记忆
  10. memory

Suppresses

  1. JW
  2. jer
  3. HERE
  4. jet
  5. APPER
  6. NewLabel
  7. strar
  8. Dion
  9. encer
  10. Crow