Neuron Explorer

Qwen3-0.6B

'could' after a noun subject

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 "could" when it follows a noun that is the subject of a clause, such as "the camera could" or "citizens could". Turned up, the model moved toward news and quiz-style text.

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

    to power outlets meant that the camera could be plugged in directly in the wall outlet in the different measurement rooms.⏎The light

  2. #2 peak

    arnessed as an energy source. Researchers had believed that keeping the donor and acceptor layers as pure as possible was the best way to ensure that the excitons could travel unimpeded, so that solar cells could capture the maximum amount of energy

  3. #3 peak

    ⏎or a theoretical ideal -it is a fact. Of it a statesman has said: 'They⏎had demonstrated to the world in actual practice that difficulties could⏎be resolved by discussion as they could not be resolved by force. Moreover,⏎

  4. #4 peak

    with endurance and resistance exercise -- to preserve lean body mass<|endoftext|>Scientists Engineer An Opioid That May Reduce Pain With Less Risk⏎Once people realized that opioid drugs could cause addiction and deadly

  5. #5 peak

    aufort , Hyde , Craven and Carteret .⏎The primary reason for establishing counties (or precincts) was to provide local seats of government where citizens could record documents, such as deeds or wills, and participate in court proceedings.

  6. #6 peak

    regeneration process in his book, The Body Electric. Dr. Royal Rife’s revolutionary work in the 1920’s documented and demonstrated how precise frequencies could destroy specific organisms such as cancer cells and viruses, while others could be used to

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: what is a new study that says that the new study has found that the first thing that has been thought about is something that says that the first thing that has been thought about something in the past that has been in the past. What is

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

    Luke 9:58 German Bible<|endoftext|>Both her and Romeo were blinded to everything except each other and did not think before doing, for example, Juliet should have ensured that Romeo found out her plan, before she did it. She loved Romeo

  2. #2 peak

    By 1738 Albemarle and Bath Counties had been dissolved and the 14 precincts then in existence became counties, a designation that has remained since the seventeenth

  3. #3 peak

    improved medical care and better nutrition will have played their part leading up to the 1950s, but after this time there were some specific factors that have contributed to not only better chances of survival post birth, but also the introduction of

  4. #4 peak

    towns there was an organized community life and Jewish public institutions. Nitra, a major administrative and economic center in the Middle Ages, had a longstanding Jewish community that is considered one of the oldest in Slovakia. A document from the year 11

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. L2:240 excites
  2. L1:2801 weakly excites
  3. L0:2025 weakly suppresses
  4. L2:1737 weakly excites
  5. L2:6 weakly excites
  6. L2:1321 weakly suppresses
  7. L1:2814 weakly excites
  8. L2:41 weakly suppresses
  9. L2:2115 weakly excites
  10. L2:209 weakly suppresses 'also' after 'not only ... but'
  11. L0:2544 weakly suppresses
  12. L1:146 weakly excites
  13. L2:1899 weakly excites
  14. L2:1302 weakly suppresses
  15. L0:840 weakly suppresses
  16. L1:391 weakly excites
  17. L1:83 weakly excites
  18. L2:2037 weakly suppresses the digit 5 in decimals like 1.25
  19. L2:497 weakly suppresses
  20. L1:1340 weakly suppresses
  21. L0:810 weakly excites
  22. L0:1217 weakly suppresses
  23. L2:1784 weakly suppresses
  24. L0:1087 weakly suppresses

Feeds later layers

  1. L7:153 excites
  2. L5:536 suppresses
  3. L4:110 excites
  4. L8:2719 suppresses
  5. L4:124 excites
  6. L17:2060 excites
  7. L6:669 excites
  8. L4:113 excites
  9. L27:2982 excites
  10. L15:2901 excites
  11. L4:1294 weakly suppresses
  12. L4:457 weakly excites
  13. L11:425 weakly excites
  14. L7:734 weakly suppresses
  15. L16:2202 weakly excites
  16. L8:2738 weakly suppresses
  17. L11:2896 weakly suppresses
  18. L11:731 weakly excites
  19. L6:535 weakly suppresses
  20. L25:1861 weakly excites
  21. L9:1550 weakly excites 'to' after 'matters' or 'means' phrases
  22. L5:2231 weakly excites
  23. L14:2949 weakly excites
  24. L13:2472 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 -3.06 and -2.921 values between -2.64 and -2.501 values between -1.65 and -1.511 values between -1.51 and -1.372 values between -1.23 and -1.084 values between -0.94 and -0.803 values between -0.80 and -0.668 values between -0.66 and -0.528 values between -0.52 and -0.3875 values between -0.38 and -0.24615 values between -0.24 and -0.102218 values between -0.10 and 0.05789 values between 0.05 and 0.19127 values between 0.19 and 0.3323 values between 0.33 and 0.478 values between 0.47 and 0.615 values between 0.61 and 0.752 values between 0.75 and 0.895 values between 0.89 and 1.035 values between 1.03 and 1.172 values between 1.17 and 1.321 values between 1.46 and 1.601 values between 1.60 and 1.741 values between 1.88 and 2.021 values between 2.02 and 2.161 values between 2.45 and 2.59offnegativepositiveoften

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. instruments
  2. →TokenName
  3. typingsSlinky
  4. stdafx
  5. leggings
  6. sourceMappingURL
  7. swore
  8. icons
  9. 出路
  10. goggles

Suppresses

  1. uted
  2. rium
  3. etration
  4. uta
  5. eten
  6. 례
  7. Citation
  8. แล
  9. acted
  10. 质