Neuron Explorer

Qwen3-0.6B

unclear: weak scattered tokens

A neuron in layer 9 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron gives only weak signals on a scattered mix of words, such as times like "4:30 p.m.", "no" and "not". Its negative signals are also unclear. Turning it up or down did not produce a clear theme.

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

    half, the winds picked up dramatically with gusts reaching nearly 44 mph as the storm rolled to the southwest.⏎By 4:30 p.m., the storm’s outflow boundary—where the the thunderstorm’s cooled air is

  2. #2 peak

    ' nearly permanently.⏎Sometimes we allow the connections to be 'forced' on us. For example, if you were to be asked where you were last Wednesday at 15.00 hours, chances are you wouldn't immediately recall. But

  3. #3 peak

    ans and journalists gathered to inaugurate the city’s new air terminal on May 3, 1961, the significance of the event was lost on no one. Though Atlanta boosters already considered their home the “capital of the South

  4. #4 peak

    relationship existed between the paired data. Furthermore, regression analysis was used to identify the exact nature<|endoftext|>Coloring pages - Edupics.com⏎All videos (0)⏎Amusement park (15)⏎Children's rights (14)⏎Space exploration

  5. #5 peak

    what is left when something happens but does not completely undo itself. Meaning that is an experience, but it cannot fully get erased

  6. #6 peak

    "I get up at five in the morning and start the day by milking the goat and taking the animals out to pasture. Only after that do I go to school," said Gyorgy Radavoi, a 1

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 thinking process of the user of the previous steps, the thinking process of the user of the previous steps, the thinking process of the user of the previous steps, the thinking process of the user of the previous steps, and so on.

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

    Alberta’s international merchandise exports last year. If we exclude oil and gas extraction and petroleum refinery exports, its share rises to 36%.⏎- Alberta exports the third largest amount of agricultural

  2. #2 peak

    is appreciated. His temperature is 99.6°F. He has no leukocytosis and CRP and ESR are normal. Radiographs of the right leg are seen in Figure A. What is the

  3. #3 peak

    its natural fleece (not shorn).⏎Chest measurement shall be done with a soft but strong measuring tape just behind the shoulders. The person measuring shall always stand on the left side of the sheep, slide the end of the tape down the right

  4. #4 peak

    ini protested that he needed to see for himself. At the seance, Houdini saw through the deception and called “Margery” on it, much to the annoyance of Bird, who angrily resisted exposing her con game. The arguments that

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. L8:191 suppresses
  2. L8:112 excites
  3. L7:512 excites
  4. L6:240 excites
  5. L7:2158 excites
  6. L6:289 suppresses
  7. L7:736 suppresses
  8. L6:1433 suppresses
  9. L8:1205 excites
  10. L7:586 suppresses
  11. L7:186 excites
  12. L7:298 suppresses
  13. L5:1128 excites
  14. L6:83 suppresses
  15. L7:347 suppresses unclear: mid-sentence function words
  16. L8:294 suppresses
  17. L7:2750 suppresses
  18. L4:297 weakly suppresses
  19. L7:177 weakly suppresses
  20. L6:936 weakly excites
  21. L5:585 weakly excites
  22. L5:2008 weakly excites
  23. L5:2163 weakly suppresses
  24. L7:84 weakly suppresses

Feeds later layers

  1. L10:766 excites
  2. L10:1073 excites
  3. L10:1863 suppresses
  4. L14:704 excites
  5. L16:1193 excites
  6. L14:995 excites
  7. L15:1333 excites
  8. L11:459 excites
  9. L10:1830 excites
  10. L13:86 suppresses
  11. L12:1467 excites numbers after dates and figures
  12. L14:924 excites
  13. L13:423 excites
  14. L14:1323 excites
  15. L19:1065 suppresses
  16. L13:1203 excites
  17. L15:2091 excites
  18. L13:858 excites
  19. L12:1426 excites
  20. L19:2653 excites "most" meaning "very" or "the most"
  21. L13:1343 excites
  22. L10:541 suppresses
  23. L13:1725 excites
  24. L13:2322 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 -2.92 and -2.801 values between -2.41 and -2.281 values between -2.02 and -1.892 values between -1.89 and -1.766 values between -1.76 and -1.643 values between -1.64 and -1.515 values between -1.51 and -1.387 values between -1.38 and -1.2514 values between -1.25 and -1.1216 values between -1.12 and -0.9921 values between -0.99 and -0.8646 values between -0.86 and -0.7350 values between -0.73 and -0.6064 values between -0.60 and -0.4895 values between -0.48 and -0.35207 values between -0.35 and -0.22462 values between -0.22 and -0.091203 values between -0.09 and 0.04775 values between 0.04 and 0.17422 values between 0.17 and 0.30205 values between 0.30 and 0.43130 values between 0.43 and 0.5671 values between 0.56 and 0.6841 values between 0.68 and 0.8125 values between 0.81 and 0.9415 values between 0.94 and 1.077 values between 1.07 and 1.205 values between 1.20 and 1.334 values between 1.33 and 1.463 values between 1.46 and 1.591 values between 2.10 and 2.23offnegativepositiveoften

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. 已久
  2. 十条
  3. ://${
  4. 很高
  5. 非常高
  6. 极高
  7. 很高的
  8. nowhere
  9. -date
  10. HasBeenSet

Suppresses

  1. sons
  2. jango
  3. 这两年
  4. Registr
  5. eft
  6. зал
  7. Tail
  8. itches
  9. Registr
  10. TAIL