Neuron Explorer

Qwen3-0.6B

"from" after hide, shield, or safe

A neuron in layer 7 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

The neuron fires on the word "from" when it follows words about hiding or protection, such as "hide from", "hidden from view", "safe from" or "shield from". Turning it up did not produce text about hiding, so the steering does not confirm the idea.

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

    is that its vast forested areas gave them ample areas to hide from man (until recent centuries).⏎Why is Africa known for its animals?⏎Africa is

  2. #2 peak

    several hours, they reach the meeting place at the head of the bay behind several small islands. It was a nice sheltered spot. It would even be safe from the occasional sudden great storms that struck without warning in that late summer season. As

  3. #3 peak

    the early 20th century, the mountains and hollers of Polk County were infamous for hosting moonshiners and bootleggers, hidden away from the long arm of the government “revenooers.” But for the past

  4. #4 peak

    This is, actually, just the most popular use of files.⏎Apart from text files, we have a range of system files whose content is hidden from the user's eye but is crucial for the proper functioning of websites and applications.

  5. #5 peak

    the victim only. A bully’s threats, lies, and put-downs can be posted online and passed along to everyone and anyone. There is no safe haven from a bully.⏎- Cyberbullying is more difficult to stop. When a bully

  6. #6 peak

    deposits.⏎Hydrophobin assemblies shield the human pathogenic fungus Aspergillus fumigatus from the immune response during infection, allowing this fungus to go on to cause life-threatening

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: https://www.abc.com/222/1000/1000/1000/1000/1000/1000/1000/10

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

    contain that have been useful in research.⏎The glands are enclosed in a capsule of and internally divided into . Blood vessels and nerves enter the glands at the and gradually branch out into the lobules.⏎In the duct system, the lumens formed by

  2. #2 peak

    <|endoftext|>A metal pole anchored down with cables juts out of the sand next to a public walkway at First Landing State Park. Four small rods stick out horizontally from the sides.⏎Research scientists from Old Dominion University installed the contraption on the d

  3. #3 peak

    movements, the neurobiologists concentrated on a group of around 20 neurons which send out their extensions from the midbrain to the spinal cord. The scientists already knew that the cells

  4. #4 peak

    lifting.⏎The spinal cord runs through the spinal canal, a tunnel that is formed by the holes in the center of each of the vertebrae. Nerves branch from the spinal cord, pass through openings (foraminae) between the vert

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. L4:192 excites
  2. L6:2657 suppresses
  3. L6:2836 weakly excites
  4. L1:1676 weakly suppresses
  5. L6:2743 weakly excites
  6. L4:2269 weakly suppresses
  7. L6:1512 weakly excites
  8. L6:1192 weakly excites
  9. L5:28 weakly excites
  10. L6:545 weakly excites
  11. L6:2595 weakly suppresses
  12. L6:1297 weakly excites
  13. L3:2791 weakly excites
  14. L2:1237 weakly excites
  15. L6:1442 weakly excites
  16. L4:3019 weakly suppresses
  17. L5:141 weakly excites
  18. L5:2520 weakly suppresses
  19. L2:1180 weakly excites
  20. L4:1469 weakly suppresses
  21. L2:2108 weakly excites
  22. L6:1799 weakly excites
  23. L0:1290 weakly suppresses
  24. L1:2488 weakly suppresses

Feeds later layers

  1. L8:2797 excites
  2. L8:2879 suppresses
  3. L12:1276 suppresses
  4. L8:427 weakly suppresses
  5. L10:852 weakly suppresses
  6. L11:2369 weakly suppresses
  7. L10:2452 weakly suppresses
  8. L20:1994 weakly suppresses
  9. L23:1822 weakly suppresses
  10. L10:3012 weakly suppresses
  11. L11:337 weakly excites
  12. L8:918 weakly suppresses
  13. L9:1619 weakly suppresses
  14. L11:2214 weakly suppresses
  15. L11:1716 weakly suppresses
  16. L17:1375 weakly suppresses
  17. L22:2193 weakly suppresses
  18. L23:125 weakly suppresses
  19. L11:2047 weakly suppresses
  20. L17:2325 weakly excites
  21. L11:422 weakly excites
  22. L8:2798 weakly suppresses
  23. L19:1931 weakly suppresses
  24. L20:1491 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 -5.19 and -4.921 values between -4.37 and -4.101 values between -2.74 and -2.465 values between -1.37 and -1.102 values between -1.10 and -0.835 values between -0.83 and -0.5692 values between -0.56 and -0.281696 values between -0.28 and -0.011998 values between -0.01 and 0.2698 values between 0.26 and 0.545 values between 0.54 and 0.811 values between 0.81 and 1.082 values between 1.36 and 1.631 values between 5.45 and 5.72offnegativepositiveoften

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. ummies
  2. UNUSED
  3. IData
  4. .GetData
  5. .READ
  6. 有色
  7. avoidance
  8. otas
  9. avadoc
  10. alloca

Suppresses

  1. adder
  2. reet
  3. arse
  4. UES
  5. aight
  6. uter
  7. anks
  8. erra
  9. idency
  10. 沿着