Neuron Explorer

Qwen3-0.6B

the word 'tropical'

A neuron in layer 5 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 lights up on the word 'tropical', as in 'tropical forests', 'tropical storms' or 'tropical deforestation'. Its negative side is weak and mostly on fragments of other words, so it is mainly a word detector. Turning it up or down did not give a clear, related change in the model's writing.

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

    unknown and unnamed by scientists as of yet. Insects, especially beetles, make up a large proportion of this undiscovered life on Earth.Experts in remote tropical countries' fauna such as the wilderness of New Guinea, Alexander Riedel of

  2. #2 peak

    1, 2, 3, and this remains the cornerstone of most biodiversity conservation strategies 4, 5, 6. “The effects of tropical deforestation on climate go well beyond carbon,” says Professor Deborah Lawrence, “[it

  3. #3 peak

    ination of Humanitarian Affairs (OCHA), which also warned that the death toll could rise in the coming days. The cyclone has been downgraded to a tropical storm and is expected to affect other southern African countries, including southern Zimbabwe, northern

  4. #4 peak

    . 1. The Issue The …⏎BONN, Germany (June 05, 2014) – Programs and policies to reduce tropical deforestation, and the global warming emissions resulting from deforestation, are … May

  5. #5 peak

    will replenish the Federal Emergency Management Agency’s depleted funds in order to aid several states still reeling in the wake of widespread flooding, wildfires, tornadoes and tropical storms. But cleanup and recovery is only one part of the disaster cycle. Fortunately

  6. #6 peak

    ANE?⏎Hurricanes are tropical cyclones that occur in the Atlantic Ocean and have winds with a sustained speed of

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 have a passion for the old world, and I am still waiting for it. The 20th Department of the People's Republic of the People's Daily. In the form of a newspaper, written in the style of

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

    commonly known as the draft. For those who are not aware, the draft is a system of conscription that enlists citizens into the military; the system is run using an online sign

  2. #2 peak

    ated five years later after the Soviet invasion of Afghanistan. Decades later the government website for the program lists the benefits of the new and “improved” conscription system. Stated, the draft is “the most equitable draft in history,”

  3. #3 peak

    itation: Formal references to statutes that describe where they are published. For instance, the citation 23 Vt. Stat. Ann. § 1185 tells us that this cited statute is

  4. #4 peak

    .⏎The project is eligible for School of Mathematics Doctoral Training Grant funding - please contact us for more information.<|endoftext|>História (S

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. L1:1372 excites
  2. L3:1436 weakly suppresses
  3. L3:117 weakly excites
  4. L4:97 weakly suppresses
  5. L4:1795 weakly suppresses "at least" phrase
  6. L2:1799 weakly excites
  7. L4:25 weakly excites
  8. L4:42 weakly suppresses
  9. L4:1640 weakly excites
  10. L3:142 weakly suppresses
  11. L2:2903 weakly suppresses
  12. L0:1650 weakly excites
  13. L3:1981 weakly excites
  14. L4:23 weakly excites
  15. L3:672 weakly excites
  16. L1:642 weakly suppresses
  17. L2:2915 weakly suppresses
  18. L0:1960 weakly suppresses unclear: mixed medical and harsh words
  19. L4:1246 weakly suppresses
  20. L2:319 weakly excites 'of' after 'pattern' or 'patterns'
  21. L2:1756 weakly excites
  22. L0:1909 weakly excites
  23. L0:2799 weakly suppresses
  24. L2:2131 weakly excites

Feeds later layers

  1. L10:2035 suppresses
  2. L10:2514 excites
  3. L10:1705 excites
  4. L9:116 excites
  5. L9:209 excites
  6. L8:2052 excites
  7. L7:2370 weakly excites
  8. L6:89 weakly excites
  9. L8:2135 weakly suppresses
  10. L16:1147 weakly excites
  11. L19:1400 weakly suppresses
  12. L10:2533 weakly excites
  13. L19:783 weakly excites
  14. L21:599 weakly suppresses
  15. L10:1316 weakly excites
  16. L9:845 weakly excites
  17. L20:2861 weakly excites
  18. L20:1196 weakly suppresses
  19. L6:2164 weakly excites second half of multi-word proper names
  20. L6:1198 weakly excites
  21. L15:1553 weakly excites
  22. L17:1042 weakly excites
  23. L12:773 weakly suppresses
  24. L26:1950 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.

4 values between -1.35 and -1.087 values between -1.08 and -0.805 values between -0.80 and -0.53127 values between -0.53 and -0.252431 values between -0.25 and 0.021283 values between 0.02 and 0.3038 values between 0.30 and 0.578 values between 0.57 and 0.851 values between 0.85 and 1.121 values between 1.12 and 1.401 values between 1.40 and 1.671 values between 4.70 and 4.981 values between 9.38 and 9.66offnegativepositiveoften

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. юр
  2. Crime
  3. greedy
  4. observational
  5. 很长
  6. avery
  7. Crime
  8. 庋
  9. Observation
  10. 忒

Suppresses

  1. atory
  2. ication
  3. ipated
  4. utory
  5. ipation
  6. illation
  7. cene
  8. 冻结
  9. bole
  10. aphore