Neuron Explorer

Qwen3-0.6B

"one" after an ordinal or superlative

A neuron in layer 2 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 "one" when it follows words like "most important", "last", or "next", as in "the last one". Its firing is weak compared with the strongest neurons. Turning it up made the text break down into loose fragments.

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

    well as a sculpture of Roman goddess Minerva commanding attention from the pediment.⏎Among the many notable events that occurred during the building’s lifetime, the most memorable one perhaps was the assassination of Lt Col NS Simpson, the infamous Inspector General of Pr

  2. #2 peak

    of different chemical products, such as carboxylic acids. The rate at which this occurs depends on a number of factors. Temperature is perhaps the most critical one, since the rate of oxidation doubles for every rise of

  3. #3 peak

    airports. The most important one: the ability to exert influence beyond the fences of the airport. Since the Rob

  4. #4 peak

    ”. This makes the readability of your PowerPoint smoother and not awkward. Headings should be the biggest of whatever point from the scale you start at, then the next one down for your dot points.You should

  5. #5 peak

    .⏎Pluto is the last one of the major planet. It's definitely different from the others. Discovered in

  6. #6 peak

    less likely to develop the disease.⏎Researchers have long been aware of an apparent link between blood-vessel health and Alzheimer’s, but this study is “the first one that shows that treating mild cognitive impairment can have results,” says Anton Porsteinsson

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 the value of the total. the total is the sum. the total is the sum. the total is the sum. the total is the sum. the total is the sum. the total is the sum. the total is the

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

    now witnessing the effect these linguistic variations are having on children born into the computer age with such a high level of access in and out of schools,” the Scotsman quoted report author Lucy Jones, a former student at Manchester University, as saying.⏎“They

  2. #2 peak

    while another group did not get an apology. The participants who received the apology felt better and viewed the offending child as nicer as well as more remorseful.⏎Another study exposed children to a more distressing event: A person knocked over a tower that

  3. #3 peak

    be to assume that diet soda is a better option. Unfortunately, research has also found a connection between diet soda and brain health.⏎One study published in Stroke found that participants who had consumed one diet soda daily were almost three times more likely to develop

  4. #4 peak

    .⏎Two of Wenzel’s findings are relevant to punishments handed down by professional sports leagues. In a 2010 study, Wenzel found that people tend to seek restorative justice rather than retributive justice when they perceive

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. L1:1322 suppresses
  2. L1:425 excites
  3. L1:551 weakly suppresses
  4. L1:405 weakly excites
  5. L1:2147 weakly suppresses
  6. L0:2509 weakly excites
  7. L1:1639 weakly suppresses
  8. L1:1691 weakly excites
  9. L0:2359 weakly suppresses
  10. L1:2425 weakly excites
  11. L1:1829 weakly suppresses
  12. L1:439 weakly excites
  13. L0:1572 weakly excites
  14. L0:2281 weakly excites
  15. L1:2170 weakly suppresses
  16. L0:1907 weakly suppresses
  17. L0:1591 weakly excites
  18. L0:1225 weakly excites
  19. L1:166 weakly excites
  20. L1:237 weakly excites
  21. L0:1264 weakly excites
  22. L1:684 weakly suppresses
  23. L1:407 weakly suppresses
  24. L1:1893 weakly excites

Feeds later layers

  1. L5:93 suppresses
  2. L7:267 excites
  3. L3:618 excites
  4. L3:93 excites
  5. L5:188 excites
  6. L13:2679 excites
  7. L4:484 excites
  8. L4:29 excites
  9. L7:2571 excites
  10. L3:30 weakly excites
  11. L5:898 weakly excites
  12. L4:443 weakly excites
  13. L8:1822 weakly excites
  14. L16:2375 weakly excites
  15. L12:30 weakly suppresses
  16. L7:34 weakly suppresses
  17. L12:2575 weakly excites
  18. L5:167 weakly excites
  19. L9:277 weakly suppresses
  20. L14:2555 weakly excites
  21. L14:625 weakly suppresses
  22. L8:969 weakly excites
  23. L14:1389 weakly excites
  24. L12:1702 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 -4.94 and -4.761 values between -2.51 and -2.341 values between -2.17 and -1.991 values between -1.99 and -1.823 values between -1.82 and -1.655 values between -1.65 and -1.487 values between -1.48 and -1.306 values between -1.30 and -1.1319 values between -1.13 and -0.9625 values between -0.96 and -0.7830 values between -0.78 and -0.6175 values between -0.61 and -0.44269 values between -0.44 and -0.26950 values between -0.26 and -0.092045 values between -0.09 and 0.08312 values between 0.08 and 0.2686 values between 0.26 and 0.4338 values between 0.43 and 0.6019 values between 0.60 and 0.7812 values between 0.78 and 0.951 values between 0.95 and 1.121 values between 1.12 and 1.301 values between 1.81 and 1.99offnegativepositiveoften

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. レーション
  2. 顺应
  3. 网讯
  4. ско
  5. opposition
  6. 这样一个
  7. ategorie
  8. south
  9. Boundary
  10. بع

Suppresses

  1. nawet
  2. tered
  3. rehe
  4. anytime
  5. tering
  6. ABCDEFGHI
  7. CERT
  8. SEE
  9. even
  10. ocha