Neuron Explorer

Qwen3-0.6B

hyphen inside numbers

A neuron in layer 10 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 a dash that sits between two numbers, like the dash in a phone number, a score range, or a date range. When it was turned up, the model's writing became scrambled and didn't show a clear theme.

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

    taught that you cannot compress liquids, so the the short answer to your question is yes, liquids have a definite volume :0).⏎Interested in booking a 1-1 lesson with me?Click here to view my profile and send me a message

  2. #2 peak

    was no simple, standardized way of describing hurricanes and their damaging effects, like the way the Richter scale is used to describe earthquakes. He created a 1-5 scale based on wind speed and sent it off to Simpson, who expanded<|endoftext|>

  3. #3 peak

    in 1942-44, with improvements based on lessons learned from the UP's 4-8-8-4 Big Boys. In service, the Challengers often complement

  4. #4 peak

    to perform basic statistical functions using both a scientific calculator and Microsoft Excel.⏎Saturday, Aug. 23, 2-4:30 p.m.⏎Basic Statistics with SPSS⏎If your major

  5. #5 peak

    and the climate conditions of Earth during that time with geochemist Tina Lüdecke.⏎You’ll find out about Tina’s research studying early humans from 1-4 million years ago. We investigate what happened during the paleolithic era. How

  6. #6 peak

    you surely will be one of those who lose. (all spiritual goodnesses)⏎THE HOLY QURAN - S39: 65⏎6- THE BRANCHES OF MON

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 time of the day, and the number of days of the week? It's one of the day of the week, or the 31-day of the month, or the 31 week-end of the day,

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

    oltage AC power supply⏎- Two capacitors, 0.1 µF each, non-polarized (Radio Shack catalog # 272-135)⏎- Two 2

  2. #2 peak

    axis cnc machining consists of linear interpolation movement of feed axes X, Y, Z and any 5 axes of X, Y, Z rotation axes A, B, and C.⏎Different processing characteristics⏎3-axis machining: The direction of the

  3. #3 peak

    The newest data sets include POLDER and GOES-7 images, soil carbon data, and information about the study sites.<|endoftext|>2. Verb. (third-person singular of surround) ¹⏎¹ Source: wiktionary.com

  4. #4 peak

    is distributed in a year;⏎n– The years to maturity; and⏎YTM– The yield to maturity.⏎Let's take Bond A, issued by Company Alpha, as an example. It has the following data:⏎- Face value: $

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. L7:1759 excites
  2. L7:828 suppresses
  3. L9:708 excites
  4. L9:867 excites
  5. L5:271 suppresses
  6. L9:802 excites
  7. L9:940 suppresses
  8. L4:290 excites
  9. L5:3047 suppresses
  10. L9:190 excites
  11. L9:145 suppresses
  12. L9:1836 excites
  13. L6:94 suppresses
  14. L9:897 excites
  15. L8:781 excites
  16. L6:1098 suppresses
  17. L8:1468 excites
  18. L6:58 excites
  19. L7:1362 excites
  20. L6:182 suppresses
  21. L9:2308 suppresses
  22. L7:223 excites
  23. L4:1861 excites
  24. L6:145 suppresses

Feeds later layers

  1. L16:1872 suppresses
  2. L15:1665 excites
  3. L11:967 excites
  4. L15:762 excites
  5. L11:112 excites
  6. L12:453 excites
  7. L11:1834 suppresses
  8. L11:25 excites
  9. L18:318 weakly excites
  10. L16:1319 weakly excites
  11. L16:1465 weakly excites
  12. L11:325 weakly excites
  13. L12:2520 weakly excites
  14. L12:315 weakly excites
  15. L15:1690 weakly excites
  16. L11:1109 weakly excites
  17. L13:458 weakly suppresses
  18. L12:916 weakly excites
  19. L15:890 weakly excites
  20. L12:1377 weakly excites
  21. L13:3054 weakly excites
  22. L16:637 weakly excites
  23. L11:453 weakly suppresses
  24. L17:22 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.

2 values between -1.93 and -1.791 values between -1.79 and -1.645 values between -1.64 and -1.494 values between -1.49 and -1.359 values between -1.35 and -1.2014 values between -1.20 and -1.0517 values between -1.05 and -0.9141 values between -0.91 and -0.7669 values between -0.76 and -0.62204 values between -0.62 and -0.47385 values between -0.47 and -0.32575 values between -0.32 and -0.18997 values between -0.18 and -0.031121 values between -0.03 and 0.12229 values between 0.12 and 0.2675 values between 0.26 and 0.4140 values between 0.41 and 0.5629 values between 0.56 and 0.7024 values between 0.70 and 0.8512 values between 0.85 and 1.009 values between 1.00 and 1.147 values between 1.14 and 1.293 values between 1.29 and 1.445 values between 1.44 and 1.583 values between 1.58 and 1.734 values between 1.73 and 1.8710 values between 1.87 and 2.023 values between 2.02 and 2.173 values between 2.17 and 2.311 values between 2.31 and 2.461 values between 2.61 and 2.752 values between 2.75 and 2.901 values between 3.05 and 3.192 values between 3.19 and 3.341 values between 3.78 and 3.92offnegativepositiveoften

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. orden
  2. ordon
  3. imit
  4. keit
  5. lear
  6. "=
  7. ead
  8. BarItem
  9. ordinates
  10. eyed

Suppresses

  1. 秘书长
  2. igli
  3. xDB
  4. 恺
  5. 窗外
  6. nerg
  7. 釜
  8. 艇
  9. tram
  10. 奶