Neuron Explorer

Qwen3-0.6B

decimal point inside numbers

A neuron in layer 22 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 responds to the dot in decimal numbers such as 0.5, 1.25 or 2.0. Its negative side fires on the word "main" in words like "mainstay". Turning it up or down gave ordinary, unrelated text, so the effect on writing is unclear.

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

    strands of the 9/10 grade band of the Minnesota Language Arts standards are addressed.⏎Credit: 1.0 (Semester A = 0.5 credit, Semester B = 0.5 credit.)⏎Honors Opportunity:

  2. #2 peak

    ,000-lb load would use the following calculation:⏎[(4,000 lbs + 6,000 lbs) × 1.25] ÷ 4 = 3,125 lbs⏎

  3. #3 peak

    rather than just during that time period.⏎If you're walking between 3 and 4 miles per hour on a hike you should multiply your weight by 0.30 to get your net calorie burn per hour, according to Runner's World

  4. #4 peak

    connecting computers and peripherals or other devices. It is capable of supporting up to 127 daisy-chained peripheral devices simultaneously. The latest USB 2.0 specification can deliver 480Mbps data transfer bandwidth. In addition,

  5. #5 peak

    ; that is, Intel had to be able to build it on the same manufacturing lines that generated the 100-MHz P5s. (0.6µm Bipolar Complementary Metal Oxide Semiconductor, or BiCMOS

  6. #6 peak

    0 (P1/P0) where P1 and P0 are two quanties that are in the same units of measure. The decibel is 0.1 bel, so the decibel value LdB is LdB = 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: if the entire $ \frac{ (m+1) }{ (m+2) } } $ $ \frac{ (m+2) }{ (m+3) } $ $ \int_{\frac{ (

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

    the NEJM column.⏎Yet hospitals and politicians alike often tout the jobs that the health care industry creates in local communities. “The health care sector is an economic mainstay, providing stability and even growth during times of recession,” the American Hospital Association

  2. #2 peak

    Gender, Complex Political Emergencies, and International Intervention Chapter 3 Gender and the Causes and Consequences of Armed Conflict Chapter 4 Gender Mainstreaming in United Nations Peacekeeping Operations: Talking the Talk, Tripping over

  3. #3 peak

    Regulations, 1992: Main aim of this legislation is establishing tour operators' responsibility towards customers and providing remedies to

  4. #4 peak

    Gopher Prairie on the train. In the drab town are three thousand dull people, in a social strata ranging from Swede farmer to bank president. Main Street has two-story brick shops flanked by Fords and lumber wagons.

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. L18:1758 suppresses
  2. L9:1854 suppresses
  3. L17:1552 suppresses
  4. L19:831 suppresses
  5. L18:791 weakly suppresses
  6. L21:763 weakly excites
  7. L20:1120 weakly excites
  8. L21:1159 weakly suppresses
  9. L18:1409 weakly suppresses
  10. L5:582 weakly excites
  11. L20:179 weakly excites
  12. L21:395 weakly excites
  13. L20:169 weakly suppresses
  14. L19:1351 weakly suppresses
  15. L5:1909 weakly suppresses
  16. L18:569 weakly excites
  17. L15:12 weakly excites
  18. L16:432 weakly suppresses period after month or initial
  19. L2:1369 weakly excites
  20. L2:75 weakly suppresses
  21. L1:670 weakly excites
  22. L20:486 weakly excites
  23. L18:737 weakly excites
  24. L20:962 weakly excites decimal points and dots in numbers

Feeds later layers

  1. L23:1279 excites
  2. L23:585 excites
  3. L27:435 weakly excites
  4. L24:256 weakly suppresses
  5. L26:760 weakly excites
  6. L24:371 weakly excites
  7. L23:950 weakly excites
  8. L26:327 weakly suppresses
  9. L23:2500 weakly excites
  10. L26:1467 weakly suppresses
  11. L23:2043 weakly excites
  12. L26:929 weakly excites
  13. L27:829 weakly excites
  14. L27:1631 weakly excites
  15. L25:985 weakly excites
  16. L24:2115 weakly excites
  17. L25:2159 weakly excites
  18. L23:2825 weakly excites
  19. L26:1307 weakly excites
  20. L23:2553 weakly suppresses
  21. L25:212 weakly excites
  22. L25:2080 weakly suppresses
  23. L24:1170 weakly excites
  24. L23:2853 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 -6.39 and -5.122 values between -3.84 and -2.572 values between -2.57 and -1.301845 values between -1.30 and -0.032045 values between -0.03 and 1.244 values between 1.24 and 2.514 values between 2.51 and 3.781 values between 3.78 and 5.061 values between 12.68 and 13.961 values between 25.40 and 26.671 values between 29.21 and 30.481 values between 43.20 and 44.47offnegativepositiveoften

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. uong
  2. annon
  3. OMP
  4. aises
  5. ética
  6. 惑
  7. aggio
  8. gebung
  9. atters
  10. орт

Suppresses

  1. pen
  2. ped
  3. opt
  4. experiment
  5. pool
  6. ades
  7. clin
  8. AxisSize
  9. men
  10. esting