Neuron Explorer

Qwen3-0.6B

second number in a pair

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 fires on the spot right before the second half of a pair of numbers, like the '95' in 'by 90 percent and 95 percent' or the '48' in 'of 36 or 48 volts'. It is tracking a number that comes next in a set of two. Turning it up made the text wander, so the steering results do not show a clear link.

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

    and HENCE v.i=10. LOCAL VARIABLE i is assigned value 0 on entry to method another() so printing these values gives 10 and 0 respectively. 3)The original object ref v is unchanged , it

  2. #2 peak

    the past 20 years, and numbers of antelope species, such as hartebeest and oryx, have declined by 90 percent and 95 percent

  3. #3 peak

    Designed as the third project in a series of Point-in-Time Surveys (PITS) (the previous two occurred in 1993-4 and 2001-2), this project surveyed Great Lakes residents over the age

  4. #4 peak

    type⏎cd example.com and it will look for⏎/var/www/example.com, stopping whenever it finds a match⏎or runs out of search locations.⏎You can have as many base search paths as you

  5. #5 peak

    climate change, the electric power industry’s emissions of carbon dioxide, CO2, have risen 5.9 percent since 2002 and 11.7 percent since 1997, the analysis shows.⏎

  6. #6 peak

    Koothu or two pairs of Koothus such as Vassai Koothu-Pukal Koothu; Vethiyal Koothu-Pothuviyal Koothu; Vari Koothu - Vari Santh

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: the first time, I have been given a number and the first number is 2 for the first time. So I have been given a number and the first number is 2. So I have been given a number and the first number

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

    a matrix⏎Find nth maximum⏎Matlab Basics II - Create a vector with a repeated entry⏎Create sequnce 1 4 9 16 25.........⏎Basics: Divide integers

  2. #2 peak

    ) and V = (- 1 , 2 , 1 ,- 1) are orthogonal vectors in R 4 . Write the vector X = (1 , 1 , 1 , 2) in the form X = a U

  3. #3 peak

    ants, six corporals, two buglers, and forty-two privates. A squadron was to be formed in each of the towns of Guerrero, Mier, Camargo, and Reinosa; and a company was to be organized at L

  4. #4 peak

    Following is the sequence:⏎- Chinta (Anxiety)⏎- Asha (Hope)⏎- Shraddha (Reverential belief, Faith, Virtue of being

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. L9:587 excites
  2. L9:622 suppresses
  3. L8:719 excites
  4. L8:150 excites
  5. L8:780 excites
  6. L7:761 weakly suppresses
  7. L5:540 weakly excites
  8. L9:1517 weakly suppresses
  9. L5:1523 weakly excites
  10. L7:737 weakly suppresses comma after a place name
  11. L8:607 weakly excites
  12. L6:775 weakly suppresses
  13. L2:2173 weakly suppresses
  14. L9:1224 weakly excites
  15. L3:1996 weakly excites
  16. L7:519 weakly suppresses
  17. L6:835 weakly excites
  18. L9:1992 weakly excites
  19. L8:47 weakly excites
  20. L7:2772 weakly suppresses
  21. L0:98 weakly suppresses
  22. L4:1139 weakly excites
  23. L0:57 weakly excites
  24. L5:640 weakly suppresses

Feeds later layers

  1. L11:1777 excites
  2. L14:988 excites the word 'two' after 'the other'
  3. L14:1629 suppresses
  4. L11:1551 excites
  5. L13:1081 suppresses
  6. L14:2455 weakly excites
  7. L12:1814 weakly suppresses
  8. L24:2829 weakly excites
  9. L18:851 weakly suppresses
  10. L15:2545 weakly suppresses
  11. L12:690 weakly suppresses
  12. L12:2981 weakly excites
  13. L23:2509 weakly excites
  14. L13:650 weakly excites
  15. L15:1097 weakly excites
  16. L11:1090 weakly suppresses
  17. L11:2842 weakly excites
  18. L21:703 weakly excites
  19. L21:1742 weakly excites
  20. L24:849 weakly suppresses
  21. L15:1492 weakly suppresses
  22. L26:2627 weakly excites
  23. L15:1067 weakly suppresses
  24. L25:2692 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 -1.84 and -1.731 values between -1.38 and -1.271 values between -1.04 and -0.932 values between -0.93 and -0.812 values between -0.81 and -0.703 values between -0.70 and -0.599 values between -0.59 and -0.4738 values between -0.47 and -0.36170 values between -0.36 and -0.24423 values between -0.24 and -0.131036 values between -0.13 and -0.021280 values between -0.02 and 0.10633 values between 0.10 and 0.21209 values between 0.21 and 0.3356 values between 0.33 and 0.4421 values between 0.44 and 0.567 values between 0.56 and 0.673 values between 0.78 and 0.906 values between 0.90 and 1.011 values between 1.01 and 1.132 values between 1.13 and 1.241 values between 1.24 and 1.351 values between 1.35 and 1.471 values between 2.04 and 2.151 values between 2.61 and 2.72offnegativepositiveoften

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. MBED
  2. irectional
  3. łam
  4. mouseleave
  5. �
  6. ustin
  7. IDL
  8. cestor
  9. EDIUM
  10. !");⏎

Suppresses

  1. @
  2. lar
  3. hostel
  4. requis
  5. 您好
  6. uja
  7. ActivityResult
  8. order
  9. repeats
  10. ObjectContext