Neuron Explorer

Qwen3-0.6B

slot before a year in dates

A neuron in layer 11 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 tends to light up on the spot just before a year, such as the space between 'November 13,' and '1850'. It also fires on a few unrelated word positions. Turning it up gave rambling text that did not clearly relate to dates, so its effect on writing is hard to read.

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

    Countess of Rosebery, died, at Dalmeny House, in Scotland, at the age of 39. As the only child of Meyer

  2. #2 peak

    daily amounts of vitamin D are:⏎19 to 70: 600 IU<|endoftext|>Stevenson was born on Nov. 13, 1850.

  3. #3 peak

    the top banner and click on the three dots or parallel lines. Scroll down to the subscription form and enter your information.)<|endoftext|>Born on December 28, 1926 to Reverend Wade Hampton McKinney and Ruth Berry McKinney

  4. #4 peak

    Vogela (later called Fanny) and Loeb (later called Levi). Levi, the youngest of the family, was born on February 16, 1829 in the Bavarian village of Buttenheim.⏎Life in

  5. #5 peak

    butterfly and sweetheart, respectively, in Visayan) were born in our shipping pens area on December 6, 2012, and they have quickly become two of my absolute favorite charges

  6. #6 peak

    you add in the details. They help you form action steps and timelines to reach your goals and bring you the success you desire.⏎<|endoftext|>Born May 3, 1791, in St

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 person is the person who is in the room. The person who is in the room is the other person. The person who is in the room is the other person. The other person is the other person. The other person 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

    0.1 bel, so the decibel value LdB is LdB = 10 log10 (P1/P0).⏎Define the mean squared error (MSE) between two monochromatic images, where one image

  2. #2 peak

    is considered to be an approximation of the other. The MSE can be described as the mean of the square of the differences in the pixel values between the corresponding pixels of the two images.⏎Express

  3. #3 peak

    liturgy celebrated?⏎- Where is the liturgy celebrated?<|endoftext|>A particle is moving along the parabola 4y = (x + 2)^2 in such a way that its x-coordinate is increasing at a constant rate of

  4. #4 peak

    ⏎Find nth maximum⏎Matlab Basics II - Create a vector with a repeated entry⏎Create sequnce 1 4 9 16 25.........⏎Basics: Divide integers to get integer outputs in all cases⏎D

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. L10:937 excites
  2. L9:1940 excites
  3. L6:2423 suppresses
  4. L7:2233 weakly excites
  5. L0:857 weakly suppresses
  6. L9:2253 weakly suppresses
  7. L10:1673 weakly suppresses
  8. L0:2620 weakly excites
  9. L10:2399 weakly suppresses
  10. L7:1084 weakly excites
  11. L9:2295 weakly excites
  12. L2:1623 weakly excites
  13. L10:278 weakly suppresses
  14. L5:2672 weakly excites comma after 'As a [role]'
  15. L7:3036 weakly excites
  16. L7:1885 weakly excites
  17. L5:1763 weakly suppresses
  18. L9:2033 weakly excites
  19. L6:709 weakly excites
  20. L6:2259 weakly excites
  21. L4:1460 weakly suppresses
  22. L5:2719 weakly suppresses
  23. L7:2269 weakly excites
  24. L7:2137 weakly suppresses

Feeds later layers

  1. L12:2189 excites
  2. L13:1628 excites
  3. L12:3045 excites
  4. L17:1762 excites
  5. L13:1112 excites
  6. L12:887 weakly excites
  7. L13:2328 weakly suppresses
  8. L12:1854 weakly excites
  9. L15:1425 weakly excites
  10. L14:2935 weakly excites
  11. L12:576 weakly excites 'of' in fixed phrases like table of contents
  12. L13:1622 weakly excites
  13. L13:1001 weakly suppresses
  14. L14:1353 weakly excites
  15. L14:1265 weakly excites
  16. L27:2842 weakly excites
  17. L13:1513 weakly excites
  18. L16:1344 weakly excites
  19. L14:2169 weakly excites
  20. L17:2768 weakly excites
  21. L17:718 weakly excites
  22. L13:2963 weakly excites
  23. L22:956 weakly suppresses
  24. L15:2628 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 -0.95 and -0.911 values between -0.58 and -0.541 values between -0.54 and -0.511 values between -0.51 and -0.473 values between -0.47 and -0.434 values between -0.43 and -0.408 values between -0.40 and -0.3622 values between -0.36 and -0.3225 values between -0.32 and -0.2954 values between -0.29 and -0.25104 values between -0.25 and -0.21141 values between -0.21 and -0.18214 values between -0.18 and -0.14262 values between -0.14 and -0.10372 values between -0.10 and -0.07451 values between -0.07 and -0.03546 values between -0.03 and 0.01447 values between 0.01 and 0.04399 values between 0.04 and 0.08283 values between 0.08 and 0.12223 values between 0.12 and 0.16132 values between 0.16 and 0.1980 values between 0.19 and 0.2370 values between 0.23 and 0.2724 values between 0.27 and 0.3021 values between 0.30 and 0.348 values between 0.34 and 0.381 values between 0.38 and 0.416 values between 0.41 and 0.453 values between 0.45 and 0.491 values between 0.49 and 0.52offnegativepositiveoften

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. аз
  2. adratic
  3. okedex
  4. thern
  5. ²
  6. 详
  7. ikk
  8. Quadr
  9. очно
  10. tık

Suppresses

  1. .Consumer
  2. ccoli
  3. .getCount
  4. getService
  5. maid
  6. 数目
  7. olding
  8. ировка
  9. 每一个
  10. _<