Neuron Explorer

Qwen3-0.6B

unclear: time spans and 'once you've'

A neuron in layer 4 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. It is featured, so it has steering outputs.

What this neuron does

This neuron lights up on a few unrelated words, mostly ones that point to a span of time, like 'period', 'year' and 'time', or the 've in 'once you've'. It switches off hard on 'when' starting a new clause. Turning it up made the model's text drift into random topics, so the steering results don't reveal much.

Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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

    "Certain trends are increasing longevity and others are reducing it," he says. "I'm trying to figure out how much each of these trends contributes."⏎Over the period Lichtenberg studied a huge number of

  2. #2 peak

    of a number, you should first know how many decimal places it falls under. You can do this by multiplying the number by the maximum possible mark. Once you’ve determined the maximum, you can also divide it by the total marks. This will

  3. #3 peak

    drive to another, we copy from a hard drive to a file. Drive images are more flexible, as you can do what you please with the data once you’ve pulled it off the source drive.⏎Your drive image is going to be a big

  4. #4 peak

    Let’s look at powder coating vs. galvanizing and when to use one or the other. Or even both!⏎Contact Minnesota Industrial Coatings when you’re ready to apply America’s best industrial coating solutions. We provide our customers with consistently high

  5. #5 peak

    hardboiled eggs or granola with fruit, nuts and seeds.⏎Eliminate or Min<|endoftext|>Venezuela’s human rights record was assessed under the UN Universal Periodic Review, whose report was adopted in March. Venezuela had accepted several of the

  6. #6 peak

    0, shows that even before the COVID-19 pandemic, majorities of citizens reported going without needed medical care and clean water at least once during the previous year. Nigerians give the government low marks for its performance in providing water and sanitation

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: I am looking for the right to be the time I have to make the decision to choose the right, I am thinking about when the time I have to be the time I have to make the decision to choose the right, I am thinking

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

    he was so wise that none could question him concerning anything but that he knew the solution. He went up and down the earth to give instruction to men; and when he came upon invitation to the abode of certain dwarves, Fjalarr

  2. #2 peak

    to attack the seemingly impregnable fortress of Fort Wagner. The odds were heavily against them; the fort had never been taken. America was watching them – and when the bugle sounded they marched towards the guns, flag held high.<|endoftext|>A

  3. #3 peak

    care and love for all whom she met—represents the best in Christian witness.⏎The poor white residents of Glynn County also trusted Deaconess Alexander. When the Depression hit the rural poor, she became the agent for government and private aid

  4. #4 peak

    :83] And when they listen to the revelation received by the Messenger, thou wilt see their eyes<|endoftext|>

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. L1:2789 suppresses
  2. L2:2166 suppresses 'in' after high or low
  3. L2:383 weakly excites
  4. L2:2451 weakly suppresses
  5. L0:178 weakly suppresses
  6. L1:1681 weakly excites
  7. L0:1301 weakly suppresses
  8. L0:1224 weakly suppresses
  9. L2:49 weakly suppresses
  10. L2:111 weakly suppresses
  11. L0:1405 weakly suppresses
  12. L1:1464 weakly excites
  13. L0:2920 weakly suppresses
  14. L1:886 weakly excites
  15. L2:537 weakly excites
  16. L0:588 weakly suppresses
  17. L0:2018 weakly suppresses
  18. L2:75 weakly suppresses
  19. L2:1710 weakly suppresses
  20. L3:691 weakly excites
  21. L1:2851 weakly excites
  22. L2:766 weakly excites
  23. L0:3009 weakly excites
  24. L1:461 weakly excites

Feeds later layers

  1. L10:833 suppresses
  2. L5:55 suppresses
  3. L8:1749 suppresses
  4. L13:2439 suppresses
  5. L17:1662 suppresses
  6. L10:456 suppresses
  7. L18:3002 suppresses
  8. L16:583 suppresses
  9. L15:2342 suppresses
  10. L17:619 suppresses
  11. L16:2083 suppresses comma after 'in this article'
  12. L7:451 suppresses
  13. L9:306 suppresses
  14. L7:1484 weakly suppresses
  15. L5:892 weakly suppresses
  16. L19:2872 weakly suppresses
  17. L8:691 weakly suppresses
  18. L16:2726 weakly suppresses
  19. L7:1182 weakly suppresses
  20. L17:2741 weakly excites
  21. L6:2369 weakly suppresses
  22. L15:513 weakly suppresses
  23. L8:1916 weakly suppresses
  24. L9:811 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.

1 values between -11.20 and -10.891 values between -10.27 and -9.961 values between -9.65 and -9.341 values between -8.41 and -8.101 values between -7.79 and -7.481 values between -6.54 and -6.231 values between -5.92 and -5.611 values between -5.61 and -5.301 values between -3.13 and -2.821 values between -2.50 and -2.193 values between -1.57 and -1.261 values between -1.26 and -0.951 values between -0.95 and -0.648 values between -0.64 and -0.33557 values between -0.33 and -0.023182 values between -0.02 and 0.29142 values between 0.29 and 0.602 values between 0.60 and 0.912 values between 0.91 and 1.22offnegativepositiveoften

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. iveness
  2. 渚
  3. скую
  4. REW
  5. insula
  6. ราม
  7. acity
  8. ifferential
  9. ucer
  10. opping

Suppresses

  1. 谈到
  2. 的时候
  3. entimes
  4. .ComponentModel
  5. indebted
  6. 大宗
  7. 谈起
  8. ึก
  9. meld
  10. (Token