Neuron Explorer

Qwen3-0.6B

'a' before a time span

A neuron in layer 24 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

Fires on the word 'a' when it comes right before an amount of time, like 'a few days' or 'a short amount of time', and it expects a word such as days or weeks to follow. Turning it up made the model write lists of time words like days, weeks, months and hours.

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

    after a preposition to refer again to something that has already been mentioned. See the full definition for<|endoftext|>I have recently been re-reading two books I purchased a few years ago, when I was still teaching languages traditionally. The books are:⏎

  2. #2 peak

    had multiple FBI sources saying they were 99% sure that Hillary Clinton’s email server had been hacked by foreign intelligence agencies (the claim was retracted a few days later), that was surely Fake News. The content built on arguably real

  3. #3 peak

    led to him receiving many scars and more importantly brain damages, landing him into a truly deep coma. Fascinatingly, he was able to recover in less than a year of his predicted “revival”. The key factor here is progesterone

  4. #4 peak

    Women’s Health Watch pointed out, the fact your urine smells stronger is not necessarily a cause for concern.⏎In fact, it’s almost always temporary and disappears in a short amount of time. Despite this, it shouldn’t be ignored that it can

  5. #5 peak

    hit your head or have a whip-lash type injury. No imaging can identify a concussion and it is purely assessed on signs and symptoms that occur immediately or within a few days following.⏎80-90% of people with a concussion make

  6. #6 peak

    to aggressively fight for a fair and just settlement.⏎What does a head injury affect quality of life?⏎Because the range of severity is so great with some symptoms lasting a few days to a week, or they could last years or even a lifetime.

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 number of my children's next Nerd's gallery is just one more than the number of my next Nerd's gallery system. Let's make a more precise system now system. Let's make a system for each one of them,

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

    water.⏎Feed tropical houseplants once a month during spring and summer, using a water-soluble houseplant fertilizer. During fall and winter, water only once every other month. Or if you prefer, you can fertilize the plant with a

  2. #2 peak

    cancer would return.⏎They looked at 6,847 women with the ER-positive version of the disease, half of whom were given tamoxifen for the standard five year period, and half who continued with the drug until the 1

  3. #3 peak

    If I told you there was a crisis that, in mortality terms, was the equivalent of a three 9/11s every week for the last 7 years⏎, would you know which one I'm talking about?⏎

  4. #4 peak

    digital gold or halving if you’ve been listening close enough. More importantly, you may have even heard about how Bitcoin or BTC prices have been hitting record highs these few days.⏎Also See: The Most Secure Hardware Wallet⏎So, you may

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. L21:1284 strongly excites
  2. L19:3065 strongly excites 'for' after a span of time
  3. L22:1074 suppresses
  4. L20:1355 suppresses
  5. L19:1917 excites
  6. L21:1797 suppresses
  7. L20:1155 excites
  8. L22:1699 excites
  9. L23:1514 suppresses
  10. L19:2563 excites
  11. L17:2544 suppresses
  12. L22:2704 excites
  13. L14:2842 suppresses
  14. L21:1729 excites
  15. L19:1712 excites
  16. L14:2425 suppresses
  17. L21:2852 suppresses spot where a number or 'time' comes next
  18. L7:1114 suppresses
  19. L20:696 excites
  20. L12:2377 excites
  21. L7:2522 suppresses
  22. L17:1512 excites
  23. L19:2872 excites
  24. L17:2964 suppresses

Feeds later layers

  1. L27:1131 excites
  2. L25:2993 excites
  3. L26:284 excites
  4. L27:1105 excites
  5. L27:1969 excites
  6. L26:2126 excites
  7. L25:565 excites
  8. L27:1939 excites
  9. L27:1743 excites
  10. L25:1286 excites
  11. L25:2035 weakly excites
  12. L25:437 weakly excites
  13. L25:2493 weakly excites
  14. L26:856 weakly excites
  15. L25:2817 weakly excites
  16. L26:1448 weakly excites
  17. L25:1377 weakly excites
  18. L27:423 weakly excites
  19. L26:1445 weakly excites
  20. L27:1233 weakly excites
  21. L26:2225 weakly excites
  22. L26:2769 weakly excites the 'inter' word beginning
  23. L26:2543 weakly excites
  24. L27:2760 weakly excites 'the' after 'between' in age ranges

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.

3 values between -13.51 and -12.262 values between -12.26 and -11.021 values between -11.02 and -9.771 values between -9.77 and -8.536 values between -7.28 and -6.043 values between -6.04 and -4.808 values between -4.80 and -3.5513 values between -3.55 and -2.3145 values between -2.31 and -1.062937 values between -1.06 and 0.18727 values between 0.18 and 1.4356 values between 1.43 and 2.6739 values between 2.67 and 3.9216 values between 3.92 and 5.1610 values between 5.16 and 6.4111 values between 6.41 and 7.654 values between 7.65 and 8.902 values between 8.90 and 10.145 values between 10.14 and 11.393 values between 11.39 and 12.634 values between 12.63 and 13.881 values between 13.88 and 15.121 values between 15.12 and 16.372 values between 16.37 and 17.613 values between 17.61 and 18.861 values between 18.86 and 20.101 values between 20.10 and 21.351 values between 21.35 and 22.591 values between 23.83 and 25.081 values between 35.04 and 36.28offnegativepositiveoften

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. months
  2. days
  3. weeks
  4. hours
  5. years
  6. days
  7. weeks
  8. Months
  9. months
  10. hours

Suppresses

  1. 第三个
  2. MatrixMode
  3. listOf
  4. istine
  5. 饧
  6. emouth
  7. nex
  8. knife
  9. 石灰
  10. -profile