Neuron Explorer

Qwen3-0.6B

spot where a number or 'time' comes next

A neuron in layer 21 of Qwen3-0.6B. Its strongest snippets are fairly consistent: several of them peak on the same token. It is featured, so it has steering outputs.

What this neuron does

This neuron fires when the next word is likely to be a number, such as an hour, a day count, or a duration, like the spot before '4:30 p.m.' or '13.7 days'. It goes strongly negative in phrases like 'without wasting time', where the next word is likely 'time'. Turning it down made the model repeat the word 'time' over and over.

Written by Claude from this neuron's activations and steering outputs. Confidence: medium. Think it is wrong? Flag it (opens an email to the maker with the neuron already filled in).

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

    By wearing glasses that block blue light allows the flow to start (2). The average time for melatonin to flow (if the person is in darkness) is 11.4 hours according to a recent study

  2. #2 peak

    .⏎Visit our Ed Clark Museum of Missouri Geology on Geologic Map Day to learn more about geologic maps. Open weekdays from 8 a.m. to 4:30 p.m., we are located at 111 Fair

  3. #3 peak

    high-voltage transients and high operating temperatures, the latter resulting in thermal runaway and over-current conditions. Protection circuitry is straightforward and easy to implement but adds to the bill-of-materials (BoM) and takes up more space

  4. #4 peak

    unter house telescope TESS, the all-sky search is on for presumably liveable planets near our photo voltaic system. TESS — orbiting Earth every 13.7 days — and ground-based telescopes are poised to search out

  5. #5 peak

    many children stay up late and sleep in, which can make it difficult to adjust to the early mornings and structured schedules

  6. #6 peak

    they first arrive. Some come to the centre hungry (it provides meals) or speaking no English. Others arrive with behaviour problems. Parents’ work schedules are often so inflexible that Federal Hill must cover basic logistics beyond school pick-up and drop-off

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: time to save the time. I have a lot of time, but I have to spend time with the time. Time is the time I have. I have a lot of time, but I have to spend time with the time. 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

    A? It’s OK. You’re going to P-R-A-C-T-I-C-E and get very comfortable with the test format so you don’t have to waste time reading instructions during the test, right? Also, this acquired familiarity is going

  2. #2 peak

    this method will help you achieve a bouncing ball animation in probably 5 minutes or less.⏎So without wasting much time, let’s get into it!⏎Here are the steps:⏎- Delete

  3. #3 peak

    this article. Of course, they can.⏎You’ll surely learn something from these moral stories at the end.⏎Read some other English moral stories.⏎Hence without wasting your valuable time, let’s start writing some of the best moral stories in English

  4. #4 peak

    70°C. [Read more…]⏎Nowadays, attend system in schools and colleges is generally based on paper. So sometimes this process causes errors and also takes more time. So this project uses RFID technology to make a note of every student

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. L15:2781 strongly suppresses
  2. L18:3067 excites
  3. L19:2792 suppresses
  4. L17:3016 excites
  5. L10:1653 suppresses
  6. L18:2776 excites
  7. L19:2896 excites
  8. L15:833 suppresses
  9. L20:1150 excites
  10. L15:1823 excites
  11. L20:1335 suppresses
  12. L18:1546 suppresses
  13. L20:1199 suppresses
  14. L8:1257 suppresses
  15. L10:1566 excites
  16. L17:2750 suppresses
  17. L18:1750 suppresses the blank slot before a year
  18. L11:1507 excites
  19. L18:569 excites
  20. L11:951 excites
  21. L19:1233 excites
  22. L19:2989 suppresses
  23. L19:2706 weakly excites "per" in rate phrases
  24. L20:2382 weakly suppresses

Feeds later layers

  1. L27:836 strongly suppresses
  2. L25:565 strongly suppresses
  3. L24:2325 strongly suppresses
  4. L22:1074 strongly suppresses
  5. L23:1549 strongly suppresses
  6. L26:2065 suppresses
  7. L22:2932 suppresses
  8. L26:2567 suppresses
  9. L22:2445 suppresses
  10. L23:2577 suppresses
  11. L24:1929 suppresses
  12. L22:2575 suppresses
  13. L23:2882 suppresses
  14. L27:934 suppresses
  15. L22:1563 suppresses
  16. L24:906 suppresses
  17. L26:2126 suppresses
  18. L23:1490 suppresses
  19. L23:2178 suppresses
  20. L22:2729 suppresses
  21. L22:2601 suppresses
  22. L24:1778 suppresses 'a' before a time span
  23. L23:2359 suppresses
  24. L24:1518 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 -12.66 and -12.071 values between -6.76 and -6.172 values between -4.99 and -4.403 values between -4.40 and -3.815 values between -3.23 and -2.646 values between -2.64 and -2.0510 values between -2.05 and -1.4632 values between -1.46 and -0.87457 values between -0.87 and -0.282747 values between -0.28 and 0.31595 values between 0.31 and 0.9021 values between 0.90 and 1.499 values between 1.49 and 2.083 values between 2.08 and 2.674 values between 2.67 and 3.263 values between 3.26 and 3.852 values between 3.85 and 4.442 values between 4.44 and 5.031 values between 5.03 and 5.621 values between 6.80 and 7.392 values between 9.74 and 10.331 values between 10.33 and 10.92offnegativepositiveoften

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. ildo
  2. -alist
  3. headquartered
  4. oidal
  5. حزب
  6. ervlet
  7. Saddam
  8. addError
  9. uitar
  10. campuses

Suppresses

  1. time
  2. 时间
  3. 时间和
  4. time
  5. (time
  6. 的时间
  7. Time
  8. Time
  9. 時間
  10. 时间