Neuron Explorer

Qwen3-0.6B

'at the beginning/start' time phrases

A neuron in layer 13 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 fires on words that mark a point in time, especially phrases like 'at the beginning' or 'at the start' of something, and 'at the time'. When turned down, the model's text got stuck in loops about a story or a girl on a bus, and turning it up pushed the model into vague, rambling sentences.

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 potentially create water savings for the school district.⏎- Many math lessons can be taught when school water use is audited. Conducting a preliminary audit at the beginning of the school year and a post audit before the end of the school year allows

  2. #2 peak

    of years after the last recorded volcanic eruption of Mount Agung dozens of miles away.⏎Wayan Lebih, 55, was a young boy at the time and remembers gathering co

  3. #3 peak

    cided lesson plans and syllabus completion plans⏎Any good integrated learning system allows the school management to set out lesson plans and decide syllabus completion deadlines at the beginning of the academic year. Once these are pre-decided, teachers can then

  4. #4 peak

    the e-learning market in the United States is expected to grow $6.8 billion by 2015, up from $2.9 billion in 2010. According to Sanjeev Ahuja, the vice president

  5. #5 peak

    of watersheds and drinking water systems across the Midwest and Southern United States. This May 2010 report summarizes scientific information that has emerged since the publication of our initial report and includes more recent monitoring data.⏎Approximately 75

  6. #6 peak

    , but also the way people will resist as part of unlearning. Snyder’s-Lance zeroed in on the ways managers were expressing resistance and encouraged it early in the process to help them “unlearn” the response patterns that were prohibiting them

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 was given a puzzle, I had a collection of a few numbers, I noticed that I had 3, I had a, I was given a, I was given the, I had a, I was told, I had a

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

    was there from the start; as a baby, Chopin is said to have been moved to tears by his mother’s singing.⏎Chopin started playing piano at age 5 under the tutelage of his sister, Ludwika, and

  2. #2 peak

    are indelibly etched in the memory of those viewing his work. Born in New York in 1882, Hopper showed early interest in art, particularly drawing, and went on to study illustration and painting. With

  3. #3 peak

    Christian emperor of Rome, ordered to build a basilica on Vatican Hill, on the site of the present St. Peter's Cathedral. Though a small shrine already existed on the site, it was replaced by a new church that was completed around

  4. #4 peak

    were excited to see them selected to support the Army Game Studio on the game."⏎Though development of the game didn't start until last year, Laughlin actually began researching the prospect of

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. L12:1930 excites
  2. L12:1313 excites
  3. L12:2898 suppresses
  4. L11:389 suppresses
  5. L12:650 suppresses
  6. L12:742 excites
  7. L12:430 excites
  8. L12:204 suppresses
  9. L11:764 suppresses
  10. L12:1516 suppresses
  11. L8:2618 suppresses
  12. L9:563 excites
  13. L11:1819 suppresses
  14. L12:2857 suppresses
  15. L12:179 suppresses
  16. L10:3015 excites
  17. L8:1228 excites
  18. L12:587 suppresses
  19. L11:1742 suppresses
  20. L12:518 weakly excites
  21. L12:1583 weakly suppresses
  22. L9:727 weakly excites
  23. L12:193 weakly suppresses
  24. L7:1104 weakly suppresses

Feeds later layers

  1. L19:457 suppresses
  2. L14:184 excites
  3. L16:885 suppresses
  4. L14:170 suppresses
  5. L15:1825 suppresses
  6. L25:1396 suppresses
  7. L15:1241 excites
  8. L18:2514 suppresses
  9. L21:1691 suppresses "at" in "at first" and "initial" words
  10. L15:193 excites
  11. L15:2918 suppresses
  12. L17:2922 suppresses
  13. L15:2013 suppresses
  14. L15:2881 suppresses
  15. L24:2617 suppresses
  16. L14:324 suppresses
  17. L14:2646 suppresses
  18. L14:2329 excites
  19. L15:1903 suppresses
  20. L14:1695 suppresses
  21. L14:1557 suppresses
  22. L18:3044 excites
  23. L15:446 excites
  24. L21:1453 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.

2 values between -4.79 and -4.642 values between -3.90 and -3.761 values between -2.87 and -2.722 values between -2.28 and -2.142 values between -2.14 and -1.992 values between -1.99 and -1.842 values between -1.84 and -1.694 values between -1.69 and -1.552 values between -1.55 and -1.404 values between -1.40 and -1.255 values between -1.25 and -1.115 values between -1.11 and -0.9611 values between -0.96 and -0.8116 values between -0.81 and -0.6623 values between -0.66 and -0.5230 values between -0.52 and -0.3786 values between -0.37 and -0.22346 values between -0.22 and -0.071595 values between -0.07 and 0.071167 values between 0.07 and 0.22480 values between 0.22 and 0.37103 values between 0.37 and 0.527 values between 0.52 and 0.666 values between 0.66 and 0.813 values between 0.81 and 0.962 values between 0.96 and 1.10offnegativepositiveoften

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. YL
  2. indemn
  3. ffee
  4. ornado
  5. 髯
  6. \""
  7. cos
  8. yl
  9. nữa
  10. độc

Suppresses

  1. initially
  2. 初始
  3. 最初
  4. 最初的
  5. initial
  6. Initial
  7. initial
  8. inicial
  9. →initial
  10. .initial