Neuron Explorer

Qwen3-0.6B

'that' after a noun, introducing a clause

A neuron in layer 4 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 fires on the word 'that' when it follows a noun and starts a describing clause, like 'the data that was being sourced'. When turned up, the model's writing drifted into rambling, repetitive text rather than anything clearly tied to this pattern.

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

    of this activity meant that we had to leave the standard classroom and venture into the school gymnasium.⏎With this is mind I wanted to ensure that the data that was being sourced within the session would be instantly displayed for the students to analyse.

  2. #2 peak

    is good for your garden soil, it’s not the best thing to add to your compost pile. This is because it takes a very long time for the bacteria that eats dog waste (and its nutrients) to be broken down. To make things

  3. #3 peak

    up to date, call Dr. Shaheedy's office.⏎As you age, your body's immune system will also change. This may mean that the antibodies that were produced by previous vaccinations may have weakened or the viruses they were designed to prevent

  4. #4 peak

    , there was not enough-if any-on daily Roman life, which, to me, was and is much more interesting. Some people might argue that all the information that you gain from museums is in books, but where do the pictures and some of

  5. #5 peak

    fiscal impact of the budget’s policies over the next ten years. The CBO does not generally look at the long-term impacts of these policies, so the numbers that ultimately shape the public debate are ten-year numbers. This introduces a serious bias into

  6. #6 peak

    The separate video card plugs into a slot on your board and provides your computer with a second sub-processor called a GPU (graphics processing unit). All the calculations that are needed to turn the computer code into the picture on the screen are given over

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 times of a system is 100. The system has a lot of factors, which can affect the performance. The number of times of a system is 100, but the system has 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

    ier has made waves with its fresh look at adolescent sexuality⏎Almost half of American adolescents report having had sex by the time they graduate from highschool — a fact that has raised concern among adults, and parents of teens in particular, about the emotional

  2. #2 peak

    the system. Incorrect data sometimes produce an erroneous result that reduces the quality of data. To handle this problem web Miner's has introduced several

  3. #3 peak

    economic stimulus packages in 2014 to bolster growth, including reforms to environmental regulations in order to spur investment in Peru’s lucrative mining sector, a move that was opposed by some environmental groups. However, in 2015,

  4. #4 peak

    ) and Maila (later called Mary), with his first wife, who died in 1822. Hirsch later married Rebecca Haas, with whom he had two children, Vogela (later called Fanny) and Loeb

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. L3:1529 excites
  2. L2:1863 weakly excites
  3. L3:2916 weakly excites
  4. L1:1562 weakly suppresses 'that' after reporting verbs
  5. L3:1968 weakly suppresses
  6. L0:1389 weakly excites
  7. L2:1564 weakly suppresses
  8. L2:1781 weakly suppresses
  9. L2:1962 weakly suppresses
  10. L3:676 weakly suppresses
  11. L1:2932 weakly suppresses
  12. L3:1485 weakly excites
  13. L3:89 weakly excites
  14. L1:2006 weakly suppresses
  15. L3:1316 weakly suppresses
  16. L1:186 weakly suppresses
  17. L1:642 weakly suppresses
  18. L2:2023 weakly suppresses
  19. L1:175 weakly excites
  20. L2:1588 weakly excites
  21. L2:258 weakly excites
  22. L1:2660 weakly excites
  23. L2:2306 weakly excites
  24. L2:552 weakly excites comma inside a stacked adjective list

Feeds later layers

  1. L5:1989 weakly excites
  2. L15:1344 weakly excites
  3. L11:429 weakly excites
  4. L15:3032 weakly excites
  5. L10:1477 weakly excites
  6. L9:75 weakly excites 'what' or 'how' after see or understand
  7. L13:354 weakly excites
  8. L24:1782 weakly excites
  9. L18:2995 weakly excites
  10. L8:742 weakly excites
  11. L7:1968 weakly excites
  12. L21:215 weakly suppresses
  13. L23:440 weakly excites
  14. L9:865 weakly excites
  15. L6:988 weakly suppresses
  16. L15:925 weakly suppresses
  17. L7:1607 weakly excites
  18. L8:1825 weakly excites
  19. L20:2651 weakly excites
  20. L16:986 weakly suppresses
  21. L10:1058 weakly excites
  22. L8:276 weakly suppresses
  23. L10:1116 weakly excites
  24. L8:465 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.

3 values between -1.89 and -1.721 values between -1.72 and -1.541 values between -1.54 and -1.371 values between -1.37 and -1.204 values between -1.20 and -1.032 values between -1.03 and -0.867 values between -0.86 and -0.6910 values between -0.69 and -0.5227 values between -0.52 and -0.35230 values between -0.35 and -0.181421 values between -0.18 and -0.011835 values between -0.01 and 0.16285 values between 0.16 and 0.3343 values between 0.33 and 0.5114 values between 0.51 and 0.685 values between 0.68 and 0.856 values between 0.85 and 1.024 values between 1.02 and 1.191 values between 1.36 and 1.532 values between 1.53 and 1.701 values between 2.90 and 3.072 values between 3.58 and 3.751 values between 3.92 and 4.091 values between 4.60 and 4.781 values between 4.78 and 4.95offnegativepositiveoften

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. 斫
  2. emens
  3. purchasers
  4. xic
  5. Configurer
  6. authorize
  7. 税务局
  8. .fasterxml
  9. pear
  10. �

Suppresses

  1. ``⏎
  2. lahoma
  3. effectiveness
  4. bole
  5. HERO
  6. ỏi
  7. ANTE
  8. Ć
  9. ​​
  10. ICC