Neuron Explorer

Qwen3-0.6B

the word 'also' before a verb

A neuron in layer 11 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

It fires on the word "also" when it sits between a subject and its verb, as in "which also focuses" or "may also be". Turning it up made the model use "also" more often in its replies, like "the sun was also shining".

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

    with fireplaces and chimneys.⏎Appropriately and reliably, this issue is also integral to NFPA 211-2019, which also focuses on the reduction of fire hazards associated with the construction and installation of these appliances

  2. #2 peak

    to hospital for IV therapy, there may not be enough staff to monitor the drip, and the child may die from receiving the wrong amount of fluid, which is also potentially fatal in cases of over-hydration. In order to help health care

  3. #3 peak

    a multiagent system. In this multiagent system, intersections use a new reservation-based approach built around a detailed communication protocol, which we also present. We demonstrate in simulation that our new mechanism has the potential to significantly out

  4. #4 peak

    lower mantle. This may change our understanding of the deep Earth.⏎"We still don't fully understand the chemistry of the H-phase," said lead author Li Zhang, also of Carnegie. "But this finding indicates that all geodynamic models need to be

  5. #5 peak

    span for adult mosquitoes is between 2 weeks to a month or more.⏎Their flight range is usually 50-100 meters. Ranges have also been expressed as 100 feet to 100 yards, and

  6. #6 peak

    technology to include software and multimedia options that encourage organizations to expand over greater geographic distances has also led to a rise in the use of virtual teams. Virtual teams may also be traditional, self-direct or leadership teams. One challenge of

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 have a problem with the first of the first, the first of the first, and the first of the first, and the first of the first, and the first... and the first..." and the first sentence is the first,

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

    , it becomes our problem too. The Aqua Dental clinic at Pearland assure you to be partners in your child

  2. #2 peak

    special character, three letters⏎- Take a word and insert numbers and special characters in specific positions⏎Those are the basic components of your new password system. Even the simplest version is a fairly difficult password to crack. If you’re really security conscious

  3. #3 peak

    stood Betsy is a better defense against the modern school system than John Taylor Gatto’s books.⏎“An older girl was reflected (in the mirror) also, near Molly, a dark-eyed, red-cheeked, sturdy little girl,

  4. #4 peak

    they have refused to host a public hearing. Are they afraid that they would again have a standing-room-only crowd of concerned citizens rejecting the HCP, as they

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. L4:1458 strongly suppresses
  2. L7:201 excites
  3. L10:696 suppresses
  4. L6:1347 suppresses
  5. L9:78 suppresses
  6. L8:845 excites 'too' meaning 'also' at clause end
  7. L9:598 weakly suppresses
  8. L5:952 weakly excites
  9. L8:7 weakly excites
  10. L5:767 weakly suppresses
  11. L8:142 weakly excites
  12. L5:628 weakly excites
  13. L8:10 weakly excites
  14. L5:295 weakly suppresses
  15. L5:872 weakly suppresses 'or' introducing an alternate name
  16. L5:540 weakly excites
  17. L8:607 weakly excites
  18. L2:942 weakly suppresses
  19. L5:158 weakly suppresses
  20. L6:7 weakly suppresses
  21. L10:1676 weakly excites
  22. L1:2058 weakly suppresses
  23. L3:172 weakly excites
  24. L7:1298 weakly excites

Feeds later layers

  1. L12:2970 excites
  2. L16:2179 excites
  3. L13:1302 excites
  4. L17:2158 excites
  5. L15:2093 excites
  6. L17:2081 weakly excites
  7. L17:2623 weakly excites
  8. L13:772 weakly excites
  9. L14:1756 weakly excites
  10. L13:1533 weakly excites
  11. L14:858 weakly excites
  12. L15:1340 weakly excites
  13. L14:900 weakly excites
  14. L13:2103 weakly excites
  15. L13:962 weakly excites
  16. L16:1688 weakly excites
  17. L13:995 weakly suppresses
  18. L15:1693 weakly excites
  19. L17:962 weakly excites
  20. L16:1068 weakly suppresses
  21. L12:2957 weakly excites
  22. L15:1948 weakly excites
  23. L12:436 weakly excites
  24. L19:406 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.

1 values between -2.00 and -1.5916 values between -1.18 and -0.781114 values between -0.78 and -0.372567 values between -0.37 and 0.04116 values between 0.04 and 0.4425 values between 0.44 and 0.8512 values between 0.85 and 1.2618 values between 1.26 and 1.6612 values between 1.66 and 2.076 values between 2.07 and 2.474 values between 2.47 and 2.882 values between 2.88 and 3.293 values between 3.69 and 4.103 values between 4.10 and 4.511 values between 4.91 and 5.322 values between 5.73 and 6.131 values between 6.13 and 6.541 values between 6.94 and 7.352 values between 11.01 and 11.411 values between 11.41 and 11.821 values between 13.85 and 14.26offnegativepositiveoften

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. �
  2. rand
  3. ます
  4. рит
  5. 竹
  6. ;r
  7. 出
  8. rend
  9. 指
  10. ately

Suppresses

  1. 却被
  2. ignet
  3. ign
  4. vig
  5. 专员
  6. Boots
  7. 只见
  8. kie
  9. 的伟大
  10. 奖学金