Neuron Explorer

Qwen3-0.6B

unclear: verbs and prepositions in news prose

A neuron in layer 16 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

The neuron fires on assorted words in news and science writing, like 'developing', 'by', or 'when' just before a clause continues. No clear pattern stands out. Turning it up gave unrelated text.

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

    graph progress of conservation measures. If monthly water use figures are provided, use can be graphed monthly.⏎- Writing, art and technology skills can be practiced in developing water conservation messages for posters, the school<|endoftext|>Agile is

  2. #2 peak

    just to evaluate where women are in terms of representation and equal pay for equal work, but also to consider the potential for jumpstarting climate negotiations and the green economy by strengthening women’s leadership in these areas.⏎Our elected

  3. #3 peak

    recalled. The study could lead to new treatments for conditions like depression, post-traumatic stress disorder (PTSD) and even addictions— all of which can involve memories that are intrusive and potentially destructive.⏎The intriguing experiment involved 42 severely

  4. #4 peak

    wandering close to a supermassive black hole and finding itself mercilessly ripped apart by this beast weighing millions to billions times more than the hapless star. It doesn’t happen very often. But when it happens, it’s pretty spectacular. And

  5. #5 peak

    Alaska’s permafrost is now essentially melting beneath our feet. We know the problem; we all drove on wavy Alaskan roads with bumps and dips caused by thawing permafrost.⏎On a larger scale, oil companies and commercial fishing

  6. #6 peak

    the speed at which your brain can function, while hours spent sitting at a desk causes back, shoulder and neck pain. Then there is the decreased sleep quality experienced when using tech too much, which causes many more health concerns. Basically, too much

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 new project that is a new one. I have a lot of questions and I need a lot of questions. It is a new one. I have a lot of questions, and it is a new one. It is 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

    .⏎Unlike medicine, there is no instant gratification with public health. It’s like Michelangelo and the Sistine Chapel. It took him 4 years to paint its ceiling. 4 years! It might not take that long for public health

  2. #2 peak

    can add in more and more challenges.<|endoftext|>A post-mortem examination, also known as an autopsy, is a step-by-step examination of the outside of the body and of the internal organs by a doctor known as a

  3. #3 peak

    , they are more prepared to grow up and obtain a decent job. When water is readily accessible, a family does not have to waste time and resources hauling buckets from the nearest stream, but

  4. #4 peak

    on obstetrics. The very first ultrasound scan during pregnancy was carried out in Glasgow in 1956, using a machine that was originally designed to detect industrial flaws in ships

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. L15:908 excites
  2. L15:45 suppresses
  3. L15:149 suppresses
  4. L15:1537 suppresses
  5. L15:121 suppresses
  6. L12:1889 excites
  7. L15:2564 weakly suppresses
  8. L5:598 weakly suppresses
  9. L13:1407 weakly suppresses
  10. L15:252 weakly excites
  11. L15:90 weakly excites
  12. L15:1369 weakly suppresses
  13. L8:1677 weakly suppresses unclear: mixed, mostly math and instructions
  14. L12:2259 weakly suppresses
  15. L14:2101 weakly excites
  16. L8:193 weakly excites
  17. L14:2133 weakly excites
  18. L15:1327 weakly suppresses
  19. L15:2198 weakly excites
  20. L15:680 weakly excites
  21. L7:191 weakly suppresses
  22. L12:684 weakly suppresses
  23. L4:909 weakly suppresses
  24. L2:1610 weakly excites

Feeds later layers

  1. L17:1151 excites
  2. L17:2907 excites
  3. L17:2057 weakly excites
  4. L18:989 weakly excites
  5. L19:783 weakly excites
  6. L21:12 weakly excites
  7. L21:296 weakly excites
  8. L20:2030 weakly suppresses
  9. L21:475 weakly suppresses
  10. L20:638 weakly excites
  11. L19:1156 weakly excites
  12. L18:1959 weakly suppresses
  13. L17:138 weakly suppresses
  14. L17:807 weakly suppresses
  15. L19:2087 weakly excites
  16. L19:2074 weakly excites
  17. L17:1315 weakly excites
  18. L26:2548 weakly suppresses
  19. L24:690 weakly excites
  20. L19:1567 weakly excites
  21. L21:715 weakly excites
  22. L19:2059 weakly excites
  23. L19:1055 weakly excites
  24. L27:2706 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 -5.11 and -4.912 values between -3.48 and -3.281 values between -3.28 and -3.081 values between -3.08 and -2.871 values between -2.87 and -2.673 values between -2.67 and -2.462 values between -2.46 and -2.266 values between -2.26 and -2.0615 values between -2.06 and -1.8530 values between -1.85 and -1.6538 values between -1.65 and -1.4565 values between -1.45 and -1.24110 values between -1.24 and -1.04184 values between -1.04 and -0.83330 values between -0.83 and -0.63489 values between -0.63 and -0.43832 values between -0.43 and -0.221092 values between -0.22 and -0.02551 values between -0.02 and 0.18100 values between 0.18 and 0.3927 values between 0.39 and 0.5914 values between 0.59 and 0.806 values between 0.80 and 1.004 values between 1.00 and 1.201 values between 1.20 and 1.412 values between 1.82 and 2.021 values between 2.83 and 3.04offnegativepositiveoften

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. 用
  2. ien
  3. indirectly
  4. palm
  5. ospace
  6. thermal
  7. cover
  8. Phys
  9. cribe
  10. Sup

Suppresses

  1. vik
  2. FRING
  3. sticks
  4. rette
  5. ]-->⏎
  6. cedures
  7. Rede
  8. 与众
  9. viders
  10. .cljs