Neuron Explorer

Qwen3-0.6B

energy and resource nouns

A neuron in layer 11 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 for things we spend or use up, like electricity, energy, money and hours. Turning it up pushed the model toward unrelated everyday and problem-style text, so the steering does not clearly show this theme.

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

    00 tons of CO2 per year. At least 70-80% of cooking time is saved and therefore the use of valuable resources such as electricity, gas, wood, paraffin, money and time is drastically reduced.⏎

  2. #2 peak

    RO), which pushes brine water through a membrane that retains the salt and other impurities.⏎Thermal desalination uses heat as well as electricity in distillation processes with saline feedwater heated to vaporise, so fresh water

  3. #3 peak

    Greenhouse Gas Emissions⏎The impact of animal agriculture on greenhouse gas emissions goes a lot further than just cows producing methane gas. Meat production requires vast amounts of energy. Not only do you have to grow the crops to feed the animals, but

  4. #4 peak

    Pioneer Institute published an in-depth financial analysis of Common Core and determined that it would cost the state of Montana about $40 million to implement. Where will that money come from, and why wasn’t the legislature the body to decide if this was

  5. #5 peak

    olshoi” is Russian for “big” or “grand,” and the name is apt: This is a large-scale computational project, consuming six million CPU hours and producing a hundred terabytes of data. And yet, when you ponder the

  6. #6 peak

    of the very few slaves who managed to gain literacy, which made him became the “monopoly source of slave narrative”.⏎Don’t waste your time!⏎Order your assignment!⏎That Douglass possessed this monopoly astonished the public and raised his credibility quickly

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: a very clean, professional, and well-structured document, with all the notes and explanations. The text is in English, with 100 words, and the notes are in a standard format. The main topics are covered in 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

    Get a clear picture of the problem (and the solutions) from ?The Coming Global Superstorm,? now only $9.95 for a hardcover signed by Whitley Strieber, click hereand scroll down.⏎NOTE: This

  2. #2 peak

    their athletic skills. They work cooperatively playing soccer, badmitton, frisbee, bean bag bowling and other games.⏎17 Views 20 Downloads<|endoftext|>In Key Stage 1 the children will:⏎The five enquiry types⏎The

  3. #3 peak

    beautiful. The rarer and more bizarre anomalies and mutations are described in this fascinating book which opens a window on this developing and complex field of inquiry.⏎Profusely

  4. #4 peak

    18 x 638 mm; watermark: Atlas holding the celestial spheres; slightly frayed and darkened edges, else in fine condition.⏎Incl. BTW € 151,25⏎Excl.

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. L4:2007 excites
  2. L10:884 suppresses
  3. L2:1899 weakly excites
  4. L4:690 weakly excites
  5. L1:2820 weakly suppresses
  6. L7:1436 weakly excites
  7. L6:808 weakly excites
  8. L2:352 weakly excites
  9. L10:1667 weakly suppresses
  10. L9:2351 weakly suppresses
  11. L4:929 weakly suppresses
  12. L3:3056 weakly excites
  13. L5:2245 weakly excites
  14. L7:817 weakly excites
  15. L10:831 weakly suppresses
  16. L8:1542 weakly suppresses
  17. L8:2663 weakly suppresses
  18. L1:2991 weakly suppresses
  19. L8:1218 weakly excites
  20. L9:1324 weakly suppresses
  21. L5:367 weakly suppresses
  22. L6:1283 weakly excites
  23. L10:391 weakly suppresses
  24. L7:1059 weakly excites

Feeds later layers

  1. L14:2981 weakly suppresses
  2. L15:1721 weakly suppresses
  3. L13:2900 weakly suppresses
  4. L13:2586 weakly excites
  5. L13:2329 weakly suppresses
  6. L17:2038 weakly suppresses
  7. L12:381 weakly excites
  8. L16:2967 weakly suppresses
  9. L19:2481 weakly suppresses
  10. L14:2084 weakly excites
  11. L22:2564 weakly excites perpendicular and geometry terms
  12. L15:761 weakly excites
  13. L17:1078 weakly suppresses
  14. L18:2353 weakly suppresses
  15. L16:1947 weakly suppresses
  16. L24:1503 weakly suppresses
  17. L25:2651 weakly excites the word 'Mount' in place names
  18. L17:1899 weakly suppresses
  19. L16:1605 weakly excites
  20. L21:2458 weakly suppresses
  21. L12:2247 weakly suppresses
  22. L24:1124 weakly excites
  23. L21:2604 weakly suppresses
  24. L14:2576 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 -2.10 and -2.011 values between -2.01 and -1.921 values between -1.74 and -1.642 values between -1.28 and -1.191 values between -1.09 and -1.001 values between -1.00 and -0.916 values between -0.82 and -0.733 values between -0.73 and -0.636 values between -0.63 and -0.5420 values between -0.54 and -0.4553 values between -0.45 and -0.36163 values between -0.36 and -0.27383 values between -0.27 and -0.18698 values between -0.18 and -0.081042 values between -0.08 and 0.01894 values between 0.01 and 0.10425 values between 0.10 and 0.19133 values between 0.19 and 0.2850 values between 0.28 and 0.3714 values between 0.37 and 0.473 values between 0.47 and 0.564 values between 0.56 and 0.652 values between 1.20 and 1.291 values between 1.29 and 1.381 values between 1.48 and 1.57offnegativepositiveoften

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. onto
  2. 第一条
  3. toggleClass
  4. ighter
  5. Concurrency
  6. .hashCode
  7. erg
  8. .onPause
  9. 停滞
  10. 伫

Suppresses

  1. assel
  2. ackage
  3. entially
  4. 完整
  5. ован
  6. matlab
  7. initely
  8. ibly
  9. andal
  10. offer