Neuron Explorer

Qwen3-0.6B

'make' before decisions or predictions

A neuron in layer 8 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 'make' when it leads into a decision, a prediction or a choice, as in 'to make informed decisions'. When it was turned up, the text drifted toward talk about making decisions.

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

    NOAA’s Fisheries Office of Aquaculture. “It is a tool that can be used when evaluating proposed or operational farming sites and gives them a factual basis to make decisions.”⏎In the Executive Summary of the report, NOAA scientists say there is an

  2. #2 peak

    driving cars.⏎One of the most exciting aspects of AI is its ability to learn and adapt. Machine learning algorithms can analyze vast amounts of data to identify patterns and make predictions, and they improve with experience. This capability has applications

  3. #3 peak

    decisions and take responsibility for their own financial well-being. Faced with an array of complex financial services and sophisticated products, many graduates lack the knowledge and skills to make rational, informed decisions on the use of their money and planning for future events

  4. #4 peak

    technology to achieve an ultraprecise lunar landing.⏎Navigation GUI, on the other hand, assists in human-machine interactions, making it possible for ground operators to make quick decisions during the descent process. It is a novel yet straightforward

  5. #5 peak

    any decision.⏎Keeps You Well-informed⏎If you consider all the facts before making any decision, then you will be well-informed of all the consequences of your decision

  6. #6 peak

    an infection.⏎Cunningham Panel™ Assists Clinicians in the Diagnosis of PANS and PANDAS⏎The Cunningham Panel of tests can assist clinicians in making the correct diagnosis. The Panel provides insight into how

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 one that I can't do, the one that I can't do. So I'm going to have to say something... So, the answer is 2, which is 22, and 22 is 2

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

    with pre-existing anemia mustn't be given garlic.⏎- Dr's Foster and Smith of peteducation.com are in agreement that garlic causes a breakdown of red blood cells in dogs though they permit that garlic in terribly tiny amounts in some industrial pet

  2. #2 peak

    by sitting down to a hot breakfast. You need to consider what you’re eating. Focus on foods that have nutrients such as calcium, B vitamins, fibre, iron and folate so that you

  3. #3 peak

    provide about 20-35% of your guideline daily calorie allowance (GDA).⏎Apart from providing energy, breakfast foods are good sources of calcium, iron and B vitamins as well as protein and fibre. Fruit and vegetables are good sources

  4. #4 peak

    it is the building block of all of our cells – and therefore our muscles, tissues and organs. And it is a tremendous source of vitamin B12 and iron. Due to its high

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. L4:1785 suppresses
  2. L4:2528 suppresses
  3. L3:1669 excites
  4. L6:1776 weakly suppresses street and urban place names
  5. L0:748 weakly excites
  6. L5:1508 weakly excites
  7. L7:1207 weakly excites
  8. L6:2633 weakly suppresses
  9. L0:2721 weakly excites
  10. L6:784 weakly suppresses
  11. L4:1419 weakly excites
  12. L2:2593 weakly suppresses
  13. L6:3007 weakly excites
  14. L3:2422 weakly excites
  15. L5:2079 weakly suppresses
  16. L2:119 weakly excites
  17. L5:521 weakly excites
  18. L1:560 weakly excites
  19. L2:777 weakly excites
  20. L7:2126 weakly excites
  21. L1:1226 weakly suppresses
  22. L4:2968 weakly excites
  23. L2:2423 weakly suppresses
  24. L0:1520 weakly excites

Feeds later layers

  1. L20:1177 excites
  2. L16:2743 excites
  3. L22:587 excites
  4. L19:2306 excites
  5. L10:2087 weakly excites
  6. L13:2628 weakly excites
  7. L19:2406 weakly excites
  8. L23:2252 weakly excites
  9. L23:1187 weakly excites
  10. L18:1468 weakly excites
  11. L10:1492 weakly excites
  12. L25:1294 weakly excites
  13. L22:361 weakly excites
  14. L24:9 weakly excites
  15. L17:3053 weakly excites
  16. L21:1494 weakly excites
  17. L20:1581 weakly excites
  18. L20:2406 weakly suppresses
  19. L14:2680 weakly excites
  20. L24:1162 weakly excites
  21. L22:487 weakly excites
  22. L23:471 weakly excites
  23. L23:2909 weakly excites
  24. L10:2761 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 -1.67 and -1.491 values between -1.49 and -1.314 values between -0.95 and -0.782 values between -0.78 and -0.6011 values between -0.60 and -0.42224 values between -0.42 and -0.241160 values between -0.24 and -0.061949 values between -0.06 and 0.12475 values between 0.12 and 0.3050 values between 0.30 and 0.4713 values between 0.47 and 0.652 values between 0.65 and 0.833 values between 0.83 and 1.013 values between 1.01 and 1.191 values between 1.19 and 1.372 values between 1.72 and 1.902 values between 2.62 and 2.801 values between 2.97 and 3.152 values between 3.69 and 3.871 values between 4.94 and 5.121 values between 5.29 and 5.47offnegativepositiveoften

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. precaution
  2. oise
  3. discoveries
  4. ylie
  5. 抉择
  6. speculation
  7. acb
  8. _callbacks
  9. _decision
  10. decisions

Suppresses

  1. time
  2. 化
  3. emia
  4. 化的
  5. -time
  6. piring
  7. 媚
  8. erg
  9. -Time
  10. ten