Neuron Explorer

Qwen3-0.6B

"Since" starting a reason clause

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 lights up on the word "Since" when it starts a sentence that explains why something is true, such as "Since they are nocturnal...". It stays quiet on most other words. Turning it up made the model's text drift into unrelated rambling about math problems, so the steering tells us little about this neuron.

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

    it can also cause laryngitis and bronchitis, and can trigger or worsen attacks if you happen to have asthma or other chronic lung problems. Since you do not stay completely immune to RSV for life once you have had it

  2. #2 peak

    . You can get garbage cans with locking lids. If you keep a dish of food outside for your dog during the day, bring that food inside at night. Since they are nocturnal, they will avoid light, so you can keep yard lights

  3. #3 peak

    no known negative side effects of MRI scans. The only major limitation of MRI machines are that, since they use magnets, patients with permanent metallic fixtures on their bodies, such as artificial

  4. #4 peak

    al neural networks (CNN). In recent years, CNN-based methods have gained popularity<|endoftext|>typedef unsigned int size_t;⏎This type is used to hold size definitions. Since sizes are always positives, this type is usually defined as an unsigned long, but

  5. #5 peak

    minutes, but may take more or less time. The actual duration of treatment depends on the individual case After root canal treatment, the tooth is usually painless. Since he is no longer supplied with blood, he has also become unstable. In most

  6. #6 peak

    ermilion is an opaque and heavy reddish orange pigment similar to the color Scarlet. It is more stable in oils than in other forms and brushes out well. Since the 1920's, it has largely been replaced by Cadmium

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 2010-1000-2010=3000 would be increased by 1000*2010=3000, where 201

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

    Elimination Systems (SPDES) permit program.<|endoftext|>The opossum is an animal most of us have only seen at the zoo. After all, they are a marsupial animal, which means the females have a pouch in which to

  2. #2 peak

    930s.⏎Images: (Left) via the National Media Museum; (Right) via Getty Images<|endoftext|>Because RNA is such an important part of how the cell builds proteins, some scientists speculated that this was its

  3. #3 peak

    variety of foods such as ready meals, tinned goods and pasta sauces. Consider using natural foods, herbs and spices to flavour your food, and give your taste buds time to get used to the new flavours.⏎As we age our digestion can become

  4. #4 peak

    mutually beneficial—after all, the ants could be doing this purely to protect themselves and their colony—but it does

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. L7:884 excites
  2. L7:754 suppresses
  3. L7:658 suppresses
  4. L7:336 suppresses
  5. L7:1183 excites
  6. L7:1425 suppresses
  7. L6:431 excites
  8. L7:1074 excites
  9. L7:1923 excites
  10. L7:2331 suppresses
  11. L7:469 excites
  12. L7:2514 suppresses
  13. L7:1112 excites
  14. L7:1964 excites
  15. L5:2239 excites
  16. L7:530 suppresses
  17. L7:61 suppresses
  18. L7:635 suppresses
  19. L7:508 excites
  20. L7:267 suppresses
  21. L7:989 suppresses
  22. L7:2387 suppresses
  23. L7:501 excites
  24. L6:2530 suppresses

Feeds later layers

  1. L13:2252 strongly excites
  2. L10:721 excites
  3. L11:1574 suppresses
  4. L12:2240 excites
  5. L16:541 suppresses
  6. L13:1431 suppresses
  7. L10:753 excites
  8. L13:350 excites
  9. L12:956 excites
  10. L13:1702 excites
  11. L11:1645 excites
  12. L12:1262 excites
  13. L12:244 excites
  14. L14:2116 excites
  15. L17:1233 excites
  16. L10:534 excites
  17. L15:1312 excites
  18. L12:162 excites
  19. L15:2499 excites
  20. L12:1830 excites
  21. L17:2403 excites
  22. L11:1792 excites
  23. L13:1130 excites
  24. L13:149 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 -2.04 and -1.911 values between -1.79 and -1.672 values between -1.67 and -1.542 values between -1.54 and -1.421 values between -1.42 and -1.301 values between -1.30 and -1.175 values between -1.17 and -1.053 values between -1.05 and -0.9310 values between -0.93 and -0.8110 values between -0.81 and -0.6814 values between -0.68 and -0.5628 values between -0.56 and -0.44104 values between -0.44 and -0.31275 values between -0.31 and -0.19675 values between -0.19 and -0.071836 values between -0.07 and 0.06565 values between 0.06 and 0.18184 values between 0.18 and 0.3074 values between 0.30 and 0.4346 values between 0.43 and 0.5521 values between 0.55 and 0.6716 values between 0.67 and 0.7912 values between 0.79 and 0.924 values between 0.92 and 1.047 values between 1.04 and 1.162 values between 1.16 and 1.292 values between 1.29 and 1.412 values between 1.53 and 1.662 values between 1.90 and 2.022 values between 2.15 and 2.271 values between 2.76 and 2.89offnegativepositiveoften

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. -Americans
  3. {i
  4. ','',
  5. LOUD
  6. due
  7. /Users
  8. stu
  9. 排毒
  10. 得太

Suppresses

  1. nels
  2. pled
  3. narr
  4. ètres
  5. angan
  6. raf
  7. setDescription
  8. verdad
  9. 描述
  10. arence