Neuron Explorer

Qwen3-0.6B

the word 'that' introducing a clause

A neuron in layer 10 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 reacts to the word 'that' when it links a phrase to a following clause, as in 'so that' or 'those that'. It goes strongly negative on 'that' used as a pointing word after a verb, like 'tell that story' or 'check out that lesson'. Turning it up did not produce a clear theme in the model's writing.

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

    Sustainability 2013, 5(7), 3006-3008; doi:10.3390/su5073006⏎I am honored to introduce this

  2. #2 peak

    possess, so that you, your children and their children after them may fear the Lord your God as long as you live by keeping all his decrees and commands that I give you, ad so that you may enjoy log life” (Deuter

  3. #3 peak

    Team⏎Our team consists of highly trained and skilled professionals that are in these sorts of children problems. We will give you the best possible advice there is, so that you could help your child as fast as you can. We are here for you

  4. #4 peak

    may be nothing about (var1 x)⏎> that changed, and there may be nothing about var11 . var1 $ x

  5. #5 peak

    We hold these Truths to be self-evident, that all Men are created equal, that they are endowed by their Creator with certain un

  6. #6 peak

    and its hardiness allows the plant to out-grow and out-compete native plants. This, in turn, affects the organisms—from insects to birds and mammals – that have evolved in tandem with Utah’s plants. Myrtle spurge is also incredibly

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 first one, the first one, the first one, the first one, the first one, the first one, the first one, the first one, the first one, 1, 1, 1, 1,

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

    in this film from start to finish. It was the first of its kind to have the depth of character, careful attention to story, original music that helped tell that story and superb artistry.”⏎Premiered at Hollywood’s Carthay Circle

  2. #2 peak

    video in their curriculum. Reading materials and lesson ideas are available. I recently used a lesson that my colleague created directly from the provided materials. You can check out that lesson here. The lesson included some background

  3. #3 peak

    smoking program "really works" and has "a 21% quit rate." These statements are contrary to findings from two recent independent studies. But try telling that to U.S. health officials.⏎According to the Centers for Disease Control (CDC

  4. #4 peak

    in policy that could make the Moon and its surrounding space a more important part of the equation.⏎And our source tells us that part of that plan is deploying RESOLVE on the lunar surface in 201

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. L4:468 excites "that" after "than" or "like"
  2. L8:14 excites
  3. L9:1752 suppresses
  4. L6:1080 excites "that" as a pointing word
  5. L4:671 excites 'that' starting a new clause
  6. L4:270 weakly suppresses
  7. L6:925 weakly excites
  8. L1:2290 weakly suppresses
  9. L9:1155 weakly suppresses
  10. L4:1762 weakly suppresses "this" in time and context phrases
  11. L9:284 weakly excites
  12. L2:181 weakly suppresses
  13. L7:1253 weakly suppresses
  14. L8:3050 weakly suppresses
  15. L0:237 weakly suppresses
  16. L9:1088 weakly suppresses
  17. L3:1196 weakly excites
  18. L0:918 weakly suppresses
  19. L7:1252 weakly suppresses
  20. L4:365 weakly excites
  21. L1:2334 weakly suppresses
  22. L5:2680 weakly excites
  23. L0:1086 weakly suppresses
  24. L7:181 weakly excites

Feeds later layers

  1. L12:521 suppresses
  2. L17:221 suppresses
  3. L15:470 suppresses the word 'this' after a preposition
  4. L11:199 suppresses
  5. L17:2011 suppresses
  6. L13:1007 suppresses
  7. L14:1807 suppresses
  8. L19:2547 suppresses
  9. L16:2861 suppresses
  10. L15:1957 suppresses
  11. L17:1821 suppresses
  12. L12:2701 suppresses
  13. L17:1816 suppresses
  14. L15:1323 suppresses
  15. L16:2466 suppresses
  16. L15:1187 suppresses
  17. L14:2263 suppresses the word 'same' as a referring word
  18. L14:2444 weakly suppresses
  19. L18:1228 weakly suppresses
  20. L14:2220 weakly suppresses "this" pointing back to a thing
  21. L19:494 weakly suppresses
  22. L12:849 weakly suppresses
  23. L16:815 weakly suppresses 'the' before a noun mid-sentence
  24. L20:1175 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 -11.46 and -11.161 values between -10.54 and -10.241 values between -8.71 and -8.401 values between -8.10 and -7.791 values between -7.18 and -6.871 values between -6.57 and -6.261 values between -5.95 and -5.651 values between -4.42 and -4.121 values between -3.81 and -3.511 values between -3.51 and -3.201 values between -3.20 and -2.891 values between -1.98 and -1.673 values between -1.67 and -1.372 values between -1.37 and -1.066 values between -1.06 and -0.758 values between -0.75 and -0.4532 values between -0.45 and -0.142249 values between -0.14 and 0.171440 values between 0.17 and 0.47156 values between 0.47 and 0.78offnegativepositiveoften

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. cứng
  2. ed
  3. care
  4. Care
  5. "/"⏎
  6. intervene
  7. ue
  8. unders
  9. 记者采访
  10. conform

Suppresses

  1. !"; ⏎
  2. ATTLE
  3. TINGS
  4. .";
  5. ."; ⏎
  6. ...";⏎
  7. hg
  8. Aws
  9. WHERE
  10. monds