Neuron Explorer

Qwen3-0.6B

"this" in time and context phrases

A neuron in layer 4 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 "this" when it starts a phrase about the current moment or situation, like "this week", "this autumn" or "this case". Turning it up didn't produce a clear theme in the written text, and the model drifted into unrelated topics.

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

    The Columbia Daily Tribune the drought has been hard on the prized specimen.⏎"The drought of '80 was different," he said. "That was hotter, but this one is longer."⏎The U.S. Drought Monitor this week said more than

  2. #2 peak

    a half century ago.⏎“Just watching that moment was pretty dog-gone exciting,” he said. “It was just like, ‘wow,’ can you believe this is actually happening?”⏎It did finally happen.⏎After years of space exploration,

  3. #3 peak

    sparking new actions that will help benefit individuals with autism, and their families, in the years to come.⏎April is World Autism Awareness Month, and three years ago this week, OHSU and more than 20 other sites across the U

  4. #4 peak

    seems as if every time Israel defends itself, somebody points an accusing finger and yells “Genocide!” Raphael Lemkin, who coined that term seventy years ago this autumn, would have been appalled by such abuse of his life’s work.⏎

  5. #5 peak

    in Seattle, who heads the UNOS living donor committee.⏎"We really do need to take a look at things again in real time to say, OK, in this current climate what are our risks?"⏎More than 90,000

  6. #6 peak

    girls have a difficult time getting an education. The sons of these workers — very much like the sons of their richer counterparts — are treated like kings, or in this case pauper-kings. They get the lion’s share of whatever little food

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 been thinking about... I have been thinking about... I have been thinking about... I have been thinking about... I have been thinking about... I have been thinking about... I have been thinking about... I have been thinking about...

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

    anam. In that the poet relates the life story of one Punithavathiar who abandoned the

  2. #2 peak

    steady progress toward the human exploration of the solar system and beyond. In that expectation, we were cruelly disappointed. Partisan politics, a collective failure of

  3. #3 peak

    joined the Loyola community in 2013 and currently serves as the Senior Associate Director in the Academic Advising and Support Center (AASC). In that role, she advises undergraduate students and works closely with the transfer student population, assisting

  4. #4 peak

    September 10, then received the name of Hurricane Humberto at the first report of the National Hurricane Center (NHC) the next day. At that time the hurricane was located about 310 mi (500 km)

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. L2:1590 weakly excites
  2. L3:108 weakly suppresses
  3. L3:2071 weakly excites
  4. L2:383 weakly excites
  5. L1:335 weakly excites
  6. L2:2476 weakly excites
  7. L0:2836 weakly excites
  8. L3:1068 weakly excites "that" pointing back to a noun
  9. L2:2677 weakly suppresses
  10. L2:146 weakly excites unclear: 'lesson' and 'lique' word pieces
  11. L0:2228 weakly excites
  12. L1:886 weakly excites
  13. L2:619 weakly excites
  14. L3:301 weakly suppresses
  15. L1:1339 weakly suppresses
  16. L2:1012 weakly excites
  17. L3:2641 weakly excites
  18. L0:1889 weakly suppresses
  19. L2:3048 weakly excites
  20. L1:2429 weakly suppresses
  21. L3:512 weakly suppresses
  22. L2:977 weakly excites
  23. L2:2511 weakly suppresses
  24. L3:745 weakly excites

Feeds later layers

  1. L5:98 strongly excites
  2. L10:2178 excites 'this' after a verb or preposition
  3. L9:82 excites
  4. L6:793 excites 'this' after 'earlier' or 'in'
  5. L5:89 suppresses
  6. L10:1104 suppresses
  7. L6:41 weakly suppresses
  8. L17:2464 weakly suppresses
  9. L10:965 weakly suppresses the word 'that' introducing a clause
  10. L19:2524 weakly suppresses
  11. L10:864 weakly excites
  12. L7:2602 weakly excites
  13. L17:2615 weakly excites
  14. L19:2916 weakly suppresses
  15. L12:1379 weakly suppresses
  16. L5:2733 weakly excites
  17. L13:1221 weakly suppresses comma or verb after a researcher or study subject
  18. L6:2404 weakly suppresses
  19. L18:2604 weakly suppresses
  20. L23:2687 weakly excites
  21. L21:1563 weakly suppresses
  22. L14:228 weakly suppresses
  23. L6:1445 weakly suppresses
  24. L18:1785 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 -3.30 and -3.131 values between -2.30 and -2.131 values between -1.80 and -1.631 values between -1.47 and -1.301 values between -1.30 and -1.141 values between -1.14 and -0.971 values between -0.97 and -0.802 values between -0.80 and -0.6411 values between -0.64 and -0.4749 values between -0.47 and -0.30430 values between -0.30 and -0.141924 values between -0.14 and 0.031249 values between 0.03 and 0.20196 values between 0.20 and 0.3623 values between 0.36 and 0.533 values between 0.53 and 0.694 values between 0.69 and 0.861 values between 0.86 and 1.031 values between 1.03 and 1.194 values between 1.36 and 1.531 values between 1.69 and 1.861 values between 2.02 and 2.191 values between 2.69 and 2.861 values between 3.19 and 3.35offnegativepositiveoften

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. cat
  2. @Web
  3. ORT
  4. 司
  5. 系列
  6. gif
  7. cat
  8. onet
  9. arti
  10. Figure

Suppresses

  1. 除外
  2. ..."); ⏎
  3. 瑕
  4. .'); ⏎
  5. balances
  6. ...');⏎
  7. 确
  8. '] ⏎
  9. 玒
  10. .existsSync