period closing a web blurb
A neuronOne of thousands of tiny dials inside each layer. It reads the text so far and produces a number, its activation, that nudges what the model writes next. in layerOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. 22 of Qwen3-0.6B. Its strongest snippets are very consistent: they mostly peak on the same token. It is featuredA neuron picked for the full treatment: steering outputs and a written explanation. About 3,000 of the 86,000 neurons are featured., so it has steering outputs.
What this neuron does
This neuron fires on the full stop that ends a closing sentence in web page boilerplate, such as a medical disclaimer, a call to contact someone, or a promotional line, often just before a new unrelated article starts. Turning it up did not make the writing show a clear 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 activationThe number a neuron produces for a token. Large positive means it fired hard; negative means it pushed the other way. was highest. Amber marks positive activation; the outlined tokenA piece of text the model reads at once: often a word, sometimes part of a word or a punctuation mark. is the peak. Hover a token for its strength.
-
not intended to be a substitute for professional medical advice, diagnosis, or treatment. Always seek the advice of qualified health providers with questions you may have regarding medical conditions.<|endoftext|>The Amendment Process and the African American Fight for Representation and Equality⏎Black History
-
for your stargazing needs. Let’s dive in!⏎What is the aperture of a telescope?⏎By definition,
-
bodies, and pigment granules. Understanding these components will not only deepen your knowledge of hair but also help you identify them and comprehend their significance. So, let's dive in!⏎The cuticle is the outermost layer of
-
access to this title, please contact your librarian.⏎To troubleshoot, please check our FAQs, and if you can't find the answer there, please contact us.<|endoftext|>Indigenous Peoples and Climate Change: from victims to change agents through decent work
-
can also visit the Gambling Help Online website here.⏎If you would like to read more about Anastasia’s research, you can read her article in The Conversation here.<|endoftext|>Continued from Part 1...⏎With the advent of this international distribution of
-
return to a shielded base.⏎To continue reading this article, subscribe to receive access to all of newscientist.com, including 20 years of archive content.<|endoftext|>Fly Created To Study How A Normal Cell Turns Cancerous⏎The study
SteeringAdding a push along a neuron while the model writes, to see how its output changes. Turning it up or down.
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.
Here is something I have been thinking about lately: The first. Core. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
, we are printing the element for index i and decrementing the value of i by 1.⏎If you run this program, it will print the below output:⏎5 4 3 2 1⏎We need to use the range
-
heading), the complete sequence is:⏎<H1>This is the Page Heading</H1>A break between paragraphs is den
-
most illuminating being the incandescent electric bulb, the phonograph and the motion-picture projector. Who was he?⏎- Match the following with their inventors:⏎(a
-
(too tired to check who it was that posted it, and I can't see just now anyway :P)⏎void genFactorial(int x, int t)⏎int total = t; // you probably wouldn't even need total now⏎void
Connections
Show on mapWhich neuronsOne of thousands of tiny dials inside each layer. It reads the text so far and produces a number, its activation, that nudges what the model writes next. in earlier layersOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. feed this one, and which later ones it feeds. Amber links excite, blue links suppress; longer bars are stronger.
Fed by earlier layers
- L20:44 strongly excites
- L20:46 strongly excites
- L17:1523 suppresses
- L18:2322 suppresses end-of-text token after a finished passage
- L21:619 suppresses
- L16:397 suppresses
- L17:1088 excites
- L15:774 excites
- L17:1029 excites
- L16:2510 excites
- L16:235 suppresses
- L20:460 excites
- L17:658 suppresses
- L17:2723 excites
- L14:667 suppresses
- L21:100 excites
- L21:660 excites
- L17:68 suppresses
- L15:101 excites
- L21:759 excites
- L16:3064 excites
- L17:135 suppresses
- L14:2826 excites
- L18:926 excites
Feeds later layers
- L24:211 strongly suppresses
- L26:449 strongly suppresses
- L25:397 strongly excites
- L27:891 strongly suppresses
- L25:808 strongly excites
- L26:864 strongly excites
- L24:11 excites
- L24:736 excites
- L26:775 excites
- L27:975 excites
- L27:204 suppresses
- L23:100 suppresses comma or full stop ending a clause
- L26:373 suppresses
- L27:366 excites
- L25:940 suppresses
- L26:828 suppresses
- L26:993 suppresses
- L25:131 suppresses
- L27:569 excites
- L27:116 excites
- L26:230 excites
- L26:34 suppresses
- L27:467 excites
- L27:2391 excites
Activation histogramA chart of how often a neuron takes each activation value.
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.
Logit lensReading which output tokens a neuron pushes toward or away from, straight off its output weights, before the later layers have their say.
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 layersOne step in the stack the text passes through. Qwen3-0.6B has 28 layers, each with 3,072 neurons. instead.
Promotes
- <|endoftext|>
- .
- 。
- .?
- %;">
- ?
- ```
- .]
- ;">
- .)
Suppresses
- ').⏎
- ").⏎
- ',)⏎
- ’.⏎⏎
- '.⏎
- ''}⏎
- ""}⏎
- ".⏎
- `.⏎
- '))⏎