unclear: opening quotes and closing brackets
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. 27 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 is hard to pin down. It fires most on quotation marks that open a title or quote right after a line break, and on a few closing brackets. When it was turned down, the model repeated closing brackets over and over.
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.
-
/MSE) = 10 log10(MAX/(MSE)^(1/2))^2 = 20 log10(MAX/(MSE)^(1/2)). Therefore, PSNR = <|endoftext|>In my current position
-
(of at most 32K), and a number of libraries (each of at most 32K).⏎GLib (pronounced ``gee-lib'') shared libraries are a way to implement shared libraries on the Pilot that differs from
-
beverage products.⏎Let’s raise a fork to the food we love and the people who produce it!<|endoftext|>Deor's Lament: a Modern English Translation⏎"Deor's Lament" is one of the truly great poems of English antiqu
-
letto Ticino in Orta, or the preachers⏎like Giovan Battista Aguggiari in Vare<|endoftext|>Photograph by Vanessa Allen⏎"Arteria" by Handiedan (Netherlands) highlights beluga whales
-
Coast — Did the Coast Change?" and calling on citizen volunteers to study aerial photographs taken before and after extreme storms to determine if or how the landscape changed.⏎"Computers cannot yet automatically identify damages and geomorphic changes to the coast from the
-
eruptions and strips the ocean of its oxygen, according to a new study.⏎And no one knows what's causing the "intermittently catastrophic" changes.⏎"Many geologists believe that geological events are random over time," said geologist Michael
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:))))))))))))))))))))))))))))))))))))))))))))))))
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
ebook on an iPhone, iPad or Android mobile device you will need the Adobe Digital Editions app, or BlueFire Reader or Txtr app. These are free, too. For more details see this article.⏎|Size: ||359
-
involves multiple systemic symptoms or appears to be rapidly worsening should be immediately referred for medical evaluation. Possible DISRs include Stevens-Johnson syndrome and toxic epidermal necrolysis.⏎4. Identify the drug, classify the type of reaction,
-
to correct its authors, title, abstract, bibliographic or download information, contact: (Michael McNulty)⏎If references are entirely missing, you can add them using this form.<|endoftext|>Farmers fight drought with move to digital age⏎21
-
keep eating fiber to maximize its many health benefits.⏎Dr. Frost is a board-certified colon and rectal surgeon and a fellow of the American Society of Colon and Rectal Surgeons. His practice offers colonoscopy, colon operations and evaluation
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
- L25:250 suppresses
- L26:712 excites
- L26:381 suppresses
- L25:36 suppresses
- L26:505 excites unclear: opening quote marks after a line break
- L16:2 excites
- L26:1602 suppresses
- L24:260 suppresses
- L23:819 excites
- L26:345 excites
- L21:2490 excites comma inside dates
- L20:1904 excites
- L19:501 excites
- L23:777 excites
- L19:223 suppresses
- L19:1875 excites
- L23:184 suppresses
- L20:79 suppresses
- L24:1539 excites
- L25:112 excites
- L18:2126 suppresses
- L19:1777 suppresses
- L21:147 suppresses
- L16:78 suppresses
Feeds later layers
None: this is the last layer.
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
- "" ⏎
- 什么呢
- '' ⏎
- ''⏎
- (){
- loquent
- toi
- же
- емся
- الة
Suppresses
- )
- ),
- )⏎
- ).
- ),⏎
- )]
- );⏎
- ).⏎⏎
- )⏎⏎
- ):⏎⏎