capital W starting a word
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
Fires on a capital W that starts a word or abbreviation, like Worms, Waste or WBMP. Turning it up pushed the model to repeat the word 'laws', a loose echo of the law-related words this neuron tends to push toward.
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.
-
also full of nutrients. Some of these nutrients are very valuable additions for your garden soil or compost pile. This is important when it comes to worm castings.⏎Worms do not eat dog waste directly. They eat the bacteria that break down
-
title, visit your library in the app or on the desktop website.⏎Chronicles of the imagination Geographical⏎- Walta Evans "School Librarian"<|endoftext|>Wing Tsun or Wing Chun?⏎or Wing Chung or Wing Tjun or V
-
course can also be counted as a 1/4 credit in American history.⏎Read with the Best (RWTB) coordinates with Write with the Best (WWTB) (by the same author, Jill Dixon). Volume 1 is
-
al., 2011). Paper wasps are likely to influence many other species in terrestrial ecosystems because they are voracious predators of invertebrates (Ward & Ramón-Laca, 2014). Four species of
-
A comprehensive approach to improving the lives of waste pickers should assist them and their families to achieve their potential and bring a positive socio-economic change in their lives.⏎Waste pickers
-
of Missions for the Pacific, located in San Francisco (WBMP), and the Woman's Board of Missions for the Pacific Islands, located in Hawaii (WBMPI). The three mainland boards cooperated with the maleled American Board of
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: whether le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le le
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
communities.⏎- Urban Forests: Urban Releaf and Friends of the Urban Forest are two grassroots organizations that beautify our neighborhoods by planting trees.⏎- Aquatic/Wetland Habitats: An inspiring example of inter-jurisdictional collaboration between
-
Historical Society, Scarborough Middle School, and Scarborough Public Library. Exhibits include the marsh, transportation and roads, shipyards and shipwrecks, clamming and lobstering, famous residents, and education.<|endoftext|>
-
, Captain Laudonniere states: “Having reconnoitered the river, I landed to talk […]⏎AN ARTIST WHO ACCOMPANIED THE FRENCH EXPEDITION TO FLORIDA UNDER LAUDONNIERE,
-
(S.W.), the New England Confederation and the Fur Trade Dutch/French (N.Y. &N.F.) is the third supporting idea. *
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
- L22:2646 excites
- L23:2441 excites
- L23:1906 suppresses
- L23:2073 excites
- L20:417 excites
- L22:369 excites
- L21:463 suppresses
- L18:533 suppresses
- L22:2236 excites
- L18:437 weakly excites
- L26:1463 weakly excites
- L16:74 weakly excites
- L21:2605 weakly excites
- L10:941 weakly suppresses
- L8:1982 weakly excites
- L2:1503 weakly excites
- L19:2719 weakly excites
- L8:14 weakly excites
- L26:1564 weakly excites
- L26:2456 weakly excites
- L16:71 weakly excites
- L20:1877 weakly excites
- L20:336 weakly excites
- L9:1588 weakly excites
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
- laws
- law
- laws
- Laws
- legislation
- Law
- -law
- 定律
- 规律
- law
Suppresses
- inal
- inals
- edBy
- INAL
- istics
- acles
- ellaneous
- setDescription
- ativo
- rapper