digit 1 inside dates and times
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. 25 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 responds to the number 1 when it appears inside a date or a clock time, such as "2012/10/05" or "12:30". When it was turned down, the model got stuck writing long strings of the digit 3.
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.
-
impact on the community in the Hollywood Gazette’s November 2019 issue — https://hollywoodgazette.com/2019/11/03/hollywood-history-the-heroes-and-the-history-of-liberia
-
|⏎|rs_class_altemira.tif||30.7 MB||1985-01-01 - 2001-12-31||2.50S To 4.00
-
as lasting? Some believe not.⏎Taken from…. The Future of Education: Tablets vs. Textbooks http://mashable.com/2012/10/05/tablets-vs-textbooks/⏎They also say…. ‘ Students
-
413967-1-768x146-1-300x57.png admin2021-10-20 23:34:59202
-
and a show. After you eat peanut butter & jelly, watch Stone Soup, performed by the Lost Caravan. Lunch is at noon; show starts at 12:30. Chesapeake Music Hall, Off Rt.
-
to know before planting⏎- Article by: Deb Brown⏎- Special to the Star Tribune⏎- January 27, 2009 - 12:32 PM⏎Q It seems like birch trees grow well around
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: 33333333333333333333333333333333333333333333333
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
in hospital equipment such medical equipment such as centrifuges and automated pipetting systems.⏎One of the most common applications of lead screws is for positioning the print head in 3D printers. They often use 8mm lead screws as demonstrated in the
-
laser scanning.⏎In the overall Scan-to-BIM process, a laser scanner is used to acquire a precise 3D scan of the project’s real-world conditions.⏎The scan data is then
-
tumour right when it starts and know its next move, maybe we can also stop it from growing at all.⏎EMBL scientists have created a new, realistic 3D testbed that could help them achieve that goal by studying cancer cells as
-
the area of the cavity. At worst, small holes or channels are completely blocked.⏎Printing with light: DLP, SLA, LCD⏎Modern resin-based 3D printing works on the principle of photopolymerization. In other words
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
- L22:1641 excites
- L22:2983 excites
- L22:934 suppresses
- L22:747 excites
- L20:815 suppresses
- L10:825 weakly suppresses 'B' in vitamin B names
- L22:1474 weakly excites
- L7:2676 weakly suppresses
- L24:1968 weakly excites
- L11:1255 weakly excites
- L0:1549 weakly suppresses
- L6:2795 weakly excites
- L5:2103 weakly excites
- L24:2412 weakly excites
- L14:1991 weakly excites
- L23:231 weakly excites
- L4:423 weakly suppresses
- L17:1228 weakly suppresses
- L23:1695 weakly excites
- L23:1166 weakly excites
- L3:2123 weakly suppresses
- L15:2377 weakly excites
- L1:2989 weakly excites
- L18:1517 weakly suppresses
Feeds later layers
- L26:463 strongly suppresses digits inside years and numbers
- L26:102 strongly excites
- L27:364 strongly suppresses
- L27:555 strongly suppresses
- L26:724 strongly suppresses
- L26:983 strongly suppresses
- L27:678 suppresses
- L26:667 suppresses
- L27:287 suppresses
- L27:113 excites
- L27:132 excites
- L27:2172 excites
- L27:992 suppresses
- L27:448 excites
- L26:476 suppresses
- L27:696 suppresses
- L27:163 excites
- L26:2953 excites
- L27:428 suppresses
- L26:611 suppresses
- L27:1093 suppresses
- L26:1083 suppresses
- L26:270 suppresses
- L26:1127 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
- .centerY
- zept
- elerik
- ubb
- veyor
- ypi
- ้อง
- alis
- Intl
- $($
Suppresses
- 3
- three
- three
- trois
- Three
- três
- 3
- THREE
- -three
- 三个