"of" in fixed phrases like plain of
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. 0 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 lights up on the word "of" in set phrases such as "Plain of Catania," "clear of fish," or "steer clear of." It is quiet on most other words. Turning it up made the model's writing wander into unrelated topics, so the link to the text is not clear from the steering.
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.
-
amounts of resources toward flood prevention and post-flood restoration. Consequently, no matter how hard the local population work, they remain the poorest of the poor.⏎The Tam Nông Hospital is an old hospital built in the early
-
almost constant state of activity. The fertile volcanic soils support extensive agriculture, with vineyards and orchards spread across the lower slopes of the mountain and the broad Plain of Catania to the south. Due to its history of recent activity and nearby population
-
island.⏎“Our livelihood depends on fishing, but in the past large ships would come in and sweep the whole ocean clear of fish. There was nothing left for us to eat,” Velasquez said.
-
ice made up of frozen nitrogen mixed with carbon monoxide and methane.⏎It covers an area four kilometers deep and a thousand kilometers wide in the Sputnik Plain of Pluto.⏎To understand the origins of this
-
the earthly to the spiritual world. The ritual decomposition is later followed by cremation and secondary burial.⏎On 6 July 2019, the Plain of
-
a farm located in the fields near the city, and there she became pregnant; the King decided to call the place "Pla del Bon Repos" ("Plain of good sleep"), a name that has been left to posterity and today is a
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 place of a group of people in a city in a month of the number of days and the number of days is the number of days in a month, the number of days, the number of days, the number of days, the
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
angles formed by the cords on the disk—the sum of the forces and the equilibrium conditions. Provenance: Lorraine collections.<|endoftext|>
-
standard contribution is used when the marginal cost is using.⏎Sales Volume Variance Formula⏎In general, the formula of Sales Volume Variance is⏎Sales Volume Variance = (Actual units sold –
-
966.⏎Yet, though many of the panellists called it a “perfect” novel, “Wide Sargasso Sea” was not the sum of Rhys’ work. Her career spanned decades, from her early novels set
-
during that period than were youth in the city's general population.24 Youth in contact with a mental health system's services, however, are not the sum of a community's youth with mental health needs because many receive no services. The results
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
None: this is the first layer.
Feeds later layers
- L20:1963 weakly suppresses
- L24:3031 weakly excites
- L3:446 weakly suppresses
- L24:49 weakly excites
- L20:1390 weakly excites
- L4:2362 weakly excites
- L8:1069 weakly excites
- L13:1601 weakly suppresses
- L15:3033 weakly suppresses
- L13:2074 weakly suppresses
- L9:2786 weakly excites
- L16:2948 weakly excites
- L22:551 weakly suppresses
- L7:1515 weakly excites
- L15:1122 weakly excites
- L2:1547 weakly excites
- L9:45 weakly suppresses
- L4:2426 weakly suppresses
- L14:786 weakly excites
- L19:1270 weakly excites
- L9:2456 weakly suppresses
- L21:2141 weakly suppresses
- L13:2998 weakly suppresses
- L10:1996 weakly 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
- '),⏎⏎
- )--
- askell
- "),⏎⏎
- -Ta
- reetings
- """),⏎
- ),⏎⏎
- )⏎⏎
- ause
Suppresses
- endant
- قلب
- ctor
- ayar
- lạc
- atively
- ilog
- 亮丽
- 走出来
- вет