vague "certain" and "such" before a noun
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. 8 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 words like "certain", "such", "various" and "specific" when they sit in front of a noun and mean some particular, unnamed one. Turning it up did not produce a clear change in topic or style.
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.
-
Every administrator and politician speaks of “party politics” in villages, and even villagers are often heard saying, “There is too much ‘party’ in such and such a village”. The coming of elections gives fresh opportunities for the crystallization of parties
-
14 in Senate primaries because of provisions in those states allowing for a second-round runoff election between the top two candidates if no candidate in the first round obtained a certain level of support.⏎Plurality Wins in Congressional Primaries, 19
-
there are perhaps more justifiable answers and less justifiable answers. Something like “Which of the following motivations best explains so-and-so’s actions in such-and-such novel?” Instead of posing this question, hoping that all your students take a moment
-
form ביון.⏎So what do לבייןand ביון mean, specifically? In mathematics there is a concept called interpolation, which involves estimating the location of various data points within a range – a precise, sophisticated game of connect the dots.
-
or bottom to find the answer. Students continue solving problems and finding answers as they move through the Fact Land. For some extra fun, students collect special tokens in certain boxes and tally them up at the bottom. Once students have made their way through
-
, subject to certain terms and conditions, the SEAI offers grants for both solar PV and EV charger
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: I have a 33-year-old, 20-year-old, 25-year-old, and 35-year-old. I am trying to get a group of 3 people together. So, how do I plan a
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
. In some cases, this step and hydrolysis, the next step, are combined into a single process.⏎In hydrolysis, the pretreated material, with the remaining solid carbohydrate fraction, primarily cellul
-
development of several species of seaweed using Lindenmayer systems (J.D. Corbit, D.J. Garbary). Generating fractals from Voronoi diagrams (K.W. Shirriff). Circles with kiss: a note
-
plane belongs to an algebraic curve if its coordinates satisfy a given polynomial equation. Basic questions involve the study of
-
ivia miniata: growing from seed can be difficult and it will be years before the plant will flower. An easier way to propagate Clivia miniata is by division, which can be done at any time of year. If you notice offsets (
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
- L4:731 excites
- L7:832 weakly excites
- L7:52 weakly excites
- L1:1912 weakly excites
- L2:2512 weakly excites
- L6:186 weakly suppresses
- L3:1050 weakly excites
- L7:476 weakly suppresses
- L2:815 weakly suppresses
- L6:3045 weakly excites
- L7:1805 weakly excites
- L6:1813 weakly excites
- L4:1236 weakly suppresses
- L4:2251 weakly excites
- L7:2515 weakly suppresses
- L3:214 weakly suppresses
- L5:1972 weakly excites
- L5:283 weakly excites
- L5:172 weakly excites
- L5:32 weakly excites
- L6:1519 weakly suppresses
- L6:2654 weakly suppresses
- L4:892 weakly excites unclear: weak math-topic words
- L5:218 weakly excites
Feeds later layers
- L10:2231 suppresses
- L10:1966 weakly suppresses
- L17:1176 weakly excites
- L13:2026 weakly excites
- L12:777 weakly excites
- L21:2215 weakly suppresses
- L10:1031 weakly excites
- L16:1351 weakly excites
- L22:1728 weakly excites unclear: code assignments and parentheses
- L16:1128 weakly excites
- L10:455 weakly excites
- L23:2298 weakly suppresses
- L13:948 weakly suppresses
- L15:1640 weakly suppresses
- L24:1656 weakly suppresses
- L12:1750 weakly excites
- L13:1982 weakly excites
- L14:1373 weakly suppresses
- L14:880 weakly excites
- L11:1846 weakly suppresses
- L16:1139 weakly suppresses 'and' in lists of places or names
- L21:2634 weakly suppresses
- L24:2492 weakly suppresses
- L13:2081 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
- _JS
- outube
- QualifiedName
- ắp
- '../../../../
- _ABI
- بوك
- ../../
- finger
- pig
Suppresses
- rale
- elve
- seule
- TEE
- 不动
- Sixth
- otte
- 二线
- 客观
- otron