sentence-starting 'This' at text start
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 reacts to the word 'This' when it begins a sentence or fragment, especially at the very start of a snippet. Turning it up did not produce a clear shift in the writing.
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.
-
. This is⏎Now an -simplex in the comma category , which is an element
-
This is a question that many consumers have when it comes to choosing plastic bottles for their
-
This is a sometimes hypothetical pre-shoot narrative document describing what happens in the film from
-
This will keep them in their new position until bones grow around the teeth to hold them
-
This can help develop a mighty plan that can make farms healthy enough to not require the
-
This is quite simple. First, compile (but don't link) each of your
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 been trying to understand the problem, and I have been thinking about it for a while. I have been thinking about it, and I have been thinking about it for a while. I have been thinking about it, and I have
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
of a meal at a restaurant? Usually you get your leftovers in a foil container with a cardboard lid⏎- styrofoam containers – again you might get this from a restaurant for leftovers or as a takeout container from a fast food place
-
Zuck, pp. 20). F. Biblical Theology "Biblical theology is the study of divine revelation as File Size: KB.⏎Get this from a library! Principles of biblical interpretation: (sacred hermene
-
get this book, download the free Human Anatomy and Physiology Pdf right now from our website.
-
(e.g., epiclesis, sanctus bells, the elevation, Mass of the Presanctified, etc.). Why is it important that we get this right? Because on our view of “real presence” in the sacrament hinges
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
- L3:27 weakly suppresses
- L11:924 weakly excites
- L18:2752 weakly suppresses
- L9:1252 weakly suppresses
- L23:959 weakly suppresses
- L9:1624 weakly excites
- L7:163 weakly suppresses
- L3:1411 weakly excites
- L26:757 weakly suppresses
- L27:1536 weakly suppresses
- L22:1911 weakly suppresses
- L22:2625 weakly excites
- L4:579 weakly excites
- L16:2889 weakly suppresses
- L23:867 weakly suppresses
- L12:2906 weakly suppresses
- L16:1540 weakly excites
- L16:1993 weakly excites
- L3:1553 weakly suppresses
- L10:413 weakly suppresses
- L13:2021 weakly suppresses
- L7:954 weakly excites
- L18:2786 weakly excites
- L25:1849 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
- denom
- then
- oks
- puis
- dent
- 玑
- 斯基
- hence
- 湜
- aku
Suppresses
- mentioned
- ши
- otify
- avigation
- 意见
- anning
- шин
- 该剧
- .addAction
- 的意见