'therm' in science words
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 fires on the word piece 'therm' as in thermometer, thermodynamic or thermophilic. Turning it up gave a mix of topics, with a lot of 'art' and 'arts' words, which is only loosely related.
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.
-
the machine configuration file for assembly binding, see How the Runtime Locates Assemblies.⏎An application configuration file contains settings that are specific<|endoftext|>Mercury is used in thermometers, fluorescent light bulbs, and some electrical switches.⏎"Mercury" is
-
out more.⏎1. Maintain the Fridge Temperature⏎A recommended fridge temperature of 5 degrees C is advised. There are thermometers available at reasonable price which can help you determine the temperature. A report suggests
-
six breast pieces, each one might cook differently based on size. Don’t forget to clean your thermometer probe between readings to avoid cross-contamination.⏎Food or meat thermometers are fairly inexpensive, and will save you the worry of “is it done
-
; the indirect techniques only calculate it from other indirect indicators.⏎Image Credit: Soonthorn Wongsaita / Shutterstock⏎Many devices for LOD measurement use the thermogravimetric principle. This uses the total loss of weight incurred by the
-
use of epoxy-based resins and may be classified as the thermosetting epoxide polymer. Exteriors that receive this coating are resistant to
-
activation energy to break covalent bonds and⏎denature/destroy molecular structure. Proteins are particularly sensitive.⏎There are exceptions, as in the case of thermophillic bacteria. There are⏎cases or heat shock-proteins or "
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: being in this lесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесесес
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
(28.2 cm.), W: 8 7/8 in. (22.5 cm.)|⏎|Location||Alte Pinakothek, Bayerische Staatsgemä
-
Museum of Alberta, Musée national des arts et traditions populaires, Département de l'organisation des systèmes d'information,<|endoftext|>A Very Short Introduction⏎Liberalism is one
-
the historical development of the West.⏎The first link is ideological. Nazism emerged in the socio-political constellation of German nationalism, which was crisscrossed by currents well represented
-
language Kasak “nomad”, “free man”, the term is first mentioned in a Ruthenian document dating from 1395.⏎In socio-political terms, from the 15th century to the present day, the
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
- L24:105 weakly suppresses
- L21:463 weakly excites
- L4:1572 weakly excites
- L20:959 weakly excites
- L0:917 weakly suppresses
- L6:2695 weakly excites
- L8:1699 weakly suppresses
- L19:1351 weakly suppresses
- L22:164 weakly suppresses
- L10:2172 weakly suppresses
- L3:2050 weakly excites
- L24:2872 weakly suppresses
- L13:481 weakly suppresses 'because' introducing a reason
- L15:1237 weakly suppresses
- L5:1301 weakly excites
- L20:390 weakly suppresses
- L7:2308 weakly suppresses comma inside stacked adjectives
- L23:1427 weakly suppresses
- L14:1304 weakly suppresses
- L7:49 weakly suppresses
- L0:801 weakly suppresses 'long' in 'how long' questions
- L8:2688 weakly excites
- L8:1487 weakly excites
- L7:1221 weakly suppresses
Feeds later layers
- L27:787 suppresses
- L26:984 weakly suppresses
- L27:1599 weakly suppresses
- L27:1233 weakly suppresses
- L27:2675 weakly excites
- L27:178 weakly excites
- L26:2948 weakly suppresses
- L27:1012 weakly suppresses
- L26:2594 weakly suppresses
- L27:2179 weakly excites
- L27:2423 weakly excites
- L27:618 weakly excites
- L27:572 weakly excites
- L27:1119 weakly excites
- L27:1055 weakly suppresses
- L26:2197 weakly suppresses
- L27:494 weakly excites
- L27:2278 weakly suppresses
- L27:2305 weakly excites
- L27:2044 weakly excites
- L26:1836 weakly excites
- L26:1626 weakly excites the word 'I' as subject
- L27:1822 weakly excites
- L26:2186 weakly suppresses
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
- art
- artin
- bow
- artisan
- indicator
- art
- иж
- Arn
- asted
- vation
Suppresses
- emat
- èmes
- erah
- ème
- etu
- eta
- eti
- eton
- eto
- ém