muscle and body movement 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. 9 of Qwen3-0.6B. Its strongest snippets are mixed: they peak on different tokens. 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 about muscles and the body moving, such as 'flex', 'muscle', 'hip flexors' and 'stretching', though it also fires on some unrelated words like 'picky'. Turning it up made the model's writing wander off into random topics like coffee and weather, so the steering shows no clear theme.
Written by Claude from this neuron's activations and steering outputs. Confidence: low. 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.
-
inal (belly) muscles support the spine from the front and sides of the body.⏎- Iliopsoas muscles support the spine and flex the hips.⏎- Erector spinae muscles support the back of the trunk
-
Just like any other sport, training in swimming is important. One of the quality training is a good nutrition. A learner also has to develop and strengthen your mental muscle son regular basis. Look for help in learning new strokes technique. In case the
-
be supplemented with a humidifier to make the air moister.⏎You might actually be the one responsible for frequent nose bleeds. Bad habits such as nose-picking can increase the chances of nose bleeds. Because the membrane and blood vessels inside
-
on purees and become picky eaters, especially if they get used to extremely smooth purees. However,
-
body.⏎Move More Muscles⏎Most of the muscles you use in running are in your lower body: your quads, hamstrings, glutes, hip flexors and calves
-
seeing among clients for less intensive fitness routines that are accessible and mesh with an on-the-go lifestyle.⏎Why gardening is exercise⏎Gardening engages all the major muscle groups, such as the arms, legs, shoulders, back and abdomen, Mandal
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 thinking about it because I have been thinking about it because I have been thinking about it. This is because I have been thinking about it because I have been thinking about it. This is because I have been thinking about it because
Most negative on
Where the activation went furthest below zero. Blue marks negative activation. Many neurons mean something different in each direction.
-
- Filed Under⏎J ennifer Skaja has seen the signs of small miracles, as children with disabilities gain their voice through American Sign Language. Individuals with hearing impairment have⏎long used sign language to communicate, but studies
-
seven octopuses tested successfully learned the task and used the strategy more often once it was mastered.⏎It's only one example of clever tricks used by cephalopods. Check out our full-length feature to find out more about these animals'
-
language.⏎The Unified Modeling Language is an open method used to specify, visualise,⏎construct and document the artifacts
-
sequential'] fruit = ['raspberry', 'pomegranite', 'bananana', 'grape', 'loganberry'] print "You are an" , random.choice (verb) , random.choice (fruit)⏎Any Tutorial Videos
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
- L7:212 excites
- L8:1254 suppresses
- L8:1926 weakly excites
- L8:1065 weakly suppresses
- L4:578 weakly suppresses
- L6:2770 weakly suppresses
- L4:680 weakly excites
- L8:2198 weakly excites
- L7:529 weakly suppresses
- L6:1498 weakly suppresses
- L6:799 weakly excites
- L8:1250 weakly suppresses
- L7:302 weakly suppresses
- L4:1287 weakly excites 'out' in 'turns out'
- L7:659 weakly excites
- L6:2490 weakly suppresses
- L6:2135 weakly excites
- L8:1798 weakly suppresses
- L7:2471 weakly excites
- L6:2334 weakly suppresses
- L3:1952 weakly excites
- L1:1923 weakly suppresses
- L0:827 weakly excites
- L6:2013 weakly suppresses
Feeds later layers
- L11:952 excites
- L10:1463 excites
- L11:1165 excites
- L11:2919 suppresses
- L17:1531 excites
- L10:1864 excites
- L10:2883 excites
- L10:1884 weakly suppresses
- L14:1034 weakly suppresses
- L24:2511 weakly suppresses
- L11:2406 weakly excites
- L20:2775 weakly suppresses
- L12:1249 weakly excites
- L16:2108 weakly suppresses
- L14:458 weakly excites
- L12:1226 weakly excites
- L11:1061 weakly suppresses
- L10:877 weakly excites
- L24:974 weakly excites
- L20:2492 weakly suppresses
- L15:2026 weakly suppresses
- L10:2637 weakly excites evaluative words: good, bad, positive
- L19:2952 weakly excites
- L14:2205 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
- agara
- grade
- kost
- ffield
- grade
- temperament
- ieber
- każdego
- atorial
- easing
Suppresses
- WithPath
- 的位置
- positions
- enqu
- lez
- mitt
- 位置
- locations
- ataires
- `_