© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 92951
    Prev
    Next
    INDEX
    Explanations

    movement and coordination

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on ordered, step-by-step movement or dance/exercise instructions (e.g. numbered “step right,” “step left behind,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     OPERATION
    -0.92
     SPEECH
    -0.88
    setts
    -0.86
    operation
    -0.86
    Operation
    -0.84
     bandoulière
    -0.84
    了他的
    -0.82
     Matériau
    -0.81
     konsumen
    -0.81
     ハンドル
    -0.81
    POSITIVE LOGITS
     movements
    1.36
     movement
    1.19
     synchronization
    1.16
     facial
    1.15
     precision
    1.12
     coordination
    1.10
     synchronized
    1.08
     coordinated
    1.04
     executed
    1.03
     interpretation
    1.01
    Activations Density 0.013%

    No Known Activations