© 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. 57459
    Prev
    Next
    INDEX
    Explanations

    making music

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

    The neuron fires on formal definitional or categorical constructions—phrases that state “X is a Y” or “list of Y” by picking up linking words like “is,” “of,” etc., in topic‐definition contexts.

    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
    揮
    -0.81
     бук
    -0.81
     choreographer
    -0.77
    nicas
    -0.75
     ballerina
    -0.74
    Dancer
    -0.72
    setImageResource
    -0.71
    tać
    -0.71
    paddingVertical
    -0.71
    挥
    -0.70
    POSITIVE LOGITS
     music
    5.44
    music
    3.53
    音楽
    3.36
    音乐
    2.91
    Music
    2.91
     Music
    2.84
     MUSIC
    2.77
     музи
    2.73
    MUSIC
    2.67
     música
    2.61
    Activations Density 0.084%

    No Known Activations