© 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. 22-GEMMASCOPE-RES-131K
    7. 15337
    Prev
    Next
    INDEX
    Explanations

    components and sections

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

    This neuron fires on nouns that denote subdivisions or sections—words like “component,” “sector,” “portion,” “part,” “side,” or “contingent.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     profesora
    -1.03
    struzioni
    -1.03
     ministra
    -0.99
    跄
    -0.99
     händer
    -0.98
    هما
    -0.96
     egz
    -0.93
    americanas
    -0.93
     directores
    -0.93
    splitContainer
    -0.93
    POSITIVE LOGITS
    ではでは
    1.16
    k
    1.13
    perp
    1.04
    ほら
    1.04
    "
    1.03
    tomy
    1.03
    さてさて
    1.02
    ﹐
    1.00
     riscos
    1.00
    ombie
    0.99
    Activations Density 0.067%

    No Known Activations