© 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. 106671
    Prev
    Next
    INDEX
    Explanations

    whoever whatever wherever whenever

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    braun
    -0.71
     Gunners
    -0.71
    уна
    -0.71
    inaire
    -0.69
    漬
    -0.68
    strukt
    -0.68
    [`
    -0.68
    締
    -0.66
     diye
    -0.66
     EntityManager
    -0.66
    POSITIVE LOGITS
     ever
    5.56
    ever
    4.34
    Ever
    3.52
    EVER
    3.39
     EVER
    3.09
     Ever
    3.09
    ver
    2.31
    soever
    2.13
     Everett
    1.77
     ooit
    1.68
    Activations Density 0.086%

    No Known Activations