© 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. 63945
    Prev
    Next
    INDEX
    Explanations

    combining roots with suffixes

    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
     veno
    -0.90
    Being
    -0.85
    erialize
    -0.84
    臟
    -0.84
    eches
    -0.84
     farina
    -0.83
    genicity
    -0.79
    onik
    -0.79
    ustus
    -0.79
     применяется
    -0.78
    POSITIVE LOGITS
    ing
    2.78
    izing
    2.70
    ising
    2.48
    ating
    2.36
    ulating
    2.11
    ifying
    1.98
    ING
    1.97
    fying
    1.87
    ting
    1.85
    cting
    1.68
    Activations Density 0.131%

    No Known Activations