© 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. 10-GEMMASCOPE-RES-131K
    7. 40011
    Prev
    Next
    INDEX
    Explanations

    happy with

    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_10/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
     Makes
    -2.06
     Believe
    -1.98
     like
    -1.98
     Here
    -1.97
     Such
    -1.95
     to
    -1.95
     There
    -1.92
     convierte
    -1.92
     Something
    -1.92
     Surely
    -1.90
    POSITIVE LOGITS
     sii
    1.96
     kwanza
    1.88
    絰
    1.87
    ug
    1.80
    uns
    1.77
     STONE
    1.74
    und
    1.70
     MUSEUM
    1.68
     WHEEL
    1.68
    as
    1.66
    Activations Density 0.005%

    No Known Activations