© 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. 112964
    Prev
    Next
    INDEX
    Explanations

    possessives, contractions, and 'once observed'

    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
     using
    -1.95
     this
    -1.87
    :
    -1.80
     even
    -1.79
    </b>
    -1.69
     both
    -1.67
    </h2>
    -1.66
     sizeable
    -1.64
     ALWAYS
    -1.62
     only
    -1.59
    POSITIVE LOGITS
     trám
    1.73
    BIDDEN
    1.62
     สม
    1.55
    ↗
    1.53
     카
    1.53
    cations
    1.52
    噥
    1.51
     .*
    1.49
     transak
    1.45
    nasi
    1.45
    Activations Density 0.045%

    No Known Activations