© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 65896
    Prev
    Next
    INDEX
    Explanations

    `isa` initialization blocks

    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
     constantemente
    -1.65
    <bos>
    -1.55
    -1.44
     but
    -1.43
     rápidamente
    -1.40
     transmisión
    -1.37
    thétique
    -1.36
     serán
    -1.35
     exclusivamente
    -1.35
     esperando
    -1.34
    POSITIVE LOGITS
    having
    1.78
    without
    1.75
    樒
    1.73
    EN
    1.71
    Auteur
    1.70
     kalın
    1.63
    ㈲
    1.63
    being
    1.62
    IN
    1.60
     髪
    1.59
    Activations Density 0.011%
    No Known Activations