© 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. 22-GEMMASCOPE-RES-131K
    7. 15128
    Prev
    Next
    INDEX
    Explanations

    phrases starting with "in"

    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_22/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
     substantial
    -0.97
     many
    -0.96
     will
    -0.95
     cannot
    -0.94
     greatly
    -0.93
    anhas
    -0.93
    Upsilon
    -0.93
    管理器
    -0.92
    <bos>
    -0.91
     people
    -0.88
    POSITIVE LOGITS
     the
    1.09
    }^{
    0.93
    Ros
    0.92
    ляция
    0.91
     singoli
    0.90
     médec
    0.89
     prü
    0.88
     första
    0.88
     fournisseurs
    0.88
    Rgb
    0.87
    Activations Density 0.068%

    No Known Activations