© 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. 129186
    Prev
    Next
    INDEX
    Explanations

    special characters and code

    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
     before
    -1.34
     when
    -1.27
     this
    -1.25
     still
    -1.24
     probably
    -1.23
     just
    -1.22
     really
    -1.16
     which
    -1.15
     on
    -1.13
     highly
    -1.13
    POSITIVE LOGITS
     predominate
    0.93
    愜
    0.92
     viime
    0.92
     seules
    0.91
     dexter
    0.91
     nœ
    0.90
    咳嗽
    0.89
    からない
    0.89
    唠
    0.88
     trous
    0.88
    Activations Density 0.013%

    No Known Activations