© 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. 22-GEMMASCOPE-RES-131K
    7. 27807
    Prev
    Next
    INDEX
    Explanations

    physical components and structures

    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
    ungkapkan
    -1.30
    逆に
    -1.26
     начинают
    -1.25
    Profesor
    -1.25
     Qualquer
    -1.22
     problém
    -1.22
     куртка
    -1.21
    ⼿
    -1.20
    🏢
    -1.19
     any
    -1.18
    POSITIVE LOGITS
    ，
    1.52
    「
    1.34
     laikā
    1.23
     has
    1.20
     will
    1.20
    https
    1.20
    如今
    1.19
     coerci
    1.14
     inaweza
    1.14
    展开
    1.13
    Activations Density 0.009%

    No Known Activations