© 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-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-65K
    7. 52536
    Prev
    Next
    INDEX
    Explanations

    short duration or extent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_65k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ibilit
    0.39
    ordat
    0.39
     insensible
    0.38
    geometry
    0.38
    orsunuz
    0.37
    덫
    0.37
    erty
    0.37
    ilon
    0.36
    quement
    0.36
     geology
    0.36
    POSITIVE LOGITS
     year
    0.43
     jaar
    0.43
    两年
    0.42
     years
    0.41
     Order
    0.41
     ห
    0.40
     tahun
    0.40
     상담
    0.40
     Laugh
    0.39
    编译
    0.38
    Activations Density 0.001%

    No Known Activations