© 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. 50335
    Prev
    Next
    INDEX
    Explanations

    using parameters

    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
     ansatz
    0.41
     consiglio
    0.39
     danno
    0.38
     ಬಂಧ
    0.38
     biota
    0.38
     乀
    0.37
     dựa
    0.37
     impactos
    0.37
     bias
    0.37
     tuổi
    0.37
    POSITIVE LOGITS
     Town
    0.41
    Town
    0.39
    Dv
    0.39
    سه
    0.38
    <u>
    0.37
    Media
    0.37
    чной
    0.37
    Setting
    0.37
    Logan
    0.37
     дво
    0.37
    Activations Density 0.000%

    No Known Activations