© 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. 41-GEMMASCOPE-2-RES-262K
    7. 65618
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post_all/layer_41_width_262k_l0_big
    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
     hazard
    0.41
    versch
    0.41
     cause
    0.38
     hazardous
    0.38
     pursued
    0.37
     Dec
    0.37
    :
    0.37
     risk
    0.37
     Hazard
    0.37
    Dec
    0.37
    POSITIVE LOGITS
    isierte
    0.40
     Seats
    0.40
    我自己
    0.40
    估
    0.40
    ಷ
    0.40
     শরীরে
    0.40
    Seats
    0.40
    但也
    0.39
     abhiv
    0.39
    avkh
    0.39
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.