© 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. 9-GEMMASCOPE-2-TRANSCODER-262K
    7. 198896
    Prev
    Next
    INDEX
    Explanations

    Arabic negative or interrogative prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_9_width_262k_l0_small_affine
    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
    ی
    1.08
    t
    1.05
    が
    0.91
    ود
    0.88
    (
    0.81
    _
    0.75
    رات
    0.73
    이가
    0.73
    <0x0D>
    0.73
     that
    0.72
    POSITIVE LOGITS
    se
    0.90
    jangan
    0.76
    ronom
    0.73
    禛
    0.71
     পরিশোধ
    0.70
     Jangan
    0.69
     iyong
    0.69
     consiglio
    0.69
     संकल्प
    0.68
    kipun
    0.68
    Activations Density 0.000%

    No Known Activations