© 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. 34-GEMMASCOPE-2-TRANSCODER-262K
    7. 167042
    Prev
    Next
    INDEX
    Explanations

    triumph of good over evil

    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_34_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
    相关
    0.37
     biến
    0.37
     bó
    0.36
     লেফটেন্যান্ট
    0.36
     जिल्ह
    0.35
     EDITOR
    0.34
     policewomen
    0.34
    停
    0.34
    RELATED
    0.34
     لیک
    0.34
    POSITIVE LOGITS
     triumph
    0.79
     prevails
    0.68
     triumphs
    0.68
     prevail
    0.67
     zwycię
    0.66
     побе
    0.64
     победи
    0.64
     preval
    0.63
     перемо
    0.62
     vencer
    0.60
    Activations Density 0.002%

    No Known Activations