© 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. 51-GEMMASCOPE-2-TRANSCODER-262K
    7. 65624
    Prev
    Next
    INDEX
    Explanations

    still

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_51_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
    Mark
    0.40
    B
    0.40
    ɓ
    0.39
     Hurley
    0.39
    licer
    0.39
    amond
    0.38
    UNUSED
    0.38
    Ду
    0.37
     池
    0.37
     consegu
    0.37
    POSITIVE LOGITS
     pr
    0.40
    ंगत
    0.40
     Braxton
    0.38
    pped
    0.37
    nič
    0.37
    ventional
    0.37
    gadas
    0.36
    KR
    0.36
     Ngh
    0.36
    🇰
    0.36
    Activations Density 0.000%

    No Known Activations