© 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. 50-GEMMASCOPE-2-TRANSCODER-262K
    7. 155731
    Prev
    Next
    INDEX
    Explanations

    S prefix

    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_50_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
    Pren
    0.46
    ាញ់
    0.39
     Pren
    0.39
    銮
    0.39
    ญ
    0.39
     partout
    0.38
    pren
    0.38
    ง
    0.38
     эт
    0.38
     ۹
    0.37
    POSITIVE LOGITS
     stockings
    0.42
    ろし
    0.42
    缟
    0.40
     stocking
    0.38
    циями
    0.37
    rinos
    0.37
     Osama
    0.37
    AO
    0.36
    REC
    0.36
    MSE
    0.36
    Activations Density 0.011%

    No Known Activations