© 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. 57-GEMMASCOPE-2-TRANSCODER-262K
    7. 39428
    Prev
    Next
    INDEX
    Explanations

    numbers

    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_57_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
     ’
    0.45
     ʼ
    0.38
     \'
    0.37
     justific
    0.34
     äh
    0.33
    Ϩ
    0.33
    ۆر
    0.33
    robespierre
    0.33
     '
    0.32
    दृ
    0.32
    POSITIVE LOGITS
     seventy
    0.64
     eighty
    0.62
     forty
    0.59
     fifties
    0.59
     fifty
    0.59
     sixty
    0.59
     nineteen
    0.58
     $=
    0.57
     twelfth
    0.57
     ninety
    0.57
    Activations Density 0.099%

    No Known Activations