© 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. 22-GEMMASCOPE-2-TRANSCODER-262K
    7. 261942
    Prev
    Next
    INDEX
    Explanations

    breakdown of key differences

    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_22_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
     экспери
    0.28
    }{$\
    0.28
    эк
    0.28
    cosystem
    0.28
     экзем
    0.28
    Donald
    0.28
     लॉग
    0.27
    ؤس
    0.27
     instanceof
    0.26
     നമഃ
    0.26
    POSITIVE LOGITS
     Politique
    0.25
    鳎
    0.25
     图片
    0.24
     ত্রাণ
    0.24
    ूरिया
    0.24
    вшейся
    0.24
     едва
    0.23
     deductible
    0.23
    rimin
    0.23
     வாய்
    0.23
    Activations Density 0.021%

    No Known Activations