© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 29-GEMMASCOPE-2-TRANSCODER-262K
    7. 80318
    Prev
    Next
    INDEX
    Explanations

    sources: followed by names/entities

    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_29_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
     yea
    0.39
     predom
    0.39
    䒨
    0.38
    逺
    0.37
     Karo
    0.37
     yüksek
    0.37
     Weights
    0.37
     pyram
    0.37
     pä
    0.36
     Stopwatch
    0.36
    POSITIVE LOGITS
    ଳ
    0.38
    ترنت
    0.38
    agli
    0.35
    Reddit
    0.34
     tenants
    0.32
     Ar
    0.32
    phere
    0.32
     Dame
    0.32
    iban
    0.31
    কারীদের
    0.31
    Activations Density 0.002%

    No Known Activations