© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26490
    Prev
    Next
    INDEX
    Explanations

    opening brackets

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ⬮
    -1.28
    にします
    -1.27
    の花
    -1.20
    を目指す
    -1.20
     that
    -1.17
     mammoth
    -1.14
     and
    -1.14
    oeuvre
    -1.13
     your
    -1.12
    同じく
    -1.11
    POSITIVE LOGITS
    cticide
    1.25
    Gmail
    1.23
    have
    1.20
    そうだ
    1.18
    hrten
    1.12
    to
    1.08
     войск
    1.08
    ͩ
    1.07
     زیرا
    1.07
    vantaged
    1.07
    Activations Density 0.069%

    No Known Activations