© 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. 72217
    Prev
    Next
    INDEX
    Explanations

    hitherto heretofore duly

    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
    んでる
    -1.53
     permettent
    -1.48
    🪳
    -1.42
     principalement
    -1.35
     principalmente
    -1.31
    wakili
    -1.29
     aprendiz
    -1.29
     každý
    -1.28
     focuses
    -1.27
     разных
    -1.26
    POSITIVE LOGITS
     duly
    1.39
     litoral
    1.37
    奬
    1.36
     hitherto
    1.30
    Otras
    1.30
     heretofore
    1.29
     blest
    1.18
     berkenaan
    1.17
     surla
    1.15
     jSON
    1.13
    Activations Density 0.205%

    No Known Activations