© 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. 10-GEMMASCOPE-RES-131K
    7. 21149
    Prev
    Next
    INDEX
    Explanations

    street, lounge, books, groove, program, butter, rock, gene, literature

    np_acts-logits-general · gemini-2.5-flash-lite

    unusual or uncommon combinations of letters in names, terms, or titles.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     nowy
    -1.56
     avto
    -1.55
     those
    -1.45
     varier
    -1.44
     exécut
    -1.42
     okna
    -1.41
     teléfonos
    -1.41
    紲
    -1.41
     estimés
    -1.38
     inglese
    -1.37
    POSITIVE LOGITS
    ?
    1.52
     Then
    1.46
     in
    1.38
     saludables
    1.34
     at
    1.31
     Different
    1.29
    颯
    1.29
    бира
    1.27
     Various
    1.26
    AISSEE
    1.26
    Activations Density 2.473%

    No Known Activations