© 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. 112824
    Prev
    Next
    INDEX
    Explanations

    classes, support, selected

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

    adverbs or adjectives ending in "-ally" or technical/formal compound terms with hyphens.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5
    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
     in
    -1.52
     of
    -1.37
    。（
    -1.33
     any
    -1.32
     its
    -1.30
    .”
    -1.29
     do
    -1.27
    ”.
    -1.27
     our
    -1.26
    }$.
    -1.26
    POSITIVE LOGITS
     Somewhat
    1.60
    唠
    1.45
     semejante
    1.35
     partiellement
    1.33
     réellement
    1.32
     nachfol
    1.31
     somewhat
    1.28
    ẠI
    1.27
    INES
    1.26
     Otras
    1.25
    Activations Density 0.164%

    No Known Activations