© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 103246
    Prev
    Next
    INDEX
    Explanations

    interjections of surprise or emphasis

    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_34/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
    OOOOOO
    -0.83
    OOOOOOOO
    -0.83
     Sünde
    -0.79
     lustige
    -0.77
    OOOOO
    -0.75
    вед
    -0.74
     Mare
    -0.73
    évêque
    -0.73
    тари
    -0.73
    ursa
    -0.72
    POSITIVE LOGITS
     ge
    3.00
     Gee
    2.56
    Gee
    2.55
     gee
    2.25
     j
    2.23
    Jeez
    2.22
     jesus
    2.14
     gosh
    2.05
    ge
    2.00
    jee
    1.96
    Activations Density 0.072%

    No Known Activations