© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 23721
    Prev
    Next
    INDEX
    Explanations

    misuse and overuse

    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
     procé
    -0.82
    穫
    -0.82
    ાં
    -0.82
    Trash
    -0.78
    ówka
    -0.78
    <0xD7>
    -0.77
    Дереккөздер
    -0.77
    ниці
    -0.75
    zes
    -0.75
     sén
    -0.75
    POSITIVE LOGITS
     misuse
    4.66
     misused
    3.73
     overuse
    2.94
     abuse
    2.91
     usage
    2.64
     use
    2.52
     abused
    2.36
     improper
    2.27
    abuse
    2.19
     used
    2.11
    Activations Density 0.081%

    No Known Activations