Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    structural punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cutter
    0.39
     stripper
    0.35
     divider
    0.34
     combine
    0.32
     theses
    0.30
     цих
    0.30
     cuff
    0.30
     hubs
    0.30
     couche
    0.30
     singularities
    0.29
    POSITIVE LOGITS
    ánchez
    0.35
     samme
    0.32
    ük
    0.32
    ators
    0.32
    enden
    0.31
    amb
    0.31
    itte
    0.31
    esség
    0.31
    aj
    0.31
    ékony
    0.31
    Act Density 0.017%

    No Known Activations