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

    Neuronpedia

    INDEX
    Explanations

    atmospheric and evocative language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     inconceivable
    0.43
     cutest
    0.43
    Adorable
    0.42
     felicidad
    0.41
    uctance
    0.41
     Beneficial
    0.41
     exaggerate
    0.40
     harmless
    0.40
     positifs
    0.40
     beneficial
    0.39
    POSITIVE LOGITS
     haunting
    1.00
     richly
    0.98
     atmospheric
    0.96
     evocative
    0.93
     lyrical
    0.92
     vivid
    0.91
     beautifully
    0.90
     poignant
    0.86
     darkly
    0.86
     compelling
    0.84
    Act Density 0.129%

    No Known Activations