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

    Neuronpedia

    INDEX
    Explanations

    first impression

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     impressed
    0.74
     impresses
    0.68
    Impress
    0.65
    impress
    0.64
     impress
    0.60
     Imp
    0.58
     impressive
    0.56
     beeindruck
    0.56
     imprint
    0.54
     imprinted
    0.54
    POSITIVE LOGITS
    cesz
    0.42
     expression
    0.41
     micro
    0.41
    թ
    0.40
     connection
    0.40
     neces
    0.39
     ente
    0.39
     mechanical
    0.38
    expression
    0.38
    micro
    0.37
    Act Density 0.001%

    No Known Activations