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

    Neuronpedia

    INDEX
    Explanations

    before understanding or considering

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     combined
    0.45
     uiteindelijk
    0.44
     podczas
    0.43
    ފ
    0.42
     alkaline
    0.41
     například
    0.41
     combiner
    0.41
     opposit
    0.40
     Podczas
    0.40
     dopo
    0.40
    POSITIVE LOGITS
    before
    0.77
    Before
    0.73
    Antes
    0.71
     before
    0.70
     Before
    0.68
    ก่อน
    0.68
     قبل
    0.67
     Trước
    0.66
    Перед
    0.65
     ก่อน
    0.64
    Act Density 0.090%

    No Known Activations