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

    Neuronpedia

    INDEX
    Explanations

    magnitude adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     iniciando
    0.86
     égaux
    0.83
     সত্যাগ্রহ
    0.80
     informatique
    0.78
     itinéraires
    0.77
    ichage
    0.77
     appartenant
    0.76
     आवृ
    0.76
     aurez
    0.75
     мъ
    0.75
    POSITIVE LOGITS
     but
    1.09
    But
    0.99
    0.95
     But
    0.93
     criteria
    0.90
     also
    0.89
     and
    0.89
     then
    0.84
     after
    0.84
     because
    0.83
    Act Density 0.493%

    No Known Activations