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

    Neuronpedia

    INDEX
    Explanations

    conditions and attributes after "with"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Being
    0.41
    是一种
    0.40
     činjen
    0.40
    是一種
    0.39
     Being
    0.38
     වීම
    0.38
     essendo
    0.38
     అనేది
    0.38
     being
    0.37
    使用
    0.37
    POSITIVE LOGITS
     disabilities
    0.71
     pretensions
    0.61
     whom
    0.59
     Disabilities
    0.59
     ambitions
    0.59
     penchant
    0.59
     acumen
    0.52
     problems
    0.52
     aspirations
    0.52
     delusions
    0.51
    Act Density 0.021%

    No Known Activations