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

    Neuronpedia

    INDEX
    Explanations

    karma and consequences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     as
    0.66
    ing
    0.61
     provide
    0.60
     rende
    0.58
     ensures
    0.57
    Ann
    0.57
    m
    0.57
    j
    0.57
    م
    0.57
     and
    0.55
    POSITIVE LOGITS
    ես
    0.55
    𝓖
    0.55
    throwing
    0.54
     théâtre
    0.53
    𝕡
    0.53
    هان
    0.52
    діть
    0.52
    情形
    0.52
    ăpadă
    0.52
     télévision
    0.51
    Act Density 0.000%

    No Known Activations