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

    Neuronpedia

    INDEX
    Explanations

    not perfect or absolute

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     directly
    0.58
    directly
    0.57
     Directly
    0.57
     directement
    0.55
     directamente
    0.55
     necesariamente
    0.54
     necessarily
    0.54
     přímo
    0.53
     напрямую
    0.52
     сразу
    0.51
    POSITIVE LOGITS
     possibilités
    0.50
     imperfect
    0.45
    可能性
    0.45
     possibilité
    0.45
     olabilir
    0.45
     quelquefois
    0.44
     algebraic
    0.44
    Poss
    0.44
    Absolute
    0.43
     immutable
    0.43
    Act Density 0.055%

    No Known Activations