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

    Neuronpedia

    INDEX
    Explanations

    first-person pronoun followed by verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     consult
    0.86
    诌
    0.80
     mislead
    0.80
     impress
    0.77
    ிழமை
    0.77
     NOT
    0.76
     elegant
    0.76
     NÃO
    0.76
     those
    0.74
    玟
    0.73
    POSITIVE LOGITS
     ever
    0.89
    νει
    0.86
    ogr
    0.81
    %。
    0.81
     constante
    0.79
    CRP
    0.78
    念念
    0.75
    ergic
    0.75
     Know
    0.74
     смог
    0.71
    Act Density 0.009%

    No Known Activations