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

    Neuronpedia

    INDEX
    Explanations

    newer research suggests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     caracteres
    0.51
     actores
    0.50
    䒾
    0.50
     nutrientes
    0.49
     utilizó
    0.48
     caracteriza
    0.47
     trabajadores
    0.47
     utiliza
    0.47
    に適
    0.46
     acteurs
    0.46
    POSITIVE LOGITS
    op
    0.50
     transcendence
    0.48
    As
    0.48
    or
    0.48
    Remark
    0.47
    Cl
    0.46
    ub
    0.46
    ⁣⁣
    0.46
    Trans
    0.45
    ability
    0.45
    Act Density 0.003%

    No Known Activations