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

    Neuronpedia

    INDEX
    Explanations

    clarification options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :
    0.77
     susceptible
    0.66
     impairs
    0.65
     it
    0.65
     alleges
    0.64
     implies
    0.63
     pig
    0.61
     predecessor
    0.61
     he
    0.61
     architecture
    0.61
    POSITIVE LOGITS
    Thank
    0.96
    Clar
    0.89
    ？
    0.89
    ؟.
    0.88
    Usuario
    0.88
    Whatsapp
    0.87
    ？）
    0.87
    Gracias
    0.86
    ؟؟
    0.85
    Todo
    0.85
    Act Density 0.079%

    No Known Activations