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

    Neuronpedia

    INDEX
    Explanations

    references to choices and selections related to parameters or strategies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    selection and choice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    gonic
    -0.44
    MessageTagHelper
    -0.42
    #+#
    -0.41
    +#+
    -0.38
     című
    -0.37
    ViewInit
    -0.36
    pinas
    -0.36
    proof
    -0.36
     NSCoder
    -0.35
    Demos
    -0.35
    POSITIVE LOGITS
     chosen
    0.56
     choix
    0.54
     выбран
    0.54
     kiezen
    0.52
     scelta
    0.52
     selected
    0.51
    脚注の使い方
    0.51
     gekozen
    0.51
     adopted
    0.50
    Choosing
    0.50
    Act Density 0.050%

    No Known Activations