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

    Neuronpedia

    INDEX
    Explanations

    medical terms and definitions

    np_acts-logits-general · gemini-2.5-flash-lite

    mathematical notation for norms, particularly indicated by double vertical bars and subscripts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi
    New Auto-Interp
    Negative Logits
    \|^{
    -0.90
    providedIn
    -0.85
     entrambi
    -0.82
    owel
    -0.81
     करे
    -0.80
    weise
    -0.78
    Therefore
    -0.78
    ylate
    -0.78
    .~\
    -0.77
    /</
    -0.77
    POSITIVE LOGITS
    ˦
    0.90
    ЛЯ
    0.86
    ♾
    0.85
    &
    0.74
     särskilt
    0.74
    Qui
    0.73
     cùng
    0.73
    urangan
    0.72
    ához
    0.72
    ůže
    0.72
    Act Density 0.007%

    No Known Activations