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

    Neuronpedia

    INDEX
    Explanations

    technical design and implementation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ]".
    1.11
    ]."
    1.06
    ⤦
    1.03
    \":
    1.01
     Якщо
    1.01
     }"
    1.00
     Dacă
    1.00
     -->'
    0.98
     nowDate
    0.98
    .'"
    0.98
    POSITIVE LOGITS
    ↵
    1.41
    .
    0.64
     implementation
    0.60
     기반
    0.59
    *
    0.59
     base
    0.58
     modul
    0.56
     module
    0.56
    ?
    0.56
     design
    0.53
    Act Density 0.317%

    No Known Activations