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

    Neuronpedia

    INDEX
    Explanations

    unsafe Rust and IRC

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    formes
    0.84
     Áng
    0.82
    ⟥
    0.80
    CCIÓN
    0.77
    D
    0.77
    সময়
    0.77
    Menurut
    0.77
     Cálculo
    0.77
     Хоккей
    0.76
    DataXYZ
    0.76
    POSITIVE LOGITS
    ↵↵
    1.09
    as
    0.76
     window
    0.75
    ↵
    0.74
     glimmer
    0.74
     peek
    0.73
    '
    0.73
    ↵↵↵
    0.72
     fast
    0.71
     pept
    0.70
    Act Density 0.008%

    No Known Activations