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

    Neuronpedia

    INDEX
    Explanations

    resulting in

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     exchanged
    0.39
     encountered
    0.37
     જો
    0.37
     Replace
    0.37
     asegurarse
    0.36
     if
    0.36
     সবসময়
    0.36
     refused
    0.36
     excavated
    0.35
     mnoho
    0.35
    POSITIVE LOGITS
    从而
    1.31
    进而
    1.06
     وبالتالي
    1.02
     resulting
    0.99
     thereby
    0.98
    resulting
    0.88
     thus
    0.87
    因而
    0.86
     hence
    0.83
     sehingga
    0.82
    Act Density 0.740%

    No Known Activations