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

    Neuronpedia

    INDEX
    Explanations

    program, safety, documentation, audit, integer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     be
    0.38
     ladr
    0.37
     atacar
    0.35
     ovat
    0.35
     čty
    0.35
    ittää
    0.35
    人都
    0.35
     joten
    0.35
     deux
    0.34
     eax
    0.34
    POSITIVE LOGITS
    9
    0.51
    P
    0.47
    7
    0.46
    2
    0.45
    p
    0.42
    6
    0.42
    on
    0.40
    post
    0.40
    DF
    0.40
    3
    0.40
    Act Density 0.161%

    No Known Activations