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

    Neuronpedia

    INDEX
    Explanations

    actions and options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    шої
    0.86
    odimensional
    0.74
    ὶ
    0.73
    ὸς
    0.72
    Эти
    0.72
    accident
    0.71
    Alfred
    0.71
    CJK
    0.70
     memperoleh
    0.69
    Performing
    0.67
    POSITIVE LOGITS
    !
    1.53
     without
    1.52
    ;
    1.38
     unless
    1.36
     until
    1.36
     before
    1.36
     while
    1.34
     via
    1.32
    !",
    1.32
    !,
    1.30
    Act Density 0.234%

    No Known Activations