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

    Neuronpedia

    INDEX
    Explanations

    what's the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     таки
    0.86
    けど
    0.80
    どちら
    0.77
    alnız
    0.77
    как
    0.76
    Either
    0.74
     yalnızca
    0.72
     apapun
    0.71
    ichever
    0.71
     solche
    0.71
    POSITIVE LOGITS
     happening
    2.05
     happen
    1.54
     happened
    1.51
     wrong
    1.49
     going
    1.40
     entailed
    1.32
     occurring
    1.31
     happens
    1.24
     Happen
    1.23
     involved
    1.22
    Act Density 0.156%

    No Known Activations