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

    Neuronpedia

    INDEX
    Explanations

    formatting symbols

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     possam
    0.52
    няка
    0.50
     यासाठी
    0.49
    色々
    0.47
    力和
    0.47
     можуть
    0.46
    的一些
    0.45
     grads
    0.45
     puedan
    0.45
     могут
    0.44
    POSITIVE LOGITS
     natomiast
    0.50
    apanicolaou
    0.48
    is
    0.48
     является
    0.48
     befindet
    0.48
     \,
    0.47
     Conversely
    0.46
    second
    0.45
     completes
    0.45
     Adjacent
    0.45
    Act Density 0.925%

    No Known Activations