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

    Neuronpedia

    INDEX
    Explanations

    assuming and stating assumptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    真正
    0.41
     literalmente
    0.39
    Nunca
    0.38
    выше
    0.38
     addirittura
    0.37
     esattamente
    0.37
    DoesNotExist
    0.37
     autrefois
    0.37
    早就
    0.36
    まさに
    0.36
    POSITIVE LOGITS
     assume
    2.03
    assume
    1.95
     assumes
    1.92
     assumed
    1.91
     Assume
    1.91
     assuming
    1.90
    Assume
    1.80
     Assuming
    1.80
    Assuming
    1.75
    假设
    1.74
    Act Density 0.089%

    No Known Activations