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

    Neuronpedia

    INDEX
    Explanations

    aspect ratio

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     RA
    0.91
    Rate
    0.87
     Rates
    0.86
     relation
    0.84
    relation
    0.84
     Rate
    0.83
    RA
    0.83
    Rates
    0.82
     rates
    0.82
    Relation
    0.81
    POSITIVE LOGITS
     patios
    0.53
    atiu
    0.52
    atia
    0.51
     Patio
    0.50
     patio
    0.49
    atius
    0.47
    āti
    0.46
    <0x81>
    0.45
    ัติ
    0.44
    atiya
    0.43
    Act Density 0.010%

    No Known Activations