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

    Neuronpedia

    INDEX
    Explanations

    abilities and capabilities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     permettant
    1.09
    IZED
    1.05
     facendo
    1.05
    层面
    1.04
     Niveau
    1.00
    k
    0.99
    ında
    0.98
     Kecamatan
    0.98
    b
    0.97
    RODUCTION
    0.95
    POSITIVE LOGITS
    да
    1.23
     CPUs
    1.13
    .³
    1.09
    𝓷
    1.05
     умолча
    1.04
    ДС
    1.04
    .¹
    1.03
     bals
    1.03
     PGE
    1.02
     positrons
    1.02
    Act Density 0.244%

    No Known Activations