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

    Neuronpedia

    INDEX
    Explanations

    trick questions about quantity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     카페
    1.25
     ansatz
    1.22
    い
    1.21
     létre
    1.17
    er
    1.17
    وٹی
    1.16
     Mete
    1.15
    𝕂
    1.15
    ından
    1.15
     Foss
    1.14
    POSITIVE LOGITS
    Interested
    1.07
    Interestingly
    1.07
    1.01
     keempat
    0.99
     cứu
    0.97
    returned
    0.96
     cope
    0.95
     succulent
    0.94
     dealing
    0.93
    ল
    0.93
    Act Density 0.000%

    No Known Activations