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

    Neuronpedia

    INDEX
    Explanations

    по

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     leite
    0.96
    িক
    0.93
    జేపీ
    0.90
    াৎ
    0.89
     hyvin
    0.85
     fels
    0.84
     veröffent
    0.83
    Whitelist
    0.81
     po
    0.81
     pati
    0.79
    POSITIVE LOGITS
     умолчанию
    1.42
    året
    1.07
     поводу
    1.06
    arity
    1.04
     ↑
    1.03
    ieurs
    1.01
    ceptible
    1.01
     persuaded
    1.01
     hearsay
    1.00
    cyano
    0.99
    Act Density 0.007%

    No Known Activations