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

    Neuronpedia

    INDEX
    Explanations

    very/incredibly + important/difficult

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     somewhat
    0.51
     oldukça
    0.48
     довольно
    0.48
     весьма
    0.46
     quite
    0.46
     notoriously
    0.45
     досить
    0.44
     fairly
    0.43
     themselves
    0.43
     bastante
    0.43
    POSITIVE LOGITS
    possible
    0.46
     möglich
    0.45
     possible
    0.43
     possibile
    0.42
     posible
    0.39
     circumst
    0.39
    ไร
    0.39
     circumstantial
    0.38
    ง่าย
    0.38
     caso
    0.38
    Act Density 0.025%

    No Known Activations