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

    Neuronpedia

    INDEX
    Explanations

    preceding phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     TRANSPORTURI
    0.51
    とても
    0.50
    SERVICIO
    0.50
    uestro
    0.50
     Primero
    0.48
     サイ
    0.47
    čila
    0.47
    احمد
    0.47
    两者
    0.46
    ⿶
    0.46
    POSITIVE LOGITS
     band
    0.44
     gamma
    0.44
     γ
    0.44
     on
    0.43
     fabric
    0.43
     full
    0.42
     radius
    0.42
     slits
    0.42
     f
    0.42
     ভীত
    0.41
    Act Density 0.007%

    No Known Activations