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

    Neuronpedia

    INDEX
    Explanations

    describing relationships or composition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    સ
    0.54
    s
    0.54
    ̀i
    0.52
    कुछ
    0.52
     cercano
    0.52
    spectra
    0.52
     fuertes
    0.51
     upsetting
    0.51
     dreadful
    0.49
    addresses
    0.49
    POSITIVE LOGITS
     McGill
    0.50
     this
    0.50
     هذا
    0.46
     تمر
    0.45
     wannan
    0.45
     userCollection
    0.44
    uncia
    0.44
     dieser
    0.44
     somit
    0.43
    蓁
    0.43
    Act Density 0.011%

    No Known Activations