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

    Neuronpedia

    INDEX
    Explanations

    code and technical terms

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     compradores
    0.59
    こな
    0.55
     chon
    0.52
    ใคร
    0.52
     restaurants
    0.51
     noticeable
    0.51
     Latinos
    0.51
     видно
    0.50
    明らかに
    0.50
    Ќ
    0.50
    POSITIVE LOGITS
    substring
    0.63
     marsh
    0.60
     погиб
    0.60
    marshalN
    0.59
     البل
    0.58
    元年
    0.57
     काळ
    0.57
    vida
    0.57
    িতা
    0.56
    蓣
    0.56
    Act Density 0.001%

    No Known Activations