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

    Neuronpedia

    INDEX
    Explanations

    technical specifications and concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Arab
    0.39
    engk
    0.38
    libr
    0.38
    ミン
    0.37
    ámica
    0.37
    Conv
    0.36
    timevals
    0.36
    †
    0.36
    min
    0.36
    Scal
    0.36
    POSITIVE LOGITS
     rade
    0.43
     DF
    0.42
     fetish
    0.41
    ドン
    0.41
     piers
    0.41
     miracle
    0.39
     Miracle
    0.39
    доо
    0.38
     hero
    0.38
     kuj
    0.38
    Act Density 0.001%

    No Known Activations