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

    Neuronpedia

    INDEX
    Explanations

    performance degradation and states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    يح
    0.44
    ること
    0.44
    是
    0.44
    kov
    0.42
    ávat
    0.41
    í
    0.41
    "#
    0.41
    "
    0.41
    ''.
    0.40
    م
    0.40
    POSITIVE LOGITS
    cosec
    0.54
    솎
    0.54
     인수
    0.52
     Олександр
    0.51
     Paise
    0.50
     AEG
    0.49
     Cargill
    0.49
     이걸
    0.49
     நன்மை
    0.49
     Список
    0.49
    Act Density 0.000%

    No Known Activations