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

    Neuronpedia

    INDEX
    Explanations

    what how questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ױ
    0.64
     selalu
    0.64
    𝓊
    0.61
    ponsor
    0.60
    ições
    0.60
     सिंपली
    0.59
    ignon
    0.58
    ]).
    0.58
    asius
    0.58
    кульп
    0.57
    POSITIVE LOGITS
     How
    3.21
     What
    3.12
     how
    3.11
    How
    3.06
    What
    3.01
     what
    2.93
    what
    2.78
    how
    2.65
     क्या
    2.61
     ماذا
    2.55
    Act Density 0.548%

    No Known Activations