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

    Neuronpedia

    INDEX
    Explanations

    sorry or apologize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    યુ
    0.84
    деть
    0.79
    𝗛
    0.77
     saker
    0.74
    জয়
    0.74
    ज्योति
    0.73
    Сейчас
    0.73
     задача
    0.72
    ЕВ
    0.71
     premio
    0.71
    POSITIVE LOGITS
     But
    0.82
    ruvic
    0.81
    ganggu
    0.80
     transgress
    0.80
    vät
    0.80
     inconven
    0.78
     shortcoming
    0.78
     shortcomings
    0.78
    üm
    0.77
    ва
    0.77
    Act Density 0.003%

    No Known Activations