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

    Neuronpedia

    INDEX
    Explanations

    choosing the best option

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Zr
    0.39
     ಅವರ
    0.39
    くい
    0.39
    вань
    0.38
     কুক
    0.38
    预期
    0.37
    rennen
    0.37
    চলে
    0.37
    呓
    0.36
    stripos
    0.35
    POSITIVE LOGITS
     Choose
    0.95
    Choose
    0.92
     choose
    0.88
    choose
    0.81
     chooses
    0.80
    選擇
    0.79
     choosing
    0.76
     pilih
    0.74
     Pilih
    0.73
    选择
    0.70
    Act Density 0.007%

    No Known Activations