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

    Neuronpedia

    INDEX
    Explanations

    read carefully

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nightclubs
    0.55
    -
    0.54
    ate
    0.53
    ide
    0.53
    utt
    0.53
    onin
    0.51
     câble
    0.51
    ga
    0.50
    AT
    0.50
    atih
    0.50
    POSITIVE LOGITS
     read
    1.09
    读
    1.02
    読
    0.92
     đọc
    0.92
     Read
    0.89
    阅读
    0.89
     읽
    0.89
    讀
    0.89
    읽
    0.88
    読む
    0.84
    Act Density 0.145%

    No Known Activations