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

    Neuronpedia

    INDEX
    Explanations

    announced

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ية
    1.67
     escrit
    1.64
     quem
    1.64
    页面存档备份
    1.63
     sorta
    1.63
     |”
    1.61
    ﻟ
    1.57
    verick
    1.52
    =”
    1.52
     essent
    1.52
    POSITIVE LOGITS
    و
    1.76
    یل
    1.62
    ငံ
    1.59
    ی
    1.57
    boxes
    1.56
    ت
    1.54
    س
    1.53
    یم
    1.51
    ごはん
    1.50
    us
    1.49
    Act Density 0.000%

    No Known Activations