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

    Neuronpedia

    INDEX
    Explanations

    extract textobjectscigarettetoy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     trex
    0.50
     Ones
    0.42
    از
    0.41
     გამოყენ
    0.41
     Использу
    0.41
     Полу
    0.40
     одному
    0.40
     الاستاذ
    0.40
     Serien
    0.40
    legen
    0.39
    POSITIVE LOGITS
     sít
    0.48
    ].”
    0.46
    рата
    0.46
    WART
    0.46
     demanded
    0.45
    ランド
    0.43
    }(\
    0.43
    哴
    0.42
     porter
    0.42
    ᱔
    0.41
    Act Density 0.001%

    No Known Activations