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

    Neuronpedia

    INDEX
    Explanations

    non-english characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     labeling
    0.73
     name
    0.65
    标题
    0.65
     labelling
    0.63
    কতা
    0.63
     shifts
    0.63
     Title
    0.62
    /**
    0.61
    shift
    0.61
     Lowest
    0.60
    POSITIVE LOGITS
    ్య
    0.90
    就要
    0.80
     harus
    0.74
     יש
    0.73
    దాయ
    0.72
     снять
    0.72
     डालेंगे
    0.71
    だけでなく
    0.71
    wrapp
    0.70
    ए
    0.70
    Act Density 0.094%

    No Known Activations