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

    Neuronpedia

    INDEX
    Explanations

    code includes and comments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    浗
    0.50
     کدام
    0.48
     납
    0.48
     کریم
    0.46
    کر
    0.46
     nobleman
    0.46
     کنم
    0.45
    <unused50>
    0.44
     Pug
    0.44
    ンドン
    0.44
    POSITIVE LOGITS
     grips
    0.55
    म
    0.52
    '
    0.50
    ।
    0.49
     workloads
    0.49
    𝙢
    0.48
    '।
    0.47
    Value
    0.46
     mosaics
    0.45
     enhancers
    0.43
    Act Density 0.003%

    No Known Activations