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

    Neuronpedia

    INDEX
    Explanations

    code comments and function descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     diast
    0.45
     stardom
    0.43
     nationalism
    0.42
     berb
    0.42
     cadres
    0.42
     solidar
    0.40
     cones
    0.39
     dunes
    0.39
     coolness
    0.39
     jurisprudence
    0.39
    POSITIVE LOGITS
     Provides
    0.54
    ↵
    0.53
    Provides
    0.48
     Creates
    0.44
     Converts
    0.43
    <0x0D>
    0.42
     获取
    0.42
    {\"
    0.42
     provides
    0.42
    在
    0.41
    Act Density 0.025%

    No Known Activations