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

    Neuronpedia

    INDEX
    Explanations

    python code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    _=
    0.91
     (_,
    0.82
    kwargs
    0.80
     elated
    0.78
     EventArgs
    0.75
    つまり
    0.75
     cleansed
    0.75
     transformed
    0.75
     args
    0.74
     sanitized
    0.73
    POSITIVE LOGITS
    ین
    0.81
    со
    0.80
    CharCode
    0.79
    Dek
    0.78
    icro
    0.77
     Gleich
    0.75
    Corpor
    0.75
    ่
    0.75
    તા
    0.74
    자가
    0.74
    Act Density 0.116%

    No Known Activations