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

    Neuronpedia

    INDEX
    Explanations

    namespace App or environment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     cline
    2.02
     dwell
    2.02
    ব
    2.01
     arresting
    2.01
    rpt
    1.84
    ことなく
    1.83
     foregoing
    1.79
     forego
    1.79
    r
    1.78
    ⺆
    1.78
    POSITIVE LOGITS
    й
    1.94
    das
    1.85
    منى
    1.74
    ppure
    1.74
    меча
    1.64
     תו
    1.58
     себя
    1.56
    ,\,\
    1.56
     রানে
    1.56
    فظ
    1.53
    Act Density 0.001%

    No Known Activations