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

    Neuronpedia

    INDEX
    Explanations

    preliminary or anticipated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     utilization
    0.44
    常用的
    0.43
     upkeep
    0.43
     oti
    0.41
     patching
    0.41
     tùy
    0.40
     vét
    0.40
     ഇടപെ
    0.40
     facilitation
    0.39
    Utilization
    0.39
    POSITIVE LOGITS
     preview
    1.00
     preliminary
    1.00
     Preliminary
    0.96
     anticipated
    0.94
     Preview
    0.92
    preview
    0.91
    Preliminary
    0.86
    anticipated
    0.84
    Preview
    0.82
     previews
    0.82
    Act Density 0.227%

    No Known Activations