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

    Neuronpedia

    INDEX
    Explanations

    quality and precision

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    间隔
    1.01
    らず
    0.99
    了一個
    0.99
    <unused967>
    0.98
    了一个
    0.96
    ookie
    0.96
     budgeted
    0.94
    ('')
    0.92
    बाय
    0.92
    Medical
    0.92
    POSITIVE LOGITS
    innen
    0.99
     patric
    0.98
     striped
    0.97
     radiating
    0.96
     locking
    0.96
     bet
    0.96
     bild
    0.96
    틱
    0.95
     velvet
    0.95
     boh
    0.94
    Act Density 0.000%

    No Known Activations