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

    Neuronpedia

    INDEX
    Explanations

    bucket of water

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ز
    0.95
    ુ
    0.94
    s
    0.93
    ého
    0.93
     cargos
    0.90
    ar
    0.88
     exfoli
    0.88
    ς
    0.88
    ισμού
    0.86
    ру
    0.86
    POSITIVE LOGITS
    ]
    1.34
     bucket
    1.27
    .
    1.26
     Bucket
    1.20
     buckets
    1.19
    AD
    1.11
    Bucket
    1.10
    bucket
    1.09
    ıyla
    0.98
    )
    0.98
    Act Density 0.002%

    No Known Activations