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

    Neuronpedia

    INDEX
    Explanations

    closing delimiters

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     அத
    0.81
    '])[
    0.70
    ("[
    0.70
     প্রা
    0.69
    ('[
    0.68
    Segoe
    0.67
    captive
    0.67
     secundarios
    0.67
    estimates
    0.66
    æ
    0.66
    POSITIVE LOGITS
    })/
    1.60
    }}/
    1.60
    )/
    1.58
    ]/
    1.52
    ))/
    1.51
    ])/
    1.47
    }/
    1.47
     "/
    1.38
    >/
    1.36
    ")/
    1.35
    Act Density 0.009%

    No Known Activations