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

    Neuronpedia

    INDEX
    Explanations

    listing inclusions and exceptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     too
    0.74
     disadvantaged
    0.73
     depress
    0.73
     Wolves
    0.72
     marginalized
    0.72
     poop
    0.70
     pre
    0.70
     ไง
    0.70
     Bulldogs
    0.69
     disenfranch
    0.69
    POSITIVE LOGITS
    including
    1.03
    Including
    1.00
     incluindo
    0.91
    except
    0.85
     including
    0.83
    .,
    0.82
    And
    0.81
    inol
    0.81
    直接
    0.81
     وذلك
    0.80
    Act Density 0.012%

    No Known Activations