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

    Neuronpedia

    INDEX
    Explanations

    breakdown of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     whatnot
    1.10
    uddled
    1.06
     stints
    1.02
    𓐄
    1.01
     purview
    1.00
     inclusivity
    0.99
     tweaks
    0.99
    0.99
     streetwear
    0.97
    goers
    0.97
    POSITIVE LOGITS
     MANY
    1.41
     ALSO
    1.33
     !!!
    1.33
     VERY
    1.32
     NEW
    1.26
     IMPORTANT
    1.20
    !!!
    1.15
     BOTH
    1.14
     FULL
    1.14
     YOU
    1.14
    Act Density 0.079%

    No Known Activations