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

    Neuronpedia

    INDEX
    Explanations

    bash commands and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    부터
    1.01
    ן
    0.86
     decouple
    0.82
    ل
    0.81
    on
    0.79
    ज
    0.77
     aos
    0.76
    ்
    0.73
     tad
    0.73
    ल
    0.72
    POSITIVE LOGITS
    borderRadius
    1.13
     elicited
    1.06
    }))
    1.03
    ylabel
    1.02
    enje
    1.01
     вино
    1.00
    asses
    0.99
     oath
    0.99
    aktiv
    0.97
    yrch
    0.97
    Act Density 0.121%

    No Known Activations