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

    Neuronpedia

    INDEX
    Explanations

    performance

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     opera
    -0.06
    -prom
    -0.06
     factions
    -0.06
    Programming
    -0.06
    /data
    -0.06
     religions
    -0.06
     cloud
    -0.06
     agua
    -0.06
     fantasy
    -0.06
    .fullName
    -0.06
    POSITIVE LOGITS
     hrad
    0.08
     επα
    0.07
    ()↵↵↵↵
    0.07
    loadModel
    0.06
    šli
    0.06
     �
    0.06
     electrom
    0.06
     equ
    0.06
    brıs
    0.06
    間
    0.06
    Act Density 0.030%

    No Known Activations