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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     prepared
    -0.07
     apologized
    -0.06
     compl
    -0.06
    大人
    -0.06
    ук
    -0.06
     konnte
    -0.06
    rect
    -0.06
    टन
    -0.06
     covered
    -0.06
     ogl
    -0.06
    POSITIVE LOGITS
    考え
    0.07
    University
    0.06
    짜
    0.06
    sticky
    0.06
    accounts
    0.06
    .stdout
    0.06
     durable
    0.06
     @_;↵
    0.06
     Till
    0.06
    ginas
    0.05
    Act Density 0.134%

    No Known Activations