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

    Neuronpedia

    INDEX
    Explanations

    @

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Dim
    -0.07
     Funny
    -0.07
     directing
    -0.06
     lotion
    -0.06
     unbiased
    -0.06
    أس
    -0.06
     Thomson
    -0.06
     brushing
    -0.06
    .Companion
    -0.06
    /'↵
    -0.06
    POSITIVE LOGITS
    äter
    0.07
    kräfte
    0.07
    جار
    0.07
    _entities
    0.07
    trys
    0.07
     ספרים
    0.07
     특정
    0.07
    (field
    0.06
    �
    0.06
    됏
    0.06
    Act Density 0.004%

    No Known Activations