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

    Neuronpedia

    INDEX
    Explanations

    note

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Dominic
    -0.07
    posé
    -0.07
    anel
    -0.07
    defer
    -0.07
     macros
    -0.06
     textbox
    -0.06
    riel
    -0.06
    LANG
    -0.06
     letra
    -0.06
    理科
    -0.06
    POSITIVE LOGITS
     STATE
    0.07
    珢
    0.07
    谁知
    0.07
    굣
    0.07
    𝕪
    0.06
     человека
    0.06
     WITH
    0.06
    (parent
    0.06
    สภาพ
    0.06
    入れ
    0.06
    Act Density 0.002%

    No Known Activations