INDEX
    Explanations

    phrases related to emotional expressions

    New Auto-Interp
    Negative Logits
    项
    -0.16
     consent
    -0.16
    uel
    -0.15
    ocol
    -0.15
    uture
    -0.15
    osp
    -0.14
    ucus
    -0.14
     Stock
    -0.14
     Îĵεν
    -0.13
    imet
    -0.13
    POSITIVE LOGITS
    krom
    0.16
    ãĢģ 
    0.15
    коз
    0.14
    enor
    0.14
    urga
    0.14
    éľĬ
    0.14
    βÎŃÏģ
    0.14
    UNET
    0.13
     Dek
    0.13
    änd
    0.13
    Act Density 0.601%

    No Known Activations