INDEX
    Explanations

    words related to emotional states and conditions

    New Auto-Interp
    Negative Logits
    jeme
    -0.16
    Sink
    -0.14
    agoon
    -0.14
     bols
    -0.14
    enden
    -0.14
     Ferd
    -0.14
    cul
    -0.14
    екаÑĢ
    -0.14
    rase
    -0.14
    False
    -0.14
    POSITIVE LOGITS
     DropIndex
    0.15
    allon
    0.14
    ÐIJÑĢÑħÑĸв
    0.14
     Dod
    0.14
    akan
    0.14
    hin
    0.14
    chers
    0.13
    urette
    0.13
    ãĥĭãĤ¢
    0.13
     pe
    0.13
    Act Density 0.039%

    No Known Activations