INDEX
    Explanations

    terms related to heat treatment processes in materials

    New Auto-Interp
    Negative Logits
    .
    -0.75
    -0.59
    except
    -0.49
    ij
    -0.49
    嬉しいです
    -0.47
     porque
    -0.46
     E
    -0.46
     racism
    -0.46
     Facebook
    -0.46
     Wegen
    -0.46
    POSITIVE LOGITS
    apimachinery
    1.10
     durchaus
    1.00
     justement
    0.99
     nemlig
    0.99
     весьма
    0.99
     ciertamente
    0.98
     destul
    0.97
     oftentimes
    0.97
     wszel
    0.96
     whatsoever
    0.94
    Act Density 4.308%

    No Known Activations