INDEX
    Explanations

    expressions of emotional depth and complexity in character descriptions

    New Auto-Interp
    Negative Logits
    elters
    -0.16
     tame
    -0.15
    ertools
    -0.15
    vae
    -0.15
    ائر
    -0.14
    leground
    -0.14
    .ind
    -0.13
    omid
    -0.13
    vas
    -0.13
    nen
    -0.13
    POSITIVE LOGITS
    -Clause
    0.14
    íĮĮ
    0.13
    ichern
    0.13
    à¸Ľà¸£à¸°à¸¡
    0.13
    isto
    0.13
    oom
    0.13
    emple
    0.13
    igar
    0.13
    ê²½
    0.13
    ãĥĬãĥ¼
    0.13
    Act Density 0.132%

    No Known Activations