INDEX
    Explanations

    quantitative measurements and data in various units

    New Auto-Interp
    Negative Logits
    ighton
    -0.18
     Malk
    -0.14
    .yahoo
    -0.14
     Dos
    -0.14
    amus
    -0.14
     Pine
    -0.14
     Saw
    -0.14
     incoming
    -0.13
     em
    -0.13
     Danger
    -0.13
    POSITIVE LOGITS
    bsub
    0.19
    ahoma
    0.17
    tracks
    0.16
    strup
    0.16
    osten
    0.15
    untu
    0.15
    agas
    0.15
    γά
    0.14
    ëĬ
    0.14
    à¸ķร
    0.14
    Act Density 0.051%

    No Known Activations