INDEX
    Explanations

    words related to high-pitched sounds

    words related to squirming or discomfort

    New Auto-Interp
    Negative Logits
    burst
    -0.84
    ially
    -0.79
    rained
    -0.76
    rons
    -0.75
    rain
    -0.74
    ora
    -0.73
    irez
    -0.72
    abad
    -0.70
     srf
    -0.69
    rav
    -0.67
    POSITIVE LOGITS
    nces
    0.84
    agin
    0.83
    teenth
    0.75
     sque
    0.70
    ential
    0.70
    amus
    0.68
    sburgh
    0.67
    Snake
    0.67
     resil
    0.66
    PUT
    0.65
    Act Density 0.021%

    No Known Activations