INDEX
    Explanations

    phrases related to starting something from the beginning

    phrases indicating the concept of building or creating something from the beginning

    New Auto-Interp
    Negative Logits
    irm
    -0.70
    olor
    -0.69
    ghai
    -0.66
    ctic
    -0.65
    linger
    -0.64
    swe
    -0.63
    leneck
    -0.62
    olson
    -0.62
    rav
    -0.62
    amount
    -0.62
    POSITIVE LOGITS
     scratch
    1.87
     scraps
    1.14
     afar
    1.07
     ashes
    0.91
     inception
    0.89
     infancy
    0.85
     conception
    0.85
     rubble
    0.84
     seeds
    0.82
     cradle
    0.80
    Act Density 0.150%

    No Known Activations