INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    :]:
    1.49
    .]:
    1.45
    :");
    1.42
    .):
    1.39
    .").
    1.33
    %:
    1.32
    ...).
    1.31
    ?).
    1.29
    …).
    1.29
    \":
    1.29
    POSITIVE LOGITS
    ↵↵
    3.17
    2.99
    ↵↵↵
    2.40
    ↵↵↵↵
    1.96
    ↵↵↵↵↵
    1.82
    1.72
    1.61
    )
    1.61
    1.59
    </h2>
    1.49
    Act Density 1.515%

    No Known Activations