INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    blockSize
    0.47
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.44
    <unused1864>
    0.41
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.41
     индивидуа
    0.41
     ե
    0.40
    <unused678>
    0.40
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.40
     Forensic
    0.39
    ↵↵↵↵↵↵↵↵↵↵↵↵↵↵↵
    0.39
    POSITIVE LOGITS
     kisses
    0.53
     translations
    0.50
     blanks
    0.50
     hangers
    0.47
     permutations
    0.47
     throws
    0.46
     fanc
    0.46
     triples
    0.46
     kits
    0.45
     hooks
    0.45
    Act Density 0.001%

    No Known Activations