INDEX
    Explanations

    The neuron consistently activates on the word stem “teleport” (as in “teleportation”).

    New Auto-Interp
    Negative Logits
     asset
    -0.07
     dataset
    -0.06
     distraction
    -0.06
    fuse
    -0.06
     order
    -0.06
    _User
    -0.06
     wrapper
    -0.06
     branches
    -0.06
     darker
    -0.06
    HOOK
    -0.06
    POSITIVE LOGITS
     teleport
    0.10
    izophren
    0.07
     طلب
    0.07
    :↵↵
    0.07
    ิทยาศาสตร
    0.07
     соврем
    0.07
     manten
    0.07
     sca
    0.06
    .setTime
    0.06
     Про
    0.06
    Act Density 0.001%

    No Known Activations