INDEX
    Explanations
    New Auto-Interp
    Negative Logits
    .controller
    -0.07
     volunteered
    -0.07
     Oct
    -0.07
    -0.06
     bullets
    -0.06
     mycket
    -0.06
     Blvd
    -0.06
    Extension
    -0.06
     있지만
    -0.06
    -0.06
    POSITIVE LOGITS
     Dep
    0.08
    _THRESH
    0.07
    FB
    0.07
    fight
    0.07
    IDDLE
    0.07
    0.07
    IMP
    0.07
     surely
    0.07
    动力
    0.07
    0.07
    Act Density 0.001%

    No Known Activations