INDEX
    Explanations

    color channels

    New Auto-Interp
    Negative Logits
     strain
    -0.08
     criminal
    -0.08
     strains
    -0.08
    -0.07
    -0.07
    Fra
    -0.07
    -0.07
     fraud
    -0.07
     Fraud
    -0.07
     fridge
    -0.07
    POSITIVE LOGITS
     Greens
    0.09
     swimsuit
    0.08
     calloc
    0.08
     bisc
    0.08
     mmap
    0.08
     eina
    0.08
    Stamped
    0.08
     greens
    0.08
    την
    0.08
     møte
    0.08
    Act Density 0.001%

    No Known Activations