INDEX
    Explanations

    Code repository links

    The neuron fires on occurrences of “lib” in names (e.g. library, balenalib, base-images), i.e. tokens indicating library or image package references.

    New Auto-Interp
    Negative Logits
     bốn
    -0.06
     Zion
    -0.06
    中國
    -0.06
     elastic
    -0.06
     mind
    -0.06
    welcome
    -0.06
    .Tween
    -0.06
    drawer
    -0.06
     multiplier
    -0.06
     Mid
    -0.06
    POSITIVE LOGITS
     aseg
    0.07
    Profit
    0.07
     roku
    0.07
    olutions
    0.07
     nedost
    0.06
    0.06
     zg
    0.06
    _pemb
    0.06
    CENT
    0.06
    قرار
    0.06
    Act Density 0.018%

    No Known Activations