INDEX
Explanations
blocking or standing in the way
New Auto-Interp
Negative Logits
cri
0.47
crippled
0.47
suppressed
0.45
Cri
0.43
Cri
0.40
strangled
0.39
cri
0.39
wr
0.38
Disable
0.38
Among
0.38
POSITIVE LOGITS
obstructing
0.95
blocking
0.95
interposed
0.91
擋
0.90
挡
0.87
standing
0.82
obstruction
0.82
Blocking
0.82
stood
0.81
blocking
0.80
Activations Density 0.011%