INDEX
Negative Logits
o
1.22
en
0.93
]
0.85
л
0.84
}
0.82
n
0.80
н
0.74
в
0.74
و
0.74
ය
0.74
POSITIVE LOGITS
brass
1.26
Brass
1.05
brass
1.02
Brass
0.96
Gujarati
0.76
ೋಜನ
0.74
ی
0.74
唢
0.74
uram
0.73
াবাদের
0.71
Activations Density 0.001%
o
en
]
л
}
n
н
в
و
ය
brass
Brass
brass
Brass
Gujarati
ೋಜನ
ی
唢
uram
াবাদের