INDEX
Explanations
visual distortion, blurriness, dancing colors
New Auto-Interp
Negative Logits
خب
0.41
함으로써
0.39
nghiệp
0.38
备份
0.38
rarement
0.38
crou
0.37
मुना
0.37
حوزه
0.37
чать
0.37
ўна
0.37
POSITIVE LOGITS
hallucinations
0.70
illusions
0.69
illusory
0.68
blurry
0.64
blurred
0.62
halluc
0.61
illusion
0.61
distorted
0.60
psychedelic
0.59
distortions
0.59
Activations Density 0.083%