INDEX
Explanations
actions of subjects concluding or changing
New Auto-Interp
Negative Logits
身
0.73
டிப்ப
0.67
宓
0.67
listen
0.66
Waver
0.66
wake
0.66
вста
0.65
sling
0.64
Wish
0.64
नमस्ते
0.64
POSITIVE LOGITS
repaired
0.92
reparations
0.90
retired
0.88
retiring
0.86
retired
0.82
repairing
0.81
বিভক্ত
0.80
collapsing
0.78
retires
0.78
分裂
0.78
Activations Density 0.006%