INDEX

Explanations

deep or shallow depth

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 albeit

-0.77

 نہ

-0.75

あるので

-0.74

wijs

-0.73

 poised

-0.73

創

-0.73

ണം

-0.73

 backside

-0.72

vish

-0.72

 envisioned

-0.72

POSITIVE LOGITS

 deep

1.03

Deep

1.02

 Deep

0.98

 derin

0.97

Shallow

0.91

 depth

0.91

depth

0.90

deep

0.88

 shallow

0.87

Depth

0.85

Activations Density 0.018%