INDEX

Explanations

past conventional practices

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

虓

-2.59

dekor

-2.20

jadikan

-2.11

nossa

-1.94

 قدیمی

-1.88

 after

-1.87

 leaſt

-1.87

鵟

-1.87

ⵌ

-1.85

fotograf

-1.84

POSITIVE LOGITS

↵

3.00

si

2.34

2.13

ti

2.09

ts

2.06

ad

1.98

 больших

1.91

op

1.89

ist

1.88

er

1.86

Activations Density 0.017%