INDEX

Explanations

computes, compares, applies methods

New Auto-Interp

Configuration

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

）.

1.25

<unused763>

1.14

<unused1962>

1.14

 GTEST

1.12

<unused292>

1.12

<unused1872>

1.11

⚋

1.08

profiles

1.08

<unused1089>

1.08

 adegu

1.08

POSITIVE LOGITS

1.05

 continues

0.90

 responded

0.88

 continue

0.86

has

0.86

 continued

0.84

 have

0.84

ha

0.83

 verb

0.79

je

0.78

Activations Density 0.014%