INDEX

Explanations

topics, issues, themes, areas, tools, factors, values, techniques

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

。

2.53

2.45

2.40

；

2.20

।

2.10

（

2.09

2.04

 when

1.97

，

1.94

POSITIVE LOGITS

ıldığı

2.04

<unused1852>

1.91

胧

1.78

<unused1693>

1.78

<unused1728>

1.76

<unused187>

1.75

לחמת

1.73

어진

1.73

шенные

1.73

raš

1.72

Activations Density 0.245%