INDEX

Explanations

No Explanations Found

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

<eos>

2.37

↵↵

2.04

<start_of_image>

1.83

<strong>

1.78

↵↵↵

1.69

<b>

1.58

↵↵↵↵

1.49

1.47

<em>

1.44

1.42

POSITIVE LOGITS

<unused1316>

2.28

<unused1324>

2.26

<unused1520>

2.24

<unused1398>

2.23

<unused1322>

2.23

<unused1525>

2.23

<unused1293>

2.22

<unused1517>

2.22

<unused1291>

2.22

<unused1333>

2.22

Activations Density 0.164%

No Known Activations