INDEX

Explanations

from python, from prior

New Auto-Interp

Configuration

Prompts (Dashboard)

238,145 prompts, 512 tokens each

Dataset (Dashboard)

lmsys + oasst1

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

↵↵

3.43

3.17

↵

3.17

↵↵↵

2.79

。

2.71

.]

2.54

↵↵↵↵

2.53

।

2.39

.}

2.34

].

2.33

POSITIVE LOGITS

<unused291>

1.55

<unused2102>

1.50

<unused491>

1.50

τώ

1.44

<unused1782>

1.42

ولز

1.41

 Gemä

1.40

<unused1876>

1.38

<unused328>

1.37

 いく

1.35

Activations Density 0.541%