INDEX

Explanations

mathematical calculations and word problems

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

которые

1.43

ční

1.30

KindOfClass

1.28

埇

1.27

𝑮

1.25

čný

1.25

𝚝

1.23

ത്

1.22

обходимо

1.22

țe

1.22

POSITIVE LOGITS

8.36

ات

5.40

sü

3.99

्स

3.93

sman

3.91

sion

3.86

sPath

3.81

ים

3.74

ség

3.72

sG

3.69

Activations Density 0.947%