INDEX

Explanations

phrases regarding permissible actions within a legal context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ľ

-1.63

ĨĴ

-1.61

®

-1.59

ľĵ

-1.57

![**

-1.51

Ĺ

-1.50

İ

-1.48

'</

-1.43

Ļª

-1.43

arer

-1.42

POSITIVE LOGITS

 alone

1.54

fulness

1.44

 fields

1.41

feit

1.40

ng

1.36

 country

1.32

 debt

1.29

 ourselves

1.29

 domains

1.28

 field

1.28

Activations Density 0.023%

phrases regarding permissible actions within a legal context

No Comments

No Known Activations