INDEX

Explanations

references to characters who take on protective roles within their families or relationships

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

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__res-sm_processed/4-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.4.hook_resid_post

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

•probe features

women

No Comments

Negative Logits

Īĺ

-3.47

*/

-3.40

\]\].

-2.99

ĸ

-2.96

ĸ´

-2.86

¯

-2.81

*/

-2.80

↵↵

-2.79

↵

-2.79

POSITIVE LOGITS

 herself

4.34

her

4.30

she

3.82

 actress

3.37

 feminist

3.21

She

3.15

 woman

3.15

 actresses

3.12

 daughter

3.08

woman

2.97

Activations Density 2.448%

references to characters who take on protective roles within their families or relationships

No Comments

No Known Activations