INDEX

Explanations

expressions of a high degree of approval or quality

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

ctigges/pythia-70m-deduped__mlp-sm_processed/0-mlp-sm

Prompts (Dashboard)

32,768 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

torch.float32

Hook Name

blocks.0.hook_mlp_out

Hook Layer

Architecture

standard

Context Size

128

Dataset

EleutherAI/the_pile_deduplicated

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

"?

-1.57

)?

-1.52

){

-1.51

ostics

-1.41

ated

-1.41

~\]

-1.40

·¸

-1.39

chÃ©

-1.37

Ã¤sident

-1.35

usted

-1.34

POSITIVE LOGITS

ascus

1.77

aceae

1.70

alus

1.47

ium

1.40

iana

1.39

charts

1.39

arman

1.38

abase

1.37

eri

1.33

worth

1.32

Activations Density 0.017%

expressions of a high degree of approval or quality

No Known Activations