INDEX

Explanations

phrases related to legal actions or consequences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating punishment or consequences related to actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

obyl

-0.74

venants

-0.72

ires

-0.68

å£«

-0.68

%%

-0.66

DragonMagazine

-0.66

atl

-0.65

eda

-0.63

ocity

-0.62

atar

-0.62

POSITIVE LOGITS

 refusing

1.33

 violating

1.30

 daring

1.23

 failing

1.20

 breaching

1.17

 exercising

1.12

 interfering

1.11

gery

1.11

 possessing

1.09

 criticizing

1.09

Activations Density 0.133%

phrases related to legal actions or consequences

phrases indicating punishment or consequences related to actions

No Comments

No Known Activations