INDEX

Explanations

words related to censorship

oai_token-act-pair · gpt-3.5-turbo

references to censorship

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

Slot

-0.82

»Ĵ

-0.78

waukee

-0.73

RV

-0.72

Var

-0.70

Fit

-0.70

 Moines

-0.69

reci

-0.68

swing

-0.68

Kal

-0.67

POSITIVE LOGITS

 censorship

3.93

 censor

3.15

 cens

2.79

 censored

2.75

cens

2.71

 blacklist

1.52

 repression

1.45

 libel

1.42

 repressive

1.37

 totalitarian

1.33

Activations Density 0.026%

words related to censorship

references to censorship

No Comments

No Known Activations