INDEX

Explanations

euphemisms related to avoiding explicit language

oai_token-act-pair · gpt-3.5-turbo

words related to euphemisms

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

yrim

-0.84

orem

-0.71

-0.70

 exhibits

-0.69

acea

-0.68

Mp

-0.66

issy

-0.62

aceutical

-0.62

:\

-0.59

DEN

-0.59

POSITIVE LOGITS

 opin

0.76

 Bulg

0.67

oÄŁ

0.66

izons

0.65

checked

0.64

enic

0.64

eners

0.64

itage

0.62

umbn

0.62

 overloaded

0.61

Activations Density 0.000%

euphemisms related to avoiding explicit language

words related to euphemisms

No Comments

No Known Activations