INDEX

Explanations

instances of people or entities being subjected to various actions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "subjected" and related concepts of oppression or experiencing something undesirable

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Helsinki

-0.78

andel

-0.72

leaf

-0.71

ARE

-0.70

 Maker

-0.67

fly

-0.66

 Clover

-0.65

wi

-0.65

 Zurich

-0.64

 Kramer

-0.63

POSITIVE LOGITS

 subjected

0.99

ynamic

0.92

Reviewer

0.87

xual

0.84

untarily

0.83

ĸļ

0.83

ivity

0.81

ĪĴ

0.80

isson

0.80

ivities

0.79

Activations Density 0.010%

instances of people or entities being subjected to various actions

instances of the word "subjected" and related concepts of oppression or experiencing something undesirable

No Comments

No Known Activations