INDEX

Explanations

mentions of eyeglasses

oai_token-act-pair · gpt-3.5-turbo

references to glasses

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

ILY

-0.90

itative

-0.72

actionDate

-0.71

Filename

-0.70

heny

-0.68

NOR

-0.68

published

-0.67

TAIN

-0.66

ESE

-0.65

States

-0.63

POSITIVE LOGITS

 glasses

1.47

 goggles

1.17

 sunglasses

1.05

creen

1.00

linger

0.97

lasses

0.94

lace

0.90

glass

0.89

ocular

0.85

 Bottle

0.83

Activations Density 0.009%

mentions of eyeglasses

references to glasses

No Comments

No Known Activations