INDEX

Explanations

images or photos in a document

oai_token-act-pair · gpt-3.5-turbo

references to image captions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

merce

-0.90

laws

-0.86

etheless

-0.84

ighter

-0.79

uckle

-0.78

unts

-0.77

unin

-0.76

manship

-0.74

rieg

-0.74

bye

-0.74

POSITIVE LOGITS

 caption

1.08

 Image

1.04

 Gallery

0.93

 Collection

0.88

gallery

0.86

 Album

0.86

 Comics

0.85

 Images

0.84

 Collections

0.84

 Thumbnails

0.83

Activations Density 0.014%

images or photos in a document

references to image captions

No Comments

No Known Activations