INDEX

Explanations

photo captions and associated press images

oai_token-act-pair · gpt-3.5-turbo

mentions of images and photographs associated with articles

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

onom

-0.86

uca

-0.81

omics

-0.75

staking

-0.69

cius

-0.64

olon

-0.63

capital

-0.62

mong

-0.62

jab

-0.62

EUR

-0.61

POSITIVE LOGITS

################################

0.81

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

0.72

ROR

0.69

ramid

0.68

 Mehran

0.67

Gaza

0.67

actionDate

0.67

ĻĤ

0.66

myra

0.66

thumbnails

0.65

Activations Density 0.055%

photo captions and associated press images

mentions of images and photographs associated with articles

No Comments

No Known Activations