INDEX

Explanations

image captions mentioning news agencies like EPA, AFP, AP, and Reuters

oai_token-act-pair · gpt-3.5-turbo

references to news agencies and image credits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 SHALL

-0.83

Reviewer

-0.76

ailable

-0.73

represented

-0.69

 appendix

-0.68

abiding

-0.67

osponsors

-0.64

 hijacked

-0.63

crafted

-0.63

FML

-0.62

POSITIVE LOGITS

 Images

1.13

 Image

1.07

 IMAGES

0.94

 Photos

0.91

 Features

0.85

allery

0.83

Images

0.82

 Shutterstock

0.79

0.78

 images

0.74

Activations Density 0.032%

image captions mentioning news agencies like EPA, AFP, AP, and Reuters

references to news agencies and image credits

No Comments

No Known Activations