INDEX

Explanations

images with captions to "Enlarge" and "toggle"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

apo

-0.72

naires

-0.71

bats

-0.68

termination

-0.62

 Schwar

-0.61

 fronts

-0.59

Sund

-0.58

gran

-0.57

 rang

-0.55

OWS

-0.55

POSITIVE LOGITS

 image

0.90

 toggle

0.87

 ARTICLE

0.85

ption

0.83

 screenshot

0.76

 photo

0.73

 slide

0.72

 thumbnail

0.72

 microscope

0.71

 Image

0.69

Activations Density 0.014%

images with captions to "Enlarge" and "toggle"

No Comments

No Known Activations