INDEX

Explanations

places where images are being mentioned or displayed

oai_token-act-pair · gpt-3.5-turbo

references to images and their sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

bid

-0.77

iqueness

-0.76

pport

-0.68

chance

-0.66

mington

-0.65

rency

-0.63

issance

-0.62

cerning

-0.62

phas

-0.62

 souls

-0.61

POSITIVE LOGITS

 Shutterstock

1.14

photos

0.86

 Courtesy

0.85

\/\/

0.82

 enlarge

0.79

 Â©

0.79

 screenshot

0.78

 Flickr

0.78

upload

0.76

 Photograph

0.76

Activations Density 0.093%

places where images are being mentioned or displayed

references to images and their sources

No Comments

No Known Activations