INDEX

Explanations

words related to illustrations or illustrators

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to illustrations and illustrators

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

etime

-0.75

ega

-0.75

ecast

-0.72

rown

-0.72

inki

-0.71

ignty

-0.70

unker

-0.69

 opio

-0.69

bies

-0.69

orne

-0.68

POSITIVE LOGITS

 illustration

1.22

 illustrations

1.08

 illust

1.03

 Illust

1.00

 illustrating

0.97

Artist

0.90

rator

0.87

 depicting

0.85

 Caption

0.81

 depicts

0.79

Activations Density 0.030%

words related to illustrations or illustrators

words and phrases related to illustrations and illustrators

No Comments

No Known Activations