INDEX

Explanations

phrases related to illustration or illustrative work

oai_token-act-pair · gpt-3.5-turbo

references to illustration and illustrators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

inki

-0.70

ega

-0.69

etime

-0.68

olics

-0.65

 opio

-0.64

UDP

-0.64

hold

-0.64

ishops

-0.63

acia

-0.62

onte

-0.62

POSITIVE LOGITS

 illustration

1.17

Artist

1.13

 illustrations

1.07

 depicting

1.05

 illustrating

1.04

 Illust

1.00

 illust

0.98

 sketches

0.97

rator

0.96

 drawings

0.90

Activations Density 0.074%

phrases related to illustration or illustrative work

references to illustration and illustrators

No Comments

No Known Activations