INDEX

Explanations

headings or titles in a text

oai_token-act-pair · gpt-3.5-turbo

punctuation marks and references in a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

hene

-0.79

orney

-0.72

earchers

-0.72

abwe

-0.71

llah

-0.71

undai

-0.70

 assurance

-0.69

Ã¼r

-0.68

urses

-0.68

MPH

-0.68

POSITIVE LOGITS

Characters

1.40

Trivia

1.38

pmwiki

1.31

Contents

1.22

Appearances

1.20

Gallery

1.19

rawdownloadcloneembedreportprint

1.18

âĶĢâĶĢâĶĢâĶĢ

1.18

References

1.12

Examples

1.06

Activations Density 0.302%

headings or titles in a text

punctuation marks and references in a text

No Comments

No Known Activations