INDEX

Explanations

editor's notes within the text

oai_token-act-pair · gpt-3.5-turbo

editorial notes and corrections in 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pload

-0.75

 manif

-0.74

ŃĶ

-0.69

 scrim

-0.65

ĪĴ

-0.65

 decom

-0.63

forms

-0.63

 sewage

-0.62

ãĥĩ

-0.62

 impe

-0.62

POSITIVE LOGITS

:]

0.78

 HuffPost

0.75

 note

0.75

BOOK

0.73

 Editors

0.73

 Edited

0.71

 Corrections

0.70

":"","

0.70

 NOTE

0.69

Activations Density 0.025%

editor's notes within the text

editorial notes and corrections in text

No Comments

No Known Activations