INDEX

Explanations

words related to clothing items, specifically hoods

oai_token-act-pair · gpt-3.5-turbo

references to hoods or hooded garments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ngth

-0.80

andum

-0.78

lihood

-0.78

FORM

-0.78

tery

-0.73

MENT

-0.69

×ķ

-0.67

yond

-0.66

TABLE

-0.65

utics

-0.65

POSITIVE LOGITS

oos

1.11

ie

1.04

sie

1.00

ed

0.98

oo

0.94

ies

0.94

tip

0.93

 ornament

0.93

elled

0.92

edo

0.86

Activations Density 0.057%

words related to clothing items, specifically hoods

references to hoods or hooded garments

No Comments

No Known Activations