INDEX

Explanations

names of places or people starting with the letter "W"

oai_token-act-pair · gpt-3.5-turbo

end-of-text markers, signifying the conclusion of a document or passage

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

 unpre

-0.80

 Prelude

-0.73

 bottleneck

-0.68

 flares

-0.66

 gratification

-0.64

 manifold

-0.64

 apprehension

-0.63

psy

-0.63

 Metallic

-0.62

 locality

-0.62

POSITIVE LOGITS

ITNESS

1.33

restling

1.29

orthy

1.28

alking

1.25

OW

1.25

atson

1.24

ORD

1.24

esley

1.24

ITCH

1.24

ALK

1.23

Activations Density 0.033%

names of places or people starting with the letter "W"

end-of-text markers, signifying the conclusion of a document or passage

No Comments

No Known Activations