INDEX

Explanations

phrases related to locations or regions

oai_token-act-pair · gpt-3.5-turbo

expressions indicating global or widespread contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

rats

-0.72

-0.70

potion

-0.69

BILL

-0.68

gee

-0.66

atari

-0.63

rences

-0.62

aciously

-0.60

ãĤº

-0.59

nery

-0.59

POSITIVE LOGITS

 Europe

1.39

 Asia

1.29

 globe

1.18

 Africa

1.17

 America

1.15

 Southeast

1.07

 Australia

1.06

 Britain

1.06

 world

1.05

 Scandinav

1.02

Activations Density 0.114%

phrases related to locations or regions

expressions indicating global or widespread contexts

No Comments

No Known Activations