INDEX

Explanations

references to geographical regions

oai_token-act-pair · gpt-3.5-turbo

references to a specific geographic region

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

glers

-0.70

ument

-0.67

urations

-0.66

gers

-0.63

terson

-0.63

CRIPTION

-0.62

tails

-0.62

bats

-0.62

thood

-0.62

ãĥīãĥ©

-0.61

POSITIVE LOGITS

ally

1.25

als

1.06

ESE

0.98

ality

0.95

naire

0.81

wide

0.80

alf

0.80

 bordering

0.79

ional

0.79

ale

0.78

Activations Density 0.038%

references to geographical regions

references to a specific geographic region

No Comments

No Known Activations