INDEX

Explanations

proper nouns related to locations, specifically those ending with "-heim"

oai_token-act-pair · gpt-3.5-turbo

mentions of specific place names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

hered

-0.70

 athlet

-0.69

ciating

-0.65

Downloadha

-0.65

VALUE

-0.64

href

-0.64

hetically

-0.63

 medically

-0.63

itu

-0.63

 Logged

-0.62

POSITIVE LOGITS

heim

1.19

stein

1.13

stadt

1.04

ers

1.02

sburg

1.00

shire

1.00

sson

0.91

lich

0.91

roth

0.87

arding

0.82

Activations Density 0.011%

proper nouns related to locations, specifically those ending with "-heim"

mentions of specific place names

No Comments

No Known Activations