INDEX

Explanations

detailed location-specific information

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of places and geographical features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 physic

-0.79

OPLE

-0.77

*/(

-0.73

 lawy

-0.72

 sanity

-0.72

 perfection

-0.72

quickShipAvailable

-0.72

ãĥīãĥ©

-0.72

 haste

-0.71

 extraord

-0.71

POSITIVE LOGITS

 River

1.11

 Hotel

1.00

 Corridor

0.99

 Mountains

0.95

 Centre

0.90

 Peninsula

0.89

 Hills

0.88

 Museum

0.87

 District

0.87

hend

0.86

Activations Density 0.240%

detailed location-specific information

proper nouns, particularly names of places and geographical features

No Comments

No Known Activations