INDEX

Explanations

specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

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

DragonMagazine

-1.00

oeuv

-0.69

ãĤ¤ãĥĪ

-0.69

 metab

-0.69

 behavi

-0.68

EStreamFrame

-0.67

ROR

-0.66

ãĥ¼ãĥĨ

-0.65

aper

-0.65

 PRESS

-0.64

POSITIVE LOGITS

 Nebraska

1.63

 Illinois

1.59

 Idaho

1.57

 Tennessee

1.57

 Nevada

1.57

 Wisconsin

1.53

 Wyoming

1.53

 Maryland

1.53

 Rhode

1.52

 Vermont

1.52

Activations Density 0.111%

specific geographical locations

No Comments

No Known Activations