INDEX

Explanations

names or mentions of individuals

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

alysed

-0.75

 apex

-0.73

inary

-0.68

align

-0.66

ĻĤ

-0.65

=>

-0.64

worldly

-0.63

cms

-0.63

polit

-0.63

ANA

-0.62

POSITIVE LOGITS

 Andersen

1.13

 Johnson

1.08

 Caldwell

1.07

 Ventura

1.05

 Greene

1.05

 Thompson

1.03

 Howell

1.03

 Shields

1.03

 Welch

1.03

 Roberts

1.02

Activations Density 0.042%

names or mentions of individuals

No Comments

No Known Activations