INDEX

Explanations

phrases related to technical or specialized terms like "individual mandate" and "GreenCollar maps"

oai_token-act-pair · gpt-3.5-turbo

references to specific political or legislative terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

âĦ¢:

-0.66

ËĪ

-0.56

 Solitaire

-0.55

 Â®

-0.55

:/

-0.55

 Lent

-0.53

 ãģ

-0.52

 âĸĪ

-0.52

imaru

-0.52

NRS

-0.49

POSITIVE LOGITS

anooga

0.75

ctrl

0.56

lege

0.56

weights

0.55

 practition

0.55

 decision

0.54

change

0.54

oward

0.53

 change

0.53

devices

0.52

Activations Density 0.879%

phrases related to technical or specialized terms like "individual mandate" and "GreenCollar maps"

references to specific political or legislative terms

No Comments

No Known Activations