INDEX

Explanations

names of countries or regions

oai_token-act-pair · gpt-3.5-turbo

the use of the definite article "the" in various contexts

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

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

aunt

-0.94

addle

-0.91

udder

-0.87

ebook

-0.86

eno

-0.83

ritch

-0.80

eson

-0.79

ayne

-0.78

oops

-0.78

Keefe

-0.77

POSITIVE LOGITS

 Ministry

1.13

 situation

1.08

 DPRK

1.05

 municipality

1.05

 Azerbai

0.99

 USSR

0.97

 pretext

0.95

 afore

0.95

 inhabitants

0.95

 ministry

0.94

Activations Density 0.492%

names of countries or regions

the use of the definite article "the" in various contexts

No Comments

No Known Activations