INDEX

Explanations

proper nouns, likely related to specific locations or organizations

oai_token-act-pair · gpt-3.5-turbo

place names 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 @ 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

onymous

-0.78

angible

-0.72

orders

-0.72

utive

-0.69

Ò

-0.69

ãĥĻ

-0.68

ãģĨ

-0.68

luster

-0.67

atorial

-0.66

aphael

-0.65

POSITIVE LOGITS

 Resort

1.40

 Farms

1.29

 Cemetery

1.26

 Elementary

1.21

 Lodge

1.17

 Armory

1.17

 Casino

1.16

 Ranch

1.15

 Neighborhood

1.15

 Mall

1.13

Activations Density 0.151%

proper nouns, likely related to specific locations or organizations

place names and geographical features

No Comments

No Known Activations