INDEX

Explanations

Proper nouns representing locations or organizations

oai_token-act-pair · gpt-3.5-turbo

references to "New York"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ONSORED

-0.83

ï¸ı

-0.80

pless

-0.79

uca

-0.74

ILA

-0.71

ascript

-0.69

abal

-0.68

OPA

-0.68

misc

-0.67

POSITIVE LOGITS

 Zealand

1.40

 York

1.32

 Orleans

1.25

bies

1.16

 Yorkers

1.14

bie

1.11

 Hampshire

1.09

 Yorker

1.07

 Testament

1.02

York

0.98

Activations Density 0.061%

Proper nouns representing locations or organizations

references to "New York"

No Comments

No Known Activations