INDEX

Explanations

proper nouns, specifically names of Chinese individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals, particularly those with the surname "Zhang."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 parole

-0.73

*/(

-0.71

ally

-0.70

 Hitchcock

-0.69

 replay

-0.65

 Normandy

-0.65

 Essex

-0.64

 Arkham

-0.64

 msec

-0.64

ords

-0.63

POSITIVE LOGITS

 Xiang

1.38

Xia

1.35

wei

1.35

jiang

1.34

 Jing

1.33

 Yuan

1.33

Cao

1.32

Bei

1.27

 Guang

1.24

 Tian

1.23

Activations Density 0.048%

proper nouns, specifically names of Chinese individuals

references to individuals, particularly those with the surname "Zhang."

No Comments

No Known Activations