INDEX

Explanations

specific patterns of characters and symbols

oai_token-act-pair · gpt-3.5-turbo

specific symbols or characters representing geographic entities and political designations

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

Assembly

-0.71

 horizont

-0.71

 bearer

-0.68

WARD

-0.67

HAM

-0.66

itarian

-0.65

 WARN

-0.64

BuyableInstoreAndOnline

-0.64

MET

-0.63

DragonMagazine

-0.62

POSITIVE LOGITS

ħ

1.52

Į

1.50

©

1.48

ĺ

1.47

Ľ

1.44

Ļ

1.43

Ī

1.43

Ĭ

1.43

ľ

1.42

Ĵ

1.41

Activations Density 0.019%

specific patterns of characters and symbols

specific symbols or characters representing geographic entities and political designations

No Comments

No Known Activations