INDEX

Explanations

titles or headings marked with special characters or symbols followed by a keyword or phrase

oai_token-act-pair · gpt-3.5-turbo

titles of reports and documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 range

-0.87

 infring

-0.71

 sear

-0.69

creen

-0.68

theless

-0.67

 anyway

-0.67

 fragmentation

-0.64

 scale

-0.64

 regardless

-0.62

 continuum

-0.62

POSITIVE LOGITS

ª

1.51

¹

1.34

Ĵ

1.31

ł

1.30

ı

1.29

¡

1.25

Ķ

1.24

³

1.24

ĳ

1.24

«

1.23

Activations Density 0.147%

titles or headings marked with special characters or symbols followed by a keyword or phrase

titles of reports and documents

No Comments

No Known Activations