INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

èĢ

-0.82

ISA

-0.80

åį

-0.73

./

-0.69

é

-0.69

æ³

-0.68

:\

-0.68

æĹ

-0.67

ãģį

-0.67

åĲ

-0.67

POSITIVE LOGITS

 Horses

0.77

eele

0.76

 Thrones

0.73

 podcasts

0.69

 Myster

0.69

ebook

0.68

imore

0.65

 Syndicate

0.64

 vans

0.64

aughs

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.