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

ãģĨ

-0.79

Cth

-0.68

ANG

-0.64

Frameworks

-0.64

ï¸

-0.63

HMS

-0.63

Ack

-0.63

BG

-0.62

 Lyndon

-0.61

POS

-0.61

POSITIVE LOGITS

ri

1.21

arnaev

0.93

negie

0.92

azeera

0.84

eele

0.76

bledon

0.76

ris

0.76

rina

0.75

rano

0.73

urst

0.73

Activations Density 0.000%

No Known Activations

This feature has no known activations.