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

uary

-0.65

 Moff

-0.64

Els

-0.62

 Chamberlain

-0.60

 Camb

-0.60

 Claud

-0.59

bourg

-0.59

 Franco

-0.58

Ely

-0.57

 Dame

-0.57

POSITIVE LOGITS

VEL

0.71

estial

0.71

ratom

0.71

ayer

0.70

\\\\\\\\

0.69

upuncture

0.68

robe

0.68

RECT

0.68

ANE

0.67

ew

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.