INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ADVERTISEMENT

-0.80

CHA

-0.73

erto

-0.70

population

-0.70

SPONSORED

-0.69

JUST

-0.69

IST

-0.69

DEV

-0.67

STAR

-0.67

foreseen

-0.67

POSITIVE LOGITS

ãĤ´ãĥ³

0.67

xp

0.64

Xer

0.61

aptic

0.60

azeera

0.59

 oxide

0.59

wip

0.59

enes

0.58

 tablets

0.58

ox

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.