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

heim

-0.70

flash

-0.70

schild

-0.70

adium

-0.68

gard

-0.67

}{

-0.65

adian

-0.65

Utah

-0.65

imation

-0.64

cock

-0.63

POSITIVE LOGITS

 receptor

0.69

ullah

0.68

NetMessage

0.68

 receptors

0.67

 behavi

0.66

lihood

0.66

 antioxid

0.63

wic

0.63

 incorpor

0.62

ollah

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.