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

illance

-0.79

ettings

-0.79

anium

-0.74

 comprom

-0.73

berus

-0.72

oute

-0.70

rification

-0.69

ufact

-0.69

chnology

-0.68

hower

-0.67

POSITIVE LOGITS

MAT

0.77

FACE

0.69

à¤

0.67

Å

0.67

Kat

0.67

XY

0.66

Å

0.64

Zak

0.64

âĢ

0.63

MN

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.