INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

helm

-0.71

Mount

-0.69

ãģ¦

-0.68

WAYS

-0.65

IDER

-0.63

 keyboards

-0.63

AIR

-0.62

fleet

-0.62

Contents

-0.61

Steel

-0.61

POSITIVE LOGITS

NetMessage

0.75

olt

0.74

nian

0.68

itsch

0.66

zech

0.65

Ã¡s

0.62

 Witnesses

0.61

atum

0.61

 Whedon

0.61

ynes

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.