INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Â¥

-0.66

aith

-0.66

mith

-0.65

kered

-0.60

Ŀ

-0.59

ħĭ

-0.58

)—

-0.58

00200000

-0.58

 perfect

-0.58

 brim

-0.57

POSITIVE LOGITS

ĨĴ

0.71

Interstitial

0.70

ONE

0.69

Russ

0.66

swick

0.62

scripts

0.61

rams

0.61

politics

0.61

oner

0.60

OVA

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.