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

Not in Any Lists

No Comments

Negative Logits

gomery

-0.83

flix

-0.75

venants

-0.74

arnaev

-0.73

emis

-0.71

uminati

-0.69

 Jonas

-0.66

ussie

-0.65

 Roose

-0.64

essee

-0.63

POSITIVE LOGITS

BACK

0.73

Altern

0.73

Repe

0.73

BSD

0.72

Poll

0.70

Seg

0.69

Bur

0.68

isson

0.66

Arg

0.66

back

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.