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

TIT

-0.71

 loudspe

-0.71

audi

-0.62

Arg

-0.62

MRI

-0.62

ãĥĺãĥ©

-0.59

Introduced

-0.59

omen

-0.59

soType

-0.58

 subtitle

-0.58

POSITIVE LOGITS

heit

0.78

asonic

0.73

wich

0.70

icial

0.70

pload

0.67

lish

0.65

lee

0.65

bable

0.64

imum

0.64

clair

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.