INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

bid

-0.70

wcsstore

-0.65

Thumbnail

-0.65

Beta

-0.64

athi

-0.63

erest

-0.63

illi

-0.63

soDeliveryDate

-0.62

thumbnails

-0.62

alli

-0.62

POSITIVE LOGITS

 territ

0.87

eous

0.70

 bang

0.70

insula

0.66

 looph

0.63

 intest

0.62

adish

0.62

TY

0.61

 ignition

0.60

APH

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.