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

Ö

-0.75

×©

-0.70

á½

-0.70

pson

-0.68

 virginity

-0.66

Ú

-0.65

ãģĭ

-0.65

MEP

-0.64

roit

-0.63

roud

-0.63

POSITIVE LOGITS

iations

0.73

ffen

0.70

berto

0.68

 Hurt

0.68

 collaborations

0.64

ussions

0.64

wx

0.62

inventory

0.62

ussed

0.62

 Dimension

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.