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

âĵĺ

-0.71

ãĤ§

-0.69

advertisement

-0.67

Ø³

-0.65

olesterol

-0.64

effects

-0.63

lez

-0.63

thumbnails

-0.61

alter

-0.60

CHA

-0.60

POSITIVE LOGITS

hower

0.74

atics

0.73

 gentlemen

0.70

erman

0.68

 warrants

0.66

yip

0.64

ow

0.63

 toler

0.62

bley

0.62

 electing

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.