INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

Hidden

-0.86

Enlarge

-0.69

Pand

-0.69

Discover

-0.68

OLOGY

-0.65

 Pandora

-0.65

 lapse

-0.64

Effective

-0.64

CEPT

-0.64

TPS

-0.63

POSITIVE LOGITS

iance

0.74

pled

0.69

 ration

0.66

 bidding

0.65

elson

0.65

hip

0.63

endor

0.63

iaz

0.62

auer

0.61

arde

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.