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

rix

-0.74

cyl

-0.72

 ourselves

-0.70

 tatt

-0.70

 snipp

-0.64

uctor

-0.63

activ

-0.63

blem

-0.62

cluding

-0.62

pat

-0.62

POSITIVE LOGITS

 Thunderbolt

0.70

 Escape

0.68

ulner

0.63

INGTON

0.62

 Couch

0.59

 Share

0.57

 Cheney

0.57

é¾įåĸļå£«

0.56

ieved

0.56

 Paddock

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.