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

endas

-0.74

eva

-0.70

 screwed

-0.67

nings

-0.66

lax

-0.65

tor

-0.64

 sleeper

-0.63

 basin

-0.63

ard

-0.63

elman

-0.63

POSITIVE LOGITS

ULT

0.74

###

0.71

 Wanted

0.68

 Cave

0.67

İĭ

0.66

 weeds

0.66

KY

0.65

 Bees

0.64

alky

0.64

 Transcript

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.