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

 Azerb

-0.73

eki

-0.73

 Mehran

-0.73

 largeDownload

-0.72

ournal

-0.72

UES

-0.68

ocalyptic

-0.68

 rake

-0.68

oÄŁan

-0.67

opian

-0.66

POSITIVE LOGITS

killers

0.74

legates

0.72

banks

0.70

 Kills

0.70

ricanes

0.70

backer

0.69

builders

0.67

holding

0.66

burn

0.66

cest

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.