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

inson

-0.81

ombie

-0.76

inous

-0.75

efer

-0.72

haar

-0.71

encrypted

-0.68

ioned

-0.67

disabled

-0.66

emis

-0.65

chain

-0.65

POSITIVE LOGITS

Reviewer

0.74

âĢ

0.70

Newsletter

0.68

 tasting

0.65

 Brus

0.64

 Atmosp

0.64

VERTISEMENT

0.63

comings

0.62

 coworkers

0.62

alon

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.