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

inos

-0.78

pires

-0.70

terday

-0.65

 superpower

-0.63

 passions

-0.62

 palms

-0.61

 automobile

-0.60

 Motors

-0.59

bun

-0.59

anan

-0.58

POSITIVE LOGITS

CSS

0.73

RFC

0.71

 Lans

0.71

ME

0.69

MW

0.69

arling

0.69

SPD

0.69

ARM

0.67

RAG

0.66

;;;;;;;;

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.