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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IE

-0.74

SPONSORED

-0.74

hing

-0.73

ciating

-0.70

ength

-0.70

ighty

-0.67

gements

-0.65

tarian

-0.64

izzle

-0.64

ney

-0.63

POSITIVE LOGITS

Pwr

0.90

 safer

0.69

better

0.67

chance

0.66

BuyableInstoreAndOnline

0.64

hu

0.64

æĸ¹

0.63

haus

0.63

worthy

0.62

çİĭ

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.