INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

aeda

-0.88

arte

-0.78

vine

-0.75

 Kamp

-0.70

abe

-0.68

={

-0.67

ande

-0.67

ihara

-0.66

 Lanc

-0.66

Kod

-0.64

POSITIVE LOGITS

agogue

0.68

 nodd

0.66

 Sabbath

0.66

 contradictory

0.65

 accommodations

0.63

 toile

0.62

 scripture

0.61

£ı

0.60

.>>

0.59

 inline

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.