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

night

-0.78

 Nightmares

-0.75

 Ples

-0.71

 Predators

-0.66

 booth

-0.64

 Lovecraft

-0.63

ctuary

-0.63

NO

-0.62

 NIGHT

-0.61

 DARK

-0.60

POSITIVE LOGITS

phabet

1.03

ollow

0.66

 behavi

0.65

SPONSORED

0.65

 tradem

0.64

baugh

0.64

 motivational

0.62

Ctrl

0.62

gall

0.62

arton

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.