INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Kuala

-0.79

 Senegal

-0.77

 Maurit

-0.77

 Malaysia

-0.71

HM

-0.70

 Emirates

-0.68

 Tasmania

-0.67

 Scotland

-0.65

 Boat

-0.63

 Sierra

-0.63

POSITIVE LOGITS

rating

0.80

roman

0.78

zsche

0.76

acly

0.76

roots

0.75

uder

0.75

hess

0.72

erker

0.71

xus

0.70

geoning

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.