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

vertisement

-0.73

 conclud

-0.71

 voic

-0.69

 distingu

-0.68

utenberg

-0.66

eele

-0.65

 fever

-0.64

onential

-0.64

 internship

-0.63

 feather

-0.63

POSITIVE LOGITS

TIT

0.79

Winged

0.76

Gam

0.76

TAG

0.76

Corp

0.74

soDeliveryDate

0.70

Gene

0.69

Prot

0.68

Cod

0.68

GMT

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.