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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

berra

-0.93

orno

-0.85

ishops

-0.72

âĦ¢:

-0.72

ansas

-0.70

[+

-0.70

wcsstore

-0.70

atown

-0.69

anse

-0.68

apa

-0.67

POSITIVE LOGITS

 Cassidy

0.64

ty

0.63

 distinction

0.61

 Laura

0.60

Liv

0.59

tie

0.58

 Laurel

0.57

Doc

0.57

lot

0.56

 supper

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.