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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Scroll

-0.86

ILA

-0.73

 LEDs

-0.68

Pi

-0.66

 colour

-0.65

 PHOTO

-0.65

Ãī

-0.64

RGB

-0.62

VIDEO

-0.62

Cro

-0.62

POSITIVE LOGITS

ruck

0.78

commit

0.71

Yel

0.66

arus

0.66

ilib

0.64

deal

0.63

 Truck

0.63

 experien

0.62

ucker

0.62

 ende

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.