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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 obser

-0.72

 caut

-0.67

 proble

-0.66

 dilig

-0.63

cci

-0.63

 ende

-0.63

 conclud

-0.62

lihood

-0.61

esley

-0.61

"]=>

-0.60

POSITIVE LOGITS

Games

0.83

itol

0.76

ivo

0.72

rix

0.71

URN

0.68

Split

0.66

bats

0.66

udos

0.65

doors

0.64

LAB

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.