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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 nonpartisan

-0.68

ICLE

-0.67

 Gutenberg

-0.65

 centrist

-0.64

 Libertarian

-0.63

 Conservative

-0.63

 Tobias

-0.63

 POLIT

-0.62

 Luther

-0.62

 sclerosis

-0.61

POSITIVE LOGITS

xia

0.75

ovi

0.74

zzo

0.73

ese

0.73

ixel

0.73

park

0.72

urs

0.69

bone

0.69

itely

0.69

Â¨

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.