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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Annotations

-0.79

 sidx

-0.78

SPONSORED

-0.76

urally

-0.74

 Canary

-0.73

uliffe

-0.69

 Commands

-0.69

 Moroc

-0.67

 ACPI

-0.66

 tremend

-0.66

POSITIVE LOGITS

wik

0.80

WP

0.71

bud

0.68

mega

0.68

Wing

0.68

netflix

0.67

ÃŁ

0.66

Ultimate

0.64

DragonMagazine

0.63

chan

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.