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

netflix

-0.70

addon

-0.68

|--

-0.66

naire

-0.66

haar

-0.66

ttp

-0.65

yrinth

-0.65

igible

-0.65

eworld

-0.65

RPG

-0.64

POSITIVE LOGITS

0.82

OUR

0.75

 presses

0.74

 VIDEOS

0.70

 proble

0.65

DOI

0.64

IDES

0.63

 Seym

0.62

Rated

0.61

 rounds

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.