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

mask

-0.68

 Order

-0.67

assian

-0.66

posts

-0.62

issue

-0.61

cuts

-0.61

cutting

-0.61

store

-0.61

yrinth

-0.60

ICE

-0.60

POSITIVE LOGITS

fman

0.76

onen

0.71

 enthusi

0.65

ftime

0.65

theless

0.63

VIDIA

0.63

 Weiner

0.63

zb

0.63

 indisc

0.62

 Dian

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.