INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

duction

-0.74

icro

-0.73

plex

-0.72

llah

-0.72

atom

-0.71

Plex

-0.70

encies

-0.70

opal

-0.69

ural

-0.69

rose

-0.68

POSITIVE LOGITS

isers

0.83

 bystanders

0.71

 proxies

0.70

 nort

0.66

ttes

0.65

 triglycer

0.65

 Ichigo

0.65

 heel

0.64

 Britons

0.64

 constit

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.