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

 Galile

-0.68

 pans

-0.66

Kov

-0.66

 Ches

-0.64

 Lewis

-0.61

cop

-0.61

 shine

-0.61

 Bronx

-0.60

 Christensen

-0.60

warn

-0.60

POSITIVE LOGITS

duction

0.84

è¦ļéĨĴ

0.82

rack

0.82

it

0.78

cffffcc

0.75

itability

0.75

abit

0.75

ubuntu

0.74

UA

0.73

acia

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.