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

wcsstore

-0.75

href

-0.73

reek

-0.69

Sel

-0.68

}{

-0.65

baugh

-0.63

ogl

-0.63

){

-0.62

woods

-0.62

gam

-0.61

POSITIVE LOGITS

afort

0.64

 Empires

0.64

ciation

0.62

ortment

0.61

 Deposit

0.61

ake

0.61

 Moff

0.60

 inval

0.60

 Statistics

0.60

 Cousins

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.