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

 Chero

-0.91

 Roots

-0.72

BuyableInstoreAndOnline

-0.69

Nar

-0.69

 Boards

-0.69

 Bleach

-0.67

 Warner

-0.66

icipated

-0.66

Kak

-0.65

66666666

-0.65

POSITIVE LOGITS

Lens

0.73

zing

0.72

resist

0.68

udi

0.62

aley

0.62

onomy

0.62

comb

0.62

 resist

0.60

 consum

0.59

uge

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.