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

Â¶

-0.74

bat

-0.73

laws

-0.68

jar

-0.67

friend

-0.66

earth

-0.63

bee

-0.62

MAG

-0.62

Batman

-0.61

 Strongh

-0.60

POSITIVE LOGITS

ront

0.76

apsed

0.72

ensical

0.67

ratulations

0.67

ittal

0.66

mercial

0.65

ritical

0.65

 IMAGES

0.64

payers

0.63

EStream

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.