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

avis

-0.72

urgical

-0.68

ption

-0.65

Mos

-0.63

Pages

-0.60

 Buckingham

-0.60

 Commons

-0.60

ORD

-0.58

otte

-0.58

ur

-0.58

POSITIVE LOGITS

 "âĢ¦

0.77

 victim

0.75

llah

0.74

vironment

0.72

lication

0.67

nat

0.66

 roommate

0.65

apego

0.65

oÄŁ

0.63

 barr

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.