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

lihood

-0.67

 Stard

-0.64

owners

-0.63

igators

-0.63

SUN

-0.62

mates

-0.62

 statutes

-0.61

sic

-0.61

 Porsche

-0.59

 salts

-0.59

POSITIVE LOGITS

Ò

0.73

atra

0.71

 outwe

0.69

animous

0.67

ilateral

0.66

conn

0.64

anooga

0.64

ĸļ

0.63

 Blossom

0.63

erest

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.