INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ĺħ

-0.85

ĵĺ

-0.69

»Ĵ

-0.68

angu

-0.67

inges

-0.67

ARE

-0.66

²¾

-0.66

ĨĴ

-0.65

ĻĤ

-0.65

 Hemp

-0.65

POSITIVE LOGITS

mone

0.77

zai

0.77

ghazi

0.73

Rez

0.70

 Giul

0.67

ths

0.66

 Metatron

0.65

Fi

0.61

slaught

0.60

 Loans

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.