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

apr

-0.73

ovember

-0.70

geries

-0.69

bare

-0.68

 secrecy

-0.67

DOC

-0.67

papers

-0.66

DOC

-0.65

staking

-0.65

 notified

-0.65

POSITIVE LOGITS

tsky

0.78

hattan

0.78

bilt

0.75

chenko

0.71

MPG

0.67

 Appearance

0.67

querque

0.66

 Thieves

0.64

 Leviathan

0.64

tg

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.