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

yrim

-0.66

yrus

-0.66

 inherit

-0.66

yss

-0.66

poral

-0.64

yr

-0.64

idth

-0.64

 Curve

-0.63

 Lynd

-0.61

 junction

-0.60

POSITIVE LOGITS

 Nanto

0.79

fried

0.76

kson

0.72

Shares

0.71

atis

0.68

acan

0.67

uador

0.67

verson

0.67

boards

0.66

anu

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.