INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

theless

-0.74

avid

-0.71

 accompl

-0.68

zinski

-0.67

anan

-0.65

asar

-0.65

Downloadha

-0.65

oys

-0.64

Rober

-0.63

aphael

-0.61

POSITIVE LOGITS

Alert

0.78

 Hipp

0.69

 Courts

0.67

atures

0.66

ONSORED

0.64

course

0.63

 Tanks

0.63

 Niagara

0.62

 Sussex

0.62

 Riverside

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.