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

earch

-0.76

ovember

-0.75

omed

-0.71

uld

-0.70

regon

-0.68

orers

-0.68

gypt

-0.67

earchers

-0.67

APTER

-0.66

eware

-0.66

POSITIVE LOGITS

 Samantha

0.67

question

0.66

 Zimmerman

0.64

 resultant

0.64

respond

0.62

 Yanukovych

0.62

llah

0.61

 bubbles

0.61

 chants

0.61

Interstitial

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.