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

ikes

-0.83

rique

-0.81

ierre

-0.72

ype

-0.72

 Verge

-0.71

orsi

-0.70

inois

-0.69

 Boone

-0.68

bians

-0.68

acca

-0.66

POSITIVE LOGITS

 womb

0.73

uphem

0.64

 relic

0.64

 corro

0.60

 glove

0.60

 moth

0.59

 Tradable

0.59

 Equip

0.59

notice

0.59

 Mercenary

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.