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

 Devices

-0.61

 Gems

-0.60

 metic

-0.60

bryce

-0.58

 culprit

-0.58

AUD

-0.58

 propositions

-0.58

 popup

-0.57

 kindred

-0.57

 cavity

-0.56

POSITIVE LOGITS

obyl

0.85

ename

0.69

lehem

0.69

enne

0.68

urally

0.66

 Melt

0.66

externalActionCode

0.65

oglu

0.65

pei

0.65

ongyang

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.