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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

FactoryReloaded

-0.65

enhagen

-0.63

grad

-0.63

alion

-0.62

itates

-0.62

 expense

-0.60

aces

-0.60

uci

-0.60

aml

-0.59

bars

-0.59

POSITIVE LOGITS

ICAN

0.69

 Vide

0.67

 Shining

0.66

 Newman

0.65

 Supporters

0.65

 Drawing

0.64

 Yesterday

0.64

 Skydragon

0.63

LLOW

0.62

 wedge

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.