INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

avia

-0.69

nect

-0.68

resa

-0.68

azo

-0.66

iman

-0.65

iyah

-0.63

ogh

-0.63

Mi

-0.63

uez

-0.62

 Latvia

-0.62

POSITIVE LOGITS

 Commentary

0.76

 Inquisition

0.71

aughs

0.68

 Vide

0.68

 customary

0.67

CHR

0.66

enance

0.62

 merry

0.62

STEP

0.61

 reserved

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.