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

cknow

-0.75

creen

-0.69

cephal

-0.66

ournal

-0.62

 Alphabet

-0.62

 cord

-0.62

igion

-0.61

irm

-0.59

elvet

-0.59

apsed

-0.59

POSITIVE LOGITS

DragonMagazine

0.81

soDeliveryDate

0.69

Sa

0.68

ipe

0.66

ime

0.66

illon

0.66

 GOODMAN

0.65

Exactly

0.65

emonium

0.64

actly

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.