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

DragonMagazine

-0.75

orative

-0.70

ucci

-0.66

inos

-0.66

ãĤª

-0.65

orians

-0.64

asso

-0.63

rand

-0.62

aic

-0.62

ĸļ

-0.61

POSITIVE LOGITS

 particular

0.89

 concess

0.70

hern

0.66

sole

0.65

unci

0.65

 person

0.63

 uncond

0.62

uling

0.62

angan

0.62

person

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.