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

mos

-0.70

untarily

-0.70

maxwell

-0.69

hops

-0.66

hiro

-0.66

ilogy

-0.65

>>>>>>>>

-0.64

omez

-0.63

hews

-0.62

EStreamFrame

-0.62

POSITIVE LOGITS

uner

0.71

Bye

0.69

 Diplom

0.69

 Appearance

0.66

 pretext

0.64

 Value

0.64

 reproduction

0.64

bourg

0.62

 programme

0.62

Ã«

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.