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

 amen

-0.69

cy

-0.64

imate

-0.63

ith

-0.63

ibel

-0.63

obin

-0.62

Loc

-0.61

ongevity

-0.61

nesota

-0.60

Shift

-0.60

POSITIVE LOGITS

 Flavoring

0.86

ensis

0.65

amaz

0.65

 ingred

0.63

)",

0.62

BuyableInstoreAndOnline

0.62

%);

0.61

duino

0.60

'),

0.59

)...

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.