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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ourge

-0.79

bara

-0.76

apego

-0.74

aspers

-0.73

cember

-0.70

asio

-0.68

ributed

-0.68

atur

-0.66

bly

-0.65

vati

-0.65

POSITIVE LOGITS

è£ħ

0.71

é¾įåĸļå£«

0.69

straight

0.65

ãĤ½

0.63

Posts

0.63

Â¶

0.62

 Akron

0.62

èª

0.61

 Odyssey

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.