INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Camera

-0.73

Mos

-0.71

 Astron

-0.70

 Observer

-0.67

edes

-0.66

jac

-0.66

Publisher

-0.65

azo

-0.65

romy

-0.64

 mole

-0.64

POSITIVE LOGITS

hement

0.73

SPONSORED

0.70

INGS

0.68

psey

0.66

irements

0.65

isons

0.65

 moot

0.63

uations

0.62

 WATCHED

0.62

ationally

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.