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

âĶĢâĶĢâĶĢâĶĢ

-0.78

--+

-0.76

-+

-0.69

otin

-0.69

 Knox

-0.68

 Siren

-0.67

 Flavoring

-0.64

wcsstore

-0.64

 Tile

-0.64

 Sierra

-0.64

POSITIVE LOGITS

ulas

0.76

ufact

0.75

inas

0.70

alities

0.70

achu

0.69

IONS

0.67

illions

0.67

doms

0.66

hops

0.66

Prince

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.