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

wcsstore

-0.72

 Japan

-0.69

oe

-0.65

 Associated

-0.63

 Tokyo

-0.63

USA

-0.63

±

-0.62

 Harper

-0.61

ril

-0.60

ike

-0.60

POSITIVE LOGITS

 equivalents

0.78

 benefic

0.73

 parity

0.68

 Ezek

0.67

avorite

0.65

etus

0.63

 Vaugh

0.62

adobe

0.61

 shotguns

0.61

++++++++++++++++

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.