INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Tile

-0.71

 Apost

-0.69

 Seah

-0.67

 forfeiture

-0.66

 tenant

-0.65

 Tide

-0.65

 Celestial

-0.64

 Vatican

-0.64

Sea

-0.62

 Droid

-0.62

POSITIVE LOGITS

govtrack

0.76

oras

0.70

Else

0.70

atus

0.66

anyon

0.66

duration

0.65

é»Ĵ

0.65

taboola

0.63

${

0.63

assed

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.