INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

amel

-0.82

 Repair

-0.75

 Mehran

-0.69

alyst

-0.69

cms

-0.67

ascript

-0.67

earances

-0.64

Peb

-0.64

 Tables

-0.64

URES

-0.64

POSITIVE LOGITS

blank

0.67

bound

0.63

 Petersen

0.63

furt

0.63

aviour

0.62

 Lange

0.62

wolf

0.62

govern

0.61

 captured

0.61

endif

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.