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

phies

-0.81

ilts

-0.76

ayers

-0.75

owicz

-0.68

icial

-0.67

utsu

-0.67

 TAMADRA

-0.66

 Izan

-0.65

arak

-0.64

enstein

-0.63

POSITIVE LOGITS

 blackout

0.65

icity

0.64

 bookstore

0.62

 lineback

0.62

 distributor

0.62

 Chero

0.61

BAT

0.60

Gay

0.60

 sued

0.59

 Contains

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.