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

ocene

-0.74

mination

-0.66

iosyn

-0.65

ners

-0.64

igon

-0.64

 Frieza

-0.63

urn

-0.62

 nominations

-0.60

otom

-0.59

 Gotham

-0.59

POSITIVE LOGITS

ACK

0.85

wagen

0.71

kj

0.69

URA

0.68

 Mandatory

0.67

ECTION

0.66

WA

0.65

 Flavoring

0.65

tl

0.64

LY

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.