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

etheless

-0.86

ADRA

-0.75

adel

-0.75

 Dragonbound

-0.75

andro

-0.72

exist

-0.72

enei

-0.67

ccording

-0.66

bernatorial

-0.66

ãĥ¥

-0.65

POSITIVE LOGITS

 Witnesses

0.76

rer

0.70

ezvous

0.68

terson

0.68

Gow

0.67

 Higgins

0.66

ertodd

0.64

 Lands

0.62

 witnesses

0.60

 Lans

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.