INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 quir

-0.68

Variable

-0.66

 Seed

-0.64

chini

-0.64

achine

-0.64

UR

-0.63

 FANT

-0.63

gdala

-0.62

URI

-0.60

URA

-0.60

POSITIVE LOGITS

ivari

0.67

 Britann

0.67

isy

0.64

 desks

0.63

iety

0.63

beit

0.62

erness

0.61

busters

0.61

geons

0.61

 Desk

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.