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

 Sapp

-0.68

icial

-0.66

soDeliveryDate

-0.65

 Thou

-0.65

æ©

-0.64

NR

-0.63

ãĥŁ

-0.62

Man

-0.62

emis

-0.62

ificial

-0.61

POSITIVE LOGITS

essor

0.91

artments

0.62

lopp

0.61

 unmarked

0.59

isters

0.59

arten

0.58

agers

0.57

abase

0.57

 predec

0.57

olitan

0.54

Activations Density 0.000%

No Known Activations

This feature has no known activations.