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

iple

-0.80

ibrary

-0.78

iliated

-0.75

isner

-0.68

ieth

-0.67

canon

-0.67

inition

-0.65

inyl

-0.64

idental

-0.64

 quir

-0.64

POSITIVE LOGITS

zh

1.16

DeL

0.73

hare

0.71

ACS

0.71

pour

0.70

cha

0.70

FIELD

0.67

JR

0.66

soType

0.66

soDeliveryDate

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.