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

 Arkham

-0.75

 accur

-0.67

 Mehran

-0.64

 mosquitoes

-0.62

fal

-0.60

è¦ļéĨĴ

-0.60

patient

-0.60

 suscept

-0.60

olitics

-0.60

 coasts

-0.59

POSITIVE LOGITS

uth

0.78

iz

0.70

media

0.69

immer

0.69

 Tape

0.69

UTH

0.68

icum

0.68

walker

0.67

glas

0.67

idden

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.