INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ECA

-0.72

gee

-0.71

INAL

-0.68

bee

-0.67

ween

-0.67

mys

-0.67

66666666

-0.65

ItemThumbnailImage

-0.65

ById

-0.64

Yah

-0.63

POSITIVE LOGITS

 fuss

0.70

 Detected

0.67

 word

0.67

ariat

0.65

uced

0.65

 moderators

0.64

lid

0.63

 Maced

0.62

ensed

0.61

versive

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.