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

 FORM

-0.73

ãĥ£

-0.73

ãĤ¡

-0.68

#$

-0.67

ãĥĨãĤ£

-0.65

cffffcc

-0.64

htt

-0.64

Pri

-0.63

ãĤ©

-0.63

âķĲâķĲ

-0.63

POSITIVE LOGITS

 freshmen

0.78

uary

0.73

 collaborations

0.67

eteen

0.65

iate

0.65

bath

0.65

taking

0.63

 cellar

0.63

eworks

0.63

 relapse

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.