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

Else

-0.74

NF

-0.74

 symp

-0.69

 Canaver

-0.65

 Remastered

-0.64

 è£ıè

-0.62

 Medline

-0.62

 Citiz

-0.62

iPhone

-0.61

 Niet

-0.60

POSITIVE LOGITS

illard

0.75

bat

0.73

unin

0.68

erie

0.67

plet

0.64

aples

0.64

uilding

0.63

oil

0.63

]"

0.61

okÃ©

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.