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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Naw

-0.73

ACP

-0.71

afety

-0.65

enfranch

-0.65

 circumcised

-0.64

pling

-0.63

icides

-0.62

RIC

-0.62

AMY

-0.62

wich

-0.62

POSITIVE LOGITS

 habitable

0.67

pmwiki

0.66

isode

0.66

tops

0.64

 Europa

0.64

past

0.62

ESA

0.60

cheat

0.59

 guiIcon

0.59

station

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.