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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.90

 Sorceress

-0.76

ãĥĬ

-0.69

 Planes

-0.69

redo

-0.68

ãĥ¼ãĥĨ

-0.68

ional

-0.67

æ©

-0.66

uminati

-0.65

grave

-0.65

POSITIVE LOGITS

rawdownloadcloneembedreportprint

0.92

bery

0.63

ulet

0.63

baugh

0.62

 Version

0.62

Pep

0.60

 special

0.57

ksh

0.57

estyle

0.57

ete

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.