INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

izen

-0.80

senal

-0.73

thur

-0.71

Mit

-0.66

urities

-0.64

å§«

-0.63

ierrez

-0.62

essen

-0.62

UME

-0.62

ersion

-0.61

POSITIVE LOGITS

tro

0.83

Reviewer

0.75

ggies

0.65

alos

0.65

wrapper

0.64

dri

0.64

cards

0.63

 blockers

0.61

 snipp

0.61

 Goth

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.