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

rero

-0.79

jri

-0.77

enegger

-0.76

ersen

-0.73

idges

-0.66

ruciating

-0.66

ocate

-0.66

entric

-0.66

cule

-0.66

ikarp

-0.66

POSITIVE LOGITS

Deal

0.76

 Overs

0.75

ync

0.71

apeake

0.70

SB

0.69

BT

0.69

 Deal

0.68

 Archdemon

0.68

Balt

0.66

 Antar

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.