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

atche

-0.81

ceptive

-0.74

sylv

-0.74

insula

-0.71

arthed

-0.70

missive

-0.70

stra

-0.68

asury

-0.67

agos

-0.66

rehens

-0.66

POSITIVE LOGITS

 twins

0.67

 conclud

0.67

 cloning

0.66

ieth

0.62

 Abel

0.61

cov

0.60

 incest

0.60

ãĥĺ

0.60

 abort

0.60

 faked

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.