INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Styles

-0.67

 Brands

-0.62

 Asians

-0.62

 Bowling

-0.60

abouts

-0.60

ources

-0.60

 Volunte

-0.59

 Pagan

-0.59

ashtra

-0.58

kWh

-0.56

POSITIVE LOGITS

ende

0.71

peg

0.71

kered

0.69

ice

0.69

zn

0.69

enei

0.68

itle

0.68

âĹ¼

0.67

TY

0.67

ublic

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.