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

ÂłÂłÂł

-0.89

Reviewer

-0.86

ÂłÂłÂłÂł

-0.76

ARCH

-0.74

alf

-0.73

Brend

-0.72

DOM

-0.71

Farm

-0.71

Chel

-0.70

tom

-0.70

POSITIVE LOGITS

ologne

0.79

weeney

0.75

TNT

0.69

 comprom

0.68

stanbul

0.67

 latex

0.67

undai

0.66

 candles

0.66

orno

0.65

irmed

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.