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

atari

-0.86

arta

-0.72

sworth

-0.71

sett

-0.69

ocumented

-0.69

zee

-0.68

zel

-0.67

sterdam

-0.67

lyn

-0.66

namese

-0.64

POSITIVE LOGITS

£ı

0.86

NRS

0.77

 fluor

0.69

ffield

0.69

 pupils

0.67

QL

0.67

 buff

0.64

obbies

0.64

CI

0.63

 Jagu

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.