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

sing

-0.80

etooth

-0.68

 Mechdragon

-0.66

occup

-0.63

Fey

-0.61

 Zombies

-0.61

cn

-0.60

 Suns

-0.59

roman

-0.58

 seeker

-0.58

POSITIVE LOGITS

olics

0.70

olesterol

0.69

merce

0.68

osterone

0.67

yrim

0.67

aldehyde

0.66

thood

0.65

iquid

0.65

jac

0.65

getic

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.